Download Prototype Feasibility Assessment

Prototype Feasibility Assessment
Senior Design PO8023
1/18/2008
In order to accurately assess the feasibility of the selected design concepts for team
PO8023, Air Muscle Artificial Limb, a prototype was assembled using the parts and
procedures listed herein.
Design and Build:
I. Design & Construction
Based on initial concept generation exercises it was determined that it would be
beneficial for the team to construct a prototype of the finger design by using a dimension
fused deposition modeler (FDM). This rapid prototyping machine used the existing
computer aided design (CAD) model to create the finger by laying down ABS plastic in
layers until the final design was complete, as shown in figure 1.
Figure 1: Prototype finger made from ABS Plastic
A custom made Lego palm and forearm, figure 2, was constructed as closely to the
final design scale as possible to allow simulation of the prototype finger design with air
muscles and verify all motion clearances.
Figure 2: Prototype Lego palm and forearm
The fingers were wired to air muscles using Berkley Steelon Nylon Coated wires and
extension ligaments were fitted using Polypropylene Covered Elastic Cord as shown in
figure 3.
Figure 3: Assembled finger with tendon cabling and extension ligaments
All completed parts were then assembled into the final prototype as shown in figure 4.
Figure 4: Final assembled prototype for demonstration
II. Parts List
Description
Quantity Units
Air Muscle Artificial Limb Assembly
Lego Forearm & Palm Assembly
1
1
Ea.
Ea.
Index Finger Assembly
Common Finger Tip (Distal Phalanx)
IF_Section 2 (Middle Phalanx)
IF_Section 1_Upper (Proximal Phalanx)
IF_Section 1_Lower (Metacarpals)
1
1
1
1
1
Ea.
Ea.
Ea.
Ea.
Ea.
Middle Finger Assembly
Common Finger Tip (Distal Phalanx)
MF_Section 2 (Middle Phalanx)
MF_Section 1_Upper (Proximal Phalanx)
MF_Section 1_Lower (Metacarpals)
1
1
1
1
1
Ea.
Ea.
Ea.
Ea.
Ea.
Ring Finger Assembly
Common Finger Tip (Distal Phalanx)
RF_Section 2 (Middle Phalanx)
RF_Section 1_Upper (Proximal Phalanx)
RF_Section 1_Lower (Metacarpals)
1
1
1
1
1
Ea.
Ea.
Ea.
Ea.
Ea.
Polypropylene Covered Elastic Cord 3/32" Diameter (50 ft.)
Double Pinch Zinc-Pltd Stl Hose & Tube Clamp 1/8" to 5/32" Clamp Diameter Range
Berkley Steelon Nylon Coated Wire
Berkley Wire Leader Connector Sleeves
SPRO Ball Bearing Swivel with Interlock Snap
25
25
30
24
9
Ft.
Ea.
Ft.
Ea.
Ea.
Air Muscle Assembly
3
Ea.
III. Prototype Issues
One of the main problems with the prototype was that the tendon cabling could not be run
through the palm the same way that it would be on the final design. Despite this
limitation of the prototype the team used alternate routing of the cables to achieve motion
of the fingers. Another drawback of the prototype was that it was difficult to service,
especially during initial assembly and when adding air muscles. One of the main reasons
for this is that there was no cable guide to help position the tendon cabling running
through the palm to the fingers. The final issue that was solved during the prototype
assembly process was that the diameter of the holes for the extension ligaments had to be
increased in order to attach the elastic cord.
IV. Influences on Final Design
Based on the building and testing conducted with the prototype Air Muscle Artificial
Limb, the team learned that serviceability must be a main focus on the final design.
Increased serviceability in the final design will allow for adjustments to be made as
necessary through the final assembly process. The team also decided to use both rubber
bands and elastic cord for the return ligaments in order to help return the finger to its rest
position. The stiffer elastic cord will begin returning the finger to its initial position while
the shorter rubber band will finish the fingers extension.
V. Conclusions
The prototype was successful in demonstrating the motion clearances in the finger design
and the feasibility of the connection points for the tendon cabling that will cause finger
motion based on air muscle contraction.
Air Muscle:
I. Design & Construction
Air Muscle Build Process
Step 1
Step 2
Step 3
Step 4
Step 5
Step 6
Step 7
Step 8
Step 9
Cut rubber tubing of desired diameter to desired length.
Thread a thin metal rod through the rubber tubing.
Insert the rubber tubing (containing the metal rod) through the mesh.
Remove the metal rod – align the edge of the tube to the edge of the
mesh.
With a heat knife, cut the mesh where the tubing ends. (Now the mesh
and tubing should be concentric and of the same length)
Place a metal crimping piece on each end of the air muscle.
Insert Eye Hook into one end of the rubber tube
Insert the appropriate airside connector in the other end of the air muscle.
Crimp both ends of the air muscle, creating a seal between the mesh,
tubing, and end fitting
Air muscles should always be tested for leaks at the pressure they will be operated at
before use.
Figure 5: Assembled Air Muscle
II. Parts List
Description
Air Muscle Assembly
Mesh – 3/8in PET (neon red)
Tubing - Super Soft Latex Rubber Tubber 1/8" ID, 3/8"
OD, 1/8" Wall Thickness
Nckl-Pltd Brass Push-to-Connect Tube Fitting
Expanding Tube Stem for 1/8" Stem OD X 1/4" Tube
OD
Double pinch zinc-pltd stl hose and tube clamp 13/64"
to 9/32" PKG=25 per pack
Eye Hook #12
Quantity
Units
8
8
ft
ft
18
Ea.
36
Ea.
18
Ea.
III. Prototype Issues
Problems with the air muscles were encountered during life cycle testing. The air side
connector failed after running for an average of 1500 cycles. During this testing, the
connector would be pulled free from the air muscle. This failure mode was the only
failure mode that appeared during air muscle testing. Of all the air side connectors tested,
the Nickel Plated Push-to-Connect fitting worked the longest.
IV. Influences on Final Design
Based on the failure mode of the air muscle, ease of serviceability is the key to the
success of the final design. As the air muscles approach failure they can be replaced by
new muscles to avoid the disassembly of the air side fitting. On the final design the air
muscles will be accessible to make serviceability simple.
V. Conclusions
The air muscle design is suitable for the final limb design. The air muscles will last
around 1500 cycles and be available for service. Improving the air side connection
method would increase the cycle life of the muscles and reduce the need for replacement.
Controls:
I. Design & Construction
For the initial prototype testing, the valves and relay board were assembled as they would
be for the final product. The valves were connected first utilizing the following
instructions.
The fasteners on the end of coil and DIN connector combination were loosened, rotated
180°, and re-tightened. The gasket was placed over the manifold and aligned with the
three-port holes of the individual station. Each valve was aligned to the manifold station
and fastened with two mounting screws. The air supply fitting was fastened to the frontface center through the hole on manifold. Plugs were also placed on the remaining two
through-holes on front-face. Plugs were placed on the back-face center through-hole and
the remaining two through- holes on back-face were left open for exhaust.
Figure 6: Valves with Air Supply Fittings
The threads of the needle valve were and valve fittings with then wrapped with thread
seal tape. The push-to-connect valve fittings were screwed into the designated
Ab/Adduction solenoid port holes and needle valves into assigned flexion solenoid port
holes. The appropriate end of the push-to-connect fitting was thread into one of the
flexion solenoid needle valves, and then the air muscle feed tube was inserted into the
push-to-connect end of the fitting ensuring that the tube is fully sealed. The DIN
connector was disassembled by unscrewing the connector from the coil and removing the
transparent shell, as shown in Figure 7.
Figure 7: DIN Connector
Terminal 1 was connected to the relay board (Measurement Computing USB SSR 24)
and terminal 2 to ground. As shown in the schematic below, terminal 2 was daisy
chained to each of the other DIN connectors with the first or last connector terminated to
the ground wire of the power supply. Similarly, the power wire was connected to a
positive terminal of the relay board and daisy chained to every other positive terminal.
Each negative relay terminal was connected directly to one solenoid valve at terminal 2,
such that each relay will control one valve.
Relay Board
+
+
1
Power Supply
+
2
+
3
+
4
-
-
-
1
1
1
+
5
-
+
6
-
+
7
-
+
8
-
+
9
-
VCC_BAR
2
2
1
2
1
2
1
2
2
Solenoid Valve DIN Connectors
2
2
1
2
1
2
1
2
1
2
1
+
10
-
1
Figure 8: Relay Board and Solenoid Valve Connections
+
11
-
12
-
-
The relay board connections are also shown in Figure 9.
Figure 9: Relay Board with Connections to Valves
The relay board was also connected to the computer via USB. The code that was used in
LabVIEW to control a single flexion air muscle consisted of a simple loop that
considered the percentage the finger moved due to one valve cycle (a single valve being
turned on and off once). The number of cycles needed to reach the target percentage was
then determined. Figure 10 provides a simple diagram of the LabVIEW program logic.
Active
Valve
Delay
Deactivate
Valve
Delay
*run loop N times (where N is # of cycles
needed to reach target
Figure 10: LabVIEW Program Logic
II. Parts List
Description
Quantity Units
Valve Assembly
Clippard 4-Way 3 Position: Double Solenoid, Closed Center, 12 VDC, 1/8" NPT
Clippard 4-Way 3 Position: Double Solenoid, Exhaust Center, 12 VDC, 1/8" NPT
Clippard 6 Valve Manifold 1/4" Ports
Clippard Needle Valves
Legris Needle Valve Fittings Mal Straight
Legris Manifold Fittings Mal Straight, 1/4" Tube x 1/4" NPT Male
Wires
Sub-D25 Slider Plug
Electrical Tape
Shielded Hood D25
Desk Top Power Supply: 2.1 x 5.5 x 10mm, Center + Straight DC Connector
1
3
3
1
9
9
1
1
1
1
1
1
Ea.
Ea.
Ea.
Ea.
Ea.
Ea.
Ea.
100 Ft.
Ea.
20 Ft.
Ea.
Ea.
Relay Assembly
Opto 22 Relays
Relay Board / DAQ
USB Cable, MaleA to MaleB
Power Supply
Sub-D25 Slider Jack
Shielded Hood D25
1
16
1
2
1
1
1
Ea.
Ea.
Ea.
Ea.
Ea.
Ea.
Ea.
Potentiometer Assembly
Poteniometer DAQ, Personal Measurement Device: PMD-1208LS
Shielded Hood D25
Sub-D25 Slider Plug
Sub-D25 Slider Jack
Slide Potentiometer Sensors: 1Kohms Travel = 60mm
1
1
2
1
1
3
Ea.
Ea.
Ea.
Ea.
Ea.
Ea.
Misc.
Labview
1
1
Ea.
Ea.
III. Prototype Issues
The primary issues that were apparent during the prototyping analysis were due to the
lack of completion in the LabVIEW coding. Although LabVIEW was appropriately
interfacing with the valves, the incremental applications of pressure were jerky and the
amount was not controllable (the user was not able to fill the air muscle, for instance,
20% of the total inflation capability). Also, the LabVIEW computer interface was not
set-up to be user-friendly; only a person that understands the software and code would be
able to use it. Furthermore, the potentiometers were not included in the prototype, so
there was no feedback during this initial testing.
IV. Influences on Final Design
Based on the initial prototype testing, it was evident that the valves were correctly
assembled, the relay board and DAQ were properly wired, and the interfacing with
LabVIEW was feasible. It was also apparent that further work on the LabVIEW code is
necessary in order for the pressurized air to be controlled more smoothly and in regulated
increments. Ultimately, the computer interface will need to be more user-friendly so that
an average costumer would be able to use the program.
V. Conclusions
From this initial testing, it was clear that the valves, relay board, and DAQ have been
correctly compiled and that they will work for this project. The prototype was also
successful in showing that LabVIEW is capable of controlling the valves and that the
interfacing works correctly. Lastly, it became apparent that there will need to be some
sort of method for holding all of the electronic equipment to make transportation more
manageable.