Download Introduction to Transmission Line Pulse

Introduction to
Transmission Line Pulse
(TLP)
Overview
1. What is TLP
2. How TLP works
3. TLP measurement
4. TLP variants
5. Interpreting TLP data
6. TLP Precision
7. VF-TLP
8. Q&A
2
History
• Transmission Line Pulse
• Introduced as a possible method to emulate energy in HBM
• Determined that for a given peak current, a TLP pulse of 100ns
carries the same energy as HBM
• Same energy is produced by different voltages:
 50V TLP = 1A
 1500V HBM = 1A
3
-50
0
50
100
150
200
250
Background
• Characterization of protection structures is important
• Predict behavior for real world events
• Important parameters:
•
•
•
•
Snapback voltage
Turn on time
RON resistance
Failure point
• Must determine parameters with techniques similar to ESD
• TLP is an excellent solution
• Controlled impedance makes measurements easier
• Low duty cycle prevents heating
4
Section 1
What is TLP
5
Section 1: What is TLP
TLP Waveform
• ESD tests simulate real world events
Vol tage
• What makes TLP different than ESD?
• HBM, MM, IEC, CDM
• TLP does not simulate any real-world event
-20
• ESD tests record failure level
• “Qualification”
• TLP tests record failure level and device behavior
• “Characterization”
6
0
20
40
60
T ime (ns)
80
100
120
Section 1: Device Characterization
 What is Device Characterization?
• Describes the resistance of a device for a given stimulus
• Resistance = Voltage / Current
• Conventionally performed by increasing amplitude until failure
2.9
2.4
I D U T (A)
1.9
1.4
0.9
0.4
-0.1
• Like Curve Tracing
7
0
5
10
V D U T (V)
15
Section 1: Device Characterization
 How is TLP different than Curve Tracing?
• TLP is a short pulse
V o lta g e
V o lta g e
• Curve Tracing is DC
T im e
• Shorter pulse
• Reduced duty cycle, less heating
• Controlled Impedance
• Allows device behavior to be observed (more on this later)
8
T im e
Section 1: Device Characterization
• How is TLP the same as Curve Tracing?
TLP
V o lta g e
V o lta g e
DC Curve Trace
Time
Tim e
• Measure resistance of device with increasing voltage
• Less heat means higher voltage before failure
9
Section 1: Device Characterization
• What can I learn from Device Characterization with TLP?
• Turn-on time
0.45
0.4
0.35
• Snapback voltage
I DU T (A)
0.3
0.25
0.2
0.15
0.1
• Performance changes with rise time
0.05
0
0
2
4
6
V DUT (V)
10
8
10
12
Section 1: TLP Waveforms
 What does a TLP waveform look
like?
Vol tage
• Square pulse
TLP Waveform
-20
0
20
• Unlike ESD waveforms, TLP does not
mimic any real world event
HBM Waveform
11
40
60
T ime (ns)
MM Waveform
80
100
120
Section 1: TLP Waveforms
 What variations are there for TLP waveforms?
Vol tage
• Pulse Width
TLP Waveform
• 100ns
• 30ns – 500ns
• VF-TLP: 1ns – 10ns
Pulse Width
-20
0
20
40
60
80
100
T ime (ns)
TLP Waveform
TLP Waveform
V o lta g e
• 0.2ns – 10ns
V o lta g e
• Rise Time
Rise Time
Rise Time
-20
0
20
40
60
T im e (ns )
12
80
100
120
-20
0
20
40
60
T im e (ns )
80
100
120
120
 How do these variations affect
the TLP test?
• Pulse Width
V o lta g e
Section 1: TLP Waveforms
• Energy under the curve
Pulse Width
-20
0
20
40
60
80
100
120
• Rise Time
• Device reaction
V o lta g e
T im e (ns )
Rise Time
-20
0
T im e (ns )
13
20
Section 1: Devices for TLP Testing
• What kinds of packages can be tested?
• Package Level
• Wafer Level
14
Section 2
How TLP Works
• How TLP pulses are generated
15
Section 2: How TLP Works
• How is a TLP waveform generated?
• Transmission Line connected to power supply
• Called Charge Line
• Length proportional
to pulse width
• Power supply charges the cable
16
Section 2: How TLP Works
• How is a TLP waveform generated?
• DUT lies at end of another transmission line
• Switch closes
• Charge exits Charge Line, propagates towards DUT
17
Section 2: How TLP Works
 How is a TLP waveform generated?
• Square waveform
V o lta g e
• Charge Line behaves as a storage device
-20
0
20
40
60
T im e (ns )
18
80
100 120
Section 2: How TLP Works
• What happens when the waveform hits the DUT?
• Recap: TLP is a short-duration pulse in a controlled-impedance environment
• Behaves like an RF signal
• RF signal behavior
• Propagates until impedance changes
19
Section 2: How TLP Works
• How is resistance measured with a TLP pulse?
• Square pulse, perceive it as a short-duration Curve Trace
DC Curve Trace
• Voltage and Current probes measure the DUT
20
TLP
Section 2: How TLP Works
 How is resistance measured with a TLP pulse?
• Plateau of waveforms are averaged
• Device allowed to settle into “quasi-static” state
Current Probe Waveform
V o lta g e
C ur r e nt
Voltage Probe Waveform
-20
0
20
40
60
T im e (ns )
21
80
100
120
-20
0
20
40
60
T im e (ns )
80
100
120
Section 2: How TLP Works
• Why use both a Voltage probe and a Current probe?
Current Probe Waveform
Volt ag e
Curre nt
Voltage Probe Waveform
-20
0
20
40
60
80
100
120
-20
Time (n s )
0
20
40
T im e (ns )
• It is possible to calculate DUT resistance with only 1 probe
• VDUT = VPulse – (IDUT * ZTLP)
• IDUT = (VPulse – VDUT) / ZTLP
• Not desirable because extremes are noisy
22
60
80
100
120
Section 2: How TLP Works
 Sequential TLP pulses produces an I/V Curve
OPEN Circuit
SHORT Circuit
0.6
1. 9
1. 4
I DU T (A )
I DU T (A )
0.5
0.9
0.4
0.3
0.2
0.4
0.1
- 0.1
0
0
5
10
15
20
25
30
0
0.2
0.4
V DU T (V )
0.4
0.35
0.2
I DU T (A )
I DU T (A )
1
Snapback Device
0.45
0.25
0.15
0.1
0.05
0.3
0.25
0.2
0 . 15
0.1
0.05
0
0
5
10
15
V DU T (V )
23
0.8
V DU T (V )
Resistor
0.3
0.6
20
25
30
0
0
2
4
6
V DU T (V )
8
10
12
Section 2: How TLP Works
• How is the Pulse Width changed?
• Length of cable
 How is the Rise Time changed?
• Low-pass filter added
LP
24
Section 3
TLP Measurement
• How devices are measured
25
Section 3: TLP Measurement
• Measurement Goals
• Capture Voltage at the DUT
• Capture Current through the DUT
26
Section 3: TLP Measurement
• Equipment to capture V and I
Voltage Probe
Oscilloscope
Current Probe
27
Section 3: TLP Measurement
• Equipment to deliver TLP pulse to DUT
Wafer Level
3
1.
DUT (on wafer)
2.
TLP Pulse delivery cable
3.
TLP Pulse delivery probe
4.
Ground Probe
5.
Ground Braid
2
5
1
Package Level
1.
DUT in socket
2.
TLP Pulse delivery cable
3.
Grounded pin
3
28
4
Section 3: TLP Measurement
• Ideally, V and I probes are directly on DUT
• Direct placement not possible
V/I Probes
29
Section 3: TLP Measurement
• Although probes are not at DUT, measurements are possible
V/I Probes
• Controlled impedance
• Waveform observable any place along path
• Time Domain Reflection (TDR)
30
Section 3: TLP Measurement
• How are measurements accomplished away from DUT?
V/I Probes
• Incident and Reflected waveforms
• Adding the waveforms reproduces DUT measurement
• Incident and Reflected waveforms are recorded separately (TDR-S)
31
Section 3: TLP Measurement
Reflected
Waveform
Polarity
DUT Ω > ZTLP
DUT Ω < ZTLP
DUT Ω = ZTLP
(Example: Open Circuit)
(Example: Short Circuit)
(Example: 50 Ω Resistor)
Positive Reflection
Negative Reflection
No Reflection
Negative Reflection
Positive Reflection
No Reflection
Voltage
Waveform
Current
Waveform
32
Section 3: TLP Measurement
• TDR-S performs waveform addition with software
V/I Probes
 Waveform addition can also be done in the TLP circuit
V/I Probes
33
Section 3: TLP Measurement
• Overlapping waveforms
V/I Probes
• Incident and Reflected overlap, add together
• Overlapped waveform plateau reproduces DUT waveform
34
Section 4
TLP Variants
35
Section 4: Importance of Impedance
 Why vary the impedance?
• Low Impedance
• More current flow for a given voltage
• More voltage amplitude
• High Impedance
• Less current flow for a given voltage
• Less voltage amplitude
36
Section 4: TLP Variants
• TLP circuit characteristics up to this point
• 50 Ω Impedance
• One stress-pin
• One ground-pin
Incident
Pulser
Reflected
• Time Domain Reflection (TDR)
• TDR-O
• TDR-S
V and I to
Scope
Absorbed
• Variations alter the system impedance and grounding style
37
D
U
T
Section 4: TLP Variants
• Time Domain Transmission (TDT)
• 25Ω system impedance
Incident
Scope Ch3
Pulser
Reflected
V and I to
Scope
38
Absorbed
Transmitted
D
U
T
Section 4: TLP Variants
• Time Domain Reflection and Transmission (TDR-T)
• DUT in series with pulse transmission path
• 100Ω Impedance
Incident
Pulser
Reflected
V and I to
Scope
39
Scope Ch3
DUT
Absorption
Transmitted
Section 4: TLP Variants
• High-Z Time Domain Reflection and Transmission (TDR-T)
• DUT in series with pulse transmission path
• 500Ω, 1kΩ Impedance
High-Z Ω
Incident
Pulser
Reflected
V and I to
Scope
40
Scope Ch3
DUT
Absorption
Transmitted
Section 5
Interpreting TLP Data
41
Section 5: Interpreting TLP I/V Curves
0.45
0.4
ID U T (A)
0.35
• What information do TLP I/V
Curves provide?
0.3
0.25
0.2
0.15
0.1
0.05
0
0
2
4
6
8
10
12
V D U T (V)
Voltage Waveform
C ur r e nt
V o lta g e
• What information do TLP
waveforms provide
-20
0
20
40
60
T im e (ns )
42
Current Waveform
80
100
120
-20
0
20
40
60
T im e (ns )
80
100 120
Section 5: Interpreting TLP I/V Curves
• Device: Resistor
• Ω is constant no matter what
Resistor
0.3
voltage is applied
I DUT (A)
0.25
0.2
0.15
0.1
0.05
0
0
43
5
10
15
V DUT (V)
20
25
30
Section 5: Interpreting TLP I/V Curves
0.5
• Breakdown Voltage
0.3
0.4
0.2
0.3
I DU T (A )
• Turn-on Voltage
0.5
I DU T (A )
• Device: Zener Diode
0.4
0.2
0.1
0.1
0
0
- 0.1
- 0.1
0
0
2
44
2
6
6
8
8
10
10
DU T (V )
VVDU
T (V )
0.05
0
- 0.05
I DU T (A )
- 0.1
- 0 . 15
- 0.2
- 0.25
- 0.3
- 0.35
- 0.4
- 15
- 10
-5
V DU T (V )
44
0
Section 5: Interpreting TLP I/V Curves
• Device: Snapback Protection
Structure
0.45
1. Low Voltage TLP Pulses
•
Open Circuit
0.35
2. Snapback Threshold
3. On Resistance
4. Failure Level
•
0.3
Begins conducting current
Peak Current
I DU T (A)
•
4
0.4
3
0.25
0.2
0.15
0.1
2
0.05
0
1
0
2
4
6
V DUT (V)
45
8
10
12
Section 5: Interpreting TLP Data
Begin Test Procedure
• How is failure detected?
• DC leakage test after
each TLP pulse
Apply stress pulse
to the DUT
Record Voltage &
Current waveforms
DC
Leakag
e test
PASS
Increase stress
pulse amplitude
NO
Max
pulse
amplitud
e
reached?
YES
Stop
46
FAIL
Section 5: Interpreting TLP Data
• How is failure detected?
SourceMeter, etc.
TLP Pulser
D
U
T
V & I to Scope
SourceMeter, etc.
TLP Pulser
D
U
T
V & I to Scope
47
Section 5: Interpreting TLP Data
• How is failure detected?
I Leakage (A)
0.0001
• DC Leak Test
0.001
0.01
0.1
1
1.2
I DU T (A)
• Force 14V
• 10mA Compliance
• >9mA Failure
1
0.8
IV Chart
0.6
Leakage (A)
0.4
0.2
0
-0.2
-2
0
2
4
6
V DUT (V)
48
8
10
12
14
• Precise amount of energy under curve
• Controlled impedance circuit
• Conservation of energy
V o lta g e
Section 5: Load Lines
-20
• Pulsed = VDUT + (IDUT * ZTLP)
• For a given TLP Amplitude, the
measured V and I will always reside
on that Amplitude’s Load Line
0
20
40
60
80
100 120
T im e (ns )
IDUT
Load Line
VDUT
49
Section 5: Load Lines
 Open Circuit:
 Short Circuit:
50
VPulse
VDUT
IDUT
50
50
0
100
100
0
150
150
0
200
200
0
250
250
0
VPulse
VDUT
IDUT
50
0
1
100
0
2
150
0
3
200
0
4
250
0
5
IDUT
VDUT
IDUT
VDUT
Section 5: Load Lines
 50Ω Circuit:
 Snapback Device:
51
VPulse
VDUT
IDUT
50
25
0.5
100
50
1
150
75
1.5
200
100
2
250
125
2.5
VPulse
VDUT
IDUT
50
50
0
100
100
0
150
150
0
200
60
2.8
250
70
3.6
IDUT
VDUT
IDUT
VDUT
Section 5: Load Lines and System Impedance
• How does system impedance affect the Load Line?
• Low Impedance
• More current flow for a given voltage
IDUT
• High Impedance
• Less current flow for a given voltage
50Ω
5 --
250Ω
4 --
• VPulse = VDUT + (IDUT * ZTLP)
• IDUT = (VPulse - VDUT) / ZTLP
3 --
2 --
1 --
52
|
|
|
|
|
50
100
150
200
250
VDUT
Section 5: System Impedance
• Why would I want to change the system impedance?
• Snapback device recap:
• Voltage triggered
• Fails by peak current
• What if Snapback device
has high trigger voltage,
low peak current?
IDUT
50Ω
5 --
250Ω
4 --
3 --
2 --
• 250V Trigger Voltage
• 5A peak current
53
1 --
|
|
|
|
|
50
100
150
200
250
VDUT
Section 5: System Impedance
• Characteristics of High Impedance
• Longer settling time
• Time Constant = ZTLP * CDUT
50 Ω
• Unstable turn-on region
• Enough Voltage to turn-on
• Not enough Current to remain on
54
500 Ω
1000 Ω
Section 5: Quasi-Static Region
• What defines the “quasi-static” region?
Quasi-Static
• After transient response has settled
Transient
• What if the device never settles?
0.11
80
80
70
70
60
60
30
30%
0.07
40%
50%
0.05
60%
70%
0.03
80%
50
50
40
I D UT (A)
0.09
40
30
0.01
-0.01
-1
4
9
V D UT (V)
55
14
Section 5: Transient Region
• Does the transient response tell me anything?
• Turn-on time
Zener Waveform (TDR-S)
Volt ag e
• Capacitance
-20
20
60
100
140
180
Time (n s )
56
220
260
300
Section 6
TLP Accuracy
• Understand limitations
• Optimize measurements
57
Section 6: TLP Accuracy
• Misconception: TLP is as precise as DC curve tracing
• What affects the accuracy of TLP measurements?
• Short duration means smaller sample size
• Measurement instrument limitations
• Parasitics and noise
• Contact resistance
• How are these limitations addressed?
58
Section 6: TLP Accuracy – How Are Limitations Addressed?
 Oscilloscope: Bandwidth
500Mhz Bandwidth
6Ghz Bandwidth
Sample Rate
-1.5
0.5
2.5
4.5
T im e (ns )
59
6.5
-1.5
0.5
2.5
4.5
Time (ns)
6.5
Section 6: TLP Accuracy – How Are Limitations Addressed?
 Oscilloscope: Sampling rate
500 MHz, 5 GS
Voltage
Vol tage
6 GHz, 20 GS
Sample Rate
-1.5
0.5
2.5
4.5
T ime (ns)
60
6.5
-1.5
0.5
2.5
4.5
Time (ns )
6.5
Section 6: TLP Accuracy – How Are Limitations Addressed?
 Oscilloscope: Signal digitization optimization
• 8 bits of resolution = 256 possible levels
TDR-O: Optimized Quasi-Static Region
15
250
225
200
175
150
125
100
75
50
25
0
-25
-50
-75
-100
10
Volt ag e
Volt ag e
TDR-O: No Optimization
0
-5
-10
-10
40
Time (n s )
61
5
90
-10
40
Time (n s )
90
Section 6: TLP Accuracy – How Are Limitations Addressed?
 Oscilloscope: Signal digitization optimization
8
7
I D UT (A)
6
5
No Adaptive Ranging
4
Adaptive Ranging
3
2
1
0
-1
-0.5
0
V D UT (V)
62
0.5
1
Section 6: TLP Accuracy – How Are Limitations Addressed?
 Oscilloscope: Signal digitization optimization
TDR-S: No Optim ization Possible
250
V o lta g e
150
50
-50
-150
-250
-20
30
80
130
180
230
T im e (ns )
• Overlapped waveform advantageous for best resolution
63
Section 6: TLP Accuracy – How Are Limitations Addressed?
• Parasitics:
• Factor out from DUT measurement
UNCORRECTED Open Circuit
0.5
0.4
0.4
I D U T (A)
I D U T (A)
CORRECTED Open Circuit
0.5
0.3
0.2
0.3
0.2
0.1
0.1
0
0
0
5
10
15
20
0
25
5
10
UNCORRECTED Short Circuit
30
0.4
I D U T (A)
I D U T (A)
25
CORRECTED Short Circuit
0.5
0.4
0.3
0.2
0.3
0.2
0.1
0.1
0
0
0
0.5
1
V D U T (V)
64
20
V D U T (V)
V D U T (V)
0.5
15
1.5
2
0
0.5
1
V D U T (V)
1.5
2
Section 6: TLP Accuracy – How Are Limitations Addressed?
• Noise:
• Multiple pulses
Voltage Waveform
-20
0
20
40
60
Time (ns)
Averaged
65
80
100
120
Section 6: TLP Accuracy – How Are Limitations Addressed?
• Contact Resistance
• Changes on each touchdown
• 4-point TLP (Kelvin)
• Contact resistance removed
Scope
V+
V-
I V
Ω
Ω
+
H_ V
DUT
66
Section 7
VF-TLP
• Similarities and differences
• Additional requirements
67
Section 7: VF-TLP
• What makes VF-TLP different?
• Short pulse width
• 1ns – 10ns
• Transient response emphasized
• Mimics CDM event
68
• Fast rise time
• 100ps – 200ps
Section 7: VF-TLP
• Additional requirements for VF-TLP?
TLP Waveform
• Oscilloscope bandwidth
• Oscilloscope sampling rate
V
• 200ps rise time = 5 GHz
• 1ns @ 20GS/sec = 20 samples
-0.25
0
0.25
0.5
Tim e (ns )
69
0.75
1
1.25
Section 7: VF-TLP
• Additional requirements for VF-TLP?
• More sensitive to parasitics
• Wafer-level only
• RF-rated probes recommended
Package Level
70
Coax Wafer Probe - 4 GHz
RF Wafer Probe - 40 GHz
Additional Topics
• How do I power up my device during TLP?
• DC BIAS supplies can be added
TLP Pulser
D
G
V & I to Scope
71
S
DC Supply
Additional Topics
• Does TLP correlate to HBM?
• Does VF-TLP correlate to CDM?
• Why use TLP instead of a TDR machine, or solid state pulse
generator?
• TLP Circuit is a closed system, will reflections continue
indefinitely?
72