Download MSE 313 2011 Course Pack[1]

MSE 313
Spring 2011
University of Washington Materials Science and Engineering MSE-­‐313 (2011) Integrated Junior Lab 3 (2 credits) Instructor: Fumio S. Ohuchi, #312 Roberts, 685-­‐8272, [email protected] Catalog description: Mechanical properties related laboratory experiments, including stress-­‐strain behavior of materials and elastic modulus of materials, effect of work hardening on stress strain behavior, and effect of surface condition of the strength of glass. Electrical properties of materials. Screen printing techniques and Impedance spectroscopy Offered: Sp. Instructor’s Description: This course is designed to introduce the student to the basic hands-­‐
on skills of laboratory experimentation and techniques used in evaluating the physical properties of the three major classes of materials (metals, ceramics, and polymers) and help the student to develop good report writing skills and Lab notebook maintenance. This course addresses the use and instruction of shop and analytical equipment, library resources for reference and literature searches, laboratory report writing, laboratory notebook recording, and data analysis. Class Schedule: Two-­‐three hours laboratory sessions weekly. 2011 Spring: Tuesday and Thursday Session A 9:30-­‐12:30 and Session B 2:30-­‐5:30 See attached schedule TA’s: Aaron Lichtner Evan Uchaker Matthew Leung Kwangsuk Park Lab Manager Tuesday Kuykendall MUE 156 221-­‐2678 Prerequisites: None Textbook and Other Required Materials: 1. Laboratory & Course Handouts, MSE 313 Course Packet, University of Washington, Seattle 2. William D. Callister, Jr., Materials Science and Engineering an Introduction, 6th Edition. 1
MSE 313
Topics Covered: Spring 2011
See lab content described in separate sheet (attached) Information Page: Prof. Ohuchi 425-­‐941-­‐6865 (Cell) 206-­‐685-­‐8272 (office) Aaron Lichtner Evan Uchaker Matthew Leung Kwangsuk Park Tuesday Kuykendall 425-­‐773-­‐2825 [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] MSE Main Office 206-­‐549-­‐2600 MSE FAX 206-­‐543-­‐3100 MSE Dept. Address: Roberts Hall 302 Box 352120 Seattle, WA 98195 MSE Web Page: http://depts.washington.edu/mse/ Mechanical Engineering Student Machine Shop: Kevin Soderlund (Lead) [email protected] Eamon McQuaide (technician) [email protected] Mechanical Engineering Main Office 206-­‐543-­‐5090 2
MSE 313
Spring 2011
Course Objectives (Student outcomes) At the end of this course, the students will be able to: • Students will become familiar with the uses and capabilities of standard shop and laboratory analytical equipment and understand the safety requirements in these environments. • Students will learn the essential skills of good report writing and proper laboratory notebook documentation • Students will be able to efficiently search the library resources for handbook and process information as well as bibliographic references. • Making students aware of the resources available at the Engineering Writing Center to enhance the student’s ability to write quality laboratory reports. • Students will learn and gain the skills of sample preparation, from basic cutting, grinding, mounting, and polishing to more advanced acid/thermal etching and sputter coating. • Students will be able to make precise temperature measurement utilizing thermocouple technology for process control. • Students will be able to identify the trends in density, hardness, and thermal conductivity for metals, ceramics, and polymeric materials. • Students will learn the use and function of a standard X-­‐ray diffractometer and proper powdered XRD sample preparation and be able to analyze single and multi-­‐
component materials. • Students will be able to grow single crystals from saturated solution method, and be able to control the growth process. • Students will be able to use the standard optical microscope for analyzing grain size in metals and ceramics through digital image capture and analysis. • Students will be able to apply the knowledge and hands on skills in this laboratory coursework to advanced processing and characterization techniques in MSE 313 and 313 as well as senior research projects. Evaluation: Grade will be based on two full lab reports, project updates, project report and project presentation. Feedback: Anonymous student evaluations (College of Engineering) Individual evaluation of items described in course objectives. 3
MSE 313
Spring 2011
MSE Laboratory Policies 1. The use of cellular phones, pagers, and other communication devices is prohibited. 2. Laboratory sessions are mandatory; otherwise the student will get a zero grade for the lab report. Makeup labs will be given only for pre-­‐arranged, excused absences. Students must contact the instructor before the absence. The only exception to the rule is absence due to illness. 3. If a student is unexpectedly detained and unable to make arrangements, she/he should make every attempt to notify the TA via phone or email. Students are expected to be on time for laboratory sessions. For every 10 minutes a student is late 10% of the lab grade will be deducted. If a student fails to show up after 30 minutes they will be issued a zero grade at the discretion of the instructor. Students are expected to be on time for laboratory sessions. Any student more than 15 minutes late without notice (phone, e-­‐mail, and any other means) will have seriously deducted from the grade of their laboratory report for that rotation. Try to avoid this situation. 4. The checking of email or surfing the web in the computer lab is not permitted during the laboratory session unless instructed to do so for lab information. Doing so could result in a grade reduction. 5. Given the quantity of students in each lab section, students are not permitted to attend a morning lab section if they are enrolled in an afternoon lab section or vice versa. 6. Safety glasses are required for all laboratory experiments so do not show up without them. 7. Lab coats are required at all times in MUE 165, 166, and 167. They can be purchased at the bookstore. Students are responsible for their cleaning and maintenance. 8. All laboratory reports shall be type written, there are no exceptions. 9. Cheating (e.g. the unauthorized copying of another persons’ laboratory write-­‐up, un-­‐
cited material from the internet or other sources) will be referred to the Dept. Chair for further action. 10. Absolutely no food or drink in the laboratory at all times. 11. Laboratory reports are due at the beginning of class at 9:30 AM for the morning lab section and 2:30 PM for the afternoon lab section. No late assignments will be accepted unless prearranged through the instructor. 4
MSE 313
Spring 2011
Detailed Guidelines for Formatting a Laboratory Report
•
Cover page (see next page) •
Margins: Left, right, top and bottom: 1 inch •
Text: Justified on both sides •
Spacing: 1.5 •
Figures and tables: Left aligned along with the figure captions and table headings Figure 1: Figure titles go below the figure Table 1: Table titles go above the table •
All pages should be numbered sequentially starting from cover page. All footnotes should be numbered sequentially •
Hindu-­‐Arabic numerals (1, 2, 3) should be used for numbering of tables. •
No color text, graphs should be drawn with black color on white, background. The only exception is color pictures or figures from lab handout. •
12 point Arial or Times New Roman font should be used for text as well as for Figure captions and table headings. •
The title should be in 14 point bold •
All section and sub-­‐section headings should be in bold. •
All sections should be numbered starting from Introduction: 1. Introduction If there are sub-­‐sections, 1.1, 1.2, etc. Further sub-­‐sections, .............................. 1.1.1,1.1.2, ... and 1.2.1,1.2.2, , etc 2. Experimental procedure For sub-­‐sections, follow the same format as introduction section 3. Results and Discussion For sub-­‐sections, follow the same format as introduction section 4. 4. Conclusions and recommendations 5. References 6. Appendices 5
MSE 313
•
Spring 2011
Appendix-­‐A, Appendix-­‐B, Appendix-­‐C, etc. The arrangement of various sections in the lab report should be as follows: Cover page with abstract Introduction Experimental procedure Results and Discussion (includes graphs and tables) Conclusions and Recommendations References
Appendices (if there are any) Answers to questions •
•
Suggestions for improvement
Page breaks: after cover page, after experimental procedure, and before Appendices. Expected length of the report: Min. 10 pages, buy not exceeding 15 pages. 6
MSE 313
Spring 2011
MSE-­‐311 Integrated Junior Lab Laboratory Report Laboratory-­‐ Rotation # Name Student # Group-­‐ Members: TA Date Abstract 7
MSE 313
Spring 2011
MSE 311 Lab Report Grading Criteria
MSE 313 Lab-III Report Grading Criteria
Cover Page (Use Cover Page template) Abstract 5 ____ • Purpose: Brief introduction to the problem • Scope: Identifies experimental approach (materials, variables, equip., etc.) • Experimental results are given (numeric when possible) • Significant conclusions are reported Introduction 5 ____ • Introductory statement: General statement of the problem • Technical background: Summary or discussion of historical and/or technical background as a foundation for present test • Purpose(s) of the present test and the report (different from the Abstract) Experimental Procedure 5 ____ • Schematic or diagram of experimental setup • Describe the procedure used for conducting tests or experiments • Various characterization methods used • Variables effecting the measurement and their ranges (capacities, sizes, etc.) Results and Discussion 40 ____ • Order of presentation is logical • Computations are complete and correct (sample calculations cited) • Data reduction is presented in tabular or graphical form (% difference, % error, etc.) Same data do not appear in both tables and graphs (unless in Appendix) • Data is interpreted correctly, is validated and compared to published work
• Good comprehension is shown of the relative meaning and importance of the results (trends are cited, averages, extremes, %change, etc.) • Discussion should concentrate on test results and correlation with analysis; no comments on test technique, lost data, etc. • Error analysis is performed, deviation from expected results is discuss Conclusions and Recommendations
5 ____
• Conclusions are presented in a numbered list • Most important conclusions are given first • Conclusions are drawn only from the body of the report • Recommendations for future studies based on conclusions Graphs. Tables. and Figures (should be a part of results and discussion) 5 ____ • Choice of information to be graphed tabulated, or illustrated is relevant and logically 8
MSE 313
•
•
•
•
Spring 2011
presented Cited and clearly explained in the appropriate sections Data are correctly transcribed into tables from data sheets Graphs are labeled, have proper scaling, and error bars (when applicable) Format is correct and appropriate: Graphs and Figures captioned (example, Figure 2, Average grain size in…), tables are titled (Table IV... Green densities of) References 5 ____ • References are cited in the text (superscript # or [#] end of sentence) • References are presented in a numbered list in the order of citation • Proper format is used for each reference Answers to Questions in the Lab and Suggestions for Improvement
5 ____
• Answers to questions should be included in this section not in the body of the lab unless otherwise indicated • Suggestions on how to improve the laboratory next time through focusing on any problems that were encountered in the lab • Should be different from Recommendations Appendices
5 ____
•
•
•
•
Arranged in order of citation in the text (A, B, C, etc.)
Raw data sheets from the tests are presented
Sample calculations and data reductions are included Should begin on a separate page
Overall arrangement and coordination
• General arrangement of ideas is logical, clear and consistent • Grammar, sentence, and paragraph structure are good • Spelling errors are absent, paper has been proof-­‐read • Page composition is appropriate, neat and pleasing • Units defined for sample calculations, tables, graphs, etc. 20 ____
Total 100 ____ 9
MSE 313
Spring 2011
How to use this manual: Reading the lab manual before lab is required. Some labs may include a pre-­‐lab quiz so it pays to be prepared. The manual is arranged by labs with a brief overview in the introduction. There is a page that has important contact information. Use the information if you cannot make it to lab for any reason. Important terms throughout the document are in italic. If any term (in italics or not), is new to you, you are responsible for looking it up. There are a number of references to ASTM standards throughout the manual. You are responsible for looking up the standard at the library. In the introduction next, there are brief descriptions of each lab followed by a list of information that should be included in your lab write-­‐up. A number of the sections in this lab manual are excerpted from a variety of sources and are cited where relevant. Students are encouraged to look up these sources to gain more in-­‐depth understanding of the subjects presented. Whenever possible electronic copies of sources are included in a folder on the desktops of the Jr. lab computers. To the best of the knowledge of the editors of this manual, no copyrighted materials are used without permission. 10
MSE 313
Spring 2011
MSE 313 Integrated Junior Laboratory (2011 Spring) Lab-­‐I. Mechanical Properties of Materials: Part 1A: Tensile Testing of Metals, Composites and Plastics The purpose of this laboratory is to determine tensile strength, elastic modulus, fracture strength, ductility and energy to fracture of metals, composites, and plastics in tension using a mechanical testing apparatus (Instron corporation, model no: 5500R). Part 1B: Compression Testing and Elastic modulus measurement The purpose of this laboratory is to measure yield strength, yield point, compressive strength and elastic modulus under compressive loading, and also to measure the elastic modulus by impulse excitation technique. Part 1C: Fracture Toughness Testing of Metallic Materials The purpose of this laboratory experiment is to determine fracture toughness of a material defined as a measure of the material’s resistance to crack propagation. Part 2: Ductile to Brittle Transition of Metals The purpose of this laboratory experiment is to demonstrate the effect of temperature on the fracture behavior of materials using the impact tester and Charpy impact specimens. Part 3: Fractography (SEM ) The purpose of fractography is to analyze fracture surfaces by SEM, and to relate the fracture surface features to the cause(s) and basic mechanisms of fracture. Lab-­‐II. Electrical Properties of Materials: Part 1: Electrical Resistivity of Metal Material : The purpose of this laboratory is to measure the electrical conductivity (resistivity) of selected conductors, insulators and semiconductors. This lab is designed to help the student understand how the electrical conductivity is related to temperature, alloying, permanent deformation and impurities. Part 2: Energy Gap in Semiconductors: The purpose of this lab is to measure temperature dependence of the electrical conductivity in a semiconductor sample. The energy gap will be calculated from data taken in an intrinsic region, and the temperature dependence of the majority carrier mobility will be deduced from measurements taken in the extrinsic region. 11
MSE 313
Spring 2011
Part 3: Terminal Device Characteristics and Diode Characterization: The objectives of this experiment are to learn methods for characterizing 2-­‐
terminal devices, such as diodes, observe some fundamental trends in the characteristics of various diode types, and to gain some familiarity with standard test bench instrumentation. Lab-­‐III. Screen Printing of Thick Film Materials and Impedance Spectroscopy: Part 1: Print/Dry/Fire Conductor The purpose of this laboratory is to screen print, dry and fire thick film circuits made of conductive ink. This lab is designed to introduce students to the technology of printing thick film and hybrid circuit using a screen printing machine. The fundamentals of the equipment, inks, and thick film materials processing will be explored. Part 2: Print/Dry Resistor The purpose of this laboratory is to screen print and dry thick film resistors. This lab is designed to allow practice screen and stage alignment to obtain patterns at precise locations on the substrates. Part 3: Firing Profiles/ Resistance Measurements / Pattern Design Principles The purpose of this laboratory is to fire thick film resistors. This lab is designed to obtain and study different firing cycles of the resistors and to determine the effect of firing temperatures on the resistors’ performance. Students will also study aspects of thick film design and layout. The students will examine the patterns of printed/fired thick film circuits obtained from Part 1 and 2. Factors such as geometry, sheet resistance, print thickness and electrical performance will be studied and students will compare calculated theoretical results with the experimental data. Jr. Lab Project The purpose of the Jr. Lab Projects will be to expose students to the various challenges involved in laboratory experimentation Each group will be expected to design, set-­‐up, and implement a lab project. Six full lab sessions will be available. 12
MSE 313
Spring 2011
MSE 313 2011 Schedule Mon Tue Wed Thu Fri Mar 28 29: All Attend: Orientation 30 31 Rotation 1 Day 1 Apr 1 4 5 Rotation 1 Day 2 6 7 Rotation 1 Day 3 8 11 12 Rotation 2 Day 1 13 14 Rotation 2 Day 2 15 18 19: Rotation 2 Day 3 20 21: COE Discovery Days Prep 22: COE DD 25 26 Rotation 3 Day 1 27 28 Rotation 3 Day 2 29 May 2 3 Rotation 3 Day 3 4 5 Rotation 4 Day 1 6 9 10 Rotation 4 Day 2 11 12 Rotation 4 Day 3 13 16 17 Rotation 5 Day 1 18 19 Rotation 5 Day 2 20 23 24 Rotation 5 Day 3 25 26 Open for project prep 27 30 31: Project Presentations June 2: Project Presentations 1 3 Finals Week Finals Week 13
MSE 313
Spring 2011
Lab 1: Mechanical Properties Lab 2: Screen Printing Lab 3: Electrical Properties Lab 4: Project Lab 5: Project Groups R1 R2 R3 R4 R5 1 & 2 Lab 1 Project Lab 2 Lab 3 Project 3 & 4 Lab 2 Project Lab 3 Project Lab 1 5 & 6 Lab 3 Lab 1 Project Lab 2 Project 7 & 8 Project Lab 2 Project Lab 1 Lab 3 9 & 10 Project Lab 3 Lab 1 Project Lab 2 14
MSE 313
Spring 2011
Blank Page 15
MSE 313
Spring 2011
MSE 313 Integrated Junior Laboratory -­‐I Mechanical Properties of Materials Part 1A: Tensile Testing of Metals, Composites and Plastics The purpose of this laboratory is to determine tensile strength, elastic modulus, fracture strength, ductility and energy to fracture of metals, composites, and plastics in tension using a mechanical testing apparatus (Instron corporation, model no: 5500R). Part 1B: Compression Testing and Elastic modulus measurement The purpose of this laboratory is to measure yield strength, yield point, compressive strength and elastic modulus under compressive loading, and also to measure the elastic modulus by impulse excitation technique. Part 1C: Fracture Toughness Testing of Metallic Materials The purpose of this laboratory experiment is to determine fracture toughness of a material defined as a measure of the material’s resistance to crack propagation. Part 2: Ductile to Brittle Transition of Metals The purpose of this laboratory experiment is to demonstrate the effect of temperature on the fracture behavior of materials using the impact tester and Charpy impact specimens. Part 3: Fractography (SEM training) The purpose of Fractography is to analyze fracture surfaces by SEM, and to relate the fracture surface features to the cause(s) and basic mechanisms of fracture. 16
MSE 313
Spring 2011
Part 1A: Tensile Testing of Metals, Composites and Plastics: Objectives: The purpose of this laboratory is to determine the mechanical properties such as tensile strength, elastic modulus, fracture strength, ductility and the energy to fracture of metals, composites, and plastics in tension using a mechanical testing apparatus (Instron corporation, model no: 5500R). READ ASTM E8 AND D638 STANDARDS BEFORE THE LAB Introduction: The Tensile Test is a common standard test which is simple to conduct and is a valuable method of determining important mechanical properties of engineering materials. A thorough understanding of a material's properties is required by the engineer if failures are to be avoided. The procedural details of the test vary for different material types, but tensile tests are generally conducted at room temperature at relatively slow loading rates although various temperatures and loading rates may be required for the determination of material behavior under specific conditions. The output of a standard tensile test is load versus displacement data. Since load-­‐
displacement characteristics are dependent on specimen size, for example, it will require twice the load to produce the same elongation if the cross-­‐sectional area of the specimen is doubled, load-­‐displacement data is routinely converted to engineering stress-­‐strain data. For axial loading, engineering stress, σ, is defined by the well known relationship, σ = P/Ao Equation 1Ai Where P is the instantaneous load applied perpendicular to the specimen cross-­‐section, in Newton’s (N), and Ao is the original cross-­‐sectional area of the specimen before any load is applied (m2). The units of stress are generally mega Pascal’s (MPa). Engineering strain, ε, along the loading axis of an uniaxial loaded sample is defined as, ε = ΔL/L Equation 1Aii Where engineering strain, ε (mm/mm), is determined by dividing the change in gauge length, ΔL (mm), by the original length of the specimen, L (mm). Engineering strain is unit less, but inches per inch or meters per meter are often used; the value of strain is clearly independent of the units system applied! Strain may also be expressed as a percentage, in which case the strain value is multiplied by 100. In Tensile Testing, the test specimen is deformed, usually until complete fracture occurs, with a gradually applied increasing tensile load that is applied uniaxially along the long axis of the specimen. When a specimen is loaded beyond its' ultimate strength 17
MSE 313
Spring 2011
the cross-­‐sectional area begins to decrease in a localized region instead of decreasing over its entire length creating a so called "neck.” Normally the test specimen is circular, but rectangular specimens can also be used. Each specimen is of a specific shape and the dimensions should be in accordance with ASTM (American Society for Testing and Materials) specifications for standardization. During testing, deformation is confined to the narrow center region, which has a uniform cross section along its length as shown in Fig. 1Ai. Figure 1Ai: Dog bone tensile testing specimens: (a) aluminum, (b) polymer, and (c) composite. By analyzing the stress-­‐strain curve of a specific material that has been tested in tension, a number of mechanical properties of the material can be determined. The list of properties that can be determined from the stress-­‐strain curves is given below. • Yield Strength (σys) -­‐ Yield Strength is defined as the stress required to produce a specified amount of plastic deformation or permanent set (strain) in a material. Below the elastic limit, the stress-­‐strain relationship in loading and unloading are identical for practical purposes. Therefore, it is not necessary to unload a specimen in order to determine the yield strength. Rather, a line parallel to the initial straight line portion of the curve is constructed. The parallel line is displaced from the origin of the curve by an amount equal to the specified permanent set. The stress at the intersection of the parallel line with the stress-­‐strain curve is called as the yield strength. The offset most commonly used is 0.2% strain or 0.002 in/in or mm/mm. The yield strength is a practical measure of the limit of elastic action of a material. It is always greater than the elastic limit and is sensitive to measurement instrument precision. • Ultimate or Tensile Strength (σul) -­‐ The Ultimate Strength, also referred to as the Tensile Strength, is calculated by dividing the maximum load sustained by the specimen by the original cross-­‐sectional area of the specimen. • Fracture or Rupture Strength (σru) -­‐ The Rupture Strength, also referred to as the Fracture Strength, is determined by dividing the load sustained at rupture by the 18
MSE 313
Spring 2011
original cross-­‐sectional area of the specimen. This load will be less than the maximum load because the cross-­‐sectional area of the specimen is reduced drastically after the maximum load is reached. The reduced cross-­‐section has an "Hour-­‐Glass" shape and the phenomenon of reduction in cross-­‐section is called as Necking or Necking Down. • Modulus of Elasticity (E) -­‐ The Modulus of Elasticity is a measure of material stiffness and is termed Young's Modulus for tensile loading. The Modulus of Elasticity, E, is the constant of proportionality between stress, σ, and strain, ε, at stresses below the proportional limit. The Modulus of Elasticity is found by measuring the slope of the straight-­‐line portion of the stress-­‐strain curve. • Toughness (UT) -­‐ The toughness of a material refers to the ability of the material to absorb energy up to the point of rupture. The Modulus of toughness is determined by measuring the area under the stress-­‐strain curve. This is not an exact indication of toughness because the specimen does not strain uniformly over its length, and hence does not absorb energy uniformly throughout its volume. The units of toughness are determined by multiplying stress with strain. • Percent Elongation (% EL) -­‐ The Percent Elongation refers to the elongation at rupture and can be expressed as: % EL = 100 (ΔLT)/Lo where ΔLT represents total elongation; ΔLT = final length – original length (Lf -­‐ Lo) • Percent Reduction in Area (% RA) -­‐ The Reduction in Area refers to the reduction in cross-­‐sectional area at rupture and can be expressed as: % RA =100 (ΔAT)/Ao where ΔAT represents total reduction in area; ΔAT = original cross-­‐sectional area-­‐ final cross-­‐sectional area (Ao -­‐ Af) • True Strain (εΤ)-­‐ The True Strain is the change in length divided by the instantaneous length and can be simply determined as: εT= ln (ε + 1) • True Stress (σT) -­‐ The True Stress is the applied load divided by the Instantaneous cross-­‐sectional Area and can be simply determined as: σT = σ (ε + 1) For some materials (e.g., concrete, gray cast iron) the initial elastic portion of the stress-­‐
strain curve is non-­‐linear where it is impractical to determine a standard modulus of elasticity. Furthermore, the modulus of elasticity is restricted to the initial linear portion of a standard stress-­‐strain diagram and is invalid beyond this region. For non-­‐
linear behavior, either the Tangent or Secant Modulus is generally utilized. 19
MSE 313
Spring 2011
Experimental Procedure: • Samples of 6061 (T6 condition) aluminum alloy, carbon fiber composites and high-­‐density polyethylene (HDPE) will be used for testing (Fig. 1A ii). • 6061 Al-­‐alloy contains Mg and Si as major alloying elements and also contains Fe, Cu, Mn, Zn, Cr and Ti. T6 condition implies that the alloy has been solution heat treated and artificially aged. • One sample of each kind will be provided to each group. The elongation of the specimen will be obtained from two sources: crosshead displacement and extensometer (more accurate). • The TA(s) will explain about the procedure and precautions to be followed while loading and testing the samples. Each student will get a chance to involve himself/herself in loading the samples, setting up the software program for testing, and testing the samples. • The load vs. displacement data of the materials will be obtained as .TXT files and these files will be distributed to the students for analysis. Figure 1Aii: Tensile Instron set-­‐up 20
MSE 313
Spring 2011
Things to be done before and after testing the samples: • Measure dimensions of the test specimen(s) [note; accurate and precise measurement of specimen dimensions is important] • For circular cross-­‐section specimens: Gage length before and after testing, diameter before and after testing. • For rectangular cross-­‐section specimens: Gage length before and after testing, width and thickness before and after testing. • It is always good to measure diameter, width and thickness at different locations along the gage length as they can be used for finding out the error bars while reporting the data. Part 1A References 1. G.E. Dieter, Mechanical Metallurgy, McGraw-­‐Hill Inc., New York, 1986. 2. F.A. McClintock and A.S. Argon, Mechanical Behavior of Materials, Addison-­‐
Wesley Inc., Reading, Mass., 1966. 3. Annual Book of ASTM Standards, Vol. 3.01, Standard E8, ASTM, Philadelphia, Pa. 2002. 4. Annual Book of ASTM Standards, Vol. 8.01, Standard D638, ASTM, Philadelphia, Pa. 2002. 21
MSE 313
Spring 2011
Part 1B : Compression Testing and Elastic modulus measurement: Objective: To measure the mechanical properties such as yield strength, yield point, compressive strength and elastic modulus under compressive loading Introduction: Mechanical property evaluation of materials under compressive loading is important in order to understand the performance of materials subjected to compressive loads. In some metal forming processes such as rolling and forging, materials undergo large plastic strains and there is a need to determine the yield stress of the materials under these severe conditions rather than those usually encountered in a normal stress-­‐strain test. In a tensile test the sample starts necking at about a strain of 0.1(10%). However, in a metal working operation such as rolling, a material undergoes plastic strain of 2 (200%) to 4(400%). Determining the yield stress (the stress at which the material starts flowing plastically) over such a large range of strain is not feasible in the case of tension test. On the other hand, compression test does not suffer from necking problem and the test can be carried out to a strain in excess of 2 (200%) in the case of ductile materials. Materials and Equipment: 6061 Al-­‐alloy solid cylinders Instron machine. Experimental Procedure: • ASTM standard E9 describes about compression testing of metallic materials. • Measure the sample diameter and mark the gage length by leaving about one diameter equivalent of length on each side of the sample. • Align the specimen properly so that the load line goes through the axis of the cylinder. The rate at which a sample is loaded is taken care by the pre-­‐set program in the Instron machine. • The load versus displacement (contraction of the sample) data is displayed on the monitor and data file is stored in the computer attached to the Instron. • The change in shape of a compression test sample is shown in fig.1. The sample bulges outwards at the sides due to the friction between the sample and the platens of the Instron. The friction at the interface between the machine platens and the sample top and bottom surfaces make the material at these regions difficult to flow plastically whereas the material at the sides is not constrained. As a result, an initial cylinder becomes a barrel shaped object at the end of testing. Moreover, there are triangular regions (see 22
MSE 313
Spring 2011
Fig.1Bi) at the top and the bottom where the material does not deform due to friction. Precisely for this reason, a length equivalent to the diameter of the sample is left on each end. As a result, the deformation in the center most portion of the sample is uniform as it is not affected by the friction. Fig.1Bi: Schematic of compression test sample before (broken lines) and after the test. The shaded triangular regions at the top and the bottom are un-­‐deformed regions due to friction. Smaller the sample (right side figure) greater the difficulty to deform as the un-­‐deformed regions extend all the way into the center of the sample and almost touch each other (Ref.: George E. Dieter, Mechanical Metallurgy). Fig. 1Bii. Compression Instron set-­‐up 23
MSE 313
Spring 2011
Part 1C: Fracture Toughness Testing of Metallic Materials Introduction Fracture toughness of a material is defined as a measure of the material’s resistance to crack propagation. The concept of fracture toughness or resistance to crack propagation has origins in linear elastic fracture mechanics. Cracks or flaws in a material concentrate (see Fig.1Ci) the applied stress and lead to the failure or fracture of the components and structures at much lower stresses than the design stress. If one intensifies the stress locally the stress intensification depends on the size of the crack and radius of curvature of the crack tip. The larger the crack length the higher the stress concentration (if design assumes that no flaws are present in the material). Fig. 1Ci: Elliptical hole in infinitely large panel produces stress concentration of 1+2a/b (Source: Reference 3) σmax/σa = (1+2 (sqrt(a/ρ)) Equation 1Ci where σmax= maximum stress at the end of major axis, σa = applied stress normal to the major axis, a= half major axis and ρ= radius of curvature (b2/a) Fracture can occur in three different modes as shown in Fig. 1Cii: Opening or tensile (Mode I), sliding or in-­‐plane (Mode II) and tearing or out of plane mode (Mode III) depending on the way the cracked surface is displaced. 24
MSE 313
Spring 2011
Fig.1Cii: Basic modes of loading involving different crack surface displacements (Source: reference 3) The crack-­‐tip stress field (based on elastic stress analysis) for each of the fracture modes is given by the following equations. Fig.1Ciii: Distribution of stresses in vicinity of crack tip (source: reference 3) K
#
# 3#
cos (1+sin sin )
2
2
2
2" r
K
#
# 3#
! x=
cos (1!sin sin )
2
2
2 2" r
K
# # 3#
$ xy=
(sin cos cos )
2 2
2
2" r
! y=
As shown in the Fig.1Ciii, the above equations give the values of stress field ahead of the crack tip. As r approaches zero, the local stresses could rise to extremely high values. However, this situation is avoided by the plastic deformation of material at the crack tip 25
MSE 313
Spring 2011
when the stress at the crack tip exceeds the yield stress of the material (there will always be a region of plastic zone at the crack tip). K in the above equations is called stress intensity factor and is a measure of the magnitude of the crack-­‐tip stress field. K = f(σ, a) Equation 1Cii The above equation states that K depends on the geometry of the cracked component and the manner in which the loads are applied to the component. The main difference between stress concentration and stress intensity factors is that the former takes into account the effect of crack size and crack-­‐tip radius whereas stress intensity factor takes into account both the geometrical terms (crack length appears in the expression whereas the crack-­‐tip radius is assumed to be sharp) and the stress level. Ideally, the crack should be atomically sharp. When the stress intensity factor exceeds a critical value called KIC, the crack propagation takes place. This critical value of stress intensity, which is a material property, is called fracture toughness. There are two scenarios that merit attention with regard to what happens to the plastic zone at the crack tip as a function of plate size. On one extreme, when the plate is too thin, the material at the crack tip is subjected to plane stress conditions and as a result the plastic zone size is larger than the thickness of the sample. Under these conditions, the fracture toughness of a material is dependent on the sample dimensions. On the other extreme, when the plate is too thick, plane strain conditions prevail and as a result the plastic zone size is relatively smaller and it is possible to reduce the plastic zone to be smaller than the smallest dimension of a fracture specimen by choosing a sample with larger dimensions. Therefore, by a proper choice of sample dimensions, it is possible to determine the fracture toughness of a metal under plane strain conditions which is independent of sample thickness and can be considered as material property. In this lab, plane strain fracture toughness testing of plain carbon steel and high speed tool steel samples will be carried out. ASTM standard E399-­‐90 describes about various specimen geometries and pre-­‐cracking procedures employed in plane strain fracture toughness testing of metallic materials. Three point bend test will be used to determine the fracture toughness of the samples in this lab. A schematic of three point bend test is shown in Fig.3. The pre-­‐crack in the specimen is introduced by notching the sample and subjecting it to fatigue loading. In general, the stress level during fatigue loading (10 to 20% of the final stress level) is such that minimal damage of the microstructure takes place during pre-­‐cracking. However, in the present lab, we will be using a notch without any pre-­‐crack. The notch itself acts as a crack. The fracture toughness can be calculated using the following equation: KQ = (
PQ L
hW
)f(
3/ 2
d
)
W Equation 1Ciii
26
MSE 313
Spring 2011
where, d
d2
(2.15 ! 3.93( ) + 2.7( 2 ))
d
d 1/ 2
d
d
W
W }]
f ( ) = 3( ) [1.99 ! ( )(1 ! ){
d
d
W
W
W
W
2(1 + 2( ))(1 ! ) 3 / 2
W
W
P W d P/
2 L/
2 L/
2 P/
2 d= W/2 Fig.1Civ: Schematic of single edge notched beam three-­‐point bend test used for fracture toughness determination. Figure 1Cv: Three-­‐point bend equipment for Instron 27
MSE 313
Spring 2011
Materials & Equipment: Plain carbon steel and high speed tool steel bars Instron machine Three point bend fixtures For the present lab, f(d/W)= 2.66 since d/W= 0.50 PQ = load (kN) h = Specimen thickness (cm) L = Span (cm) W = specimen depth (cm) d = crack or notch length (cm) The procedure followed in determining the load, PQ, will be described in the class. For more details on this topic, see calculations and interpretation of results section in ASTM standard E399. See AppendixA Part 2 References: (1) ASM handbook on metals, “Fractography”, 1986. (2) Annual Book of ASTM Standards, Vol 3.01, Standard E399-­‐90, ASTM, Philadelphia, PA. 2002. (3) Richard W. Hertzberg, “Deformation and fracture of engineering materials”, 4th edition, John Wiley & Sons, Inc., 1996. 28
MSE 313
Spring 2011
Part 2: Ductile to Brittle Transition of Metals: Lab provided by Mechanical Engineering: Written by Paul Labossiere Read ASTM E-­‐23 before the lab begins Appendix ME 354 MECHANICS OF MATERIALS LABORATORY
NAME
DATE
MECHANICAL PROPERTIES AND PERFORMANCE OF MATERIALS:
Part a) Charpy V-Notch Impact
February 2004 PEL
PURPOSE
The purpose of this exercise is to obtain a number of experimental results important for
the characterization of the mechanical behavior of materials. The Charpy V-notch impact is a
mechanical test for determining qualitative results for material properties and performance which
are useful in engineering design, analysis of structures, and materials development.
EQUIPMENT
• Charpy V-notch test specimens of 6061-T6 aluminum and 1018 (hot rolled) or A36 steel
• Charpy testing machine with 800-mm long pendulum arm and 22.6-kg impact head
• Type K thermocouple and digital readout unit
• Beakers of room-temperature water, warm water and boiling water
• Beakers of plain iced water
• Cryo-beakers of salted iced water and super cold liquids
PROCEDURE
CAUTION: When using the Charpy testing machine, stand well clear of the swinging area of the
pendulum both when the arm is cocked and for some time after the arm is released for a test while
it is still swinging. Serious injury will result from a swinging pendulum arm.
For each material repeat the following steps
• Designate a person as the "operator" of the Charpy test machine: all other persons must stand
clear during testing
• Designate a person as the "monitor and recorder" of temperatures and impact energies
• Designate a person as the "test specimen loader" who will remove test specimens from the
liquid bath, quickly placing them on the test fixture of the Charpy testing machine
• Designate a person as the "test specimen retriever" who will retrieve the broken halves of the
test specimens, will bind the halves together and will mark the test temperature on each pair of
specimen halves for later examination and inspection. Use the following procedure to conduct
tests in the order shown after exposure to the preconditions to give the approximate test
temperatures indicated:
Room temperature water (20 to 25°C)
Warm water (50-60 °C)
Boiling water (95-100°C)
Ice water (0 to 4°C)
Salted ice water (-15 to -18°C)
Acetone with some dry ice (-50 to -57°C)
Acetone with much dry ice (-80 to -85°C)
• Place the thermocouple probe in the appropriate liquid being sure to allow both the test
specimens and the thermocouple to equilibrate for at least five minutes prior to testing.
• Record the indicated temperature
29
MSE 313
Spring 2011
• "Cock" the pendulum by activating the "raise" mechanism and stand clear while the pendulum is
held in the "cocked" position.
• Using the tongs, quickly remove the test specimen from the bath and place it on the test fixture
with the notch opening facing away from the direction of the cocked pendulum
• Stand clear
• Release the pendulum
• Secure the pendulum in its rest position (i.e., hanging vertically) and retrieve the fractured
specimen halves.
• Record the impact energy (read directly from the dial on the Charpy testing machine)
• Repeat these steps for the each temperature and each material.
BACKGROUND AND ANALYSIS
Static or quasi-static properties and performance of materials are very much a function of
the processing of the material (heat treatments, cold working, etc.) in addition to design and
service factors such as stress raisers and cracks. The behaviour of materials is also dependent on
the rate at which the force is applied. For example, a polycarbonate tensile specimen which might
show a relatively low yield point but up to 200% elongation at a low loading rate may show a
much greater yield point but at only 5% elongation at an order of magnitude faster loading rate.
Low carbon steels, such as 1018, may show considerable increases in yield strength and work
hardening at high strain rates.
In quasi-static tests, the amount of energy required to deform a material is determined
from the area under the tensile stress-strain curve and is know as the modulus of toughness.
Under dynamic loading, stress-strain response is typically not recorded. Instead, the transfer of
energy from a device such as a drop weight or a swinging specimen to the deforming or breaking
specimen is equated to the "impact energy." The Charpy impact test uses a standard Charpy
impact machine to evaluate this impact energy. The machine consists of a rigid specimen holder
and a swinging pendulum hammer for striking the impact blow to a v-notched specimen as shown
in Figs. 1 and 2. Unfortunately, while the test, including machine and specimen geometry, has
been standardized, the test results do not provide definitive information about material properties
and thus are not directly applicable to design (as for example might be a yield strength); however,
the test is useful for comparing variations in the metallurgical structure of materials and in
determining environmental effects, such as temperature on the dynamic response of the material.
One of the most dramatic results of Charpy impact tests is in the form of plots of impact energy
versus temperature in which sigmoidally-shaped curves (see Fig. 3) show substantial decreases in
some materials' abilities to absorb energy below a certain transition temperature. This ductile to
brittle transition is most apparent in materials with BCC and HCP crystalline structures as for
example in steels and titanium. A classic and dramatic example of this ductile to brittle behaviour
is the low carbon steel Victory ships of WWII cracking in half under even the mild conditions of
sitting at anchor in a harbor. Materials with FCC structures (e.g., aluminum and copper) have
many slip systems and are more resistant to brittle fracture at low temperatures.
In this laboratory exercise the primary outcome will be plots of impact energy versus
temperature for two materials (FCC-606-T6 aluminum and BCC-1018 steel). Note the effects of
temperature and material type on the levels and shapes of the curves. Examine the fracture
surfaces of specimens and compare the type and degree of deformation to the impact energy and
the corresponding temperature. Consider not only the type of material, but also the effect of
notches and temperature in making design decisions.
30
MSE 313
Spring 2011
REFERENCE:
Annual Book or ASTM Standards, American Society for Testing and Materials, Vol. 3.01
E23 Standard Test Methods for Notched Bar Impact Testing of Metallic Materials
Figure 1. Schematic of Charpy Impact Testing and Izod and Charpy V-notch specimens
Figure 2 Charpy V-notch specimen used in this laboratory showing dimensions
Figure 3. Schematic of plot of impact energy versus temperature showing sigmoidal curve
31
MSE 313
Spring 2011
32
MSE 313
Spring 2011
Part 3: Fractography Objective The purpose of Fractography is to analyze fracture surfaces and to relate the fracture surface features to the cause(s) and basic mechanisms of fracture. Introduction Both visual examination and microscopic examination is used extensively to characterize the fracture surface features. Fractography is a valuable tool in identifying the mechanisms and causes of fracture of engineering components. Fracture of any material can be broadly classified into two main categories: ductile and brittle. Any material which fails after experiencing significant plastic deformation is called ductile whereas any material failing without appreciable plastic deformation is called brittle. In addition to these two broad classifications, a range of intermediate fracture modes are possible between ductile and brittle fracture behavior. In this lab we will be analyzing the fracture surface topography of the metallic and polymeric samples tested in the previous labs, using optical and scanning electron microscopes in addition to visual examination. We will also examine how scanning electron microscope has become an indispensable instrument in understanding the causes and mechanisms of failure of engineering components. Materials and Equipment: SEM Metallic and polymeric fracture samples from tension, MOR, Charpy and fracture toughness labs will be examined with the department SEM. Part 3 References: (1) ASM handbook on metals, “Fractography”, 1986. (2) Annual Book of ASTM Standards, Vol 3.01, Standard E399-­‐90, ASTM, Philadelphia, PA. 2002. (3) Richard W. Hertzberg, “Deformation and fracture of engineering materials”, 4th edition, John Wiley & Sons, Inc., 1996. 33
MSE 313
Spring 2011
Lab 1 Questions and Discussion: (Note: Don’t forget to include questions, data, discussion from part 2) • Prepare plots of Load vs. Displacement, Engineering Stress vs. Engineering Strain, and True Stress vs. True strain for each of the specimens provided. • Determine the following for the samples tested in a comparative chart: 0.2% offset Yield Strength, Ultimate Strength, Rupture Strength, Modulus of Elasticity, Modulus of Toughness, Percent Elongation, and Percent Area Reduction. Also include available literature values and compare your data with literature values. Your results must be presented in SI units. • Why are ceramics not tested in a manner similar to metals? • Find out the fracture toughness of both the samples and compare them. • Include the load versus displacement curves of both the samples and describe about the determination of load PQ. • Describe the appearance of each fracture surface. • For each fracture surface identify what type of fracture took place. • Locate and describe the indications for this type of fracture (justify your answer to #2). • How are metal failures different from polymeric failures? 34
MSE 313
Spring 2011
Blank Page
35
MSE 313
Spring 2011
MSE 313 Integrated Junior Laboratory -­‐ II Electrical Properties of Materials Part 1: Electrical Resistivity of Metal Material The purpose of this laboratory is to measure the electrical conductivity (resistivity) of selected conductors, insulators and semiconductors. This lab is designed to help the student understand how the electrical conductivity is related to temperature, alloying, permanent deformation and impurities. Part 2: Energy Gap in Semiconductors The purpose of this lab is to measure temperature dependence of the electrical conductivity in a semiconductor sample. The energy gap will be calculated from data taken in an intrinsic region, and the temperature dependence of the majority carrier mobility will be deduced from measurements taken in the extrinsic region. Part 3: Terminal Device Characteristics and Diode Characterization The objectives of this experiment are to learn methods for characterizing 2-­‐terminal devices, such as diodes, observe some fundamental trends in the characteristics of various diode types, and to gain some familiarity with standard test bench instrumentation. 36
MSE 313
Spring 2011
Part 1: Electrical Resistivity of Metal Material Objectives: The purpose of this laboratory is to measure the electrical conductivity (resistivity) of selected conductors, insulators and semiconductors. This lab is designed to help the student understand how electrical conductivity is related to temperature, alloying, permanent deformation, and impurities. Equipment: Kelvin 4 wire multimeter setup Liquid nitrogen Small furnace/cartridge heater Thermocouple LabVIEW station Materials: Chromel wire (nickel with ~10% chromium) Alumel (nickel with ~5% aluminum, manganese and silicon) Stainless steel 304 wire (Iron alloyed with chromium and nickel) Introduction and Background: 1. Electrical Resistance and Resistivity: Engineering materials are classified into three main categories: conductors, insulators and semiconductors. This classification is based on the ease with which these materials conduct electric current when an electric field (voltage) is applied. Students should review their MSE 351 notes for each of these three general material types. The electrical resistance (R) of a material through which an electric current is passing is given by Ohm’s law: Equation 1i Equation 1ii where V is the applied voltage (Volts), I is the current through the material (Amperes). Equation (1) can be rewritten as: 37
MSE 313
Spring 2011
where L is the length of the material, such that V/L is the strength of the electric field, and A Is the cross-­‐sectional area of the material, making I/A the current density. ρ is a material property termed electrical resistivity and is the inverse of the electrical conductivity σ of the material. The difference between resistance and resistivity is that resistance is dependent upon the geometry of the sample, whereas resistivity is a property of the material and is independent of the geometry of the sample. Therefore, it is important to calculate the electrical resistivity, so that it can be used as a design parameter. In order to convert the measured resistance to resistivity in this experiment, you must measure the diameter of the wire and the length of the wire between the contact points. 2. Electrical resistivity of conductors: In order to determine the resistivity, ρ, of a material, the electrical resistance, R, is measured. Samples of metals are provided in wire form. Because of the low value of the electrical resistivity of metallic materials, the connecting leads of the measuring instrument and the contacts between the probes and the sample can cause significant errors. To avoid these problems, a Kelvin (4 wire) resistance measurement is used. This method is described in detail in the website: http://www.allaboutcircuits.com/vol_1/chpt_8.html. The procedure is summarized and schematics are pasted where relevant. This resource is available in its entirety in the resource folder on the Jr. lab computer desktops. 2.1 Kelvin (4-­‐wire) resistance measurement: Suppose we measure the resistance of some component whose resistance (Rsubject) is similar to that of the wire resistance (Rwire). Although the wire resistance is very small (only a few ohms per hundreds of feet, depending primarily on the gauge, or size, of the wire), the measurement error introduced by wire resistance will be substantial in this case. A method of accurately measuring the subject resistance in a situation like this involves the use of both an ammeter and a voltmeter. From 38
MSE 313
Spring 2011
Ohm's Law (R = V/I), the resistance of the subject component can be measured from the voltage dropped across it: Because this is a series loop, current is the same at all points in the circuit. Since we are only measuring the voltage dropped across the subject resistance (and not the wire’s resistances), the calculated resistance is indicative of the subject component's resistance (Rsubject) alone. Our goal is to measure this subject resistance, but we still need to use a fairly large length of wire to make contact to our experiment. See the situation below: At first it appears that we have lost any advantage of measuring resistance this way, because the voltmeter now has to measure voltage through a long pair of (resistive) wires, introducing stray resistance back into the measuring circuit. However, upon closer inspection, it is seen that nothing is lost at all, because the voltmeter's wires carry miniscule current. Thus, those long lengths of wire connecting the voltmeter across the subject resistance will drop insignificant amounts of voltage, resulting in a voltmeter indication that is very nearly the same as if it were connected directly across the subject resistance: 39
MSE 313
Spring 2011
Any voltage dropped across the main current-­‐carrying wires will not be measured by the voltmeter, and so do not factor into the resistance calculation at all. Measurement accuracy may be improved even further by keeping the voltmeter's current at a minimum, either by using a high-­‐
quality (low full-­‐scale current) movement and/or a potentiometric (null-­‐
balance) system. This method of measurement, which avoids errors caused by wire resistance, is called the Kelvin (4-­‐wire) method. Special connecting clips, called Kelvin clips, are made to facilitate this kind of connection across a subject resistance: In regular, "alligator" style clips, both halves of the jaw are electrically common to each other, usually joined at the hinge point. In Kelvin clips, the jaw halves are insulated from each other at the hinge point, only contacting at the tips where they clasp the wire or terminal of the subject being measured. Thus, current through the "C" ("current") jaw halves does not go through the "P" ("potential," or voltage) jaw halves, and will not create any error-­‐inducing voltage drop along their length: 40
MSE 313
Spring 2011
2.2 Electrical resistivity of conductors: Most conductors are metals (all metals are conductors, a few ceramics and polymers are conductors). For metals the number of free electrons in the material is large. In addition, a pure metal has a crystalline structure with few impurities, which results in high electron mobility. The electrical conductivity of metals is on the order of 5 x 107 [Ohm-­‐m]-­‐1. See attached table. 1i The electrical resistivity of metals is influenced by temperature, alloying, and plastic (permanent) deformation. Increasing the temperature decreases conductivity (increases resistivity) because there is more electron scattering from thermal vibrations. The resistivity of metals increases due to the presence of “impurities” (often due to alloying) and defects (such as vacancies and dislocations) because they serve as scattering centers for the conducting electrons; effectively decreasing the mobility of these electrons with increasing impurity or defect concentration. 2.3 Electrical Resistivity and Temperature in Metals: One may think of the thermal component of electrical resistivity as arising from the interaction between moving electrons and atomic vibrations. Since the amplitude of atomic vibration increases with temperature and since the probability that an electron collides with a given atom depends directly on the area swept out by the vibratory motion, it can be argued that the mean free path between collisions decreases as the temperature is raised. In metals, the charge carrier density is high, even at low temperatures. The change of conductivity with temperature is mainly affected by the change of mobility with temperature. The relationship between resistivity and temperature in metals is approximately linear except at very low temperatures, ρT = aT +ρ0, where a and ρo are constants that depend on the type of metal and the imperfections in the metal (impurity atoms, dislocations, etc.) When the electrical resistivity of a metal or alloy is known at two different temperatures, the electrical resistivity at a third temperature may be obtained by linear extrapolation. The total resistivity of a conductor is given by the sum of the “thermal” component (ρth) and the “residual” component (ρr) in an equation known as Matthiessen’s Rule, ρ = ρth + ρr
Equation 1iii 41
MSE 313
Spring 2011
Figure 1i: Cartoon depiction of Matthiessen’s Rule and graphical relationship between
temperature and resistivity of a metal.
Table 1i: R vs T of various metals. 42
MSE 313
Spring 2011
Experimental Procedure: Experiment 1.1: The first experiment is to measure the resistance of conducting wires using the Kelvin 4-­‐wire probe procedure. (Fig. 1ii) Figure 1ii: Arrangement of wires • Precisely measure the length of the conducting wires (pure nickel, Chromel, Alumel, copper, and brass wires). • Measure the diameter of both wires. • Remember that you are only concerned with the length of the wire between the points where you will be making contact. • Make a loop with diameter around 2”, so that the loop of wire can be easily dipped into a liquid N2 bath. • Using the handheld meter, measure the resistance of the wires directly when the wires are subjected to various temperatures, such as room temperature (RT), furnace, liquid N2, etc. • Log the measured values for each wire, and then compute the resistivity of the wires for each temperature. • Plot the resistivity versus temperature, and discuss the behavior; (to be included in the lab report) Experiment 1.2: This experiment will be performed using LabVIEW to control the multimeter performing the Kelvin probe measurement of resistance of the wires. The output of the multimeter is fed into the computer through GPIB interface. This time, temperature is 43
MSE 313
Spring 2011
monitored by a thermocouple, and the output of which is fed into the computer through a NI-­‐DAQ device. • As in Experiment 1.2, , measure the resistance of each of the wires using LabVIEW for temperatures ranging from RT to a heated furnace and RT to liquid N2. • Compute the resistivity of the wires for each temperature and plot the resistivity versus temperature. • Discuss the behavior (to be included in report). Figure 1iii: Lab Set-up.
Experiment 1.3: Mark out 1 ft. of each of the wires to measure their resistance as a function of temperature as described above after mechanical deformation. • To induce the mechanical deformation, wrap the ends of the wire around two rods so that the wire will not slip when pulled. The length of the wire between the rods should be 1 ft. • Pull the wire apart slowly, and the wire should extend by 20-­‐30 %. You should feel an increase in the force required to elongate the wire. • Carefully measure the new length and diameter of the wire. • If the Instron is available, put the wire in the tensile apparatus taking care to ensure that the gage length is at least the area of interest. Test rate should not exceed 1.2 mm/min. Continue test until material is in the plastic region, then stop, remove the wire. • Measure the resistance of all wires after deformation at all temperatures described above. • Calculate the resistivity and conductivity. • Plot the resistivity data as a function of temperature and determine ρr from the deformed specimens. 44
MSE 313
Spring 2011
Part 2: Energy Gap in Semiconductors Objective: This lab is to measure the temperature dependence of electrical conductivity in a semiconductor sample. Energy gap will be calculated from the data taken in an intrinsic region, and the temperature dependence of the majority carrier mobility can be deduced from measurements taken in the extrinsic region. Theory: Semiconductors in the pure state are called intrinsic semiconductors. If impurities are added to the material the electrical characteristics change and the semiconductors are called extrinsic semiconductors. In either type of semiconductor there are two types of carriers: electrons, and holes. When an electron becomes free, it leaves behind a positively charged hole. Both the hole and the electron carry charge. Therefore, the conductivity is equal to the sum of the contributions of each. Equation 2i where n is the number of electrons, p is the number of holes, q is the charge of the carrier (hole or electron), and m is the mobility of the carrier. In an intrinsic semiconductor, the number of electrons is equal to the number of holes, which simplifies the expression for conductivity. Equation 2ii In extrinsic semiconductors, the semiconductor is doped with a Group IIIA element (B, Al) or a Group VA element (P, As). The Group IIIA elements need an electron to complete their outer shell and thus “steal” it from a silicon atom, leaving behind a hole, creating p type carriers. These are called p-­‐type semiconductors. In a p-­‐type semiconductor the number of holes due to doping is far greater than the number of electrons and only the second term of equation (5) is important. Equation 2iii The number of carriers, p, is simply equal to the number of impurity atoms. The group VA elements have an extra electron, which is donated to the silicon adding n type carriers. These are called n-­‐type semiconductors. In this case, when an electron becomes free, a hole is not left behind and the number of electron carriers is far greater than the number of holes. 45
MSE 313
Spring 2011
Equation 2iv Again, the number of carriers is equal to the number of electrons donated by the impurity atoms. A doped semiconductor exhibits extrinsic behavior when the dopant concentration Nd is much larger than the concentration of the electron-­‐hole pairs ni generated by thermal excitations. In this case, provided the temperature is not too low, the "free charge" concentration n = ni+Nd ≈ Nd is temperature independent, and the charge transfer is essentially due to the majority carriers (holes in p-­‐doped, or electrons in n-­‐doped samples). At even lower temperature the semiconductor enters the so called "freeze-­‐out region" and the impurity carrier concentration decreases exponentially as exp(-εd/2kT),
where εd is the impurity ionization energy (εd≈10 meV in Ge and εd≈40 meV in Si) and k is the Boltzmann constant. In the extrinsic region the current density can be simply written J = e n vD = q n µ E
Equation 2v
σ = q(nµn+pµp) = (1+b) niµn
Equation 2vi
where vD is the drift velocity, m the mobility, and e is the elementary charge. The electrical conductivity is proportional to the charge carriers concentration and to the mobility: σ = q n µ. Therefore in the extrinsic region, since n ≈ constant, the temperature dependence of the conductivity and mobility are identical. Theoretical calculations, accounting for lattice scattering of the charge carriers and neglecting contributions due to scattering with impurities, give a mobility µ ∝ T-α, with α = 3/2.
The experimentally observed α value, however, is usually larger than the predicted value 3/2, ranging from 1.6 to 2.5. The thermal generation of the electron-­‐hole pairs grows exponentially with temperature, and when the temperature is high enough the sample enters the intrinsic region where Nd becomes negligible with respect to the concentration of thermally generated electrons (ni) and holes (pi=ni). In the intrinsic region we must use the ambipolar conduction formula where the mobility ratio b = µn/µp is nearly temperature independent. We have ni = pi ∝ T3/2 exp[-Eg(T)/2kT]
Equation 2vii
where Eg(T) = Eg(0)-γT is the temperature dependent energy gap that separates the conduction band from the valence band. Here the temperature dependence of the 46
MSE 313
Spring 2011
conductivity is largely dominated by the exponential dependence of the carrier concentration, so that a semi-­‐logarithmic plot of σ versus 1/T yields a straight line with slope Eg(0)/2KB. Experimental Setup: The experiment can be performed using a constant current generator, a voltmeter with high input resistance, a thermometer, and a device for changing the sample temperature at a small rate. Figure 2i: Typical set-­‐up for this experiment The experimental apparatus will consist of a liquid nitrogen bath with a large thermally conductive mass in it, upon which will sit the copper block containing both a cartridge heater and the samples. The semiconductor sample will be affixed to the copper heating block, and temperature measurements will be taken from the built-­‐in thermocouple (inside of the cartridge heater). In this experiment, we will use a “unijunction transistor” as a Si sample (Texas Instruments 2N2160 (metal case)). The unijunction transistor is made of a bar of n-­‐type Si material with a p-­‐type junction near the center, as seen in Figure 2ii below. 47
MSE 313
Spring 2011
Figure 2ii: Unijunction transistor Since this device has the base pins 1 and 2 connected with ohmic contacts to the semiconductor substrate, we will only use the connection between base 1 and base 2 so that the unijunction transistor is nothing more than a n-­‐type Si specimen. The emitter pin is disregarded in this case. A typical run should be performed within less than 2 hours, with the temperature changing at a rate ∂T/∂t ≈ 5×10-­‐2 Ks-­‐1. Using a liquid nitrogen bath the temperature can be varied in the range 80 K < T < 430 K. The upper limit is imposed by the melting of the junction contacts on the sample to T ≈ 520 K. The smaller the temperature slope ∂T/∂t is, the smaller the thermal gradient ∂T/∂x will be within the sample. Experimental results obtained in two typical runs are shown in Figure 3, for a germanium sample (n-­‐doped, ρ ≈ 14 Ω cm) and for a unijunction transistor, respectively. Here T is the temperature in Kelvin and t is the temperature in Celsius (t = T – 273.15). Figure 2iii: The voltage drop measured across the semiconductor samples in the whole temperature range. The Germanium sample is a 2x2x10 mm bar, n-­‐doped. The Silicon sample is the base, n-­‐doped, of a 2N2160 unijunction transistor. 48
MSE 313
Spring 2011
Figure 2iv (a) shows the Ge resistance in the intrinsic region plotted versus the reciprocal absolute temperature 1/T in a semi-­‐log plot. The slope of the linear fit for 58°C < t < 143°C gives Eg(0)=0.79±0.02 eV. The same plot for the Si sample is shown in figure 2iv (b). The best fit in the temperature range 185°C <t<255°C gives Eg(0)=1.18± 0.02eV. In figure 2iv (c) the log10R for the germanium sample is plotted versus log10T and the slope of the fitting line for -­‐160°C <t<-­‐10°C gives α=1.65±0.02. The extrinsic region for the unijunction transistor (figure 2iv d) spans the temperature range -­‐100°C < t < +100°C and the exponent in the power law µ ∝ T-­‐α is α=2.30± 0.03 . The slight deviation from linearity at the lowest temperatures indicates the beginning of the "freeze-­‐out region". Figure 2iv: (a) and (b): the linear fit in the intrinsic region (high temperature); (c) and (d): linear fit in the extrinsic region (low temperature) Experimental Procedure: A unijunction transistor consists of three wires, two of which are directly connected to the n-­‐type Si bar through ohmic contact and one of which is connected to the p-­‐type region of the bar, as in figure 2v below. Since there is no indication on the transistor as to which lead is which, the student will need to test the transistor to identify all the pins. 49
MSE 313
Spring 2011
Figure 2v: Interpretation of the internal arrangement of a unijunction transistor. • Using the knowledge gained from MSE 351, first identify which two leads are connected to the n-­‐type region of the bar, and finally which one is connected to the p-­‐type region. • Once the two leads connected to the n-­‐type Si bar have been identified, connect these two leads to a constant current source. • Set the constant current source to generate 0.2 mA and measure the voltage across the unijunction transistor as a function of temperature using LabVIEW. • Generate a logR vs. 1000/T plot for intrinsic temperature region and a logR vs. logT plot for the extrinsic temperature region. • If necessary, the Fluke Digital Voltmeter can be used, the specifications for this instrument in Resistance mode is described in the Appendix. 50
MSE 313
Spring 2011
Part 3: Terminal Device Characteristics and Diode Characterization
NOTE: This lab has been modified from an EE331 written by R.B. Darling lab with permission. Objective The objectives of this experiment are to learn methods for characterizing 2-­‐terminal devices, such as diodes, observe some fundamental trends in the characteristics of various diode types, and to gain some familiarity with standard test bench instrumentation. Precautions: None of the devices used in this set of procedures are particularly static sensitive; nevertheless, you should pay close attention to the circuit connections and to the polarity of the power supplies, diodes, and oscilloscope inputs. Experiment #1: I-­‐V characteristics of diode with forward bias Set-­‐Up We will use two commercially available p-­‐n junction diode, 1N34 and 1N4004. 1N34 is a germanium based p-­‐n junction diode, whereas 1N4004 is silicon based. We start with the forward bias condition to measure the I-­‐V characteristics of the diode. Pre-­‐experiment: Verify which side of the diode is p or n using an ohm meter. Based on your knowledge learned in MSE 351, think about how to do this task. For this lab you will be using a Breadboard, which is typically a piece of white plastic with many small holes as shown below. Some of the holes are already electrically connected with each other, allowing you to insert leads from devices (such as resistors or diodes) as well as wires to create basic circuits without the need to solder or wire wrap. The holes are 0.1 inch apart, which is the standard spacing for leads on integrated circuit dual in-­‐line packages. You will test the connections in this lab. 51
MSE 313
Spring 2011
Note that each set of 5 horizontally oriented holes constitutes a tie point. The vertical columns of holes are all internally connected into a single tie point; these are normally used for power supply distribution. To attach test leads to the breadboard, you can use either the exposed end of a component lead, or you can insert a small pin into the appropriate tie point and connect the squeeze hook or oscilloscope probe to the pin. Using the breadboard, set up the following circuit. DMM1 (+)
R
1
VSS
DMM1 (-)
DC
DMM2 (+)
SUPPLY
D
1TEST DIODE
DMM2 ( )
Here, DMM refers to terminals that are to be connected to a Digital Multi Meter (DMM). You will need two DMMs to measure the voltage drop across (1) the resistor R1 and (2) the diode D1, as shown above. Create such a circuit for each of the following six resistors: R1 = 100 Ω, 1.0 kΩ, 10 kΩ, 100 kΩ, 1.0 MΩ, or 10 MΩ, 1% 1/4W To speed up this process, you may wish to insert all six resistors and the two diodes into the breadboard at once so that one end of each resistor connects to the anode of each diode. The long, horizontal tie point strips on the solderless breadboard are quite convenient for this. The proper resistor and diode can then be quickly selected by simply moving the power supply leads. Use the bench DMM to measure the DC voltage across either the resistor or diode. Measurement: For each of the diodes (1N34A and 1N4004), use the following procedure. • First, adjust the DC power supply VSS to produce +10.0 Volts across R1 and monitor the voltage on DMM1. • Measure the forward turn-­‐on voltage of the diode with DMM2. If two DMMs are not available at your lab bench, you may have to switch back and forth between the two terminals at DMM1 and DMM2. • Record the diode's current and voltage in a table in your notebook. The diode current can be calculated using Ohm’s Law (I = V / R), which is equal to 10.0 V / R1. 52
MSE 313
•
•
Spring 2011
Change the resistor to the next value and repeat. Measure the six different (I, V) pairs for each diode. Experiment #2: I-­‐V characteristics of diode with reverse bias: Set-Up :
Configure a DC power supply to produce an output voltage of VSS = +10.0 Volts. If the DC
power supply has a current limiting ability, configure the power supply to limit the current to
100 mA. Route the output of the DC power supply to your breadboard using two squeeze
hook test leads. Note: Diodes are direction dependent. If you are unsure about any set up,
have your TA look it over BEROFE turning on power.
In this section, you will measure the leakage current of the same diodes used in the experiment #1. Each diode should be connected as shown in the figure below. Use the following parts: R = 1.0 MΩ 1% 1/4 W DMM (+)
D1 = same as #1 R = 1M
VSS
DMM (-)
DC
SUPPLY
D1
TEST DIODE
Connect only one diode at a time in the circuit (above), noting that the banded end is the cathode, which corresponds to the bar on the circuit symbol. Connect the DC power supply across both R and D1 and then connect the DMM across only R using two pairs of squeeze hook test leads as shown above. The DMM should read less than +10.0 V. Measurement: Measure the reverse leakage current for each of the same diodes in Experiment #1. Do this by using the DMM to measure the voltage across R, after which divide this voltage by the resistor value (e.g., R = 1.0 MΩ) to obtain the current through R, and therefore the current through D1. Record your measurements and calculations in a table in your notebook. 53
MSE 313
Spring 2011
Experiment #3: Measurement of diode I-­‐V characteristics using the oscilloscope Set-Up:
In this procedure, you will use an oscilloscope and the laboratory variable transformer to display the current-­‐voltage (I-­‐V) characteristics of a diode. This procedure relies entirely upon the ability to float the transformer output at a potential that is different from the ground of the oscilloscope. (All oscilloscopes have each channel grounded to the 120 VAC safety or chassis ground, so an oscilloscope can only be made to float by the use of an additional isolation transformer.) This procedure can also be performed using a signal generator that produces a floating output; however, the following procedures assume that you are using the laboratory transformer. BEFORE turning on the power, have your TA verify your set-­‐up. Set up and connect the circuit shown below using the following components: R1 = 1.0 kΩ 1% 1/4 W D1 = 1N34A and 1N4004 SCOPE CHANNEL-1
BLAC
(X-INPUT)
K
D1
V
SCOPE GROUND
1
LAB XFMR
R1
1.0 k Ω
SCOPE CHANNEL-2
WHITE
(Y-INPUT)
BREADBOARD
Plug the laboratory transformer into a 120 VAC receptacle, and turn its power switch OFF. Connect one lead from the black banana jack output of the lab transformer to the diode on the breadboard, and then connect another lead from the red banana jack output of the lab transformer to the resistor R1 on the breadboard. This will establish a 60 Hz sinusoidal input to the circuit. You may set your variable transformer to around a 10V peak-­‐to-­‐peak signal. More Set-­‐Up: Next, configure an oscilloscope to display the I-­‐V characteristics as follows: Attach an oscilloscope probe to Ch-­‐1, connect the probe to the diode (the same connection point as the black output of the lab transformer), and connect its ground lead to the junction 54
MSE 313
Spring 2011
between the diode and the resistor. Attach a second oscilloscope probe to Ch-­‐2, connect the probe to the resistor (the same connection point as the red output of the lab transformer), and connect its ground lead to the junction between the diode and the resistor. Configure the oscilloscope to produce an X-­‐Y display, using Ch-­‐1 as the X-­‐axis and Ch-­‐2 as the Y-­‐axis. Set Ch-­‐2 to invert the incoming signal. Turn ON the power switch on the lab transformer to energize the circuit. At this point you should have something on the screen which resembles the I-­‐V characteristics of a diode. Adjust the position controls to center and calibrate the curve to the center point of the screen as follows: Switch both Ch-­‐1 and Ch-­‐2 input couplings to GND. Adjust the vertical position control for Ch-­‐2 and the horizontal position control to move the dot to the exact center of the oscilloscope screen. After having done so, return both the Ch-­‐1 and Ch-­‐2 input couplings to DC. You may need to decrease the intensity of the trace to remove any halo from around the dot. Also adjust the scaling on both the x and y axis such that you can take readings off the curve. The oscilloscope should now be displaying a graph of the current-­‐voltage (I-­‐V) characteristics of the device. The vertical axis or y-­‐input is proportional to the current through the diode, since it measures the voltage across R1. The voltage across R1 is proportional to the current flowing through it, and this same current flows through the diode. The horizontal axis or x-­‐input is proportional to the voltage across the diode. Thus, this circuit produces a simple, but effective and accurate curve tracer. Note that the Ch-­‐2 input to the oscilloscope must be inverted in order to account for the polarity of the voltage drop produced across R1. This then keeps the I-­‐V characteristics of a passive device within quadrants 1 and 3 of the I and V axes, as they are normally drawn. Almost all commercial curve tracers, such as the Tek-­‐576, perform their voltage sweep at a 60 Hz rate. This is usually derived directly from the AC line frequency. This feature has the advantage of making the sweep synchronous with the AC power line and therefore somewhat more robust to AC power line interference. At a different sweep frequency, the I-­‐V characteristics would otherwise flutter around as a result of beating with fluorescent light and other stray pick-­‐up coupling which might be oscillating at 60 Hz. Measurement: • Sketch the I-­‐V characteristics of each diode in your notebook (they should look like the oscilloscope trace) on the same set of axes. • Using the scaling factors from the oscilloscope, scale the x and y axes of your sketch with tick marks for current and voltage. • Graph paper is handy for this and makes the following analysis easier. • You may wish to keep the lab transformer and the oscilloscope in their present set-­‐
up configuration, since they will be used again to measure additional I-­‐V characteristics in Procedure 4. 55
MSE 313
Spring 2011
Experiment #4: Measurement of diode I-­‐V characteristics using LabVIEW Computer-­‐controlled automatic measurements are commonly used to gather data for the purpose of characterizing or testing a device or system. In this experiment, a LabVIEW curve tracer will be used to capture the characteristic I-­‐V curve for a pn-­‐
junction diode. This procedure will also use the data acquisition (DAQ) card in the computer to both create the excitation voltages and to measure the resulting test voltage responses. No other external bench instruments are needed other than the computer, the DAQ card, its cable, and the BNC-­‐2120 connector block. Software Set-­‐Up: First, log on to the computer and launch LabVIEW. From File > Open … , open the VI named “DiodeCurveTracer.vi.” For this VI to open correctly, two sub-­‐VIs named “VoutArray.vi” and “IODriver3.vi” must also exist in the same directory as “DiodeCurveTracer.vi”. Theory of Operation: The front panel window has been designed to show roughly how this diode curve tracer operates. First, the six controls inside the Scan Range box are used to set up the sequence of voltages which will be applied to the resistor and diode series combination. The forward and reverse parts of the scan are independently set up according to their starting value, their ending value, and the number of points used for each. When the START SCAN button is clicked, this sequence of voltages is passed to the power supply which first steps upward in the forward direction, then back down, then steps downward in the reverse direction, and then back up, making one complete cycle through the applied bias range for which the diode is to be tested. The cathode end of the diode (end with the bar) is grounded, and the voltage at the node between the diode and the resistor is measured, along with the voltage that is applied to the other end of the resistor by the power supply. The voltage across the diode is used to create the x-­‐
values for the I-­‐V curves, and the y-­‐values of diode current are obtained by subtracting the voltages at the two ends of the resistor and then dividing by the value of the resistor. This is typically how one accomplishes current sampling with a data acquisition system. The value of the current sampling resistor is input using the control located to the right of the resistor symbol. The diode voltage and current are then plotted as (x,y) pairs in the chart. After the scan is complete, the SAVE DATA button can be clicked to write the data to a spreadsheet file. Hardware Set-­‐Up: The power supply output and the two measured voltages are implemented through channels on the data acquisition (DAQ) card. Analog Output channel # 0 (AO-­‐0) is used to create the power supply output voltage. The voltage from the power supply is 56
MSE 313
Spring 2011
measured by Analog Input channel # 0 (AI-­‐0), and the voltage across the diode is measured by Analog Input channel # 1 (AI-­‐1). The hardware is set up by connecting the diode being tested and the current sampling resistor to the proper terminals of the BNC-­‐2120 connector block. Connect a 5.1 kΩ resistor between the center pins of the BNC connectors for AI-­‐0 and AI-­‐1. Then connect a jumper wire between the center pins of the BNC connectors for AO-­‐0 and AI-­‐0. Finally, connect a diode between the center pin of the BNC connector for AI-­‐1 and its ground shield, with the bar end of the diode (its cathode end) connected to the ground shield. Also make sure that the slide switches for AI-­‐0 and AI-­‐1 are both set to BNC inputs and that the slide switches for AI-­‐0 and AI-­‐1 are each set for grounded source measurements (the GS position). This will connect each of the BNC outer shields to the internal analog ground with a built-­‐in 4.99 kΩ resistor. However, to make the diode curve tracer work properly, each of the analog input BNC outer shields (their (−) inputs) must be shorted to ground. The outer shield of the BNC connector and the analog outputs are true analog grounds. Thus, connect the BNC outer shields of AI-­‐0 and AI-­‐1 to the BNC outer shield of AO-­‐0 to establish a proper ground system on the BNC-­‐2120. The BNC-­‐2120 connector block has far more BNC connectors than the number of BNC connector cables that a student group would normally have. To get around this, much of the wiring can be accomplished by simply inserting component or jumper wire leads into the center pins of the BNC connectors. If you choose to do this, be very careful not to damage the center pins of the BNC connectors. Most component leads by themselves are too narrow to fit securely in the BNC center pin hole. However, if you insert a pair of component leads, or a pair of #22 AWG jumper leads, this will usually provide a sufficiently tight connection that will not damage the BNC center pins. Interconnecting the BNC outer shields can be done using squeeze hook jumper leads. From the description of Hardware Set-­‐Up, schematically draw the electrical wiring diagram. By doing so, you will have a better understanding of what you are going to do. First try by yourself, then ask TA for help. Measurement: • From the front panel window, click on the ‘run’ button to start the diode curve tracer VI. • Keep the default settings for the forward and reverse bias scan ranges: forward bias: 10 steps from 0.0 Volts to +5.0 Volts, and reverse bias: 10 steps from 0.0 Volts to −5.0 Volts. • After rechecking all of the connections, click on the START SCAN button, which should start the measurement sequence and then display the resulting diode I-­‐V characteristics on the x-­‐y chart. 57
MSE 313
Spring 2011
•
•
•
•
•
Once you are happy with the measurement, click on the SAVE DATA button to save the measured diode I-­‐V characteristic data in an Excel spreadsheet format. A Save As … dialog window will open, and you can type in the name of the file that you want the data written to, for example, “Experiment1Procedure4.xls.” Click on OK to write the file. Once you have saved the data, click on the STOP button to halt the measurement VI. You should open this newly created file with Excel to verify that the data was properly written to the file. If everything was working properly, the diode voltage values (in units of Volts) should appear in the first column, and the diode current values (in units of milliamperes) should appear in the second column. It is generally a good idea to halt any running VI when you are done with it. If you wish to use other Windows programs, such as Excel, or Internet Explorer, or Windows Explorer, you will find that these programs will run rather slowly while any VIs are running at the same time. Experiment #5: Measurement of a zener diode Set-­‐Up: Replace the diode used in the Experiment #4 with a 1N4732 zener diode, keeping the banded end (the cathode) connected to the ground shield of AI-­‐1. Change the Scan Range Box settings to scan upwards from 0.0 V to +10.0 V in 20 steps, and then downwards from 0.0 V to −10.0 V in 20 steps. Just clicking on the up/down buttons is the easy way to accomplish this. You will also need to rescale the displayed I-­‐V graph by clicking on the graph and then right-­‐clicking on Properties. Go to the Scales tab and first select the X-­‐Axis from the drop-­‐down list. Change the minimum and maximum to −10 V and +10 V. Similarly, select the Y-­‐Axis from the drop-­‐down list and change the minimum and maximum to −0.5 and +0.5 (mA). Measurement: • Press the START SCAN button to begin the measurement, after which, the resulting I-­‐V characteristics of the zener diode should appear in the displayed graph. • Once you are happy with the measurement, click on the SAVE DATA button to save the measured diode I-­‐V characteristic data in an Excel spreadsheet format. • A Save As … dialog window will open, and you can type in the name of the file that you want the data written to, for example, “Experiment1Procedure5.xls.” Click on OK to write the file. • Once you have saved the data, click on the STOP button to halt the measurement VI. 58
MSE 313
•
•
Spring 2011
As a double check of the data, reverse the polarity of the zener diode, connecting its non-­‐banded end (its anode) to the shield of AI-­‐1. Press the START SCAN button again, and you should see the same characteristics as before, but now inverted about the origin of the graph (i.e. switching quadrants 1 and 3). Experiment #6: Characterization of a light-­‐emitting diode (LED) Circular LED's, as well as other small panel lamps, come in several standard sizes. A T-­‐1 size is 3 mm in diameter, and a T-­‐1 3/4 size is 5 mm dia. There are several ways of identifying which terminal is which on an LED. If the leads have not been cut, the anode or (+) lead will be the longer of the two. (This also holds true for parallel lead electrolytic capacitors.) If the leads have been cut, you will have to use the next method. Look straight down on the hemispherical dome of the LED (so that the LED would be shining toward you) and you should notice that the small lip at the bottom of the plastic has a flat side on it. The lead that is closest to this flat side is the cathode or (−) lead. Set-­‐Up: Locate a T-­‐1 red LED and replace it for the diode in the LabVIEW curve tracer of Procedure 5 or 6. Start the DiodeCurveTracer.vi by pressing the Run button on the toolbar, and set the Scan Range Box settings to scan upward from 0.0 V to 10.0 V in 20 steps, and then downward from 0.0 V to −10.0 V in 20 steps. Make certain that at least a 1 kΩ resistor is used in the circuit, to limit the current in either direction to no more than 10 mA. Measurement: • Press the START SCAN button to initiate the measurement process. • You may notice that the LED will briefly glow as the curve tracer increases the sweep voltage. The resulting I-­‐V characteristics for the LED should appear on the displayed graph. • Once you are content with the measurement, click on the SAVE DATA button to save the measured diode I-­‐V characteristic data in an Excel spreadsheet format. • A Save As … dialog window will open, and you can type in the name of the file that you want the data written to, for example, “Experiment1Procedure6.xls.” Click on OK to write the file. • Once you have saved the data, click on the STOP button to halt the measurement VI. 59
MSE 313
Spring 2011
Lab 2 Questions and Discussion: • Does the relation ρT = aT +ρ0 still hold for these materials at all temperatures examined? • Given what you know of transistors, why was a unijunction transistor used instead of any other type? • Create a semi-­‐log plot of I versus V, where I is on a log scale and V is on a linear scale. • For each decade of increase in the diode current, by how much does the diode turn-­‐on voltage increase? • Identify current ranges on your graphs that correspond ideally to diode factors of 1
and 2. Identify any other obvious trends.
• Which diode of the two you tested would be the most suitable for charging up a capacitor using a peak rectifier circuit and allowing the capacitor to keep its charge for the longest period of time (keeping in mind that you would prefer to minimize the voltage dropped across the diode and that you would want the minimal amount of energy to leak back out once the capacitor was charged)? • If the polarity of the diode was reversed (connecting its cathode to the current sampling resistor and AI-­‐1 and connecting its anode to the analog ground), what would be the expected I-­‐V curve? • If the diode was replaced by another 5.1 kΩ resistor, what would be the expected I-­‐V curve? To what would the slope of the resulting I-­‐V curve correspond? • From your sketch, extract the forward-­‐bias turn-­‐on voltage (Von) for each diode. Compare your answers to the results of the previous DMM readings. • Using the data that was collected in the spreadsheet file, compute a value for the zener resistance rz of the diode in its breakdown region. Similarly, compute a value for the forward (on) resistance rf of the diode. The easiest way to do this for both regions is to identify two strategic (I,V) points which define the best fit lines in these regions and then compute the inverse slopes of these lines. • The power rating of the 1N4732 zener diode is quoted at 1.0 Watt. Calculate the maximum current that the diode can handle in the forward (on) direction and then in the reverse (zener) direction and not exceed the 1.0 Watt limit. 60
MSE 313
Spring 2011
•
Insert the 1N4732 zener diode into the curve tracer made from the lab transformer and the oscilloscope. Compare the resulting I-­‐V characteristics with those obtained from the LabVIEW curve tracer. •
Discuss why the turn-­‐on voltage of the LED is significantly higher than that of a typical silicon switching or rectifier diode. • References: Kasap, S.O., “Principles of Electronic Materials and Devices” 3rd edition. McGraw Hill, 2006. Kuphaldt, Tony, “All About Circuits Vol. 1 DC”. On-­‐line resource: <http://www.allaboutcircuits.com/vol_1/chpt_8.html. >. Date accessed: March 18, 2011. Darling, R.B., “EE-­‐331 Laboratory Handbook; 2-­‐Terminal Device Characteristics and Diode Characterization”. Date not provided. 61
MSE 313
Spring 2011
Blank Page 62
MSE 313
Spring 2011
MSE 313 Integrated Junior Laboratory-­‐III Screen Printing of Thick Film Materials: Acknowledgement: This lab was written in part by Monika Marciniak as part of her Senior Project in 2008. Part 1: Print/ Dry/Fire Resistor Circuits The purpose of this laboratory is to screen print, dry and fire thick film circuits made of conductive ink. This lab is designed to introduce students to the technology of printing thick film and hybrid circuit using a screen-­‐printing machine. The fundamentals of the equipment, inks, and thick film materials processing will be explored. students will practice screen and stage alignment to obtain patterns at precise locations on the substrates. Part 2: Firing Profiles/Resistance Measurements / Pattern Design Principles: The purpose of this laboratory is to fire thick film resistors. This lab is designed to obtain and study different firing cycles of the resistors and to determine the effect of firing temperatures on the resistors’ performance. Students will also study aspects of thick film design and layout. The students will examine the patterns of printed/fired thick film circuits obtained from Part 1. Factors such as geometry, sheet resistance, print thickness and electrical performance will be studied and students will compare calculated theoretical results with the experimental data Part 3: Characterization: The purpose of this lab is to characterize the final printed resistor circuits using Fluke multi-­‐meter data acquisition and gain thickness measurements with a profilometer. 63
MSE 313
Spring 2011
Screen Printing Fundamentals: Background: Screen printing is a process whereby functional material is deposited in a controlled manner onto a substrate. It can be a simple, hydraulic style device or a complicated computer control system. But the same basic principles apply. Thick film screen-­‐
printing has been used in the electronics manufacturing industry since the 1940’s when it was first introduced during World War II as an alternative method to conductive wiring in circuits. It is still used today for preparing conductive pathways, resisters, and other passive components in the semiconductor industry. And, as we shall see, for manufacturing sensing components, i.e., tin-­‐dioxide sensing materials. Screen Printing Basics: There are a number of factors to consider when utilizing screen printing for circuit design and manufacturing. Choosing the type of paste for the functional components and for the conductive and resistive components is one important design consideration. One must also determine the parameters of the screen to be used. What kind of screen material, what kind of emulsion material, what mesh size, all must be considered and decisions made concerning before moving on to printing the devices. We have already done the decision making for the pastes, next we need to determine the screen design and type. There are two primary materials used in screen making. Stainless steel and nylon or polyester. Stainless steel is used in our labs for the obvious reason that it is much more durable. Nylon is used primarily when irregular surfaces are to be printed. However, Nylon tends to absorb water and because of its properties some kinds of paste will stick to the screen during snap-­‐off reducing the quality of the finished component. Stainless steel is durable and doesn’t absorb water. However, it can easily be damaged by being stretched out, bent, or broken, so regardless of the screen material care must always be used. The second consideration when designing/choosing a screen is the emulsion. The emulsion is a material that is photosensitive which is deposited on the back of the screen mesh. A pattern is ‘printed’ on the emulsion and then the screen is developed. UV light is generally used. After the development step is completed, the screen has the pattern exposed. The thickness of the emulsion and the screen will help to determine the minimum thickness of the components printed. And finally, the mesh size should be considered. The mesh count/opening as well as the weave will effect the thickness and line resolution of the final print. The thickness and rheology of the inks being used will drive the decisions about the screen mesh and visa versa. There are a number of materials used as substrates. Most common are Al2O3, Si, and Kapton, a polymer. The substrate is chosen to best fit the system overall but two 64
MSE 313
Spring 2011
important considerations will be firing temperatures required for the pastes and adhesion properties. The substrates we are using are Al2O3. These substrates are robust and fit into the screen printer well. Because of the temperatures required to fire the various pastes we are using, the alumina is the best choice. Theory: Screen-­‐printing technology is based on the formation of patterns through the use of a woven cloth ‘screens’ and viscous pastes. Some of the openings of the screen are blocked with filler and, when the paste is forced through those remaining onto the substrate, a replica of the pattern on the screen is obtained. Virtually any material that can be formulated into a viscous paste may be used, including most liquids and solids. There are many variables that may affect the quality of prints produced by the thick film process. These variables are categorized into the following main groups, and subgroups: Ink properties (viscosity, surface tension, ink composition) Screen variables (stencil, mesh, frames, emulsion) Machine variables (squeegee, snap-­‐off, flood blade) Deposition, drying and firing procedures. Description of some of these screen-­‐printing factors is provided below. Deposition, drying and firing are explored in experimental procedures. Ink/Paste Rheology: The flow behavior of thick film inks is controlled by the viscosity and surface tension of the ink and by the dependence of these properties on shear rate and time. Both of these properties are related to interatomic forces within the body of the fluid, viscosity being regarded as the resistance to motion of one layer of fluid over another, whereas the surface tension arises from the imbalance of forces at an interface between two different materials. Thick films are classified as non-­‐Newtonian pseudo-­‐plastic fluids, which means that they show decreasing resistance to flow with increasing shear stress. In other words, as the squeegee speed is increased, the ink becomes more fluid. That characteristic is very desirable, because as the squeegee shears the ink, the ink passes easily through the openings of the screen, wets the substrate, and releases cleanly from the screen a the screen peels away from the substrate. As soon as the shearing force is removed, the ink quickly recovers to nearly the original viscosity. The ink remains fluid long enough to flow over the mesh marks but not long enough to flow beyond the area where it was deposited. Paste Preparation: The constituents of a paste require careful selection and processing. Figure 1 shows an idealized paste production sequence for a thick film resistor. The general principle is common to all pastes. 65
MSE 313
Spring 2011
(a) (b) Figure 1: Fabrication of an idealized thick film resistor paste: (a) raw material preparation, (b) formation of paste. The Screen: Screen is the most important part of the screen-­‐printing equipment. It is composed of three sections -­‐ a stencil that defines the pattern to be printed and its thickness, a mesh, which supports the stencil, and a frame, which supports the mesh. Frames Screen frames for thick films are mostly made of metal, usually of aluminum alloy (see Figure 2). The load supported by each side of even a small screen frame when mesh is attached to it is in the order of 50 kg; therefore the strength and stability of metal are necessary. Figure 2: MSE metal screen frames with stencils Mesh: NOTE: See Appendix A for the dimensional characteristics of typical stainless steel meshes. There is three types of mesh material that are commonly available and they are: nylon, polyester, and stainless steel. Characteristics of the three materials are explained in Table 1. Table 1: Characteristics of mesh materials. 66
MSE 313
Material Nylon Polyester Stainless steel Spring 2011
Advantages Monofilament is strong and resilient and can be used thoroughly. It has ability to stretch, which makes it possible to print on irregular surfaces. Very stable, resilient and flexible enough to conform to the normal irregularities found on the surface of thick film circuits. Stronger and more stable than nylon and polyester. It has higher modulus of elasticity. It is less susceptible to wear, so it can last longer. Disadvantages It is not stable because it can absorb water and is affected by changes in humidity. It is also affected by changes in temperature. Low elastic limit. It has lower elastic limit, so it can be easily damaged by accidental denting or coining. There are two main rules that help choose a mesh: 1. The minimum line width which can be printed with a given mesh is 3 times the mesh thread diameter. This means that narrow lines cannot be printed with large thread diameter mesh. 2. The mesh opening should be at least 3 times the particle size of the paste being printed. In normal thick film applications, the paste particles are unlikely to be more than 5 to 20 mm in size, so the finest meshes can be used. However, solder paste has particles typically in the range of 25-­‐70 mm and this clearly restricts the choice of mesh. The open area of the mesh has a great effect on the passage of paste through the screen. For ease of passage, meshes having the greatest percentage area should be selected. Open area is calculated as: Equation 1 Apart from material, the mesh is specified by mesh count (number of threads per inch), thread diameter, and the type of weave. Printing screens are made from square weave mesh and several types are available, some of which are illustrated in Figure 3. Of these, the most widely used is the plain weave type. 67
MSE 313
Spring 2011
(a) (b) (c) Figure 3: Some common weaves for wire mesh: (a) plain, (b) twill, (c) Hollander. Screen tension is the tension necessary to stretch the mesh sufficiently to cause the screen to peel away from the substrate after printing but not to be stretched so much to cause the damage. Figure 5 shows force vs. elongation curve for stainless steel mesh. Figure 4: Mesh elongation vs. tension The elastic limit is at 1% elongation, so if a stainless steel wire is stretched by less than 1% and released, it will return to its original length. If it is stretched by more than 1% and released, it will remain somewhat stretched. If the mesh is elongated by 0.9% there is still 0.1% in reserve before elastic limit is reached. If it is elongated by 0.5%, there is still 0.5% in reserve. The advantage of using the higher of the recommended tensions is that smaller gap must be used between the screen ad the substrate. The high tension causes the screen to peel readily from the substrate even at small gaps. The disadvantage of using high tension is that the screen works at the point of being overstretched and the slightest carelessness by the operator can destroy the screen. Stencils: In classic screen-­‐printing, the stencil’s role is to define the shape of the pattern to be printed on the substrate. In thick film applications the stencil also has a secondary role of having a major influence on the print thickness. Thicker stencil results in a thicker print. Stencils can be made with metal masks, plastic sheets or photo-­‐emulsions. 68
MSE 313
Spring 2011
Emulsion: The emulsions are water-­‐soluble polymers based on polyvinyl alcohol (PVA) and a sensitizer, which, when exposed to ultra-­‐violet (UV) radiation, become insoluble. Hence, if a photo-­‐positive is placed on the photo-­‐emulsion and exposed to UV, the areas that are exposed will become insoluble while areas shadowed by the photo-­‐positive’s opaque areas will remain soluble. Washing with water reveals the desired pattern created in the stencil (Figure 6). Figure 5: A calendared mesh with emulsion imaged with (a) 50 μm lines at 100x, and (b) 40 μm lines & 100 μm vias at 220x. (Images from Sefar Printing Solutions, Ref. 8) Once the screen has been coated with the emulsion and dried, the positive is centered with its emulsion in contact on the screen (Figure 7). frame emulsio
n stencil, mesh (a) (b) Figure 6: (a) Back and (b) front of the screen with the exposed emulsion and revealed stencil pattern. Photo-­‐emulsions come in two different forms, either as a liquid (known as direct emulsion), or as a dry film coated onto a plastic backing sheet (known as indirect emulsion). Direct emulsions are applied in liquid form directly to the screen wires. Indirect emulsions are usually on a sheet of Mylar, or other type of release paper, and are applied by moistening the release paper with either direct emulsion or with water, and then applied to the wires. When dry, the release backing is peeled away. 69
MSE 313
Spring 2011
Squeegees: Squeegee is a rubber blade that performs the following three tasks: pressing the screen into line contact with the substrate; pushing paste down into the stencil and on to the substrate; cutting the paste level with the top of the screen. Angle of Attack: Most printers use squeegee blades with a Shore Durometer hardness in the range 45-­‐
90. In order to push the paste into the stencil and to shear it at the screen’s top surface, an angle of attack at the squeegee tip of around 45° is used. Much steeper angles give insufficient filling of the stencil, while much shallower angles give erratic shearing and hence poor thickness control. As the angle of attack becomes shallower, hydro-­‐dynamic pressure increases and causes an increase in print thickness. The high pressure in the fluid causes it to flow back past the squeegee, thus increasing print thickness (Figure 8). Figure 7: High squeegee pressure causing the paste to flow back under squeegee edge, increasing print thickness. The commonly used squeegee shapes include a knife-­‐edge section, the diamond section and the blade or trailing edge section (Figure 8). (a) (b) (c) Figure 8: Squeegee shapes: (a) knife-­‐edge, (b) diamond, (c) trailing edge. 70
MSE 313
Spring 2011
Part 1a: Print/ Dry/Fire Resistor Circuits Objective: The purpose of this laboratory is to screen print, dry and fire thick film circuits made of conductive ink. This lab is designed to introduce students to the technology of printing thick film and hybrid circuit using a screen printing machine. The fundamentals of the equipment, inks, and thick film materials processing will be explored. Introduction: Thick Film Conductor: Besides providing low resistance connections between circuit elements, thick film conductors can perform functions such as: providing device mounting pads for bonding, forming terminations for printed resistors, acting as pads for lead frame attachment or as printed capacitor electrodes. Thick film conductor compositions consist of three components: finely divided metals powder, small quantity of glass frits, and an organic vehicle which carries the glass material until the firing stage of thick film. The properties of an organic vehicle, when processed correctly, do not affect the properties of the fired track. It is the permanent constituents of the ink, namely, the metal, bonding oxides, and the glass frit, that play the key role in the final properties of a thick film. Microstructure of a fired conductor, shown on Figure 11 (a-­‐b) reveals a continuous array of metal particles, through which runs a secondary network of glass frit. The concentration of glass frit tends to increase near the substrate and sometimes forms a continuous layer at the interface. (a) (b) Figure 9: (a) Microstructure of a fired conductor, (b) a simplified structure of a thick film conductor. This glass reacts with the substrate materials and bonds the sintered metal particles to it, mainly by mechanical keying into the metal. The glass is therefore the main factor controlling the adhesion strength of the sintered composite. The metal, on the other hand, is responsible for electrical and metallurgical properties, whereas combination of both, the metal and the frit, controls compatibility, solderability, and bondability. 71
MSE 313
Spring 2011
Frit Material: Glass softening point: Frit materials used in thick film inks are low softening temperature glasses. Glasses do not have an ordered atomic structure, but retain, even as solid, an interatomic structure similar to that of a liquid. Consequently, the do not exhibit a definite melting point, as would a crystalline material, but soften gradually with increasing temperature until a point is reached where there is detectable flow and the glass have melted. This point is the glass softening point or its working temperature. Frit reactivity: The frit must be sufficiently reactive to form a strong chemical bond with the substrate material; however, it should have limited reactivity towards the metallic components of the ink, as modification of the surface structure of the powder particles could affect their sintering behavior and properties of the fired composite. Wetting: To keep the powder particles in place during the sintering process, it is desirable to maintain a certain amount of wetting of the metal by the molten glass. The degree of wetting and the quantity of frit should not result in complete glass coverage of the metal surfaces, as this would prevent metal-­‐to-­‐metal contact and result in a highly resistive material. Frit % content: Typical conductor composition contains up to 10% by weight of frit material, which slightly reduces the conductivity of the sintered metal compact by reducing the effective cross-­‐section of the fired track. Glass content approaching 50%, results in formation of a matrix of the composite and rapid increase in resistivity. Frit compatibility: The glass should be chemically compatible with other thick film materials such as resistors, and its thermal expansion should match that of the substrate or dielectric in the temperature range from setting point of the glass (usually about 500°C) down to the lowest temperature, at which completed circuit operates. Thermal expansion mismatch would result in a highly stressed system, with poor adhesion and poor resistance to thermal shock and thermal cycling. In worst cases, cracking or peeling of the conductor track would occur. Conductive Material: The metals used in thick film conductors are normally selected on the basis of conductivity and compatibility with the process. They must be able to sinter properly at normal thick film firing temperatures in air and in the presence of molten oxide glasses. They also must not react with the frit so that glass maintains its ability to form a strong bond to the substrate. In practice, for air firing, only noble metals such as sliver (Ag), gold (Au), palladium (Pd), and platinum (Pt) are used. 72
MSE 313
Spring 2011
Deposition of the conductor paste: Printing of thick film conductors is usually less difficult than that of resistors or dielectrics. The main challenge lies in the design of stencils and control of ink rheology to avoid loss of pattern definition and bridging of narrow gaps between conductors. Print quality can be controlled by several factors. For example, speeding up the squeegee traverse speed increases deposit thickness and improves print definition, whereas reduction of squeegee print pressure also increases deposit thickness. Screen printing of conductors below 100μm (~5mils) line width and space can be problematic; however, lower limit of 50μm (~2mils) can be achieved with extremely well controlled procedures. Drying and firing of thick film conductor: Firing reactions can be divided into three groups: (a) removal of the organic printing vehicle; (b) melting of frit and development of the bond with the substrate; and (c) alloying and sintering of the conductive metal particles. The removal of the organic binder takes place early in the firing cycle, and is followed by the frit, oxide and metal reactions which occur simultaneously in the high temperature of the process, normally at 850°C. This is shown in Figure 10. Figure 10: Thick film firing cycle showing the various stages of film formation as a function of temperature. Organic Vehicle Removal: Most of the volatile (solvent) components of the vehicle are normally removed by low temperature drying at temperatures in the range 125-­‐150°C, but a large amount of organic (resin) material remains in the printed deposit. This must be removed at relatively low temperatures to avoid excessive carbonization and retention of carbon until high temperatures. The early part of the firing cycle must, therefore, provide adequate time at low temperatures to ensure complete burn-­‐off of the organics. No organic burn-­‐out product s should remain at temperatures >500°C. Ideally, 3, 4-­‐zone belt furnace should be used in firing of thick film, as shown in Figure 11. 73
MSE 313
Spring 2011
Figure 11: Arrangement of air flow though belt furnace to ensure full burn-­‐out of organic constituents. Bonding Reactions during Firing: Frit reactions: At the optimum firing temperature, the frit should melt and react with the substrate to form a strong chemical bond, and it should form a semi-­‐continuous network within the glass/metal composite. Under-­‐firing may result in loss of adhesion due to incomplete bonding with the substrate and over-­‐firing may cause excessive removal of glass from the metal. To avoid reactions between the frit and organic binders, the furnace should be set to the upper limit of 500°C, above which the air in the furnace is free from burn-­‐
out contamination. Any carbon, retained due to improper firing, can cause reduction of the frit components and production of carbon monoxide or carbon dioxide gas, which may cause pinholes or bubbling in the composite. Metal Reactions during Firing: There are two major metallic reactions that occur during firing: (1) sintering of the individual particles to form continuous metal conduction path through the glass/metal composite, and (2) alloying of mixed metal powders. Both processes have an effect on the metallurgical and electrical properties of the conductor. These temperature-­‐
dependent reactions require the transport of materials in the solid state. The driving force is the lowering of the total energy of the system and in sintering this is mainly achieved by reduction of the surface area of the metal particles. Initially, the particles are only in point contact, but as temperature increases, the particles join together, forming necks at these points. As the necks develop, the composite shrinks and the bulk density of the material increases. As temperature increases, vacancy diffusion predominates, which is caused by the greater disorder of the crystal lattice at higher temperatures, and by the decrease in the number of grain boundaries due to grain growth. Measurements of these self-­‐diffusion rates for pure silver have shown that the change occurs at ~750°C. Reactions during Cooling: The final stage of the firing process involves cooling from the firing temperature back to the ambient one. Cooling rates should not exceed 100°C/min. as higher rates can cause thermal shock damage to the substrate or to the substrate/frit bond. 74
MSE 313
Spring 2011
Experiment: Print Conductor: NOTE: use conductive paste and the screen with the conductor pattern. Set up for Printing: • Turn main air on and check pressure regulator (should read between 60-­‐70 PSI). The printer carriage will slowly lift, as the pressure reaches the desired level. See figure 18 below for details. • Place the squeegee carriage onto bearing block and tighten the thumb screws (the squeegee carriage has two acceptance holes which should align with the “O” rings in the right-­‐hand bearing block; the squeegee blade itself will be mounted to the squeegee blade holder prior to the experiment). • Assemble screen to the screen box by attaching the four screen clamp assemblies to the print head side rails. Position the screen over the two dowel pins in the bottom of the print box and tighten the clamps securely. • Turn the vacuum pump on to approximately 1.2 SCFM vacuum at 17-­‐20”/Hg (lower value is acceptable). Place a substrate in position on the substrate plate; align it with two pins. Vacuum should keep the substrate in place (Figure 12). Figure 12: MSE Screen Printer. Vacuum attached to the substrate stage to keep the substrate in place during screen printing. 75
MSE 313
•
•
Spring 2011
Loosen the locking levers and adjust the snap-­‐off distance (see locations of the appropriate adjustment knobs and levers on figure 20). To obtain exact snap-­‐off, the screen should be brought down to contact the substrate. Then move the bezel on the dial micrometer to read zero. Move screen upwards until dial micrometer reads exact snap-­‐off required. Align substrate with the screen image by using X, Y, and φ adjustment on the stage (Figure 20). NOTE: Never move any of the adjustments with the stage locked as this could result in irreversible damage to the X-­‐Y stage. The Printer is now ready to test print. Test Printing: • Using spatula, deposit ink (approx. 1 tsp. amount) onto the screen (NOT on the pattern itself). • Press the foot pedal. The printer head will come down when the pedal is pressed, and the squeegee will deposit the ink by pressing the screen and moving across the substrate. When the front stop is reached, release the pedal. The printer head will rise simultaneously with the flood blade going back. • Check the quality of the print on the substrate. It the film is too thin or too thick, remove the substrate from stage (vacuum may be turned off for that purpose) and wipe off the paste using wipes and cleaning alcohol. Put the substrate back on the stage and turn the vacuum back on. Remove the remaining paste from the screen and place it into the ink jar. Clean the screen with wipes and alcohol by pressing gently onto the screen from both sides (follow TA’s instructions to clean the screen properly). • NOTE: The first few test prints will have to be rejected since the screen will not be thoroughly wetted and thixotropic inks will not have attained a stable viscosity. • After making test print substrate, minor additional adjustments may be required. Adjustments may include: registration of the print, proper ink deposition, print speed, stage settings (X, Y, φ), snap-­‐off distance, and squeegee level. Repeat Steps 7 through 10 until satisfactory print is received (Figure 15). •
Figure 12: Example of a satisfactory print of silver conductor. 76
MSE 313
Spring 2011
•
•
•
One person from the group should proceed to Step 11, while others move on to Experiment 1.2. Remove squeegee holder from the printer head and recover as much ink as possible (Figure 22 a-­‐c). Recover paste from the screen. Clean the screen and squeegee thoroughly with alcohol. NOTE: Cleaning of the screen is important, as any remaining ink can affect the edge definition and surface of patterns printed the next time. (a) (b) (c) Figure 13: Removal of squeegee holder (a, b), and ink recovery (c). Experiment: Settle & Dry Conductor: • Place the substrate on a flat, dust-­‐free surface (Figure 23a) and allow the ink to level for 5-­‐10 minutes. • Using thermocouple set the temperature of a hot plate to 125°C-­‐150° (Figure 23b). Allow 10 minutes for the hotplate to reach the desired temperature. Place the substrate on a hot plate or in the drying oven. and allow the ink to dry for 10-­‐
15 minutes. (a) (b) Figure 14: (a) Leveling of the wet ink. (b) Setup for drying the paste. 77
MSE 313
Spring 2011
Part 1b: Print/ Dry Resistor: Objective: The purpose of this laboratory is to screen print and dry thick film resistors. This lab is designed to allow students practice screen and stage alignment to obtain patterns at precise locations on the substrates. Introduction: Thick Film Resistor: Three most important properties of thick film resistors are: (a) sheet resistance, Rs (b) temperature coefficient of resistance, TCR, and (c) drift, the percentage change in value with time under specific conditions. Sheet resistance is the resistance of one square of material of known thickness. It equals bulk resistivity/film thickness (ρ/t=Rs) and has units of ohms, but for clarity, it is referred to as ohms/square to avoid confusion with ordinary resistance. Temperature coefficient of resistance is the slope of the line showing resistance as a function of temperature (R vs. T) over a specified range, T1 to T2. If the values are R1 at T1 and R2 at T2, then, measured in parts per million per degree Celsius: TCR = (R2-­‐R1)*106 / R1(T2-­‐T1) ppm/°C Equation 2 Drift is defined as (ΔR/Ro)% under stated conditions after a specified time, where Ro is the starting value. Almost all commercially used thick film resistor systems have contained one of the three precious metal oxides or their derivatives. These are palladium oxide (PdO), ruthenium dioxide (RuO2), and iridium dioxide (IrO2), with the second one used the most in present days. All conducting phases are degenerate semiconductors with resistivity higher than that of most metals, but a similar positive TCR. All three react with or wet suitable glasses to form the necessary reactive cermet. The glass constituents usually include silica (SiO2), boron oxide (BO), manganese dioxide (MnO2), zinc oxide (ZnO) and trace amounts of copper oxide (CuO). Deposition of the resistive paste: During the printing process in the production of thick film resistors, the aim is to deposit on the substrate a uniform layer of resistor material of the correct thickness and dimensions. Uniformity and reproducibility are very important, thus screen and stencil materials need to be chosen carefully. Stainless steels screens are used most commonly because they provide good dimensional stability. For printing resistors, screens of mesh 200, 250 and 325 are used. To print especially thick layers, coarser screens, 165 mesh, are sometimes used. Using finer screens, for example 400 mesh, provides a very good edge and shape definition. The final results of resistor printing depend on the processing are and the environmental conditions, and they should be done in the rooms at temperatures 20-­‐24°C, and humidity of 40 to 60%. 78
MSE 313
Spring 2011
Settling and Drying: Just after printing, the resistor is in the form of a matrix of small dots which flow together in a few seconds to form a uniform print. In order for this flow to complete successfully, the prints should be allowed to settle for 5-­‐10 minutes before drying. This can be done by placing the substrate lying flat in a cool, dust-­‐free place. At this stage, solvents contained in an organic vehicle are no longer needed. They are removed by drying of the print. Drying is usually done in an oven under an infra-­‐red dryer with a substrate temperature between 100 and 200°C for 15 minutes. Experiment: Print Resistor: • NOTE: use resistive paste and the screen with the resistor pattern. • Same as in Part 1. Additionally, make sure that the pattern of the resistor is aligned with the pattern of the conductor. (a) (b) Figure 15: (a) Removing the resistor ink from its container; (b) Reapplying the paste onto a wet screen. • Same as in Part 1. Printing of the resistor layer (or any layer that is deposited as the second one) can be more challenging, because the pattern needs to be correctly aligned with the first layer of conductor tracks. Therefore, it is a good practice to examine the final layout of the circuit beforehand. • If the obtained print is unsuccessful, wipe the wet resistor ink off with alcohol and wipes (Figure 27a, b). This step is allowed because the conductor tracks have been fired, and thus, they stay intact with the substrate. 79
MSE 313
Spring 2011
(a) (b) Figure 16: (a), (b) Cleaning the substrate after an unsuccessfully printed resistor. Figure 17: Example of a satisfactory print of resistor. Experiment: Settle & Dry Resistor: To level and dry resistor paste, follow Step 12-­‐13 in Part 1 Part 3a: Fire Resistor/Firing Profiles: Objective: The purpose of this laboratory is to fire thick film resistors. This lab is designed to obtain and study different firing cycles of the resistors and to determine the effect of firing temperatures on the resistors’ performance. Introduction: Firing of thick film resistor: NOTE: Measure the overall dimensions of the components after they have completely dried, but before firing to determine the % shrinkage. After printing, settling and drying, thick film resistors are fired to completely form a resistor. At the first stage of firing, the remaining vehicle components are evaporated and burn out. As the temperature reaches 500°C, the glass softens and melts, and the sintering process begins. The assistance of the melted glass is very important especially in the case of resistors with the active phase, such as ruthenium dioxide or ruthenates, which have high melting point and are resistant to sintering. The liquid phase of glass allows for sintering of that active phase. This process is responsible for creating the 80
MSE 313
Spring 2011
micro contacts between the grains and it affects final property a typical firing profile of the resistive layer. Figure 21 shows schematically a typical firing profile for resistors with peak temperature 850°C and overall firing time 60 minutes. Commonly, the heating rate is about 50°C per minute, time at peak 10 minutes, and cooling rate about 50°C per minute. Figure 18: Schematic of firing profile for resistors Experiment : Fire Resistor: Place the substrate in the furnace and run the appropriate program to fire thick film. Use Table 3 for guidelines on programming a furnace in order to match a particular firing profile. Ideally, firing of thick films should be done using a conveyor belt furnace, in which a substrate can be fired and cooled at much higher rates than in the box furnace. Using box furnace may cause delays in ramping segments. Table 2: Guidelines to program a furnace to match thick film firing profile. Segment No. 1 Duration (minutes) 1 Temperature range (°C) 0-­‐25 Ramp Rate (°C/min.) 10 2 5 at 25 0 Recommended Programming Step Raise temperature to ambient 25°C Hold for 5 minutes at 25°C 3 13 25-­‐850 63.5 Raise temperature to 850°C 4 10 at 850 0 5 6 13 18 850-­‐125 125-­‐100 55 1.4 -­‐-­‐-­‐ Total: 60 minutes -­‐-­‐-­‐ -­‐-­‐-­‐ Hold for 10 minutes at 850°C Cool down to 125°C Reset to ambient temperature -­‐-­‐-­‐ 81
MSE 313
Spring 2011
Once the cooling stage initiates, apply air connected through the vent on the back of the furnace. Slightly open the door of the furnace, allowing air and other remaining residues to escape . Remove the substrate from the furnace after the full firing cycle is complete. (a) (b) (c) (d) Figure 19: Four different thick film design layouts to be printed and tested. Resistors are located in the mid-­‐section of each pattern.
82
MSE 313
Spring 2011
Part 2: Resistance Measurements of Thick Film Patterns / Thick Film Pattern Design and Analysis:
Objective: The purpose of this laboratory is to study aspects of thick film design and layout. The students will examine the patterns of printed/fired thick film circuits obtained from Part 1. Factors such as geometry, sheet resistance, print thickness and electrical performance will be studied and students will compare calculated theoretical results with the experimental data. Introduction: Design and layout of hybrid circuits include making preliminary decisions before a detailed design is started, which are then followed by the actual layout procedures. Motivation for Use of Hybrids: A system designer who chooses to include hybrids in part of the design considers the aspects of thick film hybrids such as reduction in size and weight, uniformity of assembly, increase in reliability, choice of materials that enhance resistance to thermal, mechanical and chemical stresses, ability to make easily non-­‐standard assemblies, possibility of cost reduction, ease of addition of specialist components and other. General Layout Principles: Design of the thick film layouts may begin with the discussion from the viewpoint of the pattern production, rather than from the electrical design demands. All tracks should be laid out orthogonally and should be as short as it makes them practical. Track widths and separations should be chosen to be compatible with the screens being used. Coarser screens result in printing more paste in a thicker layer, and if the wrong combination is used, the slump may affect the yield during leveling. Using very fine screens, on the other hand, produces accurate patterns which are thinner, but they may become subject to resistive losses. Layout Procedure, Track Layout: The main objective in any layout is to turn the circuit diagram into a viable 2-­‐
dimensional representation of it with as few crossovers as possible. In the manual design, one first draws by hand non-­‐dimensional patterns to get a sense of the viable layout. From there, an accurate layout can be done using graph paper. In the other, more common approach, all manually-­‐made layers of the layout are translated into an engineering CAD program such as Autocad. Some of the layout design principles include: Confirm shape and size of the substrate. Choose type of connection (e.g. solder, wire bonding). 83
MSE 313
Spring 2011
Recalculate and decide on final resistor design (target value in ohms; list possible values for sheet resistance, length, and width). Assign layers for the substrate (the Real appearance in the product and Virtual look are part of the design procedure and are not actually fabricated). The number of resistor layers may range from 0 to 3. Crossovers may require 2 conductors and a dielectric. Additional conductors of a different metal may be needed for soldering and bonding. Master Patterns: The outcome for all real layers is the life-­‐size photopositives of final patterns that are to be screen printed. In manual approach, a two-­‐layer material is used. Its soft red layer is accurately scribed and peeled to make a master pattern at 10x magnification. This is then reduced photographically to 1x on either a high resolution film or a glass positive. In the CAD approach, the print is made on a film using photo-­‐plotter either directly at life size or by the method described above, depending on the required accuracy. Thick Film Components: Conductors: General design rules for thick film conductors are not summarized here . Dimensions of conductor tracks depend on the material used. For low-­‐cost applications 0.5 mm line and space is typical. Resistors: Resistors tend to be subject to dimensional problems. Their thickness tends to be greater near contacts and the effective value of sheet resistance may depend on the track length. For conventional resistors, the value of resistance is determined by three factors – the effective sheet resistance (usually determined by the ink manufacturer), length, and width. Sheet resistance is the resistance of one square of a material, regardless of the actual side dimensions. It is derived from the volume resistivity and is equivalent to ρ/t (see Part 1, Exp. Procedure 2), where t is the thickness of the conducting material. Thus, resistance R is defined as: R = (ρ/t) * (L/W) = Rs * (L/W) Equation 3 where ρ is material’s bulk resistivity, L is length, W is width, L/W is the aspect ratio, expressed in units of squares ( ’s) and Rs is the sheet resistance, expressed in units of ohm/square (Ω/ ). In Figure 30, the aspect ratio is approximately 1.5:1 for the over square and 1:3 for under square resistors, respectively. If the fired sheet resistance is 1000 Ω/ , then the R values become 1500 Ω and 333 Ω, respectively. 84
MSE 313
Spring 2011
Figure 20: Over square and under square thick film resistors. Resistors have a maximum film temperature for operation and this is related to the power density. The traditional value is ‘40W/square inch of track’, which is equivalent to 64mW/mm-­‐2. To avoid taking too much space, the range of aspect ratio in a printed resistor before trim is limited to approximately 0.2-­‐5.0. This range results in about 5 mm maximum length or width of the resistor. To make a feasible print, the number of different sheet resistances used on one substrate should not exceed three. Design rules for thick film resistors include the following factors: L = Length W = Width A = Printed area = L x W P = Designer’s power dissipation (mW) R = Target resistor value (Ω) RD = Design value of resistor (e.g., 0.75 x R) N = Aspect ratio printed (L/W) RS = Chosen sheet resistance The following rules apply to the resistor design: RD = 0.75 x Target RD = RS x (L/W) A = L x W P/A = 32 mW mm-­‐2 L ≥ 1 mm W ≥ 1 mm ND = range approx. 0.2-­‐5.0 RS = Range from 10, 100, 1000, 10000, 100000, 1000000 Ω/ . The design shape of the resistor is calculated from Equations (4) and (5), below. W = √{(P * RS) / (32 * RD)} Equation 4 L = √{(P * RD) / (32 * RS)} Equation 5 No resistor should be located closer to the substrate edge than 1 mm. There should be at least 1.5 mm between individual resistor stripes, and no conductor track should approach the stripe closer than 1 mm. In practice, a series of resistor test patterns is prepared and screen printed (Figure 25) in order to determine the resistor optimum dimensions for the particular resistive inks. 85
MSE 313
Spring 2011
Figure 21: Selection of typical resistor test patterns. Resistors of same width, varying length and resistors of same length, varying width. Standard resistors for loading tests. Resistors of same aspect ratio, varying area, enabling effects of resistor position on substrate to be investigated. As (b) but with conductor stripe in the center overlaid with resistor material to explore interactions. Resistors of same length but widely varying widths for detailed examination of width effects. Small resistors. 86
MSE 313
Spring 2011
Part 3: Characterization. In this lab students will test the various resistors made previously using LabVIEW and
standard data acquisition apparatus as well as determine thickness with a profilometer. These
measurements will be demonstrated by the lab instructor.
Analysis part 1: • Take the fired circuits to the profilometer to gain thickness measurements. • From this determine the cross-­‐sectional area of each resistor. Do not include areas where resistor is overlapping conductive material. Analysis Part 2: Use Fluke Meter • Carefully snap the substrate along the inscribed lines and obtain four separate substrates with various screen printed thick film circuits (Figure 26 a-­‐d). • Examine all four substrates using the following guidelines: • Substrate (a): -­‐ Determine aspect ratio of each of the five circuits and calculate their theoretical resistance values (to obtain sheet resistance, Rs use the vendor’s Data Sheet for the resistor paste used). -­‐ Using ohmmeter, measure the actual resistance values of each of the five circuits. • Note: In order to correctly measure resistor dimensions, do NOT include the area where resistor is printed over another ink. • Substrate (b): • Repeat the calculations and measurements, as for Substrate (a). • Substrate (c) – Parallel Resistors: -­‐ Using approach above, calculate total resistance of printed thick film parallel resistor. -­‐ Using ohmmeter, measure the actual resistance of the circuit. • Substrate (d) – Resistors in Series: -­‐ Using approach above, calculate total resistance of printed thick film resistor in series. -­‐ Using ohmmeter, measure the actual resistance of the circuit. Analysis Part 3: Temperature dependence: For this section, follow the basic principles outlined in Analysis Part 2 above, but choose at least five temperatures at which to conduct the measurements. From this data calculate the TCR of the resistor. Next prepare an Arrhenius plot to find the activation energy. Analysis Part 4: Conductor/Resistor interface resistance: 1. Draw a diagram of your circuit . 2. Create a circuit diagram of the resistances 87
MSE 313
Spring 2011
3. Consider what assumptions are necessary to determine the resistance of the interface vs. the resistance of the resistor only. Consider what is measured and what must be computed. 4. Find the resistance of each part. (remember ohms law!) Questions: 1. Discuss any problems encountered during print, for example, cleaning the screens, aligning the screen with the substrate, etc. 2. How did the thickness and room temperature resistivity measurements between the fired samples compare with the measurements at other temperatures? 3. By how much did the patterns shrink after firing? 4. Note any defects of unusual morphology of the fired sensors. Was there any cracking? Discoloration? , Etc Instead of a lab report students will put together a poster that lays out the principles of the lab and the results. Include information obtained from independent research regarding screen printing in the electronics industry and advances that have been made. Include a summary of all results. References: •
•
•
•
•
•
•
•
•
•
•
•
Marciniak, M., “Screen Printing of Thick Film Materials” Senior Project Report, March 2008. Kasap, S.O., “Principles of Electronic Materials and Devices” 3rd edition. McGraw Hill, 2006. ESL ElectroScience, Thick Film Materials & Ceramic Tapes, <http://www.electroscience.com/index.html>. Date accessed: July 20th, 2007. Holmes, P. J., Loasby, R. G., Handbook of Thick Film Technology, 1st ed. Electrochemical PublicationsLimited, 1976. Jones, D., Roydn, Hybrid Circuit Design and Manufacture, International Society for Hybrid Microelectronics, 1982. Koartan, Microelectronic Interconnect Materials, < http://www.koartan.com/>. Date accessed: July 20th, 2007. Missele, Carl. Screen Printing Primer, Microcircuit Products, Motorola Inc., Reprinted from Hybrid Circuit Technology, 1987. Pitt, Keith, Handbook of Thick Film Technology, 2nd ed. Electrochemical Publications Limited, 2005. Screen Printer Operation Manual for de Haart SP-­‐SA-­‐5/6 Screen Printers, 1978. Seafar Printing Solutions, Inc., <http://www.sefar-­‐screens.com/>. Date accessed: September, 2007. 88
MSE 313
Spring 2011
Blank Page 89
MSE 313
Spring 2011
Junior Lab Project: The Jr. Lab Project will be covered in the first session of MSE 313. See Appendix for relevant forms on following pages. Un-­‐numbered Appendix follows. 90
MSE 313 Team Project Assignments Section Morning Afternoon TA Aaron Lichtner Evan Uchaker Email [email protected] [email protected] Requirements – sent to respective TAs unless otherwise stated. 1. Project Milestones 2. Weekly Progress Reports Milestones: Week 1 Redefine Project and Budget Week 2 Weeks 3-­‐4 Project Plan Week 5 Intro and Background Due Weeks 6-­‐8 Week 9 Final Papers Due Week 10 Project Presentations Week 1 (Due Tuesday 3/29): • Submit altered project summary (max 1 page) • Requisition of Supplies (form) o Use proper requisition form o Send to Tuesday • Materials needed (max 1 page) o To be sent to Tuesday and your respective TAs. Make sure to use ACCURATE values for all requested purchases. Week 2 (Due Tuesday 4/5): •
Detailed Project Plan o Include breakdown of work o Project timeline, milestones Week 5 (Due Tuesday 5/3): •
First Draft of Project Introduction and Background to be used for final paper o Include cited sources Week 9 (Due Friday 5/27): •
Submit Final Paper Week 10: •
Project Presentations Weekly Progress Reports Due Weekly: Max 1-­‐2 pages Each team must complete a weekly progress report to be turned into their respective TA. Progress report authors will rotate weekly so that every team member does at least one report. Progress reports must include: 1. Work accomplished that week broken down into individual team member tasks 2. Work planned for the following week (general team) 3. Potential roadblocks that will hold up your research or progress. Examples are lead time on equipment, need for training, scheduling conflicts… 4. Summarized piece of literature found that week including cited source OR data taken that week properly formatted, captioned, and explained o Once data is being generated you must present data Suggested Progress Report Layout Date Team Member Project Work Completed: Work Planned: Potential Roadblocks: Literature or Data Review: MSE 313 2011
SUPPLY & SERVICES REQUISITION
(Please fill in blanks)
Requested by:
Authorized by:
Project Title:
Date:
X
Group & No.:
X
Group
Group
Group
Group
Group
Quantity
Requisition #:
1
2
3
4
5
Page #:
Item (please be specific)
Vendor* & Catalog
Number
Price Each
Print and fill out this form to request supplies or services. For estimates of cost of services please use the MSE User facility.
For supplies please investigate costs online, you can get this information from web pages like McMaster Carr, VWR, Lab Safety Supply,
and Alfa Aeser. Some suppliers require an account to get pricing information so assign one person in your group to do so.
Any supplies requested that the department has on hand can be provided, but cost assesment is still required.
Designation: E399 – 09´2
Standard Test Method for
Linear-Elastic Plane-Strain Fracture Toughness KIc of
Metallic Materials1
This standard is issued under the fixed designation E399; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
This standard has been approved for use by agencies of the Department of Defense.
´1 NOTE—Eq A3.4, Eq A4.4, Eq A5.4, and Eq A6.11 were editorially corrected in May 2010.
´2 NOTE—11.2 and 11.4 were editorially corrected in December 2010.
1. Scope
1.1 This test method covers the determination of fracture
toughness (KIc) of metallic materials under predominantly
linear-elastic, plane-strain conditions using fatigue precracked
specimens having a thickness of 1.6 mm (0.063 in.) or greater2
subjected to slowly, or in special (elective) cases rapidly,
increasing crack-displacement force. Details of test apparatus,
specimen configuration, and experimental procedure are given
in the Annexes.
NOTE 1—Plane-strain fracture toughness tests of thinner materials that
are sufficiently brittle (see 7.1) can be made using other types of
specimens (1).3 There is no standard test method for such thin materials.
1.2 This test method is divided into two parts. The first part
gives general recommendations and requirements for KIc
testing. The second part consists of Annexes that give specific
information on displacement gage and loading fixture design,
special requirements for individual specimen configurations,
and detailed procedures for fatigue precracking. Additional
annexes are provided that give specific procedures for beryllium and rapid-force testing.
1.3 General information and requirements common to all
specimen configurations:
Referenced Documents
Terminology
Stress-Intensity Factor
Plane-Strain Fracture Toughness
Crack Plane Orientation
Section
2
3
3.1.1
3.1.2
3.1.3
1
This test method is under the jurisdiction of ASTM Committee E08 on Fatigue
and Fracture and is the direct responsibility of Subcommittee E08.07 on Fracture
Mechanics.
Current edition approved July 1, 2009. Published August 2009. Originally
approved in 1970. Last previous edition approved in 2008 as E399 – 08. DOI:
10.1520/E0399-09E02.
2
For additional information relating to the fracture toughness testing of alumi–
inum alloys, see Practice B645.
3
The boldface numbers in parentheses refer to the list of references at the end of
this standard.
Summary of Test Method
Significance and Use
Significance
Precautions
Practical Applications
Apparatus (see also 1.4)
Tension Machine
Fatigue Machine
Loading Fixtures
Displacement Gage, Measurement
Specimen Size, Configurations, and Preparation (see
also 1.5)
Specimen Size Estimates
Standard and Alternative Specimen Configurations
Fatigue Crack Starter Notches
Fatigue Precracking (see also 1.6)
Crack Extension Beyond Starter Notch
General Procedure
Specimen Measurements
Thickness
Width
Crack Size
Crack Plane Angle
Specimen Testing
Loading Rate
Test Record
Calculation and Interpretation of Results
Test Record Analysis
Pmax/PQ Validity Requirement
Specimen Size Validity Requirements
Reporting
Precision and Bias
Section
4
5
5.1
5.1.1-5.1.5
5.2
6
6.1
6.2
6.3
6.4
7
7.1
7.2
7.3.1
7.3.2
7.3.2.2
8
8.2.1
8.2.2
8.2.3
8.2.4
8.3
8.4
9
9.1
9.1.3
9.1.4
10
11
1.4 Specific requirements related to test apparatus:
Double-Cantilever Displacement Gage
Testing Fixtures
Bend Specimen Loading Fixture
Compact Specimen Loading Clevis
Annex A1
Annex A2
Annex A2.1
Annex A2.2
1.5 Specific requirements related to individual specimen
configurations:
Bend Specimen SE(B)
Compact Specimen C(T)
Disk-Shaped Compact Specimen DC(T)
Arc-Shaped Tension Specimen A(T)
Arc-Shaped Bend Specimen A(B)
Copyright. © ASTM International. 100 Barr Harbour Drive PO box C700, West Conshohocken, Pennsylvania 19428-2959, United States
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
1
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
Annex A3
Annex A4
Annex A5
Annex A6
Annex A7
E399 – 09´2
1.6 Specific requirements related to special test procedures:
Fatigue Precracking KIc Specimens
Hot-Pressed Beryllium Testing
Rapid-Force Testing
Annex A8
Annex A9
Annex A10
1.7 The values stated in SI units are to be regarded as the
standard. The values given in parentheses are for information
only.
1.8 This standard does not purport to address all of the
safety concerns, if any, associated with its use. It is the
responsibility of the user of this standard to establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use.
2. Referenced Documents
2.1 ASTM Standards:4
B909 Guide for Plane Strain Fracture Toughness Testing of
Non-Stress Relieved Aluminum Products
B645 Practice for Linear-Elastic Plane−Strain Fracture
Toughness Testing of Aluminum Alloys
E4 Practices for Force Verification of Testing Machines
E8/E8M Test Methods for Tension Testing of Metallic
Materials
E177 Practice for Use of the Terms Precision and Bias in
ASTM Test Methods
E337 Test Method for Measuring Humidity with a Psychrometer (the Measurement of Wet- and Dry-Bulb Temperatures)
E456 Terminology Relating to Quality and Statistics
E691 Practice for Conducting an Interlaboratory Study to
Determine the Precision of a Test Method
E1820 Test Method for Measurement of Fracture Toughness
E1823 Terminology Relating to Fatigue and Fracture Testing
E1921 Test Method for Determination of Reference Temperature, To, for Ferritic Steels in the Transition Range
3. Terminology
3.1 Definitions:Terminology E1823 is applicable to this test
method:
3.1.1 stress-intensity factor, K, KI, KII, KIII [FL−3/2]—
magnitude of the ideal-crack-tip stress field (a stress-field
singularity), for a particular mode of crack displacement, in a
homogeneous, linear-elastic body.
3.1.1.1 K is a function of applied force and test specimen
size, geometry, and crack size, and has the dimensions of force
times length-3/2.
3.1.1.2 Values of K for modes I, II, and III are given as:
KI 5 lim @syy~2pr!1/2#
r→0
(1)
KII 5 lim @txy~2pr!1/2#
r→0
(2)
4
For referenced ASTM standards, visit the ASTM website, www.astm.org, or
contact ASTM Customer Service at [email protected]. For Annual Book of ASTM
Standards volume information, refer to the standard’s Document Summary page on
the ASTM website.
KIII 5 lim @tyz~2pr!1/2#
r→0
(3)
where r is the distance directly forward from the crack tip to
the location where the significant stress is calculated.
3.1.2 plane-strain fracture toughness, KIc [FL-3/2] —the
crack-extension resistance under conditions of crack-tip plane
strain in Mode I for slow rates of loading under predominantly
linear-elastic conditions and negligible plastic-zone adjustment. The stress intensity factor, KIc, is measured using the
operational procedure (and satisfying all of the validity requirements) specified in Test Method E399, that provides for the
measurement of crack-extension resistance at the onset (2% or
less) of crack extension and provides operational definitions of
crack-tip sharpness, onset of crack extension, and crack-tip
plane strain.
3.1.2.1 See also definitions of crack-extension resistance,
crack-tip plane strain, and mode in Terminology E1823.
3.1.3 crack plane orientation—identification of the plane
and direction of crack extension in relation to the characteristic
directions of the product. A hyphenated code defined in
Terminology E1823 is used wherein the letter(s) preceding the
hyphen represents the direction normal to the crack plane and
the letter(s) following the hyphen represents the anticipated
direction of crack extension (see Fig. 1).
3.1.3.1 Wrought Products—the fracture toughness of
wrought material depends on, among other factors, the orientation and propagation direction of the crack in relation to the
material’s anisotropy, which depends, in turn, on the principal
directions of mechanical working and grain flow. Orientation
of the crack plane shall be identified wherever possible. In
addition, product form shall be identified (for example,
straight-rolled plate, cross-rolled plate, pancake forging, and so
forth) along with material condition (for example, annealed,
solution treated plus aged, and so forth). The user shall be
referred to product specifications for detailed processing information.
3.1.3.2 For rectangular sections, the reference directions are
identified as in Fig. 1(a) and Fig. 1(b), which give examples for
rolled plate. The same system is used for sheet, extrusions, and
forgings with nonsymmetrical grain flow.
L = direction of principal deformation (maximum grain flow)
T = direction of least deformation
S = third orthogonal direction
3.1.3.3 Using the two-letter code, the first letter designates
the direction normal to the crack plane, and the second letter
the expected direction of crack propagation. For example, in
Fig. 1(a), the T-L specimen fracture plane normal is in the
width direction of a plate and the expected direction of crack
propagation is coincident with the direction of maximum grain
flow (or longitudinal) direction of the plate.
3.1.3.4 For specimens tilted in respect to two of the reference axes as in Fig. 1(b), crack plane orientation is identified
by a three-letter code. The designation L-TS, for example,
indicates the crack plane to be perpendicular to the principal
deformation (L) direction, and the expected fracture direction
to be intermediate between T and S. The designation TS-L
means that the crack plane is perpendicular to a direction
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
2
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
(a) Rectangular Sections—Specimens Aligned with Reference Directions
(b) Rectangular Sections—Specimens Not Aligned with Reference Directions
(c) Cylindrical Bars and Tubes
L = direction of maximum grain flow
R = radial direction
C = circumferential or tangential direction
FIG. 1 Crack Plane Identification
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
3
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
intermediate between T and S, and the expected fracture
direction is in the L direction.
3.1.3.5 For cylindrical sections, where grain flow can be in
the longitudinal, radial or circumferential direction, specimen
location and crack plane orientation shall reference original
cylindrical section geometry such that the L direction is always
the axial direction for the L-R-C system, as indicated in Fig.
1(c), regardless of the maximum grain flow. Note that this is a
geometry based system. As such, the direction of maximum
grain flow shall be reported when the direction is known.
NOTE 2—The same system is useful for extruded or forged parts having
circular cross section. In most cases the L direction corresponds to the
direction of maximum grain flow, but some products such as pancake,
disk, or ring forgings can have the R or C directions correspond to the
direction of maximum grain flow, depending on the manufacturing
method.
L = axial direction
R = radial direction
C = circumferential or tangential direction
3.1.3.6 In the case of complex structural shapes, where the
grain flow is not uniform, specimen location and crack plane
orientation shall reference host product form geometry and be
noted on component drawings.
3.1.3.7 Non-Wrought Products—for non-wrought products,
specimen location and crack plane orientation shall be defined
on the part drawing. The result of a fracture toughness test from
a non-wrought product shall not carry an orientation designation.
3.1.3.8 Discussion—when products are to be compared on
the basis of fracture toughness, it is essential that specimen
location and orientation with respect to product characteristic
directions be comparable and that the results not be generalized
beyond these limits.
3.2 Definitions of Terms Specific to This Standard:
3.2.1 crack mouth opening displacement (CMOD), Vm
[L]—component of clip gage displacement taken at the crack
starter-notch mouth.
·
3.2.2 stress-intensity factor rate, K (FL-3/2t-1)—change in
stress-intensity factor, K, per unit time.
4. Summary of Test Method
4.1 This test method covers the determination of the planestrain fracture toughness (KIc) of metallic materials by
increasing-force tests of fatigue precracked specimens. Force is
applied either in tension or three-point bending. Details of the
test specimens and experimental procedures are given in the
Annexes. Force versus crack-mouth opening displacement
(CMOD) is recorded either autographically or digitally. The
force at a 5 % secant offset from the initial slope (corresponding to about 2.0 % apparent crack extension) is established by
a specified deviation from the linear portion of the record (1).
The value of KIc is calculated from this force using equations
that have been established by elastic stress analysis of the
specimen configurations specified in this test method. The
validity of the KIc value determined by this test method
depends upon the establishment of a sharp-crack condition at
the tip of the fatigue crack in a specimen having a size adequate
to ensure predominantly linear-elastic, plane-strain conditions.
To establish the suitable crack-tip condition, the stress-
intensity factor level at which specimen fatigue precracking is
conducted is limited to a relatively low value.
4.2 The specimen size required for test validity increases as
the square of the material’s toughness-to-yield strength ratio.
Therefore a range of proportional specimens is provided.
5. Significance and Use
5.1 The property KIc determined by this test method characterizes the resistance of a material to fracture in a neutral
environment in the presence of a sharp crack under essentially
linear-elastic stress and severe tensile constraint, such that (1)
the state of stress near the crack front approaches tritensile
plane strain, and (2) the crack-tip plastic zone is small
compared to the crack size, specimen thickness, and ligament
ahead of the crack.
5.1.1 Variation in the value of KIc can be expected within
the allowable range of specimen proportions, a/W and W/B. KIc
may also be expected to rise with increasing ligament size.
Notwithstanding these variations, however, KIc is believed to
represent a lower limiting value of fracture toughness (for 2 %
apparent crack extension) in the environment and at the speed
and temperature of the test.
5.1.2 Lower values of KIc can be obtained for materials that
fail by cleavage fracture; for example, ferritic steels in the
ductile-to-brittle transition region or below, where the crack
front length affects the measurement in a stochastic manner
independent of crack front constraint. The present test method
does not apply to such materials and the user is referred to Test
Method E1921 and E1820. Likewise this test method does not
apply to high toughness or high tearing-resistance materials
whose failure is accompanied by appreciable amounts of
plasticity. Guidance on testing elastic-plastic materials is given
in Test Method E1820.
5.1.3 The value of KIc obtained by this test method may be
used to estimate the relation between failure stress and crack
size for a material in service wherein the conditions of high
constraint described above would be expected. Background
information concerning the basis for development of this test
method in terms of linear elastic fracture mechanics may be
found in Refs (1) and (3).
5.1.4 Cyclic forces can cause crack extension at KI values
less than KIc. Crack extension under cyclic or sustained forces
(as by stress corrosion cracking or creep crack growth) can be
influenced by temperature and environment. Therefore, when
KIc is applied to the design of service components, differences
between laboratory test and field conditions shall be considered.
5.1.5 Plane-strain fracture toughness testing is unusual in
that there can be no advance assurance that a valid KIc will be
determined in a particular test. Therefore, compliance with the
specified validity criteria of this test method is essential.
5.1.6 Residual stresses can adversely affect the indicated
KQ and KIc values. The effect can be especially significant for
specimens removed from as-heat treated or otherwise nonstress relieved stock, from weldments, from complex wrought
parts, or from parts with intentionally induced residual stresses.
Indications of residual stress include distortion during specimen machining, results that are specimen configuration dependent, and irregular fatigue precrack growth (either excessive
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
4
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
crack front curvature or out-of-plane growth). Guide B909
provides supplementary guidelines for plane strain fracture
toughness testing of aluminum alloy products for which
complete stress relief is not practicable. Guide B909 includes
additional guidelines for recognizing when residual stresses
may be significantly biasing test results, methods for minimizing the effects of residual stress during testing, and guidelines
for correction and interpretation of data.
5.2 This test method can serve the following purposes:
5.2.1 In research and development, to establish in quantitative terms significant to service performance, the effects of
metallurgical variables such as composition or heat treatment,
or of fabricating operations such as welding or forming, on the
fracture toughness of new or existing materials.
5.2.2 In service evaluation, to establish the suitability of a
material for a specific application for which the stress conditions are prescribed and for which maximum flaw sizes can be
established with confidence.
5.2.3 For specifications of acceptance and manufacturing
quality control, but only when there is a sound basis for
specifying minimum KIc values, and then only if the dimensions of the product are sufficient to provide specimens of the
size required for valid KIc determination. The specification of
KIc values in relation to a particular application should signify
that a fracture control study has been conducted for the
component in relation to the expected loading and environment, and in relation to the sensitivity and reliability of the
crack detection procedures that are to be applied prior to
service and subsequently during the anticipated life.
6. Apparatus
6.1 Testing Machine and Force Measurement—The calibration of the testing machine shall be verified in accordance with
Practices E4. The test machine shall have provisions for
autographic recording of the force applied to the specimen; or,
alternatively, a computer data acquisition system that may be
used to record force and CMOD for subsequent analysis.
6.2 Fatigue Precracking Machine—When possible, the
calibration of the fatigue machine and force-indicating device
shall be verified statically in accordance with Practices E4. If
the machine cannot be calibrated and verified statically, the
applied force shall otherwise be known to 62.5 %. Careful
alignment of the specimen and fixturing is necessary to
encourage straight fatigue cracks. The fixturing shall be such
that the stress distribution is uniform across the specimen
thickness and symmetrical about the plane of the prospective
crack.
6.3 Loading Fixtures—Fixtures suitable for loading the
specified specimen configurations are shown in the Annexes.
The fixtures are designed to minimize friction contributions to
the measured force.
6.4 Displacement Gage—The displacement gage electrical
output represents relative displacement (V) of two precisely
located gage positions spanning the crack starter notch mouth.
Exact and positive positioning of the gage on the specimen is
essential, yet the gage must be released without damage when
the specimen breaks. Displacement gage and knife-edge designs shall provide for free rotation of the points of contact
between the gage and the specimen. A recommended design for
a self-supporting, releasable displacement gage is shown in
FIG. 2 Double–Cantilever Clip-In Displacement Gage Showing Mounting by Means of Integral Knife Edges
(Gage Design Details are Given in Annex A1)
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
5
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
Fig. 2 and described in Annex A1. The gage’s strain gage
bridge arrangement is also shown in Fig. 2.
6.4.1 The specimen shall be provided with a pair of accurately machined knife edges to support the gage arms and serve
as displacement reference points. The knife edges may be
machined integral with the specimen as shown in Figs. 2 and 3,
or they may be separate pieces affixed to the specimen. A
suggested design for attachable knife edges is shown in Fig. 4.
This design features a knife edge spacing of 5 mm (0.2 in.).
The effective gage length is established by the points of contact
between the screw and the hole threads. For the design shown,
the major diameter of the screw is used in setting this gage
length. A No. 2 screw will permit the use of attachable knife
edges for specimens having W > 25 mm (1.0 in.).
6.4.2 Each gage shall be verified for linearity using an
extensometer calibrator or other suitable device. The resolution
of the calibrator at each displacement interval shall be within
0.00051 mm (0.000020 in.). Readings shall be taken at ten
equally spaced intervals over the working range of the gage
(see Annex A1). The verification procedure shall be performed
three times, removing and reinstalling the gage in the calibration fixture after each run. The required linearity shall correspond to a maximum deviation of 0.003 mm (0.0001 in.) of the
individual displacement readings from a least-squares-best-fit
straight line through the data. The absolute accuracy, as such,
is not important in this application, since the test method is
concerned with relative changes in displacement rather than
absolute values (see 9.1). Verification of gage calibration shall
be performed at the temperature of test 65.6°C (10°F). The
gage shall be verified during the time the gage is in use at time
intervals defined by established quality assurance practices.
Commercial gages are typically verified annually.
6.4.3 It is not the intent of this test method to exclude the
use of other types of gages or gage-fixing devices provided the
NOTE 1—Dimensions in mm.
NOTE 2— Gage length shown corresponds to clip gage spacer block
dimensions shown in Annex A1, but other gage lengths may be used
provided they are appropriate to the specimen (see 6.4.3).
NOTE 3—For starter notch configurations see Fig. 5.
NOTE 1—Dimensions are in mm.
NOTE 2—Effective gage length = 2C + Screw Thread Diameter # W/2.
(This will always be greater than the gage length specified in A1.1.)
NOTE 3—Dimension shown corresponds to clip gage spacer block
dimension in Annex A1.
Inch-Pound Units Equvalients
mm
in.
0.81
0.032
1.3
0.050
1.5
0.060
5.08
0.200
FIG. 3 Example of Integral Knife Edge Design
6.35
0.250
1.8
0.070
2.54
0.100
3.18
0.125
FIG. 4 Example of Attachable Knife Edge Design
gage used meets the requirements listed above and provided
the gage length does not exceed those limits given in the Annex
appropriate to the specimen being tested.
7. Specimen Size, Configurations, and Preparation
7.1 Specimen Size:
7.1.1 In order for a result to be considered valid according
to this test method (see also 3.1.2.1), the specimen ligament
size (W – a) must be not less than 2.5(KIc/sYS)2, where sYS is
the 0.2 % offset yield strength of the material in the environment and orientation, and at the temperature and loading rate of
the test (1, 4, 5). For testing at rates other than quasi-static see
Annex A10, Rapid Force Testing. The specimen must also be
of sufficient thickness, B, to satisfy the specimen proportions in
7.2.1 or 7.2.1.1 and meet the Pmax/PQ requirement in 9.1.3.
Meeting the ligament size and Pmax/PQ requirements cannot be
assured in advance. Thus, specimen dimensions shall be
conservatively selected for the first test in a series. If the form
of the material available is such that it is not possible to obtain
a test specimen with ligament size equal to or greater than
2.5(KIc/sYS)2, then it is not possible to make a valid KIc
measurement according to this test method.
7.1.2 The initial selection of specimen size for a valid KIc
measurement is often based on an estimated value of KIc for the
material.
7.1.3 Alternatively, the ratio of yield strength to elastic
modulus may be used for selecting a specimen size that will be
adequate for all but the toughest materials:
sYS/E
Inch-Pound Units Equivalents
mm
in.
1.5
0.060
0.0050 to 0.0057
0.0057 to 0.0062
0.0062 to 0.0065
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
6
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
Minimum Recommended
Ligament Size
mm
in.
76
64
51
3
21⁄2
2
E399 – 09´2
0.0065 to 0.0068
0.0068 to 0.0071
0.0071 to 0.0075
0.0075 to 0.0080
0.0080 to 0.0085
0.0085 to 0.0100
0.0100 or
greater
44
38
32
25
19
13
6.4
13⁄4
11⁄2
11⁄4
1
3⁄ 4
1⁄ 2
1⁄ 4
When it has been established that 2.5(KIc/sYS)2 is substantially less than the minimum recommended ligament size given
in the preceding table, then a correspondingly smaller specimen can be used.
7.2 Specimen Configurations—Recommended specimen
configurations are shown in Figs. A3.1-A6.1 and Fig. A7.1.
7.2.1 Specimen Proportions—Crack size, a, is nominally
between 0.45 and 0.55 times the width, W. Bend specimens can
have a width to thickness, W/B, ratio of 1 # W/B # 4. Tension
specimen configurations can be 2 # W/B # 4.
7.2.1.1 Recommended Proportions—It is recommended that
the thickness, B, is nominally one-half the specimen width, W
(that is, W/B = 2). Likewise, the crack size, a, should be
nominally equal to one-half the width, W (that is a/W = 1/2).
NOTE 3—Alternative W/B ratios different from the recommended ratio
in 7.2.1.1 but still meeting the requirements in 7.2.1 are sometimes useful,
especially for quality control or lot releases purposes, because they allow
a continuous range of product thicknesses to be tested using a discrete
number of specimen widths while still maintaining specimens of full
product thickness. However, because specimen width influences the
amount of crack extension corresponding to the 95 % slope, KIc obtained
with alternative W/B ratios may not agree with those obtained using the
recommended W/B ratio, particularly in products exhibiting a Type I
force-CMOD record (6). As an example, a specimen with the recommended proportion W/B = 2 would tend to yield a lower KIc than a
specimen with an alternative proportion W/B = 4. Also, because a shorter
ligament length may hinder resistance curve development, an alternative
specimen with W/B < 2 (allowed only for bend specimens) may pass the
Pmax/PQ requirement, while a specimen with the recommended W/B ratio
would fail. Conversely, an alternative specimen with W/B >2 (allowed in
both tension and bend specimens) may fail the Pmax/PQ requirement,
while a specimen with the recommended W/B would pass.
7.2.2 Alternative Specimens—In certain cases it may be
necessary or desirable to use specimens having W/B ratios
other than that specified in 7.2.1. Alternative W/B ratios and
side-grooved specimens are allowed as specified in 7.2.1.1 and
7.2.2.1. These alternative specimens shall have the same crack
length-to-specimen width ratio as the standard specimen.
(a) Starter Notches and Fatigue Cracks
Note 1—For a chevron crack starter notch the fatigue crack shall
emerge on both surfaces of the specimen.
Note 2—Fatigue crack extension on each surface of the specimen containing a straight-through notch shall be at least 0.025 W or 1.3 mm
(0.050 in.), whichever is larger.
Note 3—Fatigue crack extension on each surface of the specimen from
the stress raiser tipping the hole shall be at least 0.5 D or 1.3 mm (0.050
in.), whichever is larger.
Note 4—Crack starter notch shall be perpendicular to the specimen surfaces and to the intended direction of crack propagation within 62°.
Note 5—Notch width N need not be less than 1.6 mm (1⁄16 in.).
Note 6—From notched edge or centerline of loading holes, as appropriate.
(b) Detail of Chevron Notch
Note 1—A = C within 0.010 W.
Note 2—Cutter tip angle 90° max.
Note 3—Radius at chevron notch bottom 0.25 mm
(0.010 in.) max.
FIG. 5 Crack Starter Notch and Fatigue Crack Configurations
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
7
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
7.2.2.1 Alternative Side-Grooved Specimens—For the compact C(T) and the bend SE(B) specimen configurations sidegrooving is allowed as an alternative to plain-sided specimens.
The total thickness reduction shall not exceed 0.25 B. A total
reduction of 0.20 B has been found to work well (7) for many
materials and is recommended (10% per side). Any included
angle less than 90° is allowed. The root radius shall be 0.5 6
0.2 mm (0.02 6 0.01 in.). Precracking prior to the sidegrooving operation is recommended to produce nearly straight
fatigue precrack fronts. BN is the minimum thickness measured
at the roots of the side grooves. The root of the side groove
shall be located along the specimen centerline. Fig. 6 is a
schematic showing an example cross section of an alternative
side grooved specimen.
NOTE 4— Side-grooves increase the level of constraint with respect to
the recommended specimen. The increased constraint promotes a more
uniform stress state along the crack front and inhibits shear lip development. As a result, the KIcvalue from a side-grooved specimen is expected
to be lower than the KIc obtained from the recommended specimen,
particularly for thin products or products exhibiting Type I behavior. The
value of KIc from a side-grooved specimen may better represent the
fracture toughness of the material in structural situations where plasticity
is more highly constrained by the crack front geometry such as may be the
case for a surface or corner crack, or by structural details such as keyways,
radii, notches, etc. The value of KIc from the recommended specimen may
better represent the fracture toughness of the material in structural
situations where surface plasticity and shear lip development is not
constrained such as a through crack in a region of uniform thickness.
Side-grooving increases the likelihood of meeting the Pmax/PQ requirement, enabling a valid KIc to be obtained in products for which it would
not be possible using the recommended specimen. Side grooving after
precracking beneficially removes a portion of the non-linear crack front at
the ends of the crack front, thus increasing the likelihood of meeting crack
front straightness requirements. However, side grooving may also remove
material that influences service performance. This is often true for cast
parts and those for which thermo-mechanical working is part of the heat
treating cycle. The increased constraint also can lead to increased
likelihood of material delamination, for instance, in the plane of the
specimen, which could lead to test results different from those obtained
from plane-sided specimens.
NOTE 5—No interlaboratory ‘round robin’ test program has yet been
conducted to compare the performance of plain-sided and side-grooved
specimens. However, the results of several studies (7) indicate that KIc
from side-grooved specimens is zero to 10 % less than that of plain-sided
specimens, the difference increasing with increasing material toughness.
The within-laboratory repeatability was determined according to the
conditions in Terminology E456 and the results are presented in 11.3.
7.2.2.2 For lot acceptance testing, side-grooved specimens
shall not be used unless specifically allowed by the product
specification or by agreement between producer and user.
7.3 Specimen Preparation—All specimens shall be tested in
the finally heat-treated, mechanically-worked, and
environmentally-conditioned state. Specimens shall normally
be machined in this final state. However, for material that
cannot be machined in the final condition, the final treatment
may be carried out after machining provided that the required
dimensions and tolerances on specimen size, shape, and overall
finish are met (see specimen drawings of Figs. A3.1-A6.1 and
Fig. A7.1), and that full account is taken of the effects of
specimen size on metallurgical condition induced by certain
heat treatment procedures; for example, water quenching of
steels.
7.3.1 Fatigue Crack Starter Notch—Three fatigue crack
starter notch configurations are shown in Fig. 5. To facilitate
fatigue precracking at low stress intensity levels, the suggested
root radius for a straight-through slot terminating in a V-notch
is 0.08 mm (0.003 in.) or less. For the chevron form of notch,
the suggested root radius is 0.25 mm (0.010 in.) or less. For the
slot ending in a drilled hole, it is necessary to provide a sharp
stress raiser at the end of the hole. Care shall be taken to ensure
that this stress raiser is so located that the crack plane
orientation requirements of 8.2.4 can be met.
7.3.2 Fatigue Precracking—Fatigue precracking procedures are described in Annex A8. Fatigue cycling is continued
until a crack is produced that satisfies the requirements of
7.3.2.1 and 7.3.2.2 that follow.
7.3.2.1 Crack size (total size of crack starter plus fatigue
crack) shall be between 0.45W and 0.55W.
7.3.2.2 The size of the fatigue crack on each face of the
specimen shall not be less than the larger of 0.025W or 1.3 mm
(0.050 in.) for the straight-through crack starter configuration,
not less than the larger of 0.5D or 1.3 mm (0.050 in.) for the
slot ending in a hole (of diameter D < W/10), and need only
emerge from the chevron starter configuration.
8. General Procedure
8.1 Number of Tests—It is recommended that triplicate
tests, minimum, be made for each material condition.
8.2 Specimen Measurement—Specimen dimensions shall
conform to the drawings of Figs. A3.1-A6.1 and Fig. A7.1.
Measurements essential to the calculation of KIc are specimen
FIG. 6 Schematic of Side Groove Configration
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
8
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
thickness, B (and in the case of side-grooved alternative
specimens, BN), crack size, a, and width, W.
8.2.1 Specimen thickness, B (and in the case of sidegrooved alternative specimens, BN), shall be measured before
testing to the nearest 0.03 mm (0.001 in.) or to 0.1 %,
whichever is larger. For plain-sided specimens, B shall be
measured adjacent the notch. For side-grooved specimens, BN
shall be measured at the root of the notch and B adjacent the
notch.
NOTE 6—For plane-sided specimens the value of BN is equal to the
thickness B.
8.2.2 Specimen width, W, shall be measured, in conformance with the procedure of the annex appropriate to the
specimen configuration, to the nearest 0.03 mm (0.001 in.) or
0.1 %, whichever is larger, at not less than three positions near
the notch location, and the average value recorded.
8.2.3 Specimen crack size, a, shall be measured after
fracture to the nearest 0.5 % at mid-thickness and the two
quarter-thickness points (based on B for plain-sided specimens
and BN for side-grooved specimens). The average of these three
measurements shall be taken as the crack size, a. The difference between any two of the three crack size measurements
shall not exceed 10 % of the average. The crack size shall be
measured also at each surface. For the straight-through notch
starter configuration, no part of the crack front shall be closer
to the machined starter notch than 0.025W or 1.3 mm (0.050
in.), whichever is larger; furthermore, neither surface crack size
measurement shall differ from the average crack size by more
than 15 % and their difference shall not exceed 10 % of the
average crack size. For the chevron notch starter configuration,
the fatigue crack shall emerge from the chevron on both
surfaces; furthermore, neither surface crack size measurement
shall differ from the average crack size by more than 15 %, and
their difference shall not exceed 10 % of the average crack size.
8.2.4 The plane of the fatigue precrack and subsequent 2 %
crack extension (in the central flat fracture area; that is,
excluding surface shear lips) shall be parallel to the plane of the
starter notch to 610°. For side-grooved specimens, the plane
of the fatigue precrack and subsequent 2% crack extension
shall be within the root of the side-groove.
8.2.5 There shall be no evidence of multiple cracking (that
is, more than one crack) (8).
8.3 Loading Rate—For conventional (quasi-static) tests, the
specimen shall be loaded such that the rate of increase of
stress-intensity factor is between 0.55 and 2.75 MPa=m/s (30
and 150 ksi=in./min) during the initial elastic displacement.
Loading rates corresponding to these stress-intensity factor
rates are given in the Annex appropriate to the specimen being
tested. For rapid-force tests, loading rates are to be as specified
in Annex A10.
8.4 Test Record—A record shall be made of the output of
the force-sensing transducer versus the output of the displacement gage. The data acquisition system shall be set such that
not less than 50 % of full range is used for the test record. If an
autographic recorder is used, it shall be adjusted such that the
slope of the initial portion of the force-CMOD record is
between 0.7 and 1.5. Alternatively, if a computer data acqui-
sition system is used, it shall be programmed to capture enough
data to permit the calculations of Section 9.
8.4.1 The test shall be continued until the specimen can
sustain no further increase in applied force. The maximum
force (Pmax) shall be noted and recorded.
9. Calculation and Interpretation of Results
9.1 Interpretation of Test Record and Calculation of KIc—In
order to substantiate the validity of a KIc determination, it is
first necessary to calculate a conditional result, KQ, which
involves a construction on the test record, and then to determine whether this result is consistent with the size and yield
strength of the specimen according to 7.1. The procedure is as
follows:
9.1.1 When an autographic recorder is used, the conditional
value PQ is determined by drawing the secant line OP5, (see
Fig. 7) through the origin (point O) of the test record with slope
(P/V)5 equal to 0.95(P/V)o, where (P/V)o is the slope of the
tangent OA to the initial linear portion of the record (Note 7).
The force PQ is then defined as follows: if the force at every
point on the record which precedes P5 is lower than P5 (Fig. 7,
Type I), then P5 is PQ; if, however, there is a maximum force
preceding P5 which exceeds it (Fig. 7, Types II and III), then
this maximum force is PQ.
NOTE 7—Slight initial nonlinearity of the test record is frequently
observed, and is to be ignored. However, it is important to establish the
initial slope of the record with high precision. Therefore it is advisable to
minimize this nonlinearity by preliminarily loading the specimen to a
maximum force corresponding to a stress-intensity factor level not
exceeding that used in the final stage of fatigue cracking, then unloading.
NOTE 8—Residual stresses can adversely affect the indicated KQ and
KIc values. The applied loading is superimposed on the residual stresses,
resulting in a total crack tip stress-intensity different from that based solely
on the externally applied forces. In addition, residual stresses will likely
redistribute during machining when the specimen is extracted from the
host material. Hence, the magnitude of their influence on KQ and KIc in
the test specimen may be quite different from that in the original or finish
machined product (see also 5.1.6.)
9.1.2 When a computer data acquisition system is used, the
data reduction program shall determine the same forces (PQ
and Pmax) as above. The algorithms for doing this are discretionary.
9.1.3 The ratio Pmax/PQ, where Pmax is the maximum force
the specimen was able to sustain (see 8.4.1), shall be calculated. If this ratio does not exceed 1.10, proceed to calculate KQ
as described in the Annex appropriate to the specimen configuration. If Pmax/PQ does exceed 1.10, then the test is not a valid
KIc test and the user is referred to Test Method E1820 on
elastic-plastic fracture toughness.
9.1.4 The value 2.5(KQ/sYS)2, where sYS is the 0.2 % offset
yield strength in tension (see Test Methods E8/E8M), shall be
calculated. If this quantity is less than the specimen ligament
size, W–a then KQ is equal to KIc. Otherwise, the test is not a
valid KIc test. Expressions for calculating KQ are given in the
Annexes for each specified specimen configuration.
9.1.5 If the test result fails to meet the requirements of 9.1.3
or 9.1.4, or both, it will be necessary to use a larger specimen
to determine KIc.
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
9
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
FIG. 7 Principal Types of Force-Displacement (CMOD) Records
10. Report
10.1 The specimen configuration code shown on the specimen drawing (in the appropriate Annex) shall be reported. This
code shall be followed with the loading code (T for tension, B
for bending) and the code for crack plane orientation (see
3.1.3). The latter two codes shall appear in separate parentheses. As an example, a test result obtained using the compact
specimen (see Annex A4) might be designated as follows:
C(T)(S-T). The first letter (C) indicates the specimen to be a
compact configuration. The second letter (T) denotes the
loading as tension. The first of the two letters in the last bracket
(S) indicates the normal to the crack plane to be normal to the
direction of principal deformation. The second of these letters
(T) indicates the intended direction of crack extension to be
parallel with the direction of least deformation. For cylindrical
sections, where grain flow can be in the longitudinal, radial or
circumferential direction, the direction of maximum grain flow
shall be reported when the direction is known (see 3.1.3).
10.2 The following information shall be additionally reported for each specimen tested:
10.2.1 Characterization of the material (alloy code or chemistry and metallurgical condition) and product form (sheet,
plate, bar, forging, casting, and so forth) tested.
10.2.2 Specimen thickness, B, for plain-sided configurations. For side-grooved specimens, B, BN and (B· BN)1/2.
10.2.3 Specimen width (depth), W.
10.2.3.1 Loading hole offset, X, for the arc-shaped tension
specimen.
10.2.3.2 Outer and inner radii, r2 and r1, for arc-shaped
specimens.
10.2.4 Fatigue precracking conditions, specifically the
maximum stress-intensity factor, Kmax, stress-intensity factor
range, DKI, and number of cycles for the final 2.5 % of the
overall crack size, a (size of notch plus fatigue crack extension).
10.2.5 Crack size measurements, after fracture, at midthickness and the two quarter-thickness positions on the crack
front, as well as at the intersection of the crack front with the
specimen surface.
10.2.6 Test temperature.
10.2.7 Relative humidity as determined by Test Method
E337.
·
10.2.8 Loading rate in terms of KI (change in stressintensity factor per unit time) (3).
10.2.9 Force-versus-crack mouth opening displacement
(CMOD) record and associated calculations.
10.2.10 Yield strength as determined by Test Methods
E8/E8M.
10.2.11 KIc (or, KQ followed by the parenthetical statement
“invalid according to Sections(s) _____ of Test Method
E399”).
10.2.12 Pmax/PQ.
10.3 Fig. 8 is a convenient format for tabulating the information required in 10.1 and 10.2.
11. Precision and Bias
11.1 The precision of KIc measurements has been examined
in several interlaboratory round-robin studies. Selected aluminum alloys and high-strength steels were tested using standard
bend SE(B) (9), compact C(T) (10), and arc-shaped tension
A(T) (11) specimen configurations. The results are summarized
in 11.3 (Precision) and 11.5 (Bias) that follow. Not all of the
results reported satisfied all of the validity requirements of this
test method. Statistical analysis (10, 11, 12) was used to
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
10
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
FIG. 8 Suggested Form of Table for Reporting Information Listed in 10.1 and 10.2
exclude data that were likely influenced by deviations from the
validity requirements. No round-robin program has been conducted for the disk-shaped compact DC(T) specimen configuration, but limited data for that specimen configuration are
compared with data for other specimen configurations in
Annex A5. Round-robin studies specific to the quasi-static
testing of beryllium and the dynamic testing of a strain-rate
sensitive steel, and which involved special testing procedures,
are presented in Annex A9 and Annex A10.
11.2 It should be emphasized that the measures of precision
given in Table 1, Table 2, and Table 3 apply to alloys that
essentially exhibited no transitional fracture behavior with
temperature or strain rate under the specific test conditions of
the interlaboratory studies.
11.3 Precision—The precision of KIc determination is affected by errors in the measurement of test force and specimen
dimensions, especially the crack size. This test method specifies a precision for each measured quantity and, based on these
specifications and the round-robin results, a theoretical precision is rendered (13). Analysis of the method’s specifications
suggests that precision decreases with increasing relative crack
size, more for the bend than for the compact configuration. In
practice, the precision of KIc measurement may depend to an
unknown extent on the characteristics of the test record and
TABLE 1 Precision Using SE(B) Specimens (Nominal Crack Size-to-Specimen Width Ratio a/W =0.5)
Parameter
KIc
(MPa=m)
Material and
Yield Strength
Average
Repeatability
Standard
Deviation
Reproducibility
Standard
Deviation
Repeatability
Limit
Reproducibility
Limit
2219–T851
(353 MPa)
Maraging 18Ni
(1903 MPa)
4340–500 F
(1641 MPa)
4340–800 F
(1420 MPa)
35.94
2.27
2.54
6.37
7.11
57.02
2.15
4.03
6.03
11.27
48.55
1.86
2.17
5.20
6.07
87.76
3.03
4.13
8.49
11.56
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
11
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
TABLE 2 Precision Using C(T) Specimens (Nominal Crack Size-to-Specimen Width Ratio a/W = 0.5)
Parameter
KIc
(MPa=m)
Material and
Yield Strength
Average
Repeatability
Standard
Deviation
Reproducibility
Standard
Deviation
Repeatability
Limit
Reproducibility
Limit
2219–T851
(353 MPa)
Maraging 18Ni
(1903 MPa)
4340–500 F
(1641 MPa)
4340–800 F
(1420 MPa)
35.61
1.91
2.17
5.36
6.07
59.06
2.14
2.65
5.98
7.41
50.38
2.12
2.87
5.95
8.04
87.83
2.21
3.14
6.19
8.80
TABLE 3 Precision Precision for A(T) Specimens (Nominal Crack Size-to-Specimen Width Ratio a/W = 0.5)
Parameter
Specimen
Type
Average
Repeatability
Standard
Deviation
Reproducibility
Standard
Deviation
Repeatability
Limit
Reproducibility
Limit
KIc
(MPa=m)
X/W = 0
X/W = 0.05
102.3
101.6
4.69
2.33
7.16
4.81
13.13
6.53
20.05
13.47
analysis skills of the laboratory personnel. It is possible to
derive useful information concerning the precision of KIc
measurement from three round-robin programs (10, 11, 12) as
described below. Results for bend, compact, and arc-shaped
specimen configurations were obtained for several aluminum
alloys and high strength steels. The materials were chosen for
their reproducible, uniform composition and microstructure.
Thereby the contribution of material variability to the measurement of KIc was minimized.
11.3.1 An interlaboratory study (9) for the measurement of
plane strain fracture toughness, KIc on metallic materials, using
SE(B) specimens, was conducted among nine laboratories
using four metallic materials (one aluminum alloy and three
high-strength steels). 180 specimens were tested (5 per laboratory and material). Analyses were undertaken in accordance
with Practice E691, see ASTM Research Report No. E0810045 and Table 1.
11.3.2 A second interlaboratory study (10) for the measurement of plane strain fracture toughness, KIc on metallic
materials, using C(T) specimens, was conducted among nine
laboratories using the same four metallic materials (one aluminum alloy and three high-strength steels). 216 specimens
were tested (6 per laboratory and material). Analyses were
undertaken in accordance with Practice E691, see ASTM
Research Report No. E08-10056 and Table 2.
11.3.3 A third interlaboratory study (11) for the measurement of plane strain fracture toughness, KIc, using arc-shaped
A(T) specimens, with two different loading hole configurations
(X/W = 0 and X/W = 0.5), was conducted among eight
laboratories using one high strength steel (Ni-Cr-Mo-V
vacuum-degassed steel, yield strength sYS= 1324 MPa). 48
5
Supporting data have been filed at ASTM International Headquarters and may
be obtained by requesting Research Report: RR:E08-1004.
6
Supporting data have been filed at ASTM International Headquarters and may
be obtained by requesting Research Report: RR:E08-1005.
specimens were tested (from 3 to 5 per laboratory). Analyses
were undertaken in accordance with Practice E691, see ASTM
Research Report No.E08-10067 and Table 3.
11.3.4 The terms repeatability limit and reproducibility limit
are used as specified in Practice E177.
11.3.5 The results presented in Table 1, Table 2, and Table 3
shall not be transferred to materials or KIc levels other than
those relevant to the specific interlaboratory studies (9, 10, 11).
11.4 Alternative side-grooved specimens were tested to
determine within-laboratory limit and repeatability according
to the conditions in Terminology E456. The testing was
performed on aluminum alloy 7055–T7951 using C(T) specimens having a nominal dimensions W=50.8 (2.0 in), B =25.4
mm (1.0 in.) BN= 20.3 mm (0.80 in.) notch root angle = 45°
and notch root radius = 0.5mm (0.02 in.). The results are given
in Table 4 along with results obtained from plain-sided
specimens from manufactured the same lot of material, tested
at the same time, and under the same test conditions The
repeatability standard deviation for this test series 0.22 MP=m
(0.20 ksi=in.) for side-grooved specimens and 0.33 MPa=m
(0.30 ksi=in.) for the plane-sided specimens.
11.5 Bias—There is no accepted standard value for the
plane-strain fracture toughness of any material. In the absence
of such a true value, any statement concerning bias is not
meaningful.
7
Supporting data have been filed at ASTM International Headquarters and may
be obtained by requesting Research Report: RR:E08-1006.
TABLE 4 Repeatability Results for Side-Grooved and PlaneSided C(T) Specimens 7055–T7951
Parameter Specimen
Type
No. of
Average Repeatability Repeatability
Specimens
Standard
Limit
Deviation
Side-Grooved 11
KIc
(MPa=m) Plane-Sided
11
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
12
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
26.9
27.9
0.22
0.33
0.49
0.74
E399 – 09´2
ANNEXES
(Mandatory Information)
A1. DOUBLE-CANTILEVER DISPLACEMENT GAGE
A1.1 The displacement gage consists of two cantilever
beams and a spacer block clamped together with a single bolt
and nut (Fig. 2). Electrical-resistance strain gages are adhesively bonded to the tension and compression surfaces of each
beam, and are connected as a Wheatstone bridge incorporating
a suitable balancing resistor. The beams are made of material
with a high ratio of yield strength-to-elastic modulus. One such
material is solution treated Ti-13V-11Cr-3Al titanium alloy.
For material of different modulus, the spring constant of the
assembly is correspondingly different, but other characteristics
are unaffected. Detailed dimensions for the beams and spacer
block are given in Figs. A1.1 and A1.2. Those particular values
provide a linear (working) range from 3.8 to 7.6 mm (0.15 to
0.30 in.) and a gage length of 5.1 to 6.4 mm (0.20 to 0.25 in.).
The gage length can be adjusted by substituting a differently
sized spacer block. The gage’s required precision is stated as a
maximum deviation of 60.003 mm (0.0001 in.) from a
least-squares-best-fit straight line through its displacement
calibration data (see 6.4.2). Additional details concerning
design, construction and use of the gage are given in (14).
NOTE—Dimensions are in mm.
Inch-Pound Units Equivalents
mm
in.
mm
in.
0.10
0.15
0.25
0.48
0.51
0.53
0.64
0.76
0.99
1.04
0.004
0.006
0.010
0.019
0.020
0.021
0.025
0.030
0.039
0.041
1.52
1.65
3.6
4.72
4.78
9.40
9.47
9.52
41.15
41.28
0.060
0.065
9⁄64
0.186
0.188
0.370
0.373
0.375
1.620
1.625
FIG. A1.1 Beams for Double-Cantilever Displacement Gage
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
13
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
NOTE 1—2-mm diameter holes are for strain gage leads.
NOTE 2—Dimensions are in mm.
Inch-Pound Units Equivalents
mm
in.
mm
in.
0.80
1.14
1.27
2.00
2.21
2.36
3.18
3.60
4.72
4.78
4.83
⁄
0.045
0.050
5⁄64
0.087
0.093
0.125
9⁄64
0.186
0.188
0.190
4.95
5.21
9.47
9.52
9.55
9.60
10.16
10.21
12.45
12.70
12.70
0.195
0.205
0.373
0.375
0.376
0.378
0.400
0.402
0.490
1 ⁄2
0.500
1 32
FIG. A1.2 Aluminum-Alloy Spacer Block for Double-Cantilever
Displacement Gage
A2. TESTING FIXTURES
A2.1 Bend Specimen Loading Fixture
A2.1.1 The bend test is performed using fixtures designed to
minimize friction effects by allowing the support rollers to
rotate and translate slightly as the specimen is loaded, thereby
achieving rolling contact. A design suitable for testing standard
bend (SE(B)) and arc-shaped bend (A(B)) specimens is shown
in Fig. A2.1. While free to roll and translate during test, the
rollers are initially positioned against stops that set the span
length and are held in place by low-tension springs (such as
rubber bands).
A2.1.2 The bend fixture is aligned such that the line of
action of the applied force passes midway between the support
rollers to 61.0 % of the span, S, and is perpendicular to the
roller axes to 62° (15). The span is to be measured to 60.5 %.
A2.2 Compact Specimen Loading Clevis
A2.2.1 A loading clevis suitable for testing standard compact (C(T)), arc-shaped tension (A(T)), and disk-shaped compact (DC(T) specimens is shown in Fig. A2.2. Both ends of the
specimen are held in the clevis and loaded through pins in
order to allow rotation of the specimen during testing. The
clevis holes are provided with small flats on the loading
surfaces to provide rolling contact, thereby minimizing friction
effects (16).
A2.2.2 The size, proportions, and tolerances for the clevis
shown in Fig. A2.2 are all scaled to specimens with W/B = 2 for
B $ 13 mm (0.5 in.), and W/B = 4 for B # 13 mm (0.5 in.).
Clevis and pins made from 1930 MPa (280 ksi) yield strength
maraging steel are suitable for testing specimens of the sizes
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
14
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
and sys/E ratios of 7.1.3. For lower-strength clevis material or
substantially larger specimens at a given sys/E ratio, larger
clevises are required. As indicated in Fig. A2.2, the clevis
corners may be trimmed sufficiently to accommodate seating of
the displacement gage in specimens less than 9.53 mm (0.375
in.) thick.
A2.2.3 To minimize eccentricity in the load train, the
loading rods shall be aligned to 60.8 mm (0.03 in.) and the
specimen centered in the clevis slot to 60.8 mm (0.03 in.).
NOTE 1—Dimensions in mm, except surface finishes in µm.
NOTE 2—Support rollers and specimen contact surface of loading ram shall be parallel to each other within 0.002 W.
NOTE 3—2.54 mm = 0.100 in., 3.81 mm = 0.150 in, 1.6 µ-m = 63 µ-in.
FIG. A2.1 Loading Fixture for Standard SE(B) (shown) and Arc-Shaped A(B) (not shown) Bend Specimens
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
15
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
NOTE 1—Surface finishes in µm.
NOTE 2—Pin diameter = 0.24 W (+0.000 W/–0.005 W). For specimens with sys > 1379 ksi (200 ksi), the holes in the specimen and in the clevis may
be 0.30 W (+0.005 W/–0.000 W) and the pin diameter 0.288 W (+0.000W/–0.005 W).
NOTE 3—Corners of the clevis may be removed if necessary to accommodate the clip gage.
NOTE 4—1.6 µm = 63 µin., 3.2 µm = 125 µin.
NOTE 5—Accumulated experience indicates that subtle deviation from the recommended configuration can lead to complications. For instance,
undersized loading pins can lead to inelastic load transfer from clevis to specimen. Poorly machined flats can also cause difficulties.
FIG. A2.2 Loading Clevis for Compact C(T), Arc-Shaped A(T) and Disk-Shaped DC(T) Tension Specimens
A3. SPECIAL REQUIREMENTS FOR TESTING BEND SPECIMENS
A3.1 Specimen
A3.1.1 The standard bend specimen configuration is a
single- edge-notched and fatigue precracked beam loaded in
three-point bending. The support span, S, is nominally equal to
four times the specimen width, W. The general proportions of
the standard configuration are shown in Fig. A3.1.
A3.1.2 Alternative configurations may have 1 # W/B # 4;
however, these specimens shall also have a nominal support
span equal to 4W.
A3.2 Specimen Preparation
A3.2.1 Generally applicable specifications regarding specimen size, configuration and preparation are given in Section 7.
A3.2.2 In the interest of K-calibration accuracy, it is desirable to fatigue precrack bend specimens using the same loading
fixture to be used in subsequent testing.
A3.2.3 Bend specimens are occasionally precracked in cantilever bending, especially for reversed force cycling (see
A9.2.3.2). If the three-point bending K-calibration is used for
cantilever bending, the cantilever bending moment for a given
K value will be underestimated (8). The crack tip stress field in
cantilever bending can be distorted by excessive clamping
forces, thereby affecting fatigue crack planarity.
A3.3 Apparatus
A3.3.1 Bend Test Fixture—The loading fixture for bend
testing is illustrated in Fig. A2.1 and discussed in A2.1. The
fixture is designed to minimize friction effects by allowing the
rollers to rotate and translate slightly as the specimen is loaded,
thus providing rolling contact.
A3.3.2 Displacement Gage—Details regarding displacement gage design, calibration, and use are given in 6.4. For the
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
16
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
bend specimen, displacements are essentially independent of
gage length up to W/2.
W
a
A3.4 Procedure
A3.4.1 Measurement—Specimen width (depth), W, is measured from the notched edge of the specimen to the opposite
edge. Crack size a, is measured from the notched edge to the
crack front.
A3.4.1.1 General requirements concerning specimen measurement are given in 8.2.
A3.4.2 Bend Specimen Testing—General principles concerning the loading fixture and its setup appear in A2.1.
A3.4.2.1 Locate the specimen with the crack tip midway
between the rolls to within 1 % of the span, and square to the
roll axes within 2°. The displacement gage is seated on the
knife edges such as to maintain registry between knife edges
and gage grooves. In the case of attachable knife edges, the
gage is seated before the knife edge positioning screws are
tightened.
A3.4.2.2 The specified rate of increase of the stressintensity factor (see 8.3) ranges from 0.55 and 2.75 MPa=m/s
(30 and 150 ksi=in./min) and corresponds to a loading rate for
a standard (W/B = 2) 25.4 mm (1 in.) thick specimen between
0.30 to 1.5 kN/s (4.0 and 20 klbf/min).
A3.4.2.3 Details concerning recording of the test record are
given in 8.4.
A3.5 Calculations
A3.5.1 Interpretation of Test Record—General requirements and procedures for interpreting the test record are given
in 9.1.
A3.5.2 Validity Requirements—Validity requirements in
terms of limitation on Pmax/PQ and mandatory specimen size
are given in 9.1.3 through 9.1.4.
A3.5.3 Calculation of KQ—Bend specimen KQ is calculated
in SI or inch-pound units of Pa=m (psi=in.) as follows (see
Note A3.2):
KQ 5
= specimen width (depth) as determined in A3.4.1, m
(in.), and
= crack size as determined in 8.2.3, m (in.).
NOTE A3.1—Example: for a/W = 0.500, ƒ(a/W) = 2.66.
NOTE A3.2—This expression for a/W is considered to be accurate
within 1 % over the range 0.2 # a/W # 1 for S/W = 4 (17).
A3.5.4 Calculation of Crack Mouth Opening Compliance
Using Crack Size Measurements—Bend specimen crack mouth
opening compliance, Vm/P, is calculated in units of m/N (in./lb)
as follows (see Note A3.4):
SD
Vm
a
S
P 5 E’ BeW · q W
where:
S DF
a
6 W
S DS
DF
S DS
2
SD
a
2 2.04 W
3
S
a
1 0.66/ 1 2 W
DG
2
NOTE A3.3—Example: for a/W = 0.500, q(a/W) = 8.92.
NOTE A3.4—This expression is considered to be accurate within 1.0 %
over the entire range 0 # a/W # 1 for S/W = 4 (19). It is valid only for
crack mouth opening displacement measured at the location of the integral
knife edges shown in Fig. 3. Attachable knife edges must be reversed or
inset to effect the same measurement points.
A3.5.5 Calculation of Crack Size Using Crack Mouth
Opening Compliance Measurements—Bend specimen normalized crack size is calculated as follows (see Note A3.5):
a
W5
(A3.5)
1.00023.950 · U12.982 · U223.214 · U3151.516 · U42113.031 · U5
(A3.1)
U5
11
SD Œ
for
PQ
B
BN
a
a
0.76 2 2.28W 1 3.87 W
(A3.4)
for which:
E’ = elastic constraint modulus (E for plane stress;
E/(1 − n2) for plane strain), Pa (psi),
n = Poisson’s Ratio,
Be = B – (B– BN)2/B, and
S, B, BN, W, and a are defined in A3.5.3.
where:
a
ƒ W 53
SD
SD
a
q W 5
where:
SD
a
PQS
·ƒ W
=BBN W3/2
a
a 1.99 2 W
W·
(A3.3)
S DG
a 2
a
2.15 2 3.93 W 1 2.7 W
a
a 3/2
2 1 1 2w
12W
(A3.2)
a
12W
D
which:
= force as determined in 9.1.1, N (lbf),
= specimen thickness as determined in 8.2.1, m (in.),
= specimen thickness between the roots of the side
grooves, as determined in 8.2.1, m (in.),
S
= span as determined in A3.4.2 (see also A2.1), m (in.),
ŒS
1
E’ BeVm
P
DS D
4W
S
(A3.6)
for which:
Vm = crack mouth opening displacement, m (in.),
P
= applied force, N (lbf), and
Be = B – (B– BN)2/B, and
E’ is defined in A3.5.4 and S, B, BN, W and a are defined in
A3.5.3.
NOTE A3.5—This expression fits the equation in A3.5.4 within 0.05 %
of W in the range 0.3 # a/W # 0.9 for S/W = 4 (20). It is valid only for
crack mouth opening displacement measured at the location of the integral
knife edges shown in Fig. 3. Attachable knife edges must be reversed or
inset to effect the same measurement points.
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
17
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
NOTE
NOTE
NOTE
NOTE
NOTE
NOTE
1—Surface finishes in µm.
2—A surfaces shall be perpendicular and parallel as applicable within 0.001 W TIR.
3—Crack starter notch shall be perpendicular to specimen surfaces within 2°.
4—Integral or attachable knife edges for clip gage attachment may be used (see Figs. 3 and 4)
5—For starter notch and fatigue crack configuration see Fig. 5.
6—1.6 µm = 63 µin., 3.2 µm = 125 µin.
FIG. A3.1 Bend SE(B) Specimen—Standard Proportions and Tolerances
A4. SPECIAL REQUIREMENTS FOR TESTING COMPACT SPECIMENS
A4.1 Specimen
A4.1.1 The standard compact specimen configuration is a
single-edge-notched and fatigue precracked plate loaded in
tension. The general proportions of the standard configuration
are shown in Fig. A4.1.
A4.1.2 Alternative configurations may have 2 # W/B # 4,
but with other proportions unchanged.
A4.2 Specimen Preparation
A4.2.1 Generally applicable specifications regarding specimen size, configuration and preparation are given in Section 7.
A4.3 Apparatus
A4.3.1 Tension Testing Clevis—A loading clevis suitable for
testing compact specimens is shown in Fig. A2.2 and discussed
in A2.2. The clevis is designed to minimize friction effects by
providing for rolling contact of the loading pins and rotation of
the specimen during specimen loading.
A4.3.2 Displacement Gage—Details regarding displacement gage design, calibration, and use are given in 6.4. For the
compact specimen, displacements are essentially independent
of gage length up to 1.2W.
A4.4 Procedure
A4.4.1 Measurement—Specimen width, W, and crack size,
a, are measured from the plane of the centerline of the loading
holes. The notched edge may be used as a convenient reference
line, taking into account (that is, subtracting) the distance from
the centerline of the holes to the notched edge to arrive at W
and a.
A4.4.1.1 General requirements concerning specimen measurement are given in 8.2.
A4.4.2 Compact Specimen Testing—General principles
concerning the loading clevis and its setup appear in A2.2.
When assembling the loading train (clevises and their attachments to the tensile machine), care shall be taken to minimize
eccentricity of loading due to misalignments external to the
clevises.
A4.4.2.1 The displacement gage is seated on the knife edges
such as to maintain registry between knife edges and gage
grooves. In the case of attachable knife edges, the gage is
seated before the knife edge positioning screws are tightened.
A4.4.2.2 The specified rate of increase of the stressintensity factor is within the range 0.55 and 2.75 MPa=m/s
(30 and 150 ksi=in./min) corresponding to a loading rate for
a standard (W/B = 2) 25 mm (1.0 in.) thick specimen between
0.33 and 1.67 kN/s (4.5 to 22.5 klbf/min).
A4.4.2.3 Details concerning recording of the test record are
given in 8.4.
A4.5 Calculations
A4.5.1 General requirements and procedures for interpreting the test record are given in 9.1.
A4.5.2 Validity Requirements—Validity requirements in
terms of limitation on Pmax/PQ and mandatory specimen size
are given in 9.1.3 through 9.1.4.
A4.5.3 Calculation of KQ—Compact specimen KQ is calculated in SI or inch-pound units of Pa=m (psi=in.) as follows
(see Note A4.2):
KQ 5
SD
a
PQ
·ƒ W
=BBN=W
(A4.1)
where:
SD
SD
S D
a
ƒ W 5
S
a
21W
for
PQ
B
BN
DF
(A4.2)
SD
a 2
a
a
0.886 1 4.64W 2 13.32 W 1 14.72 W
a 3/2
12W
3
S DG
a
2 5.6 W
4
which:
= force as determined in 9.1.1, N (lbf),
= specimen thickness as determined in 8.2.1, m (in.),
= specimen thickness between the roots of the side
grooves, as determined in 8.2.1, m (in.),
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
18
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
W
a
= specimen width (depth) as determined in A3.4.1, m
(in.), and
= crack size as determined in 8.2.3 and A4.4.1, m (in.).
NOTE A4.1—Example: for a/W = 0.500, ƒ(a/W) = 9.66.
NOTE A4.2—This expression for a/W is considered to be accurate
within 0.5 % over the range 0.2 # a/W # 1 (18, 21).
A4.5.4 Calculation of Crack Mouth Opening Compliance
Using Crack Size Measurements—Compact specimen crack
mouth opening compliance, Vm/P, is calculated in units of m/N
(in./lb) as follows (see Note A4.4):
SD
Vm
a
1
P 5 E’ Be · q W
(A4.3)
NOTE A4.4—This expression is considered to be accurate to within
1.0 % for a/W $ 0.2 (22). This expression is valid only for crack mouth
opening displacement measured at the location of the integral knife edges
shown in Fig. 3. Attachable knife edges must be reversed or inset to effect
the same measurement points.
A4.5.5 Calculation of Crack Size Using Crack Mouth
Opening Compliance Measurements—Compact specimen normalized crack size is calculated as follows (see Note A4.5):
a
W5
1.00024.500·U113.157·U22172.551·U31879.944·U421514.671·U5
where:
where:
SD
SD
U5
a
q W 5
S
19.75
a
12W
D
2
F
a
a
0.5 1 0.192W 1 1.385 W
2
11
(A4.4)
SD
a
2 2.919 W
3
S DG
a
1 1.842 W
4
for which:
E’ = elastic constraint modulus(E for plane stress, Pa (psi);
E/(1 − n2) for plane strain, Pa (psi),
n = Poisson’s Ratio,
Be = B – (B– BN)2/B, and
B, BN, W and a are defined in A4.5.3.
NOTE A4.3—Example: for a/W = 0.500, q(a/W) = 54.71.
(A4.5)
1
E’ BeVm
P
Œ
(A4.6)
for which:
Vm = crack mouth opening displacement, m (in.),
P
= applied force, N (lbf), and
Be = B – (B– BN)2/B, and
E’ is defined in A4.5.4 and B, BN, W and a are defined in
A4.5.3.
NOTE A4.5—This expression fits the equation in A4.5.4 within 0.01 %
of W for 0.2 # a/W # 0.8 (22). It is valid only for crack mouth opening
displacement measured at the location of the integral knife edges shown
in Fig. 3. Attachable knife edges must be reversed or inset to effect the
same measurement points.
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
19
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
NOTE 1—Surface finishes in µm.
NOTE 2—A surfaces shall be perpendicular and parallel to within 0.002 W TIR.
NOTE 3—The intersection of the crack starter notch tips with the two specimen surfaces shall be equally distant from the top and bottom edges of the
specimen within 0.005 W.
NOTE 4—Integral or attachable knife edges for clip gage attachment to the crack mouth may be used (see Figs. 3 and 4).
NOTE 5—For starter notch and fatigue crack configuration see Fig. 5.
NOTE 6—1.6 µm = 63 µin., 3.2 µm = 125 µin.
FIG. A4.1 Compact C(T) Specimen—Standard Proportions and Tolerances
A5. SPECIAL REQUIREMENTS FOR TESTING DISK-SHAPED COMPACT SPECIMENS
A5.1 Specimen
A5.1.1 The standard disk-shaped compact specimen configuration is a single-edge-notched and fatigue precracked disk
segment loaded in tension (23). The general proportions of the
standard configuration are shown in Fig. A5.1.
A5.1.2 Alternative configurations may have 2 # W/B # 4,
but with other proportions unchanged.
A5.2 Specimen Preparation
A5.2.1 Generally applicable specifications regarding specimen size, configuration and preparation are given in Section 7.
A5.3 Apparatus
A5.3.1 Tension Testing Clevis—A loading clevis suitable for
testing disk-shaped compact specimens is shown in Fig. A2.2
and discussed in A2.2. The clevis is designed to minimize
friction effects by providing for rolling contact of the loading
pins and rotation of the specimen during specimen loading.
A5.3.2 Displacement Gage—Details regarding displacement gage design, calibration, and use are given in 6.4. For the
disk-shaped compact specimen, displacements are essentially
independent of gage length up to 0.55W.
A5.4 Procedure
A5.4.1 Measurement—Analyses of this specimen assume it
is machined from a circular blank and therefore measurements
of circularity as well as width, W, and crack size, a, must be
made.
A5.4.1.1 The specimen blank shall be checked for circularity before specimen machining. The radius shall be measured
at eight equally spaced points around the circumference, and
one of these points shall lie in the intended crack plane. The
average of these readings is taken as the radius, r. If any
measurement differs from r by more than 5.0 %, the blank is to
be machined to the required circularity. Otherwise, D = 2r =
1.35W.
A5.4.1.2 Specimen width, W, and crack size, a, are measured from the plane of the centerline of the loading holes. The
notched edge may be used as a convenient reference line taking
into account (that is, subtracting) the distance from the centerline of the holes to the notched edge to arrive at W and a.
A5.4.1.3 General requirements concerning specimen measurement are given in 8.2.
A5.4.2 Disk-Shaped Compact Specimen Testing—General
principles concerning the loading clevis and its setup appear in
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
20
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
NOTE 1—Surface finishes in µm.
NOTE 2—A surfaces shall be perpendicular and parallel to within 0.002 W TIR.
NOTE 3—The intersection of the crack starter notch tips with the two specimen surfaces shall be equally distant from the top and bottom edges of the
specimen within 0.005 W.
NOTE 4—Integral or attachable knife edges for clip gage attachment to the crack mouth nay be used (see Figs. 3 and 4).
NOTE 5—For starter notch and fatigue crack configuration see Fig. 5.
NOTE 6—1.6 µm = 63 µin, 3.2 µm = 125 µin.
FIG. A5.1 Disk-Shaped Compact DC(T) Specimen—Standard Proportions and Tolerances
TABLE A5.1 Results of KIc Tests on Disk-Shaped Compact DC(T),
Compact C(T), Arc-Shaped A(T) Tension Specimens
Laboratory 1
Ni-Cr-Mo Steel
sys = 1324 MPa
(192 ksi)
DiskShaped
Compact,
DC(T)
ArcShaped
Tension,
A(T)
Laboratory 2
Ni-Cr-Mo Steel
sys = 1289 MPa
(187 ksi)
DiskShaped
Compact,
DC(T)
Standard
Compact,
A(T)
Mean, X̄
109.4 (99.5) 109.2 (99.4) 114.7 (104.4) 116.4 (105.9)
Standard Deviation, S 4.38 (3.99) 3.76 (3.42) 1.86 (1.69)
3.56 (3.24)
A5.5 Calculations
A5.5.1 General requirements and procedures for interpreting the test record are given in 9.1.
A5.5.2 Validity Requirements—Validity requirements in
terms of limitation on Pmax/PQ and mandatory specimen size
requirements are given in 9.1.3 through 9.1.4.
A5.5.3 Calculation of KQ—Disk-shaped compact specimen
KQ is calculated in SI or inch-pound units of Pa=m (psi=in.)
as follows (see Note A5.2):
KQ 5
NOTE—Units of mean and standard deviation are MPa=m (ksi=in.).
SD
a
PQ
·ƒ W
B=W
(A5.1)
where:
A2.2. When assembling the loading train (clevises and their
attachments to the tension machine), care shall be taken to
minimize eccentricity of loading due to misalignments external
to the clevises.
A5.4.2.1 The displacement gage is seated on the knife edges
such as to maintain registry between knife edges and gage
grooves. In the case of attachable knife edges, the gage is
seated before the knife edge positioning screws are tightened.
A5.4.2.2 The specified rate of increase of the stressintensity factor is within the range 0.55 and 2.75 MPa=m/s
(30 and 150 ksi=in./min) corresponding to a loading rate for
a standard (W/B = 2) 25 mm (1.0 in.) thick specimen between
0.33 and 1.67 kN/s (4.5 to 22.5 klbf/min).
A5.4.2.3 Details concerning recording of the test record are
given in 8.4.
SD
SD
S D
a
ƒ W 5
S
a
21W
DF
a
a
0.76 1 4.8W 2 11.58 W
a
12W
2
(A5.2)
SD
a
1 11.43 W
3
S DG
a
2 4.08 W
4
3/2
for
PQ
B
W
which:
= force as determined in 9.1.1, N (lbf),
= specimen thickness as determined in 8.2.1, m (in.),
= specimen width (depth) as determined in A5.4.1, m
(in.), and
a = crack size as determined in 8.2.3 and A5.4.1, m (in.).
NOTE A5.1—Example: for a/W = 0.500, ƒ(a/W) = 10.17.
NOTE A5.2—This expression for a/W is considered to be accurate
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
21
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
within 0.3 % over the range 0.2 # a/W # 1 (24).
1.00024.459 · U12.066 · U2213.041 · U31167.627 · U42481.4 · U5
A5.5.4 Calculation of Crack Mouth Opening Compliance
Using Crack Size Measurements—Disk-shaped compact specimen crack mouth opening compliance, Vm/P, is calculated in
units of m/N (in./ lb) as follows (see Note A5.4):
SD
Vm
a
1
P 5 E’ B · q W
(A5.3)
where:
SD
SD
a
q W 5
19.75
a
12W
S D
2
F
(A5.4)
SD
S DG
a 2
a 3
a
a
0.46920.056W11.86 W 22.06 W 10.789 W
4
for which:
E’ = elastic constraint modulus (E for plane stress, Pa
(psi); E/(1 − n2) for plane strain, Pa (psi),
n = Poisson’s Ratio, and
B, W and a are defined in A5.5.3.
NOTE A5.3—Example: for a/W = 0.500, q(a/W) = 55.1.
NOTE A5.4—This expression is considered to be accurate to within
1.0 % for a/W $ 0.2 (22). This expression is valid only for crack mouth
opening displacement measured at the location of the integral knife edges
shown in Fig. 3. Attachable knife edges must be reversed or inset to effect
the same measurement points.
A5.5.5 Calculation of Crack Size Using Crack Mouth
Opening Compliance Measurements—Disk-shaped compact
specimen normalized crack size is calculated as follows (see
Note A5.5):
a
W5
(A5.5)
where:
U5
11
1
E’ BVm
P
Œ
(A5.6)
for which:
Vm = crack mouth opening displacement, m (in.),
P
= applied force, N (lbf), and
E’ is defined in A5.5.4 and B, W and a are defined in A5.5.3.
NOTE A5.5—This expression fits the equation in A5.5.4 within 0.01 %
of W for 0.2 # a/W # 0.8 (22). This expression is valid only for crack
mouth opening displacement measured at the location of the integral knife
edges shown in Fig. 3. Attachable knife edges must be reversed or inset to
effect the same measurement points.
A5.6 Precision and Bias (see also Section 11)
A5.6.1 There has been no round-robin test program for the
disk-shaped compact specimen. However, the results of two
testing programs (23) designed to compare the results of the
disk-shaped compact DC(T) specimen with those of the
compact C(T) and arc-shaped tension A(T) specimens are
summarized in Table A5.1. Based on the results in Table A5.1
and the geometric similarity of the specimens, there is no
reason to suspect that the precision for the disk-shaped
compact specimen would differ from that for the standard
compact specimen. The arc-tension specimen has been shown
(11) to have essentially the same grand mean and standard
deviation as the standard compact specimen.
A6. SPECIAL REQUIREMENTS FOR TESTING ARC-SHAPED TENSION SPECIMENS
A6.1 Specimen
A6.1.1 The standard arc-shaped tension specimen configuration is a single-edge-notched and fatigue precracked ring
segment loaded in tension. The general proportions of (two
variants of) the standard configuration are shown in Fig. A6.1.
The value of the radius ratio r1/r2 is unspecified, so specimens
may be taken from any cylindrical geometry. It should be
noted, however, that specimens with r1/r2 = 0 (that is, from a
solid cylinder) do not make efficient use of test material,
because W for the arc-shaped tension specimen applies to
hollow cylinders. The disk-shaped specimen shall be used for
tests of solid cylinders (see Annex A5).
A6.1.2 The arc-shaped tension specimen measures toughness only for a crack whose normal is circumferential and
propagation direction is radial, designated C-R (see 3.1.3). For
other crack plane orientations and propagation directions the
bend (Annex A3) or compact (Annex A4) specimen are to be
used.
A6.1.3 The specimen depicted in Fig. A6.1(a) with X/W =
0.5 represents a half-ring segment. The specimen with X/W =
0 (Fig. A6.1(b)) is the smallest specimen of this configuration
that can be cut from a ring.
A6.1.4 Alternative configurations may have 2 # W/B # 4,
but with other proportions unchanged. The use of alternative
specimen proportions is advantageous when a specimen can be
extracted from a ring segments without machining the inner
and outer radii; that is, with no change in W.
A6.2 Specimen Preparation
A6.2.1 Generally applicable specifications regarding specimen size, configuration and preparation are given in Section 7.
A6.3 Apparatus
A6.3.1 Tension Testing Clevis—A loading clevis suitable for
testing arc-shaped tension specimens is shown in Fig. A2.2 and
discussed in A2.2. The clevis is designed to minimize friction
effects by providing for rolling contact of the loading pins and
rotation of the specimen during specimen loading.
A6.3.2 Displacement Gage—Details regarding displacement gage design, calibration, and use are given in 6.4. For the
arc-shaped tension specimen, displacements are essentially
independent of gage length up to W/2.
A6.3.2.1 An alternative means for measuring displacement
is permitted for the specimen with X/W = 0.5. Conical
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
22
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
NOTE 1—Surface finishes in µm.
NOTE 2—For starter notch and fatigue crack configurations see Fig. 5.
NOTE 3—Alternative displacement gage reference points (see A6.4.1.1 for calculation of (a)).
NOTE 4—Axis of holes to be tangent to inner radius within 0.005 W.
NOTE 5—A surfaces to be perpendicular parallel as applicable within 0.002 W TIR. D surfaces to be perpendicular or parallel as applicable to A surfaces
within 0.02 W TIR (see A6.4.1).
NOTE 6—1.6 µm = 63 µin, 3.2 µm = 125 µin.
FIG. A6.1 Arc-Shaped Tension A(T) Specimen Designs—Standard Proportions and Tolerances
center-punch-type indentations are provided on the inner surface of the specimen at mid-thickness and in the plane of the
centerline of the loading holes as shown in Fig. A6.1(a).
Load-point displacement is measured at these points using a
displacement gage fitted with points and meeting the requirements of 6.4.
A6.4 Procedure
A6.4.1 Measurement—Before testing, (r2 − r1) is measured
to the nearest 0.03 mm (0.001 in.) or to 0.1 %, whichever is
greater, at mid-thickness positions on both sides of, and
immediately adjacent to, the crack starter notch mouth. The
average of these two readings is taken as W. Measurement of
(r2 − r1) is also made at four additional positions, two as close
as possible to the loading holes and two at approximately
one-half the circumferential distance between the loading holes
and the crack plane. If any of these four measurements differ
from W by more than 10 %, the specimen shall be discarded or
reworked. The distance between the loading-hole centers and
the outside surface of the specimen at the notch plane is
measured to the nearest 0.03 mm (0.001 in.) or to 0.1 %,
whichever is greater. This measurement is made on both sides
of the specimen by referencing the loading holes. Specimen
width is subtracted from the average of these two measurements and the difference recorded as the quantity X. The
distance g between the crack mouth opening displacement
measurement reference points is measured to within 5.0 %. [It
should be recognized that g may be equal to the crack slot
width, N, (for example, g = 6.4 mm (0.25 in.) in Fig. 3) or
larger than N if machined knife edges are used.] The outer
radius r2 is measured, if possible, to within 5.0 %. If not
possible, then an average value of r2 is be calculated (see Note
A6.1) from the measured (within 5.0 %) length, L, of the chord
of the outer surface, which chord passes through the loading
hole centers (see Fig. A6.2), using the following relationship:
r2 5
L2
W1X
1 2
8~W 1 X!
(A6.1)
then:
FIG. A6.2 Measurement of Outer Radius (r2) and Crack Size for
the Arc-Shaped Tension A(T) Specimen (see A6.4.1)
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
23
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
r1
W
r2 5 1 2 r2
(A6.2)
NOTE A6.1—A10 % variation in the ratio r1/r2 will affect the value of
stress-intensity factor by 1.0 % or less, providing that the relative crack
size a/W is not less than 0.3. This, however, is based on the assumption
that specimens are cut from stock of uniform, axisymmetric cross section.
If inspection shows that the stock deviates from axisymmetry by more
than 10 %, it should be reworked to within this tolerance.
A6.4.1.1 Post-test crack size measurement (in accordance
with 8.2.3) involves a special procedure due to the specimen’s
curvature. A size measurement, m, is made from a reference
point on the curved inner surface, adjacent to the crack mouth,
to a point on the crack front. That size is greater than the
corresponding distance from the virtual point of intersection
between the crack plane and the inside circumference of the
specimen (see Fig. A6.2). Error, e, is computed from the
following expression:
e 5 r1 2
Œ
g2
r21 2 4
(A6.3)
where g is the distance between the crack mouth opening
displacement measurement reference points. If the relative
error e/m < 0.01, then m is taken as the crack size; otherwise e
is subtracted from m and the result recorded as the crack size.
A6.4.2 Arc-Shaped Tension Specimen Testing—General
principles concerning the loading clevis and its setup appear in
A2.2. When assembling the load train (clevises and their
attachments to the tension machine), care shall be taken to
minimize eccentricity of loading due to misalignments external
to the clevises.
A6.4.2.1 The displacement gage is seated on the knife edges
such as to maintain registry between knife edges and gage
grooves. In the case of attachable knife edges, the gage is
seated before the knife edge positioning screws are tightened.
A6.4.2.2 The specified rate of increase of the stressintensity factor is within the range 0.55 and 2.75 MPa=m/s
(30 and 150 ksi=in./min) corresponding to a loading rate
between 0.21 and 1.04 kN/s (2.8 to 14.0 klbf/min) for a
standard (W/B = 2) 25 mm (1.0 in.) thick specimen with X/W
= 0.5, and between 0.33 and 1.67 kN/s (4.5 to 22.5 klbf/min)
for a standard (W/B = 2) 1 in. thick specimen with X/W = 0.
A6.4.2.3 Details concerning recording of the test record are
given in 8.4.
A6.5 Calculations
A6.5.1 Interpretation of Test Record—General requirements and procedures for interpreting the test record are given
in 9.1.
A6.5.2 Validity Requirements—Validity requirements in
terms of limitation on Pmax/PQ and mandatory specimen size
are given in 9.1.3 through 9.1.4.
A6.5.3 Calculation of KQ—Arc-shaped tension specimen
KQ is calculated in SI or inch-pound units of Pa=m (psi=in.)
as follows (see Note A6.3):
KQ 5
S
X
a
P
3 1 1.9 1 1.1W
B=W W
where:
DF
(A6.4)
S
a
1 1 0.25 1 2 W
DS
2
r1
12r
2
DG S D
a
·ƒ W
S D SŒD F
a
ƒ W 5
a
W
a
12W
3/2
SD
a
a
3.74 2 6.30W 1 6.32 W
2
S DG
a
2 2.43 W
3
(A6.5)
for which:
PQ
= force as determined in 9.1.1, N (lbf),
B
= specimen thickness as determined in 8.2.1, m (in.),
X
= loading hole offset as determined in A6.4.1, m (in.),
W
= specimen width (depth) as determined in A6.4.1, m
(in.),
a
= crack size as determined in 8.2.3 and A6.4.1.1, m
(in.), and
r1/r2 = ratio of inner-to-outer radii as determined in A6.4.1.
NOTE A6.2—Example: for a/W = 0.500, ƒ(a/W) = 3.73.
NOTE A6.3—The accuracy of this expression for a/W for all values of
r1/r2 is considered to be as follows: (1) within 1.0 % for 0.45 # a/W #
0.55 and X/W of 0 or 0.5, (2) within 1.5 % for 0.2 # a/W # 1 and X/W
of 0 or 0.5, and (3) within 3.0 % for 0.2 # a/W # 1 and 0 # X/W # 1 (25).
A6.5.4 Calculation of Crack Mouth Opening Compliance
Using Crack Size Measurements—Arc-shaped tension specimen crack mouth opening compliance, Vm/P, is calculated in
units of m/N (in./lb) as follows (see Note A6.5):
for the specimen with X/W = 0:
SDF S
a
Vm P 1 W
a
r1
P 5 E’ B · 0.43 1 2 r2 1 q1 W
D S DG
(A6.6)
where:
SD S
a
P1 W 5
a
11W
a
12W
D
(A6.7)
2
and:
SD
SD
a
a
a
q1 W 5 0.542 1 13.137W 2 12.316 W
2
SD
a
1 6.576 W
3
(A6.8)
or, for the specimen with X/W = 0.5:
SDF S
a
Vm P 2 W
a
r1
P 5 E’ B · 0.45 1 2 r2 1 q2 W
D S DG
(A6.9)
where:
SD S
a
P2 W 5
a
21W
a
12W
D
(A6.10)
2
and:
SD
SD
a
a
a
q2 W 5 0.399 1 12.63W 2 9.838 W
2
SD
a
1 4.66 W
3
(A6.11)
for which:
E’ = elastic constraint modulus (E for plane stress, Pa
(psi); E/(1 − n2) for plane strain, Pa (psi)),
n = Poisson’s Ratio, and
X, B, W, a, and (r1/r2) are defined in A6.5.3.
NOTE A6.4—Example: for a/W = 0.500, p1(a/W) = 6.00, q1(a/W) =
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
24
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
4.85, p2(a/W) = 10.00, and q2(a/W) = 4.84.
NOTE A6.5—These expressions are considered to be accurate within
1.4 % (X/W = 0) or 1.6 % (X/W = 0.5) for 0.2 # a/W # 0.8 and (r1/r2) $
0.4 (22). These expressions are valid only for crack mouth opening
displacement measured at the location of integral knife edges comparable
to that shown in Fig. 3. Attachable knife edges must be reversed or inset
to effect the same measurement points.
A6.5.5 Calculation of Crack Size Using Crack Mouth
Opening Compliance Measurements—Arc-shaped tension
specimen normalized crack size is calculated as follows (see
Note A6.6):
for the specimen with X/W = 0:
a
W5
(A6.12)
0.98923.463 · U20.171 · U2124.354 · U3272.805 · U4184.375 · U5
where:
U5
11
1
E’ BVm
r1
1 1 0.101 1 2 r
P
2
Œ
F
S
DG
(A6.13)
or, for the specimen with X/W = 0.5:
a
W5
(A6.14)
0.98624.082 · U25.065 · U2186.819 · U32313.338 · U41429.101 · U5
where:
U5
11
1
E’ BVm
r1
1 1 0.108 1 2 r
P
2
Œ
F
S
DG
(A6.15)
for which:
Vm = crack mouth opening displacement, m (in.),
P
= applied force, N (lbf), and
E’ is defined in A6.5.4 and B, W, a and (r1/r2) are defined in
A6.5.3.
NOTE A6.6—This expression fits the equations in A6.5.4 within 0.003W
for 0.2 # a/W # 0.8, (r1/r2) $ 0.4, and X/W = 0 or 0.5 (21). This
expression is valid only for crack mouth opening displacement measured
at the location of the integral knife edges comparable to that shown in Fig.
3. Attachable knife edges must be reversed or inset to effect the same
measurement points.
A7. SPECIAL REQUIREMENTS FOR TESTING ARC-SHAPED BEND SPECIMENS
A7.1 Specimen
A7.1.1 The standard arc-shaped bend specimen configuration (26) is a single-edge-notched and fatigue precracked ring
segment loaded in bending. The general proportions of the
standard configuration are shown in Fig. A7.1. The value of the
radius ratio r1/r2 is limited to the range 0.6 to 1.0 when the
span-to-width ratio S/W is 4, and from 0.4 to 1.0 when S/W is
3. For cylinders with radius ratios less than these limits, the
arc-shaped tension-loaded specimen or the disk-shaped specimen shall be used.
A7.1.2 The arc-shaped bend specimen measures toughness
only for a crack whose normal is circumferential and propa-
NOTE
NOTE
NOTE
NOTE
NOTE
NOTE
gation direction is radial, designated C-R (see 3.1.3). For other
crack plane orientations and propagation directions the bend
(Annex A3) or compact (Annex A4) specimen are to be used.
A7.1.3 Alternative configurations may have 2 # W/B # 4,
but with other proportions unchanged. The use of alternative
specimen proportions is advantageous when a specimen can be
extracted from a ring segment without machining the inner and
outer radii.
A7.2 Specimen Preparation
A7.2.1 Generally applicable specifications regarding specimen size, configuration and preparation are given in Section 7.
1—Surface finishes in µm.
2—A surfaces shall be perpendicular and parallel as applicable within 0.0001 W TIR.
3—Crack starter notch shall be perpendicular to specimen surfaces within 6 2°
4—Integral or attachable knife edges for clip gage attachment (see Figs. 3 and 4) shall be provided for displacement gage attachment.
5—For starter notch and fatigue crack configuration, see Fig. 5.
6—1.6 µm = 63 µin., 3.2 µm = 125 µin.
FIG. A7.1 Arc-Shaped Bend A(B) Specimen—Standard Proportions and Tolerances
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
25
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
reworked. The distance in the crack plane between the chord
that connects the two machined flat surfaces and the outer
radius is measured to the nearest 0.03 mm (0.001 in.) or to
0.1 %, whichever is greater. This measurement is made on both
sides of the specimen referencing each machined flat surface.
Specimen width W is subtracted from the average of these two
measurements and the difference recorded as the quantity Z.
The distance g between the crack mouth opening displacement
measurement reference points is measured to within 5.0 %. [It
should be recognized that g may be equal to the crack slot
width, N, (for example, g = 6.4 mm (0.25 in.) in Fig. 3) or
larger than N if machined knife edges are used.] The outer
radius r2 is measured, if possible, to within 5.0 %. If not
possible, then an average value of r2 is calculated (see Note
A6.1) from the measured (within 5.0 %) length, L, of the chord
of the outer surface (that is, the chord established by the
machined flat surfaces (see Fig. A7.2)) using the following
relationship:
r2 5
L2
W1Z
1 2
8~W 1 Z!
(A7.1)
then:
r1
W
r2 5 1 2 r2
FIG. A7.2 Measurement of Outer Radius (r2) and Crack Size for
the Arc-Shaped Bend A(B) Specimen (see A7.4.1.1)
A7.3 Apparatus
A7.3.1 Bend Test Fixture—The loading fixture for standard
bend specimen testing in Annex A3 is equally suitable for the
arc-shaped bend specimen. The fixture is designed to minimize
friction effects by allowing the rollers to rotate and translate
slightly as the specimen is loaded, thus providing rolling
contact.
A7.3.2 Displacement Gage—Details regarding displacement gage design, calibration, and use are given in 6.4. For the
standard bend specimen, displacements are essentially independent of gage length up to W/2. It is presumed that for the
cylindrical bend specimen, displacements are essentially independent of gage length up to W/2 as well.
A7.4 Procedure
A7.4.1 Measurement—Before testing, (r2 − r1) is measured
to the nearest 0.03 mm (0.001 in.) or to 0.1 %, whichever is
greater, at mid-thickness positions on both sides of, and
immediately adjacent to, the crack starter notch mouth. The
average of these two readings is taken as W. Measurement of
(r2 − r1) is made also at four additional positions, two as close
as possible to the intersection of the inside radius with the
machined flat surfaces, and two at approximately one-half the
circumferential distance between the machined flat surfaces
and the crack plane. If any of these four measurements differ
from W by more than 10 %, the specimen shall be discarded or
(A7.2)
NOTE A7.1—A10 % variation in the ratio r1/r2 will affect the value of
the stress-intensity factor by 1.2 % or less, providing that the relative
crack length a/W is not less than 0.3. This, however, is based on the
assumption that the specimen is cut from stock of uniform, axisymmetric
cross section. If inspection shows that the stock deviates from axisymmetry by more than 10 %, it should be reworked to within this tolerance.
A7.4.1.1 Post-test crack size measurement (in accordance
with 8.2.3) involves a special procedure due to the specimen’s
curvature. A size measurement, m, is made from a reference
point on the curved inner surface, adjacent to the crack mouth,
to a point on the crack front. That size is greater than the
corresponding distance from the virtual point of intersection
between the crack plane and the inside circumference of the
specimen (see Fig. A7.2). Error, e, is computed from the
following expression:
e 5 r1 2
Œ
g2
r21 2 4
(A7.3)
where g is the separation of the crack mouth opening
displacement measurement reference points. If the relative
error e/m < 0.01, then m is taken as the crack size; otherwise e
is subtracted from m and the result recorded as the crack size.
A7.4.2 Arc-Shaped Bend Specimen Testing—General principles concerning the loading fixture and its setup appear in
A2.1.
A7.4.2.1 The displacement gage is seated on the knife edges
such as to maintain registry between knife edges and gage
grooves. In the case of attachable knife edges, the gage is
seated before the knife edge positioning screws are tightened.
A7.4.2.2 The specified rate of increase of the stressintensity factor (see 8.3) ranges from 0.55 to 2.75 MPa=m/s
(30 to 150 ksi=in./min) and corresponds to a loading rate
between 0.33 and 2.37 kN/s (4.5 to 32.0 klbf/min) for the
standard (W/B = 2) 25 mm (1.0 in.) thick specimen with S =
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
26
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
3W, and between 0.24 and 1.71 kN/s (3.2 to 23.0 klbf/min) for
the standard (W/B = 2) 25 mm (1.0 in.) thick specimen with S
= 4W.
A7.4.2.3 Details concerning recording of the test record are
given in 8.4.
A7.5 Calculations
A7.5.1 Interpretation of Test Record—General requirements and procedures for interpreting the test record are given
in 9.1.
A7.5.2 Validity Requirements—Validity requirements in
terms of limitation on Pmax/PQ and mandatory specimen size
are given in 9.1.3 through 9.1.4.
A7.5.3 Calculation of KQ—Arc-shaped bend specimen KQ
is calculated in SI or inch-pound units of Pa=m (psi=in.) as
follows (see Note A7.3):
For S = 4W:
KQ 5
F S
D S DG S D
a
PQS
r1
· h1 W
3/2 1 1 1 2 r
BW
2
a
· ƒ1 W
(A7.4)
where:
SD
SD
a
a
a
h1 W 5 0.29 2 0.66W 1 0.37 W
2
(A7.5)
and:
SD F
SD
S D
S DG
a 2
a
a
0.677 1 1.078W 2 1.43 W 1 0.669 W
a
ƒ1 W 5
a 3/2
12W
3
(A7.6)
for S = 3W:
KQ 5
F S
D S DG S D
a
PQS
r1
· h2 W
3/2 1 1 1 2 r
BW
2
a
· ƒ2 W
(A7.7)
where:
SD
SD
a
a
a
h2 W 5 0.20 2 0.32W 1 0.12 W
2
(A7.8)
and:
F
SD
S D
S DG
a 2
a
a
0.644 1 1.11W 2 1.49 W 1 0.73 W
a
ƒ2 W 5
a 3/2
12W
SD
3
(A7.9)
for
PQ
B
S
W
a
r1
r2
which:
= force as determined in 9.1.1, N (lbf),
= specimen thickness as determined in 8.2.1, m (in.),
= span as determined in A7.4.2 and A2.1, m (in.),
= specimen width as determined in A7.4.1, m (in.),
= crack size as determined in 8.2.3 and A7.4.1.1, m
(in.),
= inner radius as determined in A7.4.1, m (in.), and
= outer radius as determined in A7.4.1, m (in.).
NOTE A7.2—Example: for a/W = 0.500, h1(a/W) = 0.0525, f1(a/W) =
2.66, h2(a/W) = 0.0700, and f2(a/W) = 2.60.
NOTE A7.3—These expressions are considered to be accurate to within
1.0 % for 0.2 # a/W # 1.0, 0.6 # r1/r2 # 1.0, and S = 4W; and 1.5 % for
0.2 # a/W # 1.0, 0.4 # r1/r2 # 1.0, and S = 3 (26).
A8. FATIGUE PRECRACKING KIc FRACTURE TOUGHNESS SPECIMENS
A8.1 Introduction
A8.1.1 Experience has shown that even the narrowest practical machined notch cannot simulate a natural crack well
enough to provide a satisfactory measurement of KIc. Recourse
is made to an artifice consisting of a narrow notch from which
extends a comparatively short fatigue crack, called the precrack. The dimensions of the notch and the precrack, and the
sharpness of the precrack, must meet certain conditions which
can be readily met with most engineering materials. There are,
however, some materials that are too brittle to be fatigue
cracked; they fracture at the onset of fatigue crack initiation.
These are outside the scope of this test method. An exception
is beryllium, which requires special fatigue precracking procedures that are described in Annex A9.
A8.1.2 The objective of fatigue precracking is to produce a
sharp crack which is unaffected by the precracking procedure.
In what follows, guidance is offered on the production of
satisfactory fatigue precracks. Associated requirements to ensure a valid KIc test are also given.
A8.1.3 A fatigue precrack is produced by cyclically loading
the notched specimen at a ratio of minimum-to-maximum
stress between −1 and +0.1 for a number of cycles, usually
between about 104 and 106 depending on specimen size, notch
preparation, and cyclic stress- intensity factor level. The
maximum stress-intensity factor, Kmax, during any stage of
fatigue crack growth shall not exceed 80 % of the KQ value
determined in the subsequent test if KQ is to qualify as a valid
KIc result. For the terminal stage of fatigue precracking (2.5 %
of crack size a), Kmax shall not exceed 60 % of KQ. Some
fraction of the total number of cycles required to produce the
fatigue precrack is consumed in the initiation of the crack at the
notch root; the remainder represents growth of the crack to the
specified size. If the total number of cycles is excessive, the
cause is usually an excessive number of cycles required for
initiation rather than subsequent crack growth. Crack initiation
can be hastened by: (1) increasing the acuity of the notch tip;
(2) using a chevron starter notch (see Fig. 5) in place of a
straight-through starter notch; (3) applying a static preload to
the specimen such that the notch tip is compressed in a
direction normal to the intended crack plane, but without
allowing the nominal compressive stress to exceed the compressive yield strength of the material; and (4) using a negative
fatigue stress ratio.
A8.2 Equipment
A8.2.1 The fixtures recommended for fracture testing are
also suitable for fatigue precracking at positive stress intensity
ratios. K-calibration for the specimen using the fixtures shall be
known with an error not exceeding 5.0 %. K-calibration is the
relation between the stress- intensity factor K and either the
force or some prescribed displacement and the specimen
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
27
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
dimensions (1). If different fixtures are used, the appropriate K
calibration shall be determined experimentally with those
fixtures (8). The advantage of experimental K calibration,
compared to numerical methods of analysis, is that accurate
modeling of the boundary conditions with the actual fixtures is
assured. It is important to bear in mind that if the fatigue cycle
involves reversal of force, the K calibration can be very
sensitive to the distribution of clamping forces necessary to
grip the specimen.
A8.2.2 The fatigue cracking setup shall be such that the
stress distribution is uniform through the specimen thickness;
otherwise the crack will not grow uniformly. The stress
distribution shall also be symmetrical about the plane of the
prospective crack; otherwise the crack will deviate unduly
from that plane and the test result will be significantly affected,
possibly invalidated (8). A single obvious exception to these
requirements is that of cantilever bending used only for fatigue
precracking beryllium (see A3.2.3 and A9.2.3.2).
A8.3 Specimen Requirements
A8.3.1 Fatigue precracking shall be done with the specimen
in the finally heat-treated, mechanically-worked, or
environmentally-conditioned state in which it is to be tested.
A8.3.2 The combination of starter notch and fatigue precrack shall conform to the requirements of Fig. 5. The standard
specified crack size ranges from 0.45W to 0.55W and is the
total size of the starter notch slot plus fatigue crack. To
facilitate fatigue precracking at a low level of stress intensity,
the notch root radius of a straight-across notch should be no
more than 0.08 mm (0.003 in.). The chevron notch (see Fig. 5)
root radius can be as much as 0.25 mm (0.010 in.) because of
the compound stress intensification at the point of the chevron.
Crack initiation in either specimen variety can be accelerated
by precompressing the notch tip region, as stated in A8.1.3.
A8.3.3 It is suggested that two pencil lines be marked on
each side of the specimen normal to the anticipated crack-path
surface traces. The line most distant from the notch tip shall
indicate the minimum required size of fatigue crack; the other
(at a lesser distance) the terminal part of that size equal to not
less than 2.5 % of the overall crack size of notch plus fatigue
crack; that is, 0.0125W. During the final stage of fatigue crack
extension, for at least this distance, the ratio of maximum
stress-intensity factor of the fatigue cycle to the Young’s
modulus of the material, Kmax/E, shall not exceed 0.0003 =m
(0.002 =in.). Furthermore, Kmax must not exceed 60 % of the
KQ value determined in the subsequent test if KQ is to qualify
as a valid KIc result.
A8.4 Precracking Procedure
A8.4.1 Fatigue precracking normally shall be done at room
temperature with the specimen in the finally heat-treated,
mechanically-worked, or environmentally-conditioned state in
which it is to be tested. Different fatigue precracking temperatures and intermediate thermal/mechanical/environmental
treatments between fatigue precracking and testing shall be
used only when such treatments are necessary to simulate the
conditions for a specific structural application and required
dimensions and tolerances on specimen size and shape can be
maintained.
A8.4.2 Fatigue precracking may be conducted under either
force control or displacement control provided that the appropriate K-calibration is known with requisite accuracy for the
specimen and fixture (see A8.2.1). If the force range is
maintained constant, Kmax and the K range (DK) will increase
with crack size; if the displacement range is maintained
constant, the opposite will happen. The initial value of the
maximum fatigue force or displacement shall be calculated
from the K calibration and the specimen and notch dimensions.
It is suggested that this force be selected such that the
maximum stress-intensity factor in the initial portion of the
fatigue cycle does not exceed 80 % of the estimated KIc value
of the material. Higher Kmax values may result in undesirably
high crack growth rates. The minimum is then selected so that
the stress ratio is between −1 and +0.1. The more negative the
stress ratio, the faster the fatigue precrack will be completed,
but this advantage is offset by the need for more elaborate
fixtures than are required when the stress ratio is positive.
A8.4.3 The specimen shall be accurately located in the
loading fixture and secured as required so that the boundary
conditions correspond to the applicable K calibration. Fatigue
cycling is then begun, usually with a sinusoidal waveform and
near to the highest practical frequency. There is no known
marked frequency effect on fatigue precrack formation up to at
least 100 Hz in the absence of adverse environments. The
specimen shall be carefully monitored until crack initiation is
observed on one side. If crack initiation is not observed on the
other side before appreciable growth is observed on the first,
then fatigue cycling should be stopped to try to determine the
cause and remedy for the unsymmetrical behavior. Sometimes,
simply turning the specimen end for end in relation to the
fixture will solve the problem. When the most advanced crack
trace has almost reached the first scribed line corresponding to
97.5 % of the final crack size, the maximum force or displacement, as appropriate, shall be reduced so that the terminal value
of Kmax is unlikely to exceed 60 % of the estimated minimum
value of KIc of the material, and also that the terminal value of
Kmax/E will not exceed 0.0003 =m (0.002 =in.). The minimum setting is then adjusted so that the stress ratio is between
−1 and +0.1. Fatigue cycling is then continued until the surface
traces on both sides of the specimen indicate that the overall
size of notch plus crack will meet the requirements of 7.3.2.1
and 7.3.2.2, and Fig. 5 of this test method.
A8.4.4 When fatigue cracking is conducted at temperature
T1 and testing at different temperature T2, Kmax(T1) shall not
exceed 0.6[(sys (T1)/sys (T2)] [KQ(T2) = KIc(T2)], where sys (T1) and
sys (T2) are the yield strengths at the respective temperatures T1
and T2.
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
28
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
A9. SPECIAL REQUIREMENTS FOR TESTING HOT-PRESSED BERYLLIUM
A9.1 Scope
A9.1.1 This Annex describes special requirements for determining the plane-strain fracture toughness of hot pressed
beryllium. With only few exceptions, the provisions of Test
Method E399 are applicable to the fracture toughness testing of
beryllium. However, certain modifications to specimen preparation and record analysis, as described in this Annex, arise
because of beryllium’s potential toxicity, inherent brittleness
associated with cleavage fracture, high elastic modulus,
nonlinear-elastic behavior, and very high fatigue crack growth
rates (27, 28).
bending is used as a conservative approximation of Kmax for
cantilever bending (substituting, of course, maximum fatigue
force for PQ). An approximation (32) obtained by curve-fitting
the compliance calibration data of (8) for a cantilever bend
specimen with L/W = 2, is (see Note A9.3) (in units of Pa=m
(psi=in.):
NOTE A9.1—Inhalation of dust or fumes from metallic beryllium,
beryllium oxide, or soluble beryllium compounds can result in systemic
disease. Machining and testing of beryllium require special precautions
and an industrial hygienist familiar with OSHA Standards should be
consulted before a beryllium test program is started.
(A9.2)
A9.2 Specimen Size, Configuration and Preparation
A9.2.1 Specimen Size—The thickness of hot-pressed beryllium specimens shall be 13 mm (0.50 in.) or greater to avoid
excessive nonlinearity in the elastic portion of the forceCMOD record.
A9.2.2 Specimen Configuration—Standard bend SE(B) or
compact C(T) specimens may be used. A straight-through
notch (see Fig. 5) shall be used to provide sufficient fatigue
crack extension in the required reversed loading.
A9.2.3 Specimen Preparation:
A9.2.3.1 Machining—Beryllium is easy to machine. Nonetheless, machining damage is frequently encountered and
tensile test specimens are therefore etched to remove the
damaged layer. Experience has shown, however, that such is
not required in the preparation of beryllium fracture toughness
specimens (29).
A9.2.3.2 Fatigue Cracking—Fatigue cracking is done in
reverse loading, with the compression cycle 2 to 3 times that of
the tension cycle (−3 < R < −2). Under such loading, the
fatigue crack growth rate decreases with crack extension, and
it is necessary to gradually increase the tension cycle level to
develop sufficiently long cracks. Generally, for the final 2.5 %
of crack growth, tension force exceeding 60 % of the anticipated KIc value will be required. To prevent the specimen from
breaking, values of Kmax greater than 80 % of the anticipated
KIc shall be avoided. As a guideline, KIc at room temperature
and in normal laboratory environments may be assumed to be
between 10 and 11 MPa=m (9 and 10 ksi=in.). Fatigue crack
progress is to be observed on both sides of the specimen. It has
proven helpful to use a dye solution (such as those used for
penetrant inspection) to delineate the crack since crack opening
is relatively small due to the high elastic modulus of this metal.
Fatigue cracking of compact specimens in tension-compression
loading is especially difficult. A special gripping arrangement
is described in (30). Fatigue cracking SE(B) specimens has
been successfully accomplished in cantilever bending (27, 31).
The expression in A3.5.3 for KQ applicable to three-point
Kmax 5
SD
a
PL
ƒ
BW3/2 W
(A9.1)
where:
SD
SD
a
a
a
ƒ W 5 0.326 1 30.318W 2 59.905 W
2
SD
a
1 68.889 W
3
for which:
P = maximum cyclic force, N (lbf),
L = S/2 = one-half span, m (in.), and
S, B, W, and a are as defined in A3.5.3 or A4.5.3.
NOTE A9.2—Example: for a/W = 0.500, ƒ(a/W) = 9.12.
NOTE A9.3—This expression is considered to be accurate within 5.0 %
for a/W # 0.6 (8).
A9.2.3.3 When using cantilever bending, excessive clamping forces will produce cracks at the specimen edges that will
invalidate the test.
A9.3 Testing and Record Analyses
A9.3.1 Forces and displacements will be relatively low, and
the production of a satisfactory test record will require high
gain in the clip gage circuit. It is advantageous to use a
relatively slow loading rate corresponding to about 0.18
MPa=m/s (10 ksi=in./min) in order to provide sufficient time
to unload the specimen if the recording gain controls require
adjustment to achieve the slope range specified by this test
method. When the elastic portion of the force-versus-CMOD
record is nonlinear, an initial slope is determined by drawing a
straight line between two points on the force-CMOD record;
one point at 20 % of maximum force, the other at 80 % of
maximum force.
A9.4 Precision and Bias (see also Section 11)
A9.4.1 Hot pressed beryllium from two suppliers was tested
in six laboratories in accordance with the procedures of this
Annex with the following results:
sys
–
Grand Mean, X
Standard Deviation, S
Batch 1
236 MPa (34.3 ksi)
10.7 (9.72)
Batch 2
197 MPa (28.6 ksi)
10.4 (9.50)
0.93 (0.85)
0.78 (0.71)
NOTE A9.4—Units of grand mean and standard deviation are MPa=m
(ksi=in.).
A9.4.2 The tensile elongation of beryllium depends on
temperature and strain rate, but the magnitude of such variability on KIc is not known. However, the results of an
interlaboratory program (29) did not appear influenced by
loading rates which varied from 0.20 to 2.62 MPa=m/s (11 to
143 ksi=in./min).
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
29
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
A10. SPECIAL REQUIREMENTS FOR RAPID-FORCE, PLANE-STRAIN FRACTURE TOUGHNESS KIc (t) TESTING
A10.1 Scope
A10.1.1 This Annex specifies the method for determining
plane-strain fracture toughness (KIc) of metallic materials at
loading rates exceeding those for conventional (quasi-static)
testing [that is, rates exceeding 2.75 MPa=m/s (150 ksi=in./
min)].
A10.2 Summary of Requirements
A10.2.1 The special requirements described in this Annex
for plane-strain fracture toughness testing at loading rates
exceeding those for conventional (quasi-static) plane-strain
fracture toughness tests do not apply to impact or quasi-impact
testing (free-falling or swinging masses). They apply only to
rapid loading of conventional fracture toughness specimens to
the measurement point in not less than one millisecond. Force
versus time, crack mouth opening displacement (CMOD)
versus time, and force versus CMOD curves are recorded. The
initial linear portion of the force versus CMOD record must
define PQ unambiguously. The test time and an optionally
·
calculated average stress-intensity factor rate K characterize
the rapid-force load test. The yield strength used in analysis of
the test data can be measured directly or estimated for the
loading time of the fracture test. All criteria for quasi-static KIc
determination apply equally to the rapid-force test. The rapidforce, plane-strain fracture toughness property is denoted by
KIc( ), where the time to reach the force corresponding to KQ
is indicated in milliseconds within the brackets ( ).
A10.3 Significance and Use
A10.3.1 The significance of conventional (quasi-static) KIc
applies also to rapid-force KIc (t). The plane-strain fracture
toughness of certain materials may be sensitive to the loading
rate and decreased toughness may be noted as the loading rate
increases.
A10.4 Terminology
A10.4.1 Definitions:
A10.4.1.1 The definitions given in Terminology E1823 and
Section 3 apply to this Annex.
·
A10.4.1.2 stress-intensity factor rate, K (FL-3/2 t-1)—change
in stress-intensity factor, K, per unit time.
A10.4.2 Description of Terms Specific to This Annex:
A10.4.2.1 rapid force—any force in fracture testing that
results in an average stress-intensity factor rate in excess of
2.75 MPa=m/s (150 ksi=in./min).
A10.4.2.2 rapid-force plane-strain fracture toughness, KIc
(t) (FL-3/2)—the crack extension resistance under conditions of
crack-tip plane strain at average loading rates exceeding 2.75
MPa=m/s (150 ksi=in./min). The time, t, in milliseconds to
reach PQ is indicated in the brackets ( ) following KIc.
A10.5 Apparatus
A10.5.1 Loading—Hydraulic machines with rapid-acting
servo-controlled valves are generally used. Depending on the
compliance of the loading system and the pump capacity, an
accumulator may be required.
A10.5.2 Fixtures—Fixtures used for quasi-static, planestrain fracture toughness tests are generally suitable for rapidforce tests, except rapid-force fixtures are to be fabricated from
materials unaffected by rapid loading.
A10.5.3 Force and Displacement Transducers—The transducers used for quasi-static, plane-strain fracture toughness
tests are generally suitable for rapid-force tests. However, these
transducers must have response characteristics without inertial
effects that could contaminate the force and displacement
signals.
NOTE A10.1—While not required, the resonant frequencies of these
transducers may be determined by suitably exciting them and observing
the wave characteristic on an oscilloscope. If ringing (high frequency
oscillation) is observed within the time period required to reach force PQ,
the stiffness of the transducers is to be increased or the mass reduced.
Force transducers are usually stiff and are unlikely to be problematical at
the loading time minimum of 1 ms. On the other hand, the displacement
transducer might be cause for concern depending on its design. The
cantilever beam displacement gage described in Annex A1 has been used
successfully at loading times slightly lower than 1 ms (33). The resonant
frequency of the gage when mounted on a specimen in a conventional
manner and excited by tapping is about 3300 Hz. The free-arm resonant
frequency is about 750 Hz. Other gages of the same type, but having
different dimensions, should operate satisfactorily if their free arm
resonance is at least 750 Hz. The following equation may be used to
estimate the free-arm resonant frequency of such gages:
ƒ 5 C~0.162!
Œ
b2 Eg
rl4
(A10.1)
where:
ƒ = resonant frequency, Hz,
C = dimensional constant, 0.319 for SI and 1.0 for inchpound,
b = arm thickness, m (in.),
E = elastic modulus of the arms, Pa (psi),
g = gravitational acceleration, 9.807 m/s2 (386 in./s2),
r = density of the arm material in, kg/m3 (lbm/in.3), and
l = length of the uniform-thickness section of the arm, m
(in.).
A10.5.4 Signal Conditioners—Amplification or filtering of
the transducer signals may be necessary. Such signal conditioning devices are to have frequency response from dc to at
least 20t-1 (kHz) where t is the test time in ms as defined in
A10.7.2. Conventional mechanical recording devices may not
have sufficient frequency response to permit direct plotting of
the force versus time and the displacement versus time signals.
A10.6 Procedure
A10.6.1 Loading Rate—The rate of loading is discretionary,
but the time to reach the force corresponding to KQ shall be not
less than 1 ms. A preload is permitted to eliminate ringing in
the force or displacement transducers associated with the
closing of clearances in the load train at the start of rapid
loading.
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
30
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
A10.6.2 For every test, force versus time, crack mouth
opening displacment (CMOD) versus time, and force versus
CMOD records shall be obtained. The time scale of the records
shall be accurately determined, as the time is used to characterize the test. The time-dependent records are to be examined
for the presence of ringing before reaching the PQ force.
Ringing can result from the inertial effects described in Note
A10.1. The special record analysis procedure described in
A10.7.2 may be helpful in assessing the magnitude of such
effects.
NOTE A10.2—It should be recognized that some materials may exhibit
a burst of crack extension at forces less than PQ, sufficiently abrupt to
produce ringing in the displacement transducer signal. Such an abrupt
advance of the crack may be associated with material inhomogeneities
local to the fatigue crack tip. If the ringing is severe, it may not be possible
to unambiguously determine PQ. The presence of such bursts of crack
extension should be recorded for those tests having analyzable force
versus CMOD records.
NOTE A10.3—Test data may be directly recorded if the recording
device has sufficient frequency response. Generally, it is advantageous to
use a storage device that will capture the data and permit playing it out at
a sufficiently slow speed that a pen recorder can be used to produce the
required record. Such storage devices are commonly available in the form
of digital storage oscilloscopes having pen recorder outputs. Separate
storage instruments are also available. In general, these digital storage
devices have performance characteristics that are more than adequate to
capture, store, and replay the transducer signals from a 1 ms test.
Calculations show, for example, that for a typical fracture test as described
in (33), the crack mouth opening displacement (CMOD) resolution would
be approximately 0.76 µm/sample (0.030 mils/sample) and the force
resolution would be approximately 712 N/sample (160 lbf/sample). It
should be possible to obtain at least 1000 simultaneous samples of force
and CMOD during such a test. A digital storage scope capable of at least
this performance would have the following characteristics: maximum
digitizing rate 1 MHz, maximum sensitivity 6 100 mV, resolution
0.025 %, and memory of 4096 words by 12 bits. It may be necessary to
amplify the output of the clip gage moderately, and possibly that of the
force transducer depending on its capacity in terms of the range required.
FIG. A10.1 Special Requirements for Analysis of ForceDisplacement Records (5 % Secant Line Not Shown)
The above values of resolution are based on a total noise figure of
approximately 50 µV.
A10.7 Calculation and Interpretation of Results
A10.7.1 Special requirements are placed on the analysis of
the rapid-force versus CMOD record, because experience (33)
has shown these records to be frequently not as smooth in the
linear range as those obtained from quasi-static tests. The
special requirements of this annex are designed to ensure that
an unambiguous value of PQ can be determined.
A10.7.1.1 The rapid-force versus CMOD record is illustrated in Fig. A10.1. It is analyzed as follows: Straight line OA
is constructed to best represent the initial portion of the test
record, which ideally should be linear but may not be smooth.
Line OP5 is then constructed as described in 9.1.1 (see Fig. 7)
to determine PQ. A vertical line is drawn at nP passing through
PQ. Pn is defined at the point of intersection of this line with the
line OA. Lines BC and DE are drawn parallel to OA, with BC
passing through (Pn + 0.05Pn), and DE passing through PQ
(Pn − 0.05Pn). A horizontal line is drawn at P = 0.5PQ. For the
test to be valid, the rapid-force versus CMOD curve up to PQ
must lie within the envelope described by these parallel lines
for that portion of the record with P $ 0.5PQ.
A10.7.2 Test time t in milliseconds is determined from the
record of force versus time shown schematically in Fig. A10.2.
The best straight line OA is drawn through the most nearly
linear portion of the record. Time t is represented as the span
from the intersection of this line with the time axis, to the
intersection with the time axis of a vertical line from PQ. This
time t is reported in the brackets ( ) following the KIc value. An
·
average stress intensity rate K is calculated by dividing KQ or
KIc by t, the result being expressed in MPa=m/s or ksi=in./s.
Minor errors in determining the loading time are not important
because significant changes in toughness require several orders
of magnitude change in loading rate.
A10.7.3 The 0.2 % offset tensile yield strength sys is used in
determining satisfaction of the specimen size requirements
FIG. A10.2 Determination of Test Time from Force-Time Record
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
31
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
described in 9.1.4 for test validity. If the rapid-force KQ is valid
as KIc using a quasi-static yield strength value determined at a
temperature at or above that of the rapid-force test, no further
yield strength considerations is necessary.
A10.7.3.1 If the test is invalid using the quasi-static yield
strength, it will be necessary to conduct a supplementary
tension test on the test material at the temperature and loading
time of the rapid-force toughness test, with the time to reach
the yield force in the tension test approximately equal to the
time t defined in A10.7.2.
A10.7.3.2 In the absence of rapid-force load sys values as
defined in A10.7.3.1, the dynamic yield strength sYD of certain
steels may be estimated using the following equation (34, 35):
sYD 5 sYS 1
A
2B
Tx · Log~2 · 107t!
(A10.2)
where:
sYS = 0.2 % offset room temperature quasi-static yield
strength,
t
= loading time in ms (see A10.7.2), and
= temperature of rapid-force toughness test.
Tx
Units:
For sYS in MPa, A = 1 198 860 and B = 187 MPa
For sYS in ksi, A = 174 000 and B = 27.2 ksi
For T in °F, Tx = (T + 460)
For T in °K, Tx = 1.8(T)
NOTE A10.4—The equation in A10.7.3.2 has been found useful only in
estimating the low temperature dynamic yield strength of constructional
steels having room temperature yield strengths below 483 MPa (70 ksi).
A10.8 Report
A10.8.1 The test report shall include the following additional information:
A10.8.1.1 Test time (in milliseconds) written in ( ) after KQ
or KIc.
A10.8.1.2 Method by which sYD of A10.7.3 was determined.
A10.8.1.3 Indications of ringing, before PQ is reached, in
the force versus time or displacement versus time record.
A10.9 Precision and Bias
A10.9.1 Precision—Eighteen valid values of KIc (t) at
−51°C (−60°F) have been reported (33), with sYD determined
by extrapolation of dynamic tensile yield strength values
obtained at strain rates from 0.01 s-1 to 1.0 s-1 at temperatures
from room to −40°C (−40°F). No statistical analysis of the
dynamic tensile yield strength data was made. The rapid-force,
plane-strain fracture toughness tests represented standard bend
SE(B) and compact C(T) specimens tested in three thicknesses
by seven laboratories. Not all laboratories tested all thicknesses. Statistical tests for outliers and for the differences
between means indicated that the data should be pooled.
Considering all the valid data, the grand mean X = 61.14
MPa=m (55.64 ksi=in.), the standard deviation S = 8.68
MPa=m (7.90 ksi=in.) and the coefficient of variation = 14 %
of the average.
A10.9.2 Bias—There is no accepted “standard” value for
the plane-strain fracture toughness of any material. In the
absence of such a true value, any statement concerning bias is
meaningless.
REFERENCES
(1) Brown, W. F., Jr., and Srawley, J. E., “Plane Strain Crack Toughness
Testing of High Strength Metallic Materials,” ASTM STP 410, 1966.
(2) Merkle, J. G., Wallin, K., and McCabe, D. E., “Technical Basis for an
ASTM Standard on Determining the Reference Temperature To for
Ferritic Steels in the Transition Range,” NUREG/CR-5504 (ORNL/
TM-13631), November 1998.
(3) Srawley, J. E.. “Plane Strain Fracture Toughness,” Fracture, Vol 4,
Ch. 2, p. 45–68.
(4) Wessel, E. T., “State of the Art of the WOL Specimen for KIc Fracture
Toughness Testing,” Engineering Fracture Mechanics, Vol 1, No. 1,
January 1968.
(5) Srawley, J. E., Jones, M. H., and Brown, W. F., Jr., “Determination of
Plane Strain Fracture Toughness,” Materials Research and Standards, ASTM, Vol 7, No. 6, June 1967, p. 262.
(6) Jones, M. H., and Brown, W. F., Jr. “The Influence of Crack Length
and Thickness in Plane Strain Fracture Toughness Tests,” ASTM STP
463, 1970, p. 63.
(7) Bray, G. H., “Literature Review on Use of Side Grooves in LinearElastic Fracture Toughness Testing: Supporting Document for E08.07
Ballot Item 6 (April 2006) Detailed references added Jan. 2007”.
Contact Dr. Mark James. [email protected]
(8) Fisher, D. M., and Repko, A. J., “Note on Inclination of Fatigue Cracks
in Plane Strain Fracture Toughness Test Specimens,” Materials
Research and Standards, ASTM, Vol 9, No. 4, April 1969.
(9) Heyer, R. H., and McCabe, D. E., “Evaluation of a Test Method for
Plane-Strain Fracture Toughness Using a Bend Specimen,” ASTM
STP 463, 1970, p. 22.
(10) McCabe, D. E., “Evaluation of the Compact Tension Specimen for
Determining Plane-Strain Fracture Toughness of High Strength
Materials,” Journal of Materials, Vol 7, No. 4, December 1972, p.
449.
(11) Underwood, J. H., and Kendall, D. P., “Cooperative Plane Strain
Fracture Toughness Tests with C-Shaped Specimens,” Journal of
Testing and Evaluation, Vol 6, No. 5, September 1978, p. 296.
(12) McCabe, D. E., “Evaluation of the Compact Tension Specimen for
Plane Strain Fracture Toughness of High Strength Materials,” Journal of Materials, Vol 7, No. 4, December 1972, p. 449.
(13) Orange, T. W., “Some Effects of Experimental Error in Fracture
Testing,” Fracture Analysis, ASTM STP 560, 1974, pp. 122–133.
(14) Fisher, D. M., Bubsey, R. T., and Srawley, J. E., “Design and Use of
a Displacement Gage for Crack Extension Measurements,” NASA
TN-D-3724, Nat. Aeronautics and Space Administration, 1966.
(15) Baratta, F. I. and Fett, T., “The Effect of Load and Crack Misalignment on Stress Intensity Factors for Bend Type Fracture Toughness
Specimens,” Journal of Testing and Evaluation, Vol 28, No. 2,
March 2000, pp. 96-102.
(16) Jones, M. H., Bubsey, R. T., and Brown, W. F., Jr., “Clevis Design for
Compact Tension Specimens Used in KIc Testing,” Materials
Research and Standards, ASTM, Vol 9, No. 5, May 1969.
(17) Freese, C. E. and Baratta, F. I., “Single Edge-Crack Stress Intensity
Factor Solutions,” Engineering Fracture Mechanics, Vol 73, 2006,
pp. 616-625.
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
32
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E399 – 09´2
(18) Srawley, J. E., “Wide Range Stress Intensity Factor Expressions for
ASTM E399 Standard Fracture Toughness Specimens,” International Journal of Fracture, Vol 12, June 1976, p. 475.
(19) Tada, H., Paris, P. C., and Irwin, G. R., “ The Stress Analysis of
Cracks Handbook,” Del Research Corporation, Hellertown, PA,
1973.
(20) Wu, S. X., “Crack Length Calculation Formula for Three-Point Bend
Specimens,” International Journal of Fracture, Vol 24, 1984, pp.
R33–R35.
(21) Newman, J. C., “Stress Analysis of Compact Specimens Including
the Effects of Pin Loading,” ASTM STP 560, 1974, p. 105.
(22) Kapp, J. A., “Improved Wide Range Expressions for Displacements
and Inverse Displacements for Standard Fracture Mechanics Specimens,” Journal of Testing and Evaluation, JTEVA, Vol 19, No. 1,
January 1991, pp. 45–54.
(23) Underwood, J. H., Newman, J. C., Jr., and Seeley, R. R., “A Proposed
Standard Round Compact Specimen for Plane Strain Fracture Toughness Testing,” Journal of Testing and Evaluation, Vol 8, No. 6,
November 1980, p. 308–313.
(24) Newman, J. C., Jr., “Stress Intensity Factors and Crack Opening
Displacements for Round Compact Specimens,” International Journal of Fracture, Vol 17, No. 6, December 1981, pp. 567–578.
(25) Kapp, J. A., Newman, J. C., Jr., and Underwood, J. H., “A Wide
Range Stress Intensity Factor Expression for the C-Shaped Specimen,” Journal of Testing and Evaluation, Vol 8, No. 6, November
1980, pp. 314–317.
(26) Underwood, J. H., “Proposed Standard Arc-Bend Chord-Support
Fracture Toughness Specimens and K Expressions,” Journal of
Testing and Evaluation, JTEVA, Vol 17, No. 4, July 1989, pp.
230–233.
(27) Jones, M. H., Bubsey, R. T., and Brown, W. F. Jr., “Crack Toughness
Evaluation of Hot Pressed and Forged Beryllium,” Journal of
Testing and Evaluation, JTEVA, Vol 1, No. 2, March 1973, pp.
100–109.
(28) Lemon, D. D., and Brown, W. F., Jr., “Fracture Toughness of Hot
Pressed Be,” Journal of Testing and Evaluation, JTEVA, Vol 13, No.
2, March 1985, p. 152.
(29) Conrad, H., and Sargent, G. A., “To Establish a Standard ASTM
Method for Fracture Toughness Testing of Beryllium,” NASA Grant
NSG3013, Oct. 1977, ASTM Research Report No. RR:E24-1005.
(30) Shabbits, W. O., and Logsdon, W. A., “S-200 Grade Beryllium
Fracture Toughness Properties,” Journal of Testing and Evaluation,
JTEVA, Vol 1, No. 2, March 1973, pp. 110–118.
(31) Conrad, H., Sargent, G. A., and Brown, W. F., Jr., “A Joint Fracture
Toughness Evaluation of Hot Pressed Beryllium,” Beryllium Conference, The Royal Society, London, 1977, Paper 21.
(32) Baratta, F. I., Private Communication, 21 Nov 1989. ASTM Research
Report RR:E24-1015.
(33) Shoemaker, A. K., and Seeley, R. R., “Summary Report of RoundRobin Testing by the ASTM Task Group E24.01.06 on Rapid
Loading Plane-Strain Fracture Toughness KIc Testing,” Journal of
Testing and Evaluation, JTEVA Vol 11, No. 4, July 1983 pp.
261–272.
(34) Madison, R. B., and Irwin, G. R., “Dynamic Kc Testing of Structural
Steel, ” Journal of the Structural Division, ASCE, Vol 100, No. ST
7, Proceedings paper 10653, July 1974, p. 1331.
(35) Irwin, G. R., Krafft, J. M., Paris, P., and Wells, A. A., “Basic Aspects
of Crack Growth and Fracture,” NRL Report 6598, Naval Research
Laboratory, November 1967.
(36) Petti, J. and Dodds, R. H., “Input on Side-Grooved Specimen
Discussion for E399” Department of Civil and Environmental Engineering University of Illinois at Urbana-Champaign, Oct. 24, 2003.
Contact Dr. Mark James. [email protected]
ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned
in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk
of infringement of such rights, are entirely their own responsibility.
This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM International Headquarters. Your comments will receive careful consideration at a meeting of the
responsible technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should
make your views known to the ASTM Committee on Standards, at the address shown below.
This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or [email protected] (e-mail); or through the ASTM website
(www.astm.org). Permission rights to photocopy the standard may also be secured from the ASTM website (www.astm.org/
COPYRIGHT/).
Copyright by ASTM Int'l (all rights reserved); Mon Mar 21 14:01:38 EDT 2011
33
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
Designation: E2248 – 09
Standard Test Method for
Impact Testing of Miniaturized Charpy V-Notch Specimens1
This standard is issued under the fixed designation E2248; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
1. Scope
1.1 This test method describes notched-bar impact testing of
metallic materials using Miniaturized Charpy V-Notch
(MCVN) specimens and test apparatus. It provides: (a) a
description of the apparatus, (b) requirements for inspection
and calibration, (c) safety precautions, (d) sampling, (e) dimensions and preparation of specimens, (f) testing procedures, and
(g) precision and bias.
1.2 This standard concerns Miniaturized Charpy V-Notch
specimens, for which all linear dimensions, including length
and notch depth, are reduced with respect to a type A standard
impact test specimen in accordance with Test Methods E23.
These are not the same as sub-size specimens, described in
Annex A3 of Test Methods E23, for which length, notch angle
and notch depth are the same as for the standard type A Charpy
specimen. See also 1.5 below.
1.3 Comparison of the MCVN data with conventional
Charpy V-Notch (CVN) data or application of the MCVN data,
or both, to the evaluation of ferritic material behavior is the
responsibility of the user of this test method and is not
explicitly covered by this test method.
1.4 The values stated in SI units are to be regarded as
standard. No other units of measurement are included in this
standard.
1.5 This standard does not address testing of sub-size
specimens as discussed in Test Methods E23. The reader
should understand the distinction between miniature and subsize. Miniature specimens are shorter that sub-size specimens
so that more tests can be conducted per unit volume of
material. Moreover, miniature specimens are designed so that
the stress fields which control fracture are similar to those of
conventional Test Methods E23 specimens.
1.6 The MCVN test may be performed using a typical Test
Methods E23 test machine with suitably modified anvils and
striker or using a smaller capacity machine.
1.7 This standard does not purport to address all of the
safety concerns, if any, associated with its use. It is the
responsibility of the user of this standard to establish appro1
This test method is under the jurisdiction of ASTM Committee E28 on
Mechanical Testing and is the direct responsibility of Subcommittee E28.07 on
Impact Testing.
Current edition approved April 1, 2009. Published April 2009. DOI: 10.1520/
E2248-09.
priate safety and health practices and determine the applicability of regulatory limitations prior to use.
2. Referenced Documents
2.1 ASTM Standards:2
A370 Test Methods and Definitions for Mechanical Testing
of Steel Products
E23 Test Methods for Notched Bar Impact Testing of
Metallic Materials
E177 Practice for Use of the Terms Precision and Bias in
ASTM Test Methods
E691 Practice for Conducting an Interlaboratory Study to
Determine the Precision of a Test Method
E2298 Test Method for Instrumented Impact Testing of
Metallic Materials
2.2 ISO Standards:3
ISO 148 Metallic materials -- Charpy pendulum impact test
-- Part 1: Test method
ISO 14556 Steel -- Charpy V-notch pendulum impact test -Instrumented test method
3. Summary of Test Method
3.1 The essential features of the MCVN impact test are: (a)
a suitable miniature three point bend specimen, (b) anvils and
supports on which the test specimen is placed to receive the
blow of the moving mass, (c) a moving mass (striker) that has
been released from a sufficient height to cause the mass to
break the specimen placed in its path, (d) a device for
determining the energy absorbed by the broken specimen, and
optionally (e) instrumentation for measuring applied force as a
function of time during specimen loading (refer to Test Method
E2298).
3.2 The test consists of breaking the miniaturized specimen,
notched in the middle, and supported at each end, with one
blow from a swinging pendulum under conditions defined
hereafter.
2
For referenced ASTM standards, visit the ASTM website, www.astm.org, or
contact ASTM Customer Service at [email protected]. For Annual Book of ASTM
Standards volume information, refer to the standard’s Document Summary page on
the ASTM website.
3
Available from International Organization for Standardization (ISO), 1, ch. de
la Voie-Creuse, Case postale 56, CH-1211, Geneva 20, Switzerland, http://
www.iso.org.
Copyright (C) ASTM International. 100 Barr Harbor Drive P.O. Box C-700 West Conshohocken, Pennsylvania 19428-2959, United States
Copyright by ASTM Int'l (all rights reserved); Tue Mar 22 16:46:00 EDT 2011
1
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E2248 – 09
4. Significance and Use
4.1 There are cases where it is impractical or impossible to
prepare conventional CVN specimens. MCVN specimens are
an alternative approach for characterizing notched specimen
impact behavior. Typical applications include MCVN specimens prepared from the broken halves of previously tested
specimens, from thin product form material, or from material
cut from in-service components.
4.2 This standard establishes the requirements for performing impact tests on MCVN specimens fabricated from metallic
materials. Minimum requirements are given for measurement
and recording equipment such that similar sensitivity and
comparable measurements, as compared to conventional CVN
tests, are achieved. The user should be aware that the transition
region temperature dependence data obtained from MCVN
specimens are not directly comparable to those obtained from
full-size standard Charpy-V specimens and suitable correlation
procedures have to be employed to obtain ductile-to-brittle
transition temperature (DBTT) data equivalent to those obtained using CVN specimens. In all instances, correlations will
have to be developed to relate upper shelf energy (USE) data
from MCVN test to CVN comparable energy levels. Application of MCVN test data to the evaluation of ferritic material
behavior is the responsibility of the user of this test method.
MCVN test data should not be used directly to determine the
lowest allowable operating temperature for an in-service material. The data must be interpreted within the framework of a
fracture mechanics assessment.
4.3 While this Test Method treats the use of an instrumented
striker as an option, the use of instrumentation in the impact
test is recommended and is fully described in Test Method
E2298. In order to establish the force-displacement diagram, it
is necessary to measure the impact force as a function of time
during contact of the striker with the specimen. The area under
the force-displacement curve is a measure of absorbed energy.
As an alternative, absorbed energy may be evaluated directly
from machine dial reading. Whenever possible, an optical
encoder shall be used in place of the machine dial because an
encoder has better resolution than a dial.
5. Test Machine
5.1 The test shall be carried out with a pendulum-type
impact testing machine which is (optionally) instrumented to
determine force-time curves. The test machine shall have
sufficient capacity to break the specimen in one blow while
losing not more than 80 % of the initial potential energy.
Provided energy measurements can be obtained with a resolution better than or equal to 0.1 J, the same test machine used for
CVN testing may be used to test MCVN specimens.
5.2 The MCVN specimen has to be suitably supported so
that the centerline of the specimen coincides with the center of
strike of the pendulum. If the same machine used for CVN
testing is used for MCVN specimens, refer to Appendix X3 of
E23 for changing the specimen support height by manufacturing new supports or adding shims.
5.3 The impact velocity (tangential velocity) of the pendulum at the center of the strike shall not be less than 1 nor more
than 6 m/s.
NOTE 1—Impact velocities above 4 m/s are not advisable for instrumented MCVN tests, since excessive oscillations are then superimposed
on the initial portion of the test diagram and errors in the evaluation of the
force-displacement trace may occur. For the same reason (ease of
interpretation of the instrumented curve), lower velocities are allowed for
MCVN tests than required by E23 (not less than 3 m/s).
5.4 It is recommended that the scalability of the stress fields
is maintained. This is accomplished by scaling the striker
FIG. 1 Scaled 8 mm and 2 mm Strikers for Use in Miniaturized Charpy Impact Test
Copyright by ASTM Int'l (all rights reserved); Tue Mar 22 16:46:00 EDT 2011
2
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E2248 – 09
radius, anvil radii, and the span of the anvils with respect to a
specimen size that is proportional to the CVN specimen. Fig. 1
shows the dimensions of 8 and 2 mm strikers (3.86 mm and
0.96 mm) scaled for use with the nominal 1⁄2-scale MCVN
(4.83 by 4.83 by 24.13 mm) specimen shown in Fig. 2. For
both of these scaled strikers, the anvil radius is scaled to 0.48
6 0.025 mm, and the span is 19.3 6 0.025 mm.
5.5 A non-scaled 2 mm striker can be used to test the 4 by
3 by 27 mm MCVN specimen described in Annex D of ISO
14556. The anvil radius and span, in this case are 1 0+0.50 mm
and 220+0.10 mm respectively.
NOTE 2—This particular test is allowed because a substantial amount of
data exists for this specimen and test geometry. This MCVN specimen is
not proportional to the CVN specimen, so scaling is not appropriate.
5.6 The testing machine shall be a pendulum type of rigid
construction. All general requirements for apparatus and calibration specified in Test Methods E23 shall be satisfied.
5.7 For instrumented force measurements using optional
force measuring instrumentation, the requirements given in
Test Method E2298 regarding striker instrumentation, data
acquisition, and data analysis shall be satisfied.
6. Hazards
6.1 Safety precautions should be taken to protect personnel
from electric shock, the swinging pendulum, flying broken
specimens, and hazards associated with specimen warming and
cooling media. See also 1.6.
7. Test Specimens
7.1 The recommended proportional specimen configuration
is the square cross section notched bar shown in Fig. 2. The
cross sectional dimension is slightly under 5 mm to enable
machining from a previously tested CVN. Information on
additional specimen geometries that have been successfully
used is provided in Appendix X1.
NOTE 3—In case MCVN specimens are extracted from broken CVN
specimens of highly ductile materials, the user should ensure that the
severe plastic deformation occurred during fracture of the CVN specimens
does not affect the impact behavior of the miniaturized samples.
7.2 Microstructural considerations dictate that only V-notch
specimens with cross sectional dimensions sufficient to ensure
a representative volume of material is tested may be used. In
order to satisfy this requirement, the size scale and mean
separation distance of inhomogeneities that exist in the material must be known. The cross sectional dimension must be at
least five times greater than the largest inhomogeneity. Posttest metallography may be performed in order to confirm that
the requirement has been met.
7.3 Stress field similitude dictates that if the miniaturized
specimens (such as the one shown in Fig. 2) do not satisfy the
microstructural considerations, specimens with a larger cross
section may be used. For the square cross section specimen in
Fig. 2, all the remaining specimen dimensions (length, notch
depth, etc.) shall be scaled by appropriate ratio with the
conventional CVN dimensions. This has the advantage of
standardization of approach and scalability of previously
calculated finite element solutions.
7.4 Side grooving of the MCVN specimens (see also Appendix X2) is optional. Investigations (1) have shown that the
use of side grooves on MCVN specimens provides a larger
volume of material which is sampled at plane strain conditions.
This results in less downward shift in temperature due to loss
of constraint caused by miniaturization, and thereby reduces
the need for correction factors to simulate CVN transitional
fracture temperature dependence.
7.5 The choice of specimen depends on the application.
NOTE 1—The notch root radius shall be 0.13 mm.
NOTE 2—Permissible variations shall be as follows:
notch length to edge
radius of notch
adjacent sides at
notch depth
cross section dimensions
finish requirements
90° 6 2°
60.025 mm
90° 6 10 min
60.025 mm
60.025 mm
2 [im notched surface/opposite
4 \im other surfaces
+0, −0.12 mm
60.12 mm
61°
length of specimen
centering of notch
angle of notch
FIG. 2 Miniaturized Charpy Impact Specimen
Copyright by ASTM Int'l (all rights reserved); Tue Mar 22 16:46:00 EDT 2011
3
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E2248 – 09
NOTE 4—Although this test method specifically addresses impact tests
performed on notched specimens, the use of unnotched samples may be
advantageous when testing refractory metals or materials produced by
powder metallurgy methods. For such materials, machining an accurate
notch without producing significant damage is extremely difficult. The use
of unnotched specimens, however, is outside the scope of this test method.
8. Test Procedure
8.1 The test procedure may be summarized as follows: the
test specimen is heated/cooled in situ (that is, at the impact
location) or it is removed from its cooling (or heating) medium,
and positioned on the specimen supports; the pendulum is
released with minimum vibration; and the absorbed energy is
recorded from the machine dial or, preferably, from the optical
encoder. For instrumented tests, the force-time curve is measured and evaluated to give the total absorbed energy.
8.2 The temperature of the specimen at impact must be
within 62°C of the nominal test temperature. Due to the small
size of the specimen, in tests below or above room temperature
(RT), special attention must be devoted to temperature control
within the above mentioned tolerance. It is recommended that
in-situ heating/cooling be used. If a bath transfer system is
used, it will be necessary to transfer the specimen to the
supports and strike the specimen within a very short period of
time (~ 1 s or less). If a thermal bath transfer system is not
used, dummy specimens (with internal thermocouples) or test
specimens (with surface thermocouples) shall be used to
demonstrate that the 62°C requirement has been met. If in-situ
heating/cooling is used, dummy specimens (with internal
thermocouples) or test specimens (with surface thermocouples)
shall be used to calibrate the system and to demonstrate that the
62°C requirement has been met.
8.3 The specimen shall be placed on the supports against the
anvils to ensure that the notch is centered to within 0.25 mm.
9. Lateral Expansion and Percent Shear Determination
9.1 The measurement of lateral expansion shall satisfy the
requirements of Test Methods E23. The uncertainty of the
measurement shall be determined by using precision machined
reference blocks.
9.2 The fracture appearance, characterized as percent shear
area, may be measured directly or determined using a correlation of characteristic values (see 9.2.2).
9.2.1 Direct measurement of fracture appearance shall satisfy the requirements of Test Methods E23.
9.2.2 Fracture Appearance Correlation—The equations described in Test Method E2298 may be used to estimate the
shear fracture area. These equations relate characteristic force
measurements with the percentage of shear fracture area. The
applicability and accuracy of the correlation for a particular
material shall be demonstrated.
10. Inspection, Verification, and Preparation of
Apparatus
10.1 Machine inspection and verification shall be performed
in accordance with the requirements of this test method, Test
Methods E23, and Test Method E2298 as appropriate.
10.2 In cases where the MCVN specimens are tested on a
large capacity Test Methods E23 test machine, the test machine
shall be indirectly verified using CVN verification specimens
in accordance with the requirements of Test Methods E23. In
particular, the anvils and striker for CVN specimens shall be
used to verify the test machine. MCVN anvils and striker shall
then be put on the machine and the machine shall be further
checked by testing MCVN specimens which are prepared from
a material with a microstructure that produces small scatter in
the fracture test results and/or for which a large experimental
database is available (such as round-robin results, see for
example references (1, 2, 3)).
10.3 MCVN test machines of small capacity, which are not
capable of testing CVN verification specimens, shall be
checked by testing MCVN specimens which are prepared from
a material with a microstructure that produces small scatter in
the fracture test results and/or for which a large experimental
database is available (such as round-robin results, see for
example references (1, 2, 3)). In the case of materials with a
microstructure that produces small scatter in the fracture test
results, it is not possible to compare MVCN results with known
certified values. In such cases, it is recommended that a large
batch of test specimens be prepared and used to establish the
mean and standard deviation at various energy levels. The
batch of test specimens can be used in the future to perform
yearly test machine verification and to verify the performance
of other MCVN test machines.
10.4 Prior to testing a group of specimens, and before a
specimen is placed in position to be tested, check the machine
by a free swing of the pendulum. With the dial indicator (if
used) at the maximum energy position, a free swing of the
pendulum shall indicate zero energy within at least 0.1 J on
machines reading directly in energy, and which are compensated for frictional losses. On machines using optical encoders,
the indicated values, when converted to energy, shall be
compensated for frictional losses and the free swing of the
pendulum shall indicate zero energy within 60.1 J.
10.5 For instrumented testing, the calibration and verification procedures of Test Method E2298 shall be satisfied.
11. Report
11.1 For all tests, report the following information:
11.1.1 Specimen type and dimensions,
11.1.2 Test machine characteristics including anvil spacing,
anvil radius, span, and striker geometry,
11.1.3 Test temperature of specimen and method of heating
or cooling, and
11.1.4 Energy absorbed as measured by dial or optical
encoder.
11.2 Optional variables which may be reported include:
11.2.1 Lateral expansion,
11.2.2 Fracture appearance (shear),
11.2.3 Specimen orientation, and
11.2.4 Specimen location within the plate or weld.
11.3 For instrumented tests, additional information in accordance with Test Method E2298 shall be reported.
12. Precision and Bias
12.1 Precision—MCVN impact data from two interlaboratory studies have been analyzed in accordance with Practice
E691 in order to establish the precision of this Test Method.
Copyright by ASTM Int'l (all rights reserved); Tue Mar 22 16:46:00 EDT 2011
4
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E2248 – 09
The terms repeatability limit and reproducibility limit are used
as specified in Practice E177.
12.1.1 An interlaboratory study (1) was conducted using
miniaturized Charpy V-notch specimens with square cross
section (Fig. 2) of A533B Cl.1 (tested at room temperature and
150°C) and of two reference materials produced by the
National Institute of Standards and Technology, Boulder CO
(low energy and high energy). The ILS was conducted in
accordance with Practice E691 in six laboratories, each one
obtaining up to six test results for the absorbed energy
measured by the machine dial or encoder (Table 1). See ASTM
Research Report No.E28-1039.4
12.1.2 Another interlaboratory study (2, 3) of characteristic
instrumented impact forces, displacements an d energies was
conducted using KLST miniaturized Charpy V-notch speci4
Supporting data have been filed at ASTM International Headquarters and may
be obtained by requesting Research Report RR:E28-1039.
TABLE 1 Absorbed energy (J) from MCVN specimens (2)
Material
Average
A553B RT
A533B 150°C
4340 (low en)
4340 (high en)
11.09
10.94
10.92
10.93
Repeatability Reproducibility
Repeatability Reproducibility
Standard
Standard
Limit
Limit
Deviation
Deviation
0.40
0.36
0.36
0.23
0.40
0.40
0.38
0.34
1.13
1.00
1.00
0.64
1.13
1.11
1.07
0.95
mens with 334mm rectangular cross section (see Appendix
X1) of A533B Cl.1. The ILS was conducted in accordance with
Practice E691 in thirteen laboratories with fourteen test machines, each one obtaining up to five test results for the
absorbed energy (Table 2). See ASTM Research Report No.
E28-1037.5
12.2 Bias—Bias cannot be defined for MCVN absorbed
energy. The physical simplicity of the pendulum design is
complicated by complex energy loss mechanisms within the
machine and the specimen. Therefore, there is no absolute
standard to which the measured values can be compared.
13. Keywords
13.1 fracture appearance; impact test; instrumented impact
test; lateral expansion; miniaturized Charpy test; notched
specimens; pendulum machine
5
Supporting data have been filed at ASTM International Headquarters and may
be obtained by requesting Research Report RR:E28-1037.
TABLE 2 Absorbed energy (J) from MCVN specimens of A533B
cl. 1 (2, 3)
Average
Repeatability
Standard
Deviation
Reproducibility
Standard
Deviation
Repeatability
Limit
Reproducibility
Limit
8.22
0.43
0.71
1.19
1.98
APPENDIXES
(Nonmandatory Information)
X1. ALTERNATIVE MCVN SPECIMEN CONFIGURATIONS
X1.1 MCVN specimen described in Annex D of
ISO 14556—This 4 3 3 3 27 mm3 specimen has been used
extensively in Europe (2, 3). We refer the user to Annex D of
ISO 14556 for details of the specimen geometry.
X2. SIDE GROOVED MINIATURE CHARPY V-NOTCH SPECIMENS
X2.1 Side Grooved Miniature Charpy V-notch Specimens:
X2.1.1 This test method recommends the use of a proportional specimen with a square cross section, like that shown in
Fig. 2. A square cross section side grooved specimen can also
be used.
X2.1.2 The use of the side grooves on MCVN specimens
will provide a larger volume of material which is sampled in
plane strain conditions. This results in less downward shift in
temperature due to loss of constraint as a result of miniaturization, and thereby reduces the need for correction factors to
simulate CVN transitional fracture temperature dependence.
The choice of specimen depends on the application and it is
important to note that some side grooved specimens and test
procedures associated with them are patented technologies (4).
Copyright by ASTM Int'l (all rights reserved); Tue Mar 22 16:46:00 EDT 2011
5
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
E2248 – 09
REFERENCES
(1) Manahan, M. P., Sr., Martin F. J., and Stonesifer, R. B., “Results of the
ASTM Instrumented/Miniaturized Round Robin Test Program”, Pendulum Impact Testing: A Century of Progress, ASTM STP 1380, T. A.
Siewert and M. P. Manahan, Sr.,Eds., American Society for Testing
and Materials, West Conshohocken, PA, 1999.
(2) Lucon, E., 9Round-Robin on Instrumented Impact Testing of Sub Size
Charpy-V Specimens: Results of Phase 19, ESIS TC5, Final Report, 2
April 1998.
(3) Lucon, E., 9European Activity on Instrumented Impact Testing of
Subsize Charpy V-Notch Specimens (ESIS TC5)9, Pendulum Impact
Testing: A Century of Progress, ASTM STP 1380, T. A. Siewert, and
M. P. Manahan, Sr., Eds., American Society for Testing and Materials,
West Conshohocken, PA, 1999, p. 242-252.
(4) United States Patent numbers 4,864,867 and 5,165,287, M.P. Manahan
inventor Battelle Development Corporation assignee, filed 1988.
ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned
in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk
of infringement of such rights, are entirely their own responsibility.
This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM International Headquarters. Your comments will receive careful consideration at a meeting of the
responsible technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should
make your views known to the ASTM Committee on Standards, at the address shown below.
This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or [email protected] (e-mail); or through the ASTM website
(www.astm.org). Permission rights to photocopy the standard may also be secured from the ASTM website (www.astm.org/
COPYRIGHT/).
Copyright by ASTM Int'l (all rights reserved); Tue Mar 22 16:46:00 EDT 2011
6
Downloaded/printed by
University of Washington pursuant to License Agreement. No further reproductions authorized.
8842A
Instruction Manual
3. Allow the displayed reading to settle.
4. Press the OFFSET button.
5. Remove the resistor.
6. Proceed with the desired measurement.
Example:
Measure a 1.5V source with 1 M! source impedance, correcting for input bias current.
1. Connect a 1 M! resistor between the INPUT HI and INPUT LO terminals.
2. Select the VDC function and the 2V range.
3. Allow the display to settle.
4. Press OFFSET. (This zeroes the input bias current error.)
5. Remove the 1 M! resistor.
6. Measure the voltage of the circuit under test.
Note that this procedure does not correct for circuit loading error. Also note that if input
bias current error is not corrected for, it may be added to the circuit loading error.
4-5. RESISTANCE MEASUREMENT
The 8842A allows you to measure resistance in both 2-wire and 4-wire configurations.
Each has its benefits.
4-6.
2-Wire Ohms
Two-Wire ohms measurements are simple to set up and yield good results for most
measurement conditions. Measurements are made as shown in Figure 4-3. An internal
current source (the "ohms current source") passes a known test current (Itest) through the
resistance being tested (Runknown). The 8842A measures the voltage drop across
Runknown, calculates Runknown using Ohm’s law (Runknown = Vtest/Itest), and
displays the result.
f4-03.wmf
Figure 4-3. Wire Ohms Measurement
4-4
Measurement Tutorial
RESISTANCE MEASUREMENT
4
The test current and full-scale voltage for each resistance range are shown in Table 4-1.
Since the HI INPUT test lead is positive with respect to the LO INPUT lead, these test
leads are not interchangeable when a semiconductor device is being measured.
4-7.
Correcting for Test Lead Resistance in 2-Wire Ohms
In 2-wire ohms, the resistance of the test leads can introduce error when measuring low
resistances. Typical test leads may add as much as 0.5Q to 2-wire ohms readings.
With the 8842A, it is easy to correct for this error using the OFFSET button:
1. Select the 2-wire ohms function.
2. Touch the test leads together. The 8842A should indicate the resistance of the test
leads.
3. With the test leads still touching, press the OFFSET button.The 8842A should read
0!.
4-8.
4-Wire Ohms
Four-Wire ohms measurements provide the highest accuracy for low resistance
measurements. The 4-wire configuration automatically corrects for both test lead
resistance and contact resistance. Contact resistance (the resistance between the test probe
tips and the circuit being tested) is unpredictable, and therefore cannot be reliably
corrected with a fixed offset.
Four-Wire ohms measurements are especially important when using long test leads. In a
typical automated test system, for example, the test leads could be connected through
four or five switching relays, each with 2! of resistance!
NOTE
Instability of the test lead’s resistance can cause significant error on low
ohms ranges, particularly on the 20! and 200! ranges. Therefore, only 4wire ohms measurement is permitted in the 20! range.
The 8842A makes 4-wire ohms measurements as shown in Figure 4-4. The HI and LO
INPUT leads apply a known, internal current source to the unknown resistance, just as in
2-wire ohms. (See Table 4-1.) However, the voltage drop across the unknown resistance
is measured with the SENSE leads rather than the INPUT leads. Since the current flow in
the SENSE leads is negligible, the error caused by the voltage drop across the leads is
also negligible.
4-5
8842A
Instruction Manual
f4-04.wmf
Figure 4-4. Wire Ohms Measurement
Table 4-1. Ohms Test Current
4-6
RANGE
TEST CURRENT
FULL SCALE VOLTAGE
20!
1 mA
0.02V
200!
1 mA
0.2V
2 k!
1 mA
2.0V
20 k!
100 µA
2.0V
200 k!
10 µA
2.0V
2000 k!
5 µA
10.0V
20 M!
500 nA
10.0V
Measurement Tutorial
RESISTANCE MEASUREMENT
4
NOTE
!
and
20
M
!
ranges
of
4-wire ohms, the voltage across the
In the 2 M
unknown resistance is sensed between the HI SENSE and LO INPUT
terminals. Accuracy is not affected as long as the resistance of the LO
INPUT lead is less than 10! in the 2 M! range, and less than 100! in the
20 M! range.
4-9.
Applications of the Ohms Functions
The 2-wire and 4-wire ohms functions can be used for a variety of purposes in addition to
measuring resistance, as the following applications show.
4-10. TESTING DIODES
The 2-wire ohms function can also be used to test diodes.
1. Select the 2-wire ohms function and the 2 k! range.
2. Measure the resistance of the diode. If the diode is good, when forward-biased it will
measure about 0.6 k! to 0.7 k! for silicon (0.25 k! to 0.3 k! for germanium), and
when reverse-biased it will cause the 8842A to indicate overrange. (The forwardbiased reading depends upon the range used.)
The 2 k! range is used because its 1 mA test current provides a typical operating point,
and its 2V full-scale voltage is sufficient to turn on most diodes (even two diodes in
series).
4-11. TESTING ELECTROLYTIC CAPACITORS
The 2-wire ohms function can also give a rough test of an electrolytic capacitor’s leakage
and dielectric absorption. This test works well for capacitors 0.5 µF and larger.
1. Select the 2-wire ohms function, the 2 k! range, and the medium reading rate.
2. Connect the test leads to the capacitor (with the INPUT HI lead to the + lead and the
INPUT LO lead to the - lead). The 8842A attempts to charge it to the open-circuit
voltage of the 2 k! range (about 6V).
3. Disconnect the + test lead.
4. To test for leakage, select the VDC function and the 20V range (leave the 8842A in
the medium reading rate), and measure the voltage that was stored on the capacitor
during step 2.
a. If the capacitor is good, the voltage across the capacitor will be about 6V, and
will be relatively stable.
b. If the capacitor is leaky, the voltage across the capacitor will be much less than
6V, and the voltage will be decreasing. The rate of change depends on how leaky
the capacitor is.
c. With some electrolytic capacitors, the reading will increase. This usually
indicates the capacitor is defective.
5. To test the capacitor’s dielectric absorption, briefly short the capacitor’s leads together
and then measure the voltage across the capacitor.
a. If the dielectric is good (i.e., has low dielectric absorption), the voltage across the
capacitor will be nearly zero volts.
b. If the dielectric is poor (i.e., has high dielectric absorption), the voltage across the
capacitor will be significantly above zero.
4-7
8842A
Instruction Manual
4-12. A PRECISION CURRENT SOURCE
The ohms current source (the internal current source used in the ohms functions) makes a
useful troubleshooting tool in itself. It has excellent linearity and temperature stability. Its
compliance voltage is typically 5V in the lower five ohms ranges, and 12V in the upper
two ohms ranges. The inputs are protected against accidental applications of voltage up to
300V rms.
To use the ohms current source, connect the test leads to the HI and LO INPUTS, and
select either the 2-wire or 4-wire ohms function. Press the range buttons to select any of
the current levels shown in Table 4-1.
The ohms current source can be used to troubleshoot circuits by injecting current into
selected nodes, forcing the circuits to be in a specific test state. For example, the ohms
current source can be used to set or modify the bias of amplifier circuits. The current
level can be changed simply by changing range.
The ohms current source can also be used to test mA or µA panel meters. The accuracy
of the current source is more than enough to verify panel meters, whose accuracy is
typically 1% to 5%. To test an analog panel meter, simply connect the current source
across the meter movement (as though measuring its resistance). A 1 mA meter should
show full scale when the ohms function is set on the 2 k! range. The same technique also
works with digital panel meters.
4-13. DC CURRENT MEASUREMENT
To get the best accuracy using the mA DC function, it is important to understand the
concept of burden voltage error.
When a meter is placed in series with a circuit to measure current, error can be caused by
the small voltage drop across the meter (in this case, across the protective fuses and
current shunt). This voltage drop is called the burden voltage, and it is highest for fullscale measurements. The full-scale burden voltage for the 8842A is typically less than
1V.
The burden voltage can present a significant error if the current source being measured is
unregulated (i.e., not a true current source) and if the resistance of the fuse and shunt is a
significant part of the source resistance. If burden voltage does present a significant error,
the percentage of error can be calculated and corrected for using the formulas in Figure 45.
4-8
MSDS # 09912­K
REV..# 0811-B
Replaces 10/27/2005
MATERIAL SAFETY DATA SHEET
Page 1
I. PRODUCT IDENTIFICATION
TRADE NAME: 9912­K
EMERGENCY PHONE NUMBER
:
"FOR CHEMICAL EMERGENCY"
Silver paste
MANUFACTURER:
Electro­Science Laboratories, Inc.
BY: John Moore 416 E. Church Road , King of Prussia, PA 19406
Phone: (610) 272­8000
DATE PREPARED:
Nov 18, 2008
TYPE:
Spill, Leak, Fire,Exposure or Accident
Call CHEMTREC R ­ Day or Night
(800)­424­9300
International (call collect):703­527­3887
II. HAZARDOUS INGREDIENTS
CHEMICAL NAMES
CAS NUMBERS
EXPOSURE LIMITS IN AIR
ACGIH OSHA PEL
Silver
7440-22-04
0.1
MG/M3
3-hydroxy-2,2,4-trimethylpentyl Isobutyrate
25265-77-4
NONE ESTABLISHED
0.01
OTHER (SPECIFY)
­ The specific chemical identity is withheld as a Trade Secret.
III. PHYSICAL PROPERTIES
* Physical Properties for primary solvent only.if present.
* VAPOR DENSITY (air=1):
7.45
MELTING POINT OR RANGE, °C:
N/A
* SPECIFIC GRAVITY:
0.95
* BOILING POINT OR RANGE, °C:
255
* EVAPORATION RATE (Butyl Acetate=1):
* SOLUBILITY IN WATER: Insoluble
* VAPOR PRESSURE, mm Hg:
0.002
Metallic colored paste with a mild odor.
APPEARANCE AND ODOR:
1 @ 87C
IV. FIRE AND EXPLOSION
* FLASH POINT, °C: 120
* FLAMMABLE LIMITS IN AIR, VOLUME %
* AUTOIGNITION TEMPERATURE, °C:
393
LOWER (LEL):
0.52 @149C
UPPER (LEL): 4.24 @201C
Foam, CO2, Dry Chemical.
FIRE EXTINGUISHING MATERIALS:
Wear self­contained breathing apparatus and protective clothing to prevent contact with skin and eyes. Do not use water to extinguish flame.
None known to ESL.
SPECIAL FIRE­FIGHTING PROCEDURE:
UNUSUAL FIRE AND EXPLOSION HAZARDS:
V. HEALTH HAZARD INFORMATION
SYMPTOMS OF OVEREXPOSURE:
INHALED:
H.M.I.S.
Flammability
1
Health 2
Reactivity 1
Headache.
CONTACT WITH SKIN OR EYES:
ABSORBED THROUGH SKIN:
SWALLOWED:
May cause irritation to eyes, skin & mucous membranes.
None known to ESL.
May cause nausea and vomiting, diarrhea, spasms, abdominal pain, toxic, see below
HEALTH EFFECTS OR RISKS
May cause headache, nausea, vomiting & irritation to eyes, skin and mucous membranes.
ACUTE:
CHRONIC:
Personal Protection
May cause eczema, gastroenteritis.
FIRST AID EMERGENCY PROCEDURES:
Immediately flush with water for 15 minutes; call physician.
EYE CONTACT:
SKIN CONTACT: Wash thoroughly with soap and large amounts of water.
Remove from exposure, treat symptomatically.
INHALED:
SWALLOWED:
Induce vomiting; call physician.
G
MATERIAL SAFETY DATA SHEET
MSDS # 09912­K
SUSPECTED CANCER AGENT?
NO
YES
Page 2
This product's ingredients are not found in the lists below.
FEDERAL OSHA
MEDICAL CONDITIONS AGGRAVATED BY EXPOSURE:
NTP
IARC
Allergies and Asthma.
VI. REACTIVITY DATA
STABLE
STABILITY:
UNSTABLE
Contact with eyes, skin and clothing.
CONDITIONS TO AVOID:
None foreseeable for normal use.
INCOMPATABILITY:
Carbon Monoxide, Carbon Dioxide, smokes when burned.
HAZARDOUS DECOMPOSITION PRODUCTS
(including combustion products):
Will not occur.
HAZARDOUS POLMERIZATION:
VII. SPILL, LEAK AND DISPOSAL PROCEDURES
Avoid prolonged contact with skin and prolonged breathing of vapors. Collect spills with absorbent materials. Residue may be washed with soap and water. Prevent run­off
from entering drains.
PREPARING WASTES FOR DISPOSAL: Disposal method must comply with federal, state and local regulations.
SPILL RESPONSE PROCEDURES:
VIII. SPECIAL HANDLING INFORMATION
VENTILATION AND ENGINEERING CONTROLS:
Good general ventilation. 10 air changes per hour.
When used with adequate ventilation, none should be needed.
RESPIRATORY PROTECTION:
EYE PROTECTION:
NIOSH approved safety glasses.
GLOVES:
Rubber or Latex.
OTHER CLOTHING AND EQUIPMENT:
All equipment should be well vented to prevent decomposition products and vapors
from accumulating.
WORK PRACTICES, HYGENIC Good industrial hygiene practices should be followed to prevent skin and eye contact and PRACTICES: inhalation of vapors. Wash hands thoroughly after handling and before eating and smoking.
Keep away from heat, sparks and open flames.
OTHER HANDLING AND STORAGE REQUIREMENTS:
PROTECTIVE MEASURES DURING MAINTENANCE: All of the above.
IX. OTHER REGULATORY INFORMATION
This product contains the following chemicals subject to the reporting requirements of Section 313 of the Emergency Planning
and Community Right­To­Know Act of 1986 ( 40 CFR 372).
TOXIC CHEMICAL
Silver
Transportation non-regulated
CAS NUMBERS
% CONCENTRATION (Upper Limit)
90
7440­22­04
Volatile Organic Content
TSCA Inventory and Canadian DSL: All ingredients listed unless exempt.
?
g/l
NOTICE FROM ESL, INC.
The information contained in this MSDS is believed to be accurate and represents the best information currently available to ESL. The data in this MSDS relates only to the specific product(s) designed herein and does not relate to its use in combination with any other material or process.
MSDS # 09912­K FL
REV..# 0706-NEW
Replaces
MATERIAL SAFETY DATA SHEET
Page 1
I. PRODUCT IDENTIFICATION
TRADE NAME: 9912­K FL
EMERGENCY PHONE NUMBER
:
"FOR CHEMICAL EMERGENCY"
TYPE:
Electro­Science Laboratories, Inc.
416 E. Church Road , King of Prussia, PA 19406
Phone: (610) 272­8000
DATE PREPARED:
Jun 7, 2007
Spill, Leak, Fire,Exposure or Accident
Call CHEMTREC R ­ Day or Night
(800)­424­9300
International (call collect):703­527­3887
II. HAZARDOUS INGREDIENTS
MANUFACTURER:
BY: John Moore
CHEMICAL NAMES
CAS NUMBERS
EXPOSURE LIMITS IN AIR
ACGIH OSHA PEL
Silver
7440-22-04
0.1
MG/M3
0.01
Terpineol
8000-41-7
None Established
OTHER (SPECIFY)
­ The specific chemical identity is withheld as a Trade Secret.
III. PHYSICAL PROPERTIES
* Physical Properties for primary solvent only.if present.
* VAPOR DENSITY (air=1):
6.5
MELTING POINT OR RANGE, °C:
­10
* SPECIFIC GRAVITY:
0.94
* BOILING POINT OR RANGE, °C:
214 ­ 255
* EVAPORATION RATE (Butyl Acetate=1):
* SOLUBILITY IN WATER: @ 20ÝC ­ Negligible
* VAPOR PRESSURE, mm Hg:
APPEARANCE AND ODOR:
< 1
* FLASH POINT, °C: 88 Cleveland Open Cup
* AUTOIGNITION TEMPERATURE, °C:
1.0
Metallic colored paste with a mild odor.
IV. FIRE AND EXPLOSION
* FLAMMABLE LIMITS IN AIR, VOLUME %
Not determined
LOWER (LEL):
UPPER (LEL): N/A
N/A
Water Spray, Foam, CO2, Dry Chemical.
FIRE EXTINGUISHING MATERIALS:
Wear self­contained breathing apparatus and protective clothing to prevent contact with skin and eyes.
May self­ignite at upper explosion limit conditions.
SPECIAL FIRE­FIGHTING PROCEDURE:
UNUSUAL FIRE AND EXPLOSION HAZARDS:
V. HEALTH HAZARD INFORMATION
SYMPTOMS OF OVEREXPOSURE:
INHALED:
H.M.I.S.
Flammability
2
Health 1
Reactivity 1
Personal Protection
Headache, coughing
CONTACT WITH SKIN OR EYES:
ABSORBED THROUGH SKIN:
May cause irritation to eyes, skin & mucous membranes.
Low hazard per usual industrial handling.
May cause severe gastrointestinal tract irritation. Aspiration may cause pulmonary edema ad chemical pneumonitis.
HEALTH EFFECTS OR RISKS
May cause headache, nausea, vomiting & irritation to eyes, skin and mucous membranes.
ACUTE:
SWALLOWED:
CHRONIC:
May cause eczema, gastroenteritis.
FIRST AID EMERGENCY PROCEDURES:
Immediately flush with water for 15 minutes; call physician.
EYE CONTACT:
SKIN CONTACT: Wash thoroughly with soap and large amounts of water.
Move to fresh air immediately.
INHALED:
SWALLOWED:
Do NOT induce vomiting! Call a physician; give lots of water.
G
MATERIAL SAFETY DATA SHEET
MSDS # 09912­K FL
SUSPECTED CANCER AGENT?
NO
YES
Page 2
This product's ingredients are not found in the lists below.
FEDERAL OSHA
MEDICAL CONDITIONS AGGRAVATED BY EXPOSURE:
NTP
IARC
Allergies and Asthma.
VI. REACTIVITY DATA
STABLE
STABILITY:
UNSTABLE
Contact with eyes, skin and clothing.
CONDITIONS TO AVOID:
None foreseeable for normal use.
INCOMPATABILITY:
Carbon Monoxide, Carbon Dioxide, smokes when burned.
HAZARDOUS DECOMPOSITION PRODUCTS
(including combustion products):
Will not occur.
HAZARDOUS POLMERIZATION:
VII. SPILL, LEAK AND DISPOSAL PROCEDURES
Avoid prolonged contact with skin and prolonged breathing of vapors. Collect spills with absorbent materials. Residue may be washed with soap and water. Prevent run­off
from entering drains.
PREPARING WASTES FOR DISPOSAL: Disposal method must comply with federal, state and local regulations.
SPILL RESPONSE PROCEDURES:
VIII. SPECIAL HANDLING INFORMATION
VENTILATION AND ENGINEERING CONTROLS:
Good general ventilation. 10 air changes per hour.
When used with adequate ventilation, none should be needed.
RESPIRATORY PROTECTION:
EYE PROTECTION:
NIOSH approved safety glasses.
GLOVES:
Rubber or Latex.
OTHER CLOTHING AND EQUIPMENT:
All equipment should be well vented to prevent decomposition products and vapors
from accumulating.
WORK PRACTICES, HYGENIC Good industrial hygiene practices should be followed to prevent skin and eye contact and PRACTICES: inhalation of vapors. Wash hands thoroughly after handling and before eating and smoking.
Keep away from heat, sparks and open flames.
OTHER HANDLING AND STORAGE REQUIREMENTS:
PROTECTIVE MEASURES DURING MAINTENANCE: All of the above.
IX. OTHER REGULATORY INFORMATION
This product contains the following chemicals subject to the reporting requirements of Section 313 of the Emergency Planning
and Community Right­To­Know Act of 1986 ( 40 CFR 372).
TOXIC CHEMICAL
Silver
Transportation non-regulated
CAS NUMBERS
% CONCENTRATION (Upper Limit)
90
7440­22­04
Volatile Organic Content
TSCA Inventory and Canadian DSL: All ingredients listed unless exempt.
?
g/l
NOTICE FROM ESL, INC.
The information contained in this MSDS is believed to be accurate and represents the best information currently available to ESL. The data in this MSDS relates only to the specific product(s) designed herein and does not relate to its use in combination with any other material or process.
MSDS # 02602-310A
REV..# 0510-A
Replaces 9/25/2003
MATERIAL SAFETY DATA SHEET
Page 1
I. PRODUCT IDENTIFICATION
TRADE NAME: R-310-A, R-311-A, R-312-A
EMERGENCY PHONE NUMBER
:
"FOR
CHEMICAL EMERGENCY"
Resistor
TYPE:
Electro-Science Laboratories, Inc.
416 E. Church Road , King of Prussia, PA 19406
Phone: (610) 272-8000
DATE PREPARED:
10/27/05
Spill, Leak, Fire,Exposure or Accident
Call CHEMTREC R - Day or Night
(800)-424-9300
International
(call collect):703-527-3887
II. HAZARDOUS INGREDIENTS
MANUFACTURER:
BY: Robert Fox
CHEMICAL NAMES
CAS NUMBERS
EXPOSURE LIMITS IN AIR
ACGIH OSHA PEL
Glass containing lead
7439-92-1
0.05
MG/M3
0.05
Silver
7440-22-04
0.1
MG/M3
0.01
Cadmium compound in glass
7440-43-9
0.05
MG/M3
0.005
3-hydroxy-2,2,4-trimethylpentyl Isobutyrate
25265-77-4
NONE ESTABLISHED
OTHER (SPECIFY)
- The specific chemical
identity is withheld as a
Trade Secret.
III. PHYSICAL PROPERTIES
* Physical Properties for primary solvent only.if present.
* VAPOR DENSITY (air=1):
* SPECIFIC GRAVITY:
7.45
0.95
N/A
* BOILING POINT OR RANGE, °C:
255
* EVAPORATION RATE (Butyl Acetate=1):
* SOLUBILITY IN WATER: Insoluble
* VAPOR PRESSURE, mm Hg:
MELTING POINT OR RANGE, °C:
0.002
Paste with a mild odor.
APPEARANCE AND ODOR:
1 @ 87C
IV. FIRE AND EXPLOSION
* FLASH POINT, °C: 120
* FLAMMABLE LIMITS IN AIR, VOLUME %
* AUTOIGNITION TEMPERATURE, °C:
393
LOWER (LEL):
0.52 @149C
UPPER (LEL): 4.24 @201C
Foam, CO2, Dry Chemical.
FIRE EXTINGUISHING MATERIALS:
SPECIAL FIRE­FIGHTING PROCEDURE:
UNUSUAL FIRE AND EXPLOSION HAZARDS:
Wear self-contained breathing apparatus and protective clothing to prevent
contact with skin and eyes. Do not use water to extinguish flame.
None known to ESL.
V. HEALTH HAZARD INFORMATION
SYMPTOMS OF OVEREXPOSURE:
INHALED:
Flammability
1
Health 2
Reactivity 1
Personal Protection
G
Headache and irritation.
CONTACT WITH SKIN OR EYES:
ABSORBED THROUGH SKIN:
SWALLOWED:
H.M.I.S.
May cause irritation and discoloration of eyes, skin & mucous membranes.
None known to ESL.
May cause nausea and vomiting, diarrhea, spasms, abdominal pain, toxic, see below
HEALTH EFFECTS OR RISKS
May cause headache, nausea, vomiting & irritation to eyes, skin and mucous membranes.
ACUTE:
May cause eczema, gastroenteritis. Lead may damage liver, kidney, blood, nervous and reproductive systems and is
a suspected carcinogen.
FIRST AID EMERGENCY PROCEDURES:
Immediately flush with water for 15 minutes; call physician.
EYE CONTACT:
SKIN CONTACT: Wash thoroughly with soap and large amounts of water.
Move to fresh air immediately.
INHALED:
SWALLOWED:
Induce vomiting; call physician.
CHRONIC:
MATERIAL SAFETY DATA SHEET
MSDS # 02602-310A
SUSPECTED CANCER AGENT?
NO
YES
Page 2
This product's ingredients are not found in the lists below.
FEDERAL OSHA
MEDICAL CONDITIONS AGGRAVATED BY EXPOSURE:
NTP
IARC
Allergies and Asthma.
VI. REACTIVITY DATA
STABLE
STABILITY:
UNSTABLE
Contact with eyes, skin and clothing.
CONDITIONS TO AVOID:
None foreseeable for normal use.
INCOMPATABILITY:
Carbon Monoxide, Carbon Dioxide, smokes when burned.
HAZARDOUS DECOMPOSITION PRODUCTS
(including combustion products):
Will not occur.
HAZARDOUS POLMERIZATION:
VII. SPILL, LEAK AND DISPOSAL PROCEDURES
Avoid prolonged contact with skin and prolonged breathing of vapors. Collect spills
with absorbent materials. Residue may be washed with soap and water. Prevent run-off
from entering drains.
PREPARING WASTES FOR DISPOSAL: Disposal method must comply with federal, state and local regulations.
SPILL RESPONSE PROCEDURES:
VIII. SPECIAL HANDLING INFORMATION
VENTILATION AND ENGINEERING CONTROLS:
Good general ventilation. 10 air changes per hour.
When used with adequate ventilation, none should be needed.
RESPIRATORY PROTECTION:
EYE PROTECTION:
NIOSH approved safety glasses.
GLOVES:
Rubber or Latex.
OTHER CLOTHING AND EQUIPMENT:
All equipment should be well vented to prevent decomposition products and vapors
from accumulating.
WORK PRACTICES, HYGENIC Good industrial hygiene practices should be followed to prevent skin and eye contact and
PRACTICES: inhalation of vapors. Wash hands thoroughly after handling and before eating and smoking.
Keep away from heat, sparks and open flames.
OTHER HANDLING AND STORAGE REQUIREMENTS:
PROTECTIVE MEASURES DURING MAINTENANCE: All of the above.
IX. OTHER REGULATORY INFORMATION
This product contains the following chemicals subject to the reporting requirements of Section 313 of the Emergency Planning
and Community Right-To-Know Act of 1986 ( 40 CFR 372).
TOXIC CHEMICAL
Glass containing Lead
Transition metal compounds
silver
Cadmium compound
Transportation non-regulated
CAS NUMBERS
% CONCENTRATION (Upper Limit)
35
10
35
2
7440-22-04
Volatile Organic Content
TSCA Inventory and Canadian DSL: All ingredients listed unless exempt.
?
g/l
NOTICE FROM ESL, INC.
The information contained in this MSDS is believed to be accurate and represents the best information currently available to ESL. The data in this MSDS relates only to the
specific product(s) designed herein and does not relate to its use in combination with any other material or process.
ESL ELECTROSCIENCE
CERAMIC TAPES &
THICK-FILM MATERIALS
416 EAST CHURCH ROAD
KING OF PRUSSIA, PA 19406-2625, U.S.A
T: 610-272-8000
F: 610-272-6759
www.electroscience.com
9912-K
CERMET SILVER CONDUCTOR
Lead, Cadmium and Nickel-Free*
ESL 9912-K is a silver conductor having a wide range of applications, for example chip
resistors, consumer hybrids, potentiometers and heaters. Due to the wide firing temperature
range, this conductor may be processed onto a variety of substrates including glass,
Porcelain Enamelled Steel (PES), alumina and special ceramics. Additionally, 9912-K also
exhibits excellent gold wire bondability.
PASTE DATA
Rheology:
Thixotropic, screen-printable paste
Viscosity:
(Brookfield RVT, 10rpm,
ABZ Spindle, 25.5 ± 0.5 °C)
200 ± 25 Pa.s
Bonding Mechanism:
Mixed-bonded
Shelf Life (20 - 25 °C):
6 months
PROCESSING
Screen Mesh, Emulsion:
325 S/S, 25 µm
Levelling Time (at 20°C):
5 - 10 min
Drying Time (at 125°C):
Firing Temperature Range:
10 - 15 min
On alumina/beryllia/ceramics:
850 On Porcelain Enamelled Steel (PES):
Optimum (alumina):
Optimum (beryllia):
Time at peak:
930°C in air
625°C in air
850°C in air
930°C in air
10 min
Total Firing Cycle:
1 hour
Substrate for Calibration:
96% alumina
Thinner:
ESL 401
(Note: furnace air must be clean, dry and oil-free)
ESL Europe 9912-K 0601-E
ESL Affiliates
ESL Europe (Agmet Ltd) • 8 Commercial Road • Reading • Berkshire • England • RG2 0QZ • Tel: +44 (0) 118 918 2400 • Fax: +44 (0) 118 986 7331 • [email protected]
th
ESL Nippon • Sukegawa Bldg. • 6 floor • 3-4 Yanagibashi 1-chome • Taito-ku • Tokyo 111, Japan • Tel: +81-3-3864-8521 • Fax: +81-3-3864-9270 • [email protected]
ESL China • Room #1707, Tower A, City Center of Shanghai • 100 Zunyi Road • Shanghai, China 200051 • Tel: +86-21-6237-0336 and 0337 • Fax: +86-21-6237-0338
[email protected]
See Caution and Disclaimer on other side.
TYPICAL PROPERTIES
(measurement on alumina after firing at 850°C)
Fired Thickness:
(measured on a 2 mm x 2 mm pad on 96% alumina)
Approximate Coverage:
Resistivity:
(measured on a 100 mm x 0.25 mm
conductor track at 12.5 µm fired thickness)
Printing Resolution:
(line/space)
11.5 ± 2.5 µm
100 - 125 cm²/g
< 2.5 mΩ/
0.200 mm / 0.200 mm
Adhesion:
(90° pull, 2 mm x 2 mm pads, 62Sn/36Pb/2Ag)
Initial pull strength:
(on most ceramic substrates)
Aged 48 hours at 150°C:
> 7.0 kg
> 4.5 kg
Thermosonic Au Wire Bond:
(25 µm wire; bond length 1 mm;
100% wire breaks; on alumina; 850°C firing)
>8g
Aged Au Wire Bond:
(24 hours at 200°C)
>7g
ESL Europe 9912-K 0601-E
*Complies with RoHS, ELV, WEEE and CHIP 3 EC directives.
CAUTION: Proper industrial safety precautions should be exercised in using these products. Use with adequate ventilation. Avoid prolonged contact with skin or inhalation of any vapours
emitted during use or heating of these compositions. The use of safety eye goggles, gloves or hand protection creams is recommended. Wash hands or skin thoroughly with soap and water
after using these products. Do not eat or smoke in areas where these materials are used. Refer to appropriate MSDS sheet.
DISCLAIMER: The product information and recommendations contained herein are based on data obtained by tests we believe to be accurate, but the accuracy and completeness thereof is
not guaranteed. No warranty is expressed or implied regarding the accuracy of these data, the results obtained from the use hereof, or that any such use will not infringe any patent.
ElectroScience assumes no liability for any injury, loss, or damage, direct or consequential, arising out of its use by others. This information is furnished upon the condition that the person
receiving it shall make his own tests to determine the suitability thereof for his particular use, before using it. User assumes all risk and liability whatsoever in connection with his intended use.
ElectroScience’s only obligation shall be to replace such quantity of the product proved defective.
Electro-Science Laboratories, Inc.
416 East Church Road • King of Prussia, PA 19406-2625, U.S.A
610-272-8000 • Fax: 610-272-6759 • www.ElectroScience.com • [email protected]
R-300-A/B Resistor Series
Tolerant to Processing and Design Variations
Excellent Printing Characteristics
Low TCR’s
High Performance
Low Cost
The ESL R-300-A and R-300-B Resistor Series are economical, high performance
materials for the manufacture of hybrid circuits and resistors networks. Features of the
ESL R-300-A and R-300-B Series include excellent printability and low sensitivity to
processing conditions.
The dependence of resistance and TCR on blending follows the usual curves for
resistor materials. Adjacent members of the Series can be blended. The R-300-A
Series members can not be blended with members of the R-300-B Series.
The resistors are calibrated with ESL 9693-S PdAg conductor terminations. Other
silver-based and gold-based conductors can be used; however, TCR and resistivity
shifts may be observed.
R-300-A/B 9910-F
ESL Affiliates
Japan: ESL-Nippon Company, Ltd. • Sukegawa Bldg. • 6 floor • 3-4 Yanagibashi 1-chome • Taito-ku • Tokyo 111, Japan • Tel: (011-81)-3-3864-8521 • Fax: (011-81)-3-3864-9270
[email protected]
th
China: Shanghai Agmet Electro-Science Laboratory Ltd. • Second Floor Bldg. 12A1 • #223 North Fe Te Road • Waigaoqiao Free Trade Zone • Shanghai, China
Tel: (011-86)-21-5866-0497 • Fax: (011-86)-21-5866-0497 • [email protected]
Europe: Agmet, Ltd. • 8 Commercial Road • Reading, Berkshire, England RG2 0QZ • Tel: (011-44)-118-987-3139 • Fax: (011-44)-118-986-7331 • [email protected]
See Caution and Disclaimer on other side.
Typical 850°C Firing Profile
R-300-A/B
Temperature (°C)
1000
900
800
700
600
500
400
300
200
Belt Direction
Exit
100
0
0
5
10
15
20
25
30
35
40
Time (minutes)
EFFECT OF OVERGLAZING ON R-300-A/B RESISTORS
R-316-B
R-315-B
R-314-B
R-314-A
R-313-A
R-312-A
R-311-A
R-310-A
RESISTANCE CHANGE AFTER OVERGLAZING WITH G-471
0
Resistance Change(%)
-1
-2
-3
-4
-5
-6
-7
-8
o
Fired at 500 C, 30 minute cycle, 2 minutes
at peak temperature
-9
-10
R-300-A/B 9910-F
CAUTION: Proper industrial safety precautions should be exercised in using these products. Use with adequate ventilation. Avoid prolonged contact with skin or inhalation of any vapors emitted during use or
heating of these compositions. The use of safety eye goggles, gloves or hand protection creams is recommended. Wash hands or skin thoroughly with soap and water after using these products. Do not eat or
smoke in areas where these materials are used. Refer to appropriate MSDS sheet.
DISCLAIMER: The product information and recommendations contained herein are based on data obtained by tests we believe to be accurate, but the accuracy and completeness thereof is not guaranteed. No
warranty is expressed or implied regarding the accuracy of these data, the results obtained from the use hereof, or that any such use will not infringe any patent. Electro-Science assumes no liability for any injury,
loss, or damage, direct or consequential arising out of its use by others. This information is furnished upon the condition that the person receiving it shall make their own tests to determine the suitability thereof for
their particular use, before using it. User assumes all risk and liability whatsoever in connection with their intended use. Electro-Science’s only obligation shall be to replace such quantity of the product proved
defective.
<8
COEFFICIENT OF VARIATION (%)
436
NA
NA
MAX RATED POWERf (mW/mm2)
QUAN-TECH NOISE (dB)
LASER TRIM (%∆R)
(1000 hours at 150°C)
≤ 0.3
NA
871
2.95
7.38
≤ 0.3
≤ -10
300
54.8
137
0 ± 50
ESL 401
<7
± 10
10 k
R-314-A
ESL 9693-S
≤ 0.3
≤ -10
≤ -10
≤ 0.3
944
30.7
76.8
0 ± 50
968
9.84
24.6
0 ± 100
<8
± 10
1k
R-313-A
≤ 0.3
≤2
360
60
150
0 ± 100
<5
± 10
10 k
R-314-B
R-300-A/B 9910-F
b
3
CALIBRATION: Resistor size used for tests; A—1.25 mm square; B—1.0 mm square at dried thickness shown.
VISCOSITY: Brookfield RVT, ABZ Spindle, 10 rpm, 25.5°C±0.5°C.
c
CTCR: -55°C to +25°C. HTCR: +25°C to +125°C.
d
STOL: Voltage required, 5 second duration, to induce a resistance change of ±0.1% at 25°C. Resistor size as in 1.
e
STANDARD WORKING VOLTAGE: 0.4 x STOL Voltage.
f
MAXIMUM RATED POWER: (Standard Working Voltage)2/Resistance.
a
The R-314-B is used as a blending member with R-315-B. For use as a 10 kΩ/sq. resistor, R-314-A is recommended.
TERMINATION OF CALIBRATION
0.66
STD. WORKING VOLTAGE (V/mm)
1.65
STOLd (V/mm)
e
50 ± 100
TCR (ppm/°C)
c
THINNER
22.5 ± 2.5
DRIED THICKNESS (µm)
<8
225 ± 25
0 ± 100
<8
± 10
100
R-312-A
VISCOSITY (Pa•s)
b
± 30
SHIPPING SPECIFICATION (%)
± 10
10
1
RESISTIVITY (Ω/square)
a
R-311-A
R-310-A
PROPERTIES
TYPICAL RESISTOR PROPERTIES
R-300-A/B RESISTOR SERIES
≤ 0.4
≤5
190
140
350
0 ± 100
20.0 ± 2.0
300 ± 50
<5
± 10
100 k
R-315-B
≤ 0.5
NA
17
130
330
0 ± 100
<8
± 10
1M
R-316-B
4
4
5
5
0.8
1.0
1.2
1.4
1.6
0.4
R-300-A/B 9910-F
0.4
2
3
Resistor Length (mm)
R-314-A
2
3
Resistor Length (mm)
0.4
1
1
0.6
0.8
1.0
1.2
1.4
1.6
0.6
0
0
R-310-A
0.6
0.8
1.0
1.2
1.4
1.6
0.4
0.6
0.8
1.0
1.2
1.4
1.6
Relative Resistivity
Relative Resistivity
0
0
1
1
2
3
Resistor Length (mm)
R-314-B
2
3
Resistor Length (mm)
R-311-A
4
4
5
5
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0.4
0.6
0.8
1.0
1.2
1.4
1.6
4
0
0
1
1
2
3
Resistor Length (mm)
R-315-B
2
3
Resistor Length (mm)
R-312-A
RESISTANCE VERSUS LENGTH
R-300-A/B Resistor Series
Relative Resistivity
Relative Resistivity
Relative Resistivity
Relative Resistivity
4
4
5
5
Relative Resistivity
Relative Resistivity
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0
0
1
1
2
3
Resistor Length (mm)
R-316-B
2
3
Resistor Length (mm)
R-313-A
4
4
5
5
R-300-A/B Resistor Series
TCR VERSUS PEAK FIRING TEMPERATURE
(10 minutes at peak temperature)
R-310-A
R-311-A
175
R-312-A
175
175
Cold TCR
125
125
75
75
Hot TCR
125
Hot TCR
75
Hot TCR
25
25
25
Cold TCR
-25
-25
-25
-75
-75
-75
-125
-125
-125
Cold TCR
-175
825
-175
850
Peak Firing Temperature
875
-175
825
R-313-A
850
Peak Firing Temperature
875
825
R-314-A
175
R-314-B
175
125
175
125
125
Hot TCR
Hot TCR
75
75
75
25
25
25
-25
-25
-25
-75
-75
-75
-125
-125
-125
825
-175
850
Peak Firing Temperature (°C)
Hot TCR
Cold TCR
Cold TCR
Cold TCR
-175
875
-175
825
850
Peak Firing Temperature (°C)
R-315-B
875
825
850
Peak Firing Temperature (°C)
R-316-B
175
175
125
125
75
Hot TCR
25
75
25
-25
Hot TCR
-25
Cold TCR
-75
-75
-125
-125
-175
825
R-300-A/B 9910-F
Cold TCR
-175
850
Peak Firing Temperature
875
850
Peak Firing Temperature
875
825
5
850
Peak Firing Temperature (°C)
875
875
R-300-A/B Resistor Series
TCR VERSUS TIME AT PEAK FIRING TEMPERATURE (850°°C)
R-310-A
R-311-A
R-312-A
350
350
150
300
300
100
250
250
Hot TCR
50
200
200
0
150
150
Cold TCR
100
Hot TCR
Cold TCR
-50
100
Hot TCR
50
-100
50
Cold TCR
0
0
8
10
12
Time at Peak Firing Temperature (minutes)
-150
8
10
12
Time at Peak Firing Temperature (minutes)
R-313-A
8
10
12
Time at Peak Firing Temperature (minutes)
R-314-A
R-314-B
150
150
150
100
100
100
50
50
Hot TCR
Hot TCR
50
Hot TCR
0
0
0
Cold TCR
-50
-50
-50
Cold TCR
-100
-100
-150
12
8
10
Time at Peak Firing Temperature (minutes)
-150
Cold TCR
-150
8
10
12
Time at Peak Firing Temperature (minutes)
R-315-B
8
10
12
Time at Peak Firing Temperature (minutes)
R-316-B
150
150
100
100
50
-100
Hot TCR
50
0
0
-50
-50
Hot TCR
Cold TCR
-100
-100
-150
-150
12
8
10
Time at Peak Firing Temperature (minutes)
8
10
12
Time at Peak Firing Temperature (minutes)
R-300-A/B 9910-F
Cold TCR
6
R-300-A/B Resistor Series
R-300-A SERIES - RELATIVE RESISTIVITY vs. PEAK FIRING
TEMPERATURE
1.60
Relative Resistivity
R-314-A
1.40
R-313-A
R-312-A
R-311-A
1.20
R-310-A
1.00
0.80
0.60
820
830
840
850
860
870
880
Peak Firing Temperature (°C)
R-300-B SERIES - RELATIVE RESISTIVITY vs. PEAK
FIRING TEMPERATURE
1.60
R-316-B
Relative Resistivity
1.40
R-315-B
R-314-B
1.20
1.00
0.80
0.60
0.40
820
830
840
850
860
870
880
Peak Firing Temperature (°C)
Time at peak temperature = 10 minutes.
Relative Resistivity = resistivity at specified peak firing temperature/resistivity at 850°C.
R-300-A/B 9910-F
7
R-300-A/B Resistor Series
R-300-A - RELATIVE RESISTIVITY vs. TIME AT PEAK
TEMPERATURE
R-314-A
1.20
R-313-A
Relative Resistivity
R-312-A
1.10
R-311-A
R-310-A
1.00
0.90
0.80
6
8
10
12
14
Time at Peak Temperature (minutes)
R-300-B - RELATIVE RESISTIVITY vs. TIME AT PEAK
TEMPERATURE
Relative Resistivity
1.20
R-316-B
1.10
R-315-B
R-314-B
1.00
0.90
0.80
6
8
10
12
14
Time at Peak Temperature (minutes)
Peak temperature = 850°C
Relative Resistivity = resistivity at specified time at peak temperature/resistivity at 10 minutes.
R-300-A/B 9910-F
8
1987 Gordon and Breach Science Publishers, Inc.
Printed in Great Britain
Active and Passive Elec. Comp., 1987, Vol. 12, pp. 231-237
Photocopying permitted by license only
STABILITY AND PERFORMANCE
CHARACTERIZATION OF THICK FILM
MICRORESISTORS
SHOBHA C. RAVI
Roa& Kanata, Canada
Epitek Electronics, 100 Schnieder
In recent years, interest in the performance of small dimension resistors has increased primarily due to the need
to reduce parasitic resistance in high frequency applications. This paper presents results on the characterization
of thick film microresistors, i.e. resistors of 80 80 mil to 10 x 10 mil dimension.
The thickness, sheet resistance and temperature coefficient of resistance were dependent on the resistor
length, whereas, thermal ageing drift and the thermal cycling drift values did not exhibit any such dependence.
With reasonable precautions and optimization of the manufacturing conditions highly stable and good
quality microresistors can be fabricated.
1.
INTRODUCTION
The general trend towards hybrid miniaturization, the increasing customer’s demand for
high density circuits and the need to reduce parasitic resistance in high frequency circuits
motivated this investigation on characterization of thick film microresistors. These have
reduced dimensions compared with those generally used in hybrids and that are
recommended by the ink vendor. It is known that the resistor performance is effected by its
dimension.
The purpose of this study was not to introduce any major changes to our thick film
manufacturing line, nor to investigate any particular ink source or any new material. The
properties of as-fired resistors, terminated on gold, have been measured as a function of
resistor length and resistance value. The post-trim drift, temperature coefficient resistance
and stability have been considered.
To the author’s knowledge this is one of the first investigations on performance of thick
film microresistors having dimensions smaller than 40 40 mils. Bellardo et.al and Naguib 2
have reported on resistors of larger dimensions, i.e. 50 mils.
2.
EXPERIMENTAL METHOD
The inks used in this work were DuPont 1711 (10 ohm/sq) and the DuPont 1600 series (100
ohm/sq to 1 Mohm/sq). All resistors were terminated with gold conductors (4119) and
overglazed with 3563. Ceramic substrates of 96% alumina and 2 in. 2 in. were used.
The test pattern, as illustrated in figure 1, consisted of 20 parts (specimens) having nine
resistors each. The resistors were of equal aspect ratio (AR 1.0) but of different lengths,
i.e. from 80 x 80 mils to 10 x 10 mils. The specimens were prepared under the normal thick
film production processing conditions; conductors and resistors were printed with a 325 mesh
screen to a fired average thickness of 20-25 m. Prior to the final specimen preparation
initial work was done to establish the optimum fabrication parameters. All resistors were
laser trimmed and encapsulated before subjecting them to stability tests. The resistors were
trimmed to approximately 2.3 times the as-fired value with conventional plunge cuts. Laser
trimming conditions were those normally used on our production line.
231
SHOBHA C. RAVI
232
FIGURE 1. Test pattern: (Scale in inches)
80
50
30
R1
R4
R7
80 mils; R2
50 mils; R5
30 mils; R8
70
40
20
70 mils; R3
40 mils; R6
20 mils; R9
60
30
10
60 mils;
30 mils;
10 mils.
RESULTS AND DISCUSSION
3.
The post-trim drift at room temperature was negligible (average (/kR/R)% < 0.1) over a
two month period. The pre-trimmed average resistance values are tabulated in Table I. It is
observed that there is a strong dependence of resistance on the resistor length. The sheet
resistance is constant down to 50 mils and decreases with further decrease in length (Figure
2). This is due to the larger thickness of the smaller dimension resistors. Analysis of
thickness profiles of resistors is illustrated in Figure 3. The results indicate that films of length
Pre-trimmed
TABLE
Average Resistance vs. Resistor Length
Average Resistance (ohm)
length
(mil)
1711
1621
1631
1641
1651
1660
80
14.04
155.63
1410
10640
83750
1912000
60
13.60
150.62
1340
10270
81920
1857000
30
12.35
136.46
1180
9110
73350
1524000
20
11.26
116.52
1080
8770
65150
1268000
10
8.61
93.3
790
7010
50420
888000
CHARACTERIZATION OF THICK FILM MICRORESISTORS
233
160----
120
80
40
1621
0
80
60
30
20
10
Length (mils)
FIGURE 2. Dcpcndcnce of shcet resistance on resistor length
less than 50 mils are thicker. The thickness is constant down to 50 mils and increases with
further decrease in length. This behavior of sheet resistance and thickness is independent of
the resistance value of the paste.
4.
TEMPERATURE COEFFICIENT OF RESISTANCE
’Hot’ and ’Cold’ temperature co-efficient of resistance (TCR) were calculated from the
measurements taken between +25C and -55C respectively. As expected, the TCR
3O
E 25
=
o
20 a
10
o
20
o
o
30
40
50
Length (mils)
FIGURE 3. Thickness variation with resistor length
60
70
80
SHOBHA C. RAVI
234
TABLE II
Average TCR (ppm/C)
Rcs.
Length
(mils)
C
H
1631
1621
1711
80
+71.35 +25.74
H
C
H
1641
C
H
69.93 -170.10
50.01 -141.09 -4.59
1651
C
H
1660
C
H
C
82.85 -29.17 -118.58 -41.98 -125.81
70
+77.65 +30.98
68.16 -171.82
51.88 -145.61 -4.91
83.89 -28.93 -117.65 -43.82 -128.93
60
+80.71 +33.14
73.27 -179.06
51.58 -146.98 -7.24
88.04 -30.12 -120.06 -45.10 -129.86
50
+87.85 +38.28
75.69 -181.04
53.05 -147.60 -9.11
91.41 -31.39 -121.65 -44.68 -128.18
40
+78.81 +29.85
80.51 -186.98
55.62 -151.01 -1.41
96.82 -33.67 -124.81 -45.96 -130.05
30
+74.05 +29.82
87.61 -197.38
66.41 -167.27 -1.85 -102.89 -37.59 -128.02 -44.88 -128.18
30
+73.32 +30.74
86.93 -191.59
68.02 -175.34 -1.78 -104.21 -36.93 -136.43 -43.63 -132.16
20
+73.06 +32.93 -103.14 -214.74
79.72 -181.07 -3.43 -122.51 -51.60 -147.39 -46.22 -133.28
10
+87.41 +48.07 -144.77 -270.76 -142.42 -260.95 -7.43 -181.07 -88.24 -195.11 -69.15 -157.11
behavior trends for the two series, 1711 and 1600, were quite different. Both the ’Hot’ and
’Cold’ TCR values are positive for 1711 paste, whereas, it is negative for 1600 series (Table
II). This could be due to their different rheological properties and also due to the higher
metal content of the 10 ohm/sq paste. However, in both cases the TCR values do not change
sign with decrease in length.
All the higher resistivity (100 ohm/sq to 100K ohm/sq) resistors have TCR strongly
dependent on resistor length. Our results indicate that the magnitude is constant down to 50
mils and increases with further decrease in resistor length. This behavior is independent of
the paste resistance value. Thus, this is a consequence of the observed decrease in sheet
resistance with length.
0’I
80
-40 0"
-80
70
Resistor Length (mil)
60
50
30
40
_I------l--
20
_ii \
*--’’*-’----*--__""
oo=,a
M
100 K
K
-120
100ohm/sq
-160
FIGURE 4.
10
Dependence of temperature co-efficient of resistance on resistor length
CHARACTERIZATION OF THICK FILM MICRORESISTORS
235
Resistor
Termination
FIGURE 5. Illustration of thick film resistor
A simple formula for the total resistance is (3):
R
Os (l/w) + 0’s (2d/w) (1
+ 0’s/0s) -1 + Rc
(1)
where
0s and 0’s are the sheet resistivities of the thick film resistor and the conductive termination
material, and w are the length and width of the resistor and d is the overlap length of the
resistor on the termination (Figure 5).
The first term represents the resistor sheet resistance, the second is that of the parallel
combination of resistor and conductor overlap and the third term, RC, is the contact
resistance at the resistor-conductor interface. Rc can be expressed as (3):
Rc
2/w
x (0s/G) 1/2
(2)
where
G is the conductance of the resistor-conductor interface.
From Eqns. 1 and 2 the effective sheet resistance is given by
R
0s (l/w)
+ 0’s (2d/w) x 1/(1 + 0’s/0s) + 2/w x (0s/G) 1/2
The second term can be neglected since 0s ">
R
0s (l/w)
0’s.
+ 2/w x (0s/G)1/2
(3)
Thus,
(4)
The resistance is thus a function of geometry and the interface conductance. This explains
the decrease in resistance and the subsequent increase in TCR with decrease in resistor
length.
It is also observed that the magnitude of TCR is dependent on the resistivity of the paste.
As shown in Figure 6, the TCR decreases with increase in paste resistivity, attains a
minimum value at 10 Kohm/sq and then increases with further increase in resistivity. This
behavior is independent of the resistor length. The first part of the curve can be explained as
due to the increase in resistor resistivity and consequently a decrease in TCR. The increase
in TCR observed in the latter half of the curve may be due to the large change in resistance
(A R) with temperature. It is well known that higher resistivity materials undergo a larger
change in resistance than low resistivity materials.
SHOBHA C. RAVI
236
1621
Paste
#
1631
1641
1651
1660
1621--100 ohrn/sq
163 I--I Kohm/sq
1641-- 10 Kohm/sq
1651--100 Kohm/sq
1660-- Mohm/sq
FIGURE 6. Variation of temperature co-efficient of resistance with resistivity of the paste
5. THERMAL AGEING
Thermal ageing of encapsulated, trimmed resistors was carried out at 125C for 100 hrs with
no load applied. The average percentage change in resistance (A R/R, %) is tabulated as a
function of resistor length in Table III. The results can be summarized as follows:
1. Not all the drifts were positive
2. The 10 ohm/sq paste resistors have a relatively higher drift value
3. There is no dependence on the resistor length nor on the resistivity
TABLE III
@ 125 C for 100 hrs.
Thermal ageing
(AR/R)%
length
(mil)
1711
1621
1631
1641
1651
1660
80
-0.31
+0.02
+0.04
-0.06
+0.42
-0.09
60
-0.33
+0.03
-0.11
-0.09
+0.50
-0.14
30
-0.41
-0.10
-0.04
-0.07
+0.17
-0.17
20
-0.36
-0.07
+0.13
-0.20
+0.12
-0.11
10
-0.39
-0.09
-0.07
+0.12
+0.42
-0.15
CHARACTERIZATION OF THICK FILM MICRORESISTORS
237
TABLE IV
Pcrccntagc resistance drift after 20 thermal cycles
(AR/R%)
length
(mil)
1711
1621
1631
1641
165l
1660
80
0.14
0.06
0.04
0.04
0.01
0.05
0. (18
0. (/2
(/. 01
(I. 02
0.09
0.06
0.08
60
30
0. (19
0.02
0.00
0.07
0.05
20
0.10
0.04
0.01
0.11
0.16
0.06
10
0.16
0.04
0.06
(I. 14
0.22
0.06
6.
THERMAL CYCLING
The thick film resistors were subjected to twenty thermal cycles between -55C and + 125C,
and the percentage resistance drift was measured. As shown inTable IV, the geometry and
resistivity of the resistors do not effect the stability. All the resistors exhibit excellent
stability: A R/R < 0.20% after twenty cycles. The drifts were all positive.
7.
CONCLUSIONS
The resistor thickness increases with decrease in resistor length below 50 mils. The post-trim
drift is negligible (less than 0.1% over a two months period) and is independent of the
resistor length and resistivity. The temperature coefficient of resistance is minimum for the
10 Kohm/sq paste (1641). The ’Cold’ TCR value is in general larger than the ’Hot’ TCR
value. TCR is length independent down to 50 mils, increases gradually at 20 mils and sharply
for a 10 mil resistor. The thermal ageing drift is less than 0.2% and the thermal cycling drift
(20 cycles) is less than 0.20%
The performances of these microresistors having dimensions as small as 10 10 mils are
acceptable and have qualified the MIL-STD-202 specifications. However, good control
during manufacture is necessary. All the above data presented in this paper has been
averaged over 500 thick film resistors for each test.
ACKNOWLEDGEMENT
The author would like to thank Rob Collier for conducting most of the measurements.
REFERENCES
1.
2.
3.
A. Bellardo, and G. Lovati, Hybrid Circuits. no. 4, pp. 26-31 (1984).
H.M. Naguib, Proc. International Microelectronics Symposium, pp. 48-59 (1977).
H.S. Fisher, and P.M. Hall, Proc. IEEE, vol. 59, p. 1418 (1971).