Burn-In & Test Socket Workshop: Archive
Burn-In & Test Socket Workshop: Archive
ARCHIVE
Burn-in & Test Socket
Workshop
TM
COPYRIGHT NOTICE
• The papers in this publication comprise the proceedings of the 2005
BiTS Workshop. They reflect the authors’ opinions and are reproduced
as presented , without change. Their inclusion in this publication does
not constitute an endorsement by the BiTS Workshop, the sponsors,
BiTS Workshop LLC, or the authors.
• There is NO copyright protection claimed by this publication or the
authors. However, each presentation is the work of the authors and
their respective companies: as such, it is strongly suggested that any
use reflect proper acknowledgement to the appropriate source. Any
questions regarding the use of any materials presented should be
directed to the author/s or their companies.
• The BiTS logo and ‘Burn-in & Test Socket Workshop’ are trademarks
of BiTS Workshop LLC.
Burn-in & Test
tm
Socket Workshop Technical Program
BiTS Tutorial
Sunday 3/06/05 1:00PM
Outline
9 Who cares?
9 What’s important: signal integrity, power integrity
9 Common vocabulary
9 Insertion loss: what is and is not important?
9 Loop inductance: what is and is not important?
Electrical Performance in
Perspective
Constraints:
• Performance • Vendors
9 Compliance • Corporate Culture
• Compatibility: Industry, Legacy
9 Pitch
9 Cycle lifetime
9 Time between cleaning Cost:
9 Electrical $$$, TCOO,
– DC resistance Schedule, Risk
– Hi Frequency
– Signal Integrity
» Bandwidth Partitioning:
» Insertion loss • Pin electronics
» Return loss
» SPICE models
• Wiring/cabling
– Power integrity • Loadboards
» Loop inductance • Sockets
• Good news:
9 Always the final test
9 100% certainty for that situation
• Bad news
9 What about the next socket?
9 What do you tell your supplier?
9 How does supplier evaluate quality?
9 If it doesn’t work, where do you look to re-design?
9 Can you afford the time for multiple iterations?
• The only way to know if the system will work before building and
testing is system level simulation with accurate component
models
Lpower
Clk1 Gate2
Cconn Cconn
Cpin
Cpin
model
Lgnd
simulate
Lgnd
Clk1
• Bandwidth
• Insertion loss
• dB
• Why -1 dB
• What design features influence this performance
Is this acceptable?
incident amplitude
phase
transmitted
Bandwidth
10
Bandwidth (GHz)
0.1
.
0.01
0.01 0.1 1 10
10% to 90% Rise Time
(nsec)
0.35 5 1
Tperiod = 15 ≈ Tperiod =
BW BW Fclock
Key Assumptions
incident
transmitted
reflected
amplitude amplitude
phase phase
Terminology
incident
transmitted
Vtransmitted
What’s important: at each frequency
Vincident
• Also called:
9 Insertion loss
9 S21
9 Transfer function
There is a magnitude and a phase at each frequency
Source Termination
impedance impedance
= 50 Ohms = 50 Ohms
incident
transmitted
0.9 good
0.8
0.7
0.6
0.5
0.4 Is this good?
0.3
1.0
freq, GHz
• Is there a difference between
9 good
9 good enough
9 better ?
100% ?
⎛ Pb ⎞
ratio[dB ] = 10 x log⎜⎜ ⎟⎟ 0.000001 watt
= 10 −6 = −60 dB
⎝ Pa ⎠ 1 watt
If we have the ratio of the powers, take the exponent of the power 10 and multiple by 10
to get the dB
If we have the number of dB, divide by 10 and put to the power of 10 and this is the ratio
of the powers
-10 dB = 0.1 -20 dB = 0.01 -40 dB = 0.01%
V
P ~ V2
⎛ Pb ⎞ ⎛ Vb2 ⎞ ⎛V ⎞ ⎛V ⎞
ratio[dB ] = 10 x log⎜⎜ ⎟⎟ = 10 x log⎜⎜ 2 ⎟⎟ = 10 x 2 x log⎜⎜ b ⎟⎟ = 20 x log⎜⎜ b ⎟⎟
⎝ Pa ⎠ ⎝ Va ⎠ ⎝ Va ⎠ ⎝ Va ⎠
⎛ Vb ⎞
ratio[dB ] = 20 x log⎜⎜ ⎟⎟ When measuring the ratio of voltages, we use a 20
⎝ Va ⎠ When measuring the ratio of powers, we use a 10
0.1
-20 dB = _________ 0.95
-0.5 dB = ______
0.01
-40 dB = _________ -1 dB = _______
0.90
0.001
-60 dB = _________ 0.80
-2 dB = _______
-80 dB = _________
0.0001 0.70
-3 dB = _______ When dB is small,
Magnitude ~1 + dB/10
1
0 dB = __________
0.3
-10 dB = _________
-25 dB = _________
0.05
From magnitude to dB
If we have the ratio of the amplitudes,
⎛ Vout ⎞
1. Write it to the power of 10 ratio[dB ] = 20 x log⎜⎜ ⎟⎟
2. Take the exponent (or take the log of the number) ⎝ Vin ⎠
3. Multiply by 20
4. This is the dB
10% = ________
-20 dB
-40 dB
1% = _________
0.1 volt
= 10 −1 = −20 dB -60 dB
0.001 = ________
1 volt
0.01 volt -1 dB
90% = _________
= 10 −2 = −40 dB When magnitude is close to 1,
1 volt 80% = _________
-2 dB dB ~ (mag – 1) x 10
-0.2 dB
98% = _________
100% ?
Insertion 0 dB + 0 dB + 0 dB + -6 dB = -6 dB
losses:
0 dB + -1 dB + -1 dB + -6 dB = -8 dB
• Multiple approximations:
9 Bandwidth of the signal
9 Is the system a 50 Ohm system?
9 Total system budget
9 Allocation to the socket
1. Matched Impedance
2. Controlled impedance
3. Discontinuities of load board
4. Length
5. Dielectric loss
6. Conductor loss
7. DC contact resistance
Length
Signal path
GROUND
Return path
The Signal
Vsignal
V
Signal path
Vin V
Return path
GROUND
signal
ε
return
12 inches 12 inches
v= n sec
= n sec
= 6 inches
n sec
4 2
Instantaneous Impedance
Vsignal
V
Signal path
Return path
• Characteristic impedance
w = 10 mils
h = 5 mils
1
Z0 ~
CL
Z1 Z2
Z is instantaneous impedance
V incident V reflected V
transmitted
Z1 Z2
Vtransmitted 2 x Z2
S21 = =
Vincident Z2 + Z1
V incident V reflected V
transmitted
Vtransmitted 2 x Z2
S21 = =
Vincident Z2 + 50Ω
Insertion Loss
Magnitude
and phase
Detector
DUT Z0 = 50Ω
50Ω Z 0 = 50Ω
V
~ magnitude/
source 50Ω phase
detector
+
+
S21 ~ Z0 sinh ( γ x len)
• Depends on:
9 Change in characteristic
Vtransmitted 2 x Z2
S21 = =
impedance from 50 Ohms
9 Time Delay
9 Highest frequency Vincident Z2 + Z1
Z2 1st order:
2 x Z2 2 x 50 200 x Z2
IL = x =
50 + Z2 50 + Z2 (50 + Z 2 )2
IL1 IL2
2st order: x 2 to account for the second reflection
2 x Z2 2 x 50
IL1 = IL2 =
Z2 + 50 50 + Z2
© Eric Bogatin 2005
Signal Integrity of Sockets-Simplified Slide - 44
Rough Estimates
0
-1 1st order
-2 2nd order Estimates
Insertion Loss (dB)
-3 Z0 Insertion Loss
-4 70 Ohms -0.5 dB
-5
-6 30 Ohms -1 dB
-7 20 Ohms -3.2 dB
-8 2nd order
-9
-10
0
0 10 20 30 40 50 60 70 80 90 100
Impedance (Ohms)
Insertion Loss, dB
-1
-2
-3
2nd order estimate: for > -1 dB insertion loss,
keep 30 Ohms < Z0 < 80 Ohms -4
-5
0 2 4 6 8 10 12 14 16 18 20
Z0 = 80 Ohms
Insertion Loss, dB
-1
Z0 = 30 Ohms
Z0 = 20 Ohms
-2
-3
-4
Simulated with Agilent ADS
-5
0 2 4 6 8 10 12 14 16 18 20
freq, GHz
0
Insertion Loss, dB
-1
-2
Total length = 0.2 inches
-3 30Ω 80Ω
-4
-5
0 2 4 6 8 10 12 14 16 18 20
freq, GHz
• Low frequency behavior is related to ~ average impedance- can be better than either one
• Highest insertion loss can be much worse than either discontinuity (> 3x)
-2
-3 30Ω 80Ω
-4
Total length = 0.2 inches
-5
-6
-7 30Ω 80Ω 30Ω
-8
0 2 4 6 8 10 12 14 16 18 20
freq, GHz
• Low frequency behavior is related to ~ average impedance- can be better than either one
• Highest insertion loss can be much worse than either discontinuity (> 7x)
2.5 nH L discontinuity
10%-90% RT = 2.2 x L/R = 2.2 x L/(2 x Z0) ~ L/Z0 = 2.5 nH/50 ~ 50 psec
0
-1
C = 0.5 pF
Insertion Loss, dB
-2
-3
-4
-5
C = 1 pF
-6
-7
-8
-9
Simulated with Agilent ADS
-10
0 2 4 6 8 10 12 14 16 18 20
freq, GHz
Z0 = 50 Ohms wo caps
Z0 = 50 Ohms w caps
0
-1
Insertion Loss, dB
-2
-3
-4
-5
-6
-7
-8
-9 Simulated with Agilent ADS
-10
0 2 4 6 8 10 12 14 16 18 20
• Pad capacitance can easily dominate freq, GHz
insertion loss measurements
• For matched socket, best performance
is with no pad capacitance
0
Insertion Loss, dB
-1
Z0 = 30 Ohms wo caps
Z0 = 80 Ohms w caps
Z0 = 50 Ohms w caps
-2 Z0 = 30 Ohms w caps
freq, GHz
• Lower the socket impedance, the greater the impact from pad capacitance
• An optimized socket will be degraded by pad capacitance
0
-1
Insertion Loss, dB
-2
-3
-4
-5
-6
-7
-8
-9 Simulated with Agilent ADS
-10
0 2 4 6 8 10 12 14 16 18 20
freq, GHz
Insertion Loss, dB
-1
-2
-3
-4
-5
9 Reflections from front and back, 180 deg out of phase freq, GHz
9 No reflection Simulated with Agilent ADS
9 All transmitted waves in phase and add
9 Max transmission
• When Len = ¼ λ
9 Reflected waves from front and back add
9 Maximum reflected signal
9 Transmitted waves 180 deg out of phase
9 Minimum transmitted signal
Z0 = 20 Ohms
Len = 10 mils (0.25 mm), 100 mils (2.5 mm), 500 mils (12.5 mm)
0
Insertion Loss, dB
-1
-2 If Len << ¼ λ,
-1 dB insertion loss BW will
-3 be higher
-4
freq, GHz
• If worst case insertion loss is less than -1 dB, TD may not be important
• If worst case insertion loss is greater than -1 dB, keep length << ¼ λ
• Minimize insertion loss by keeping length << ¼ λ
• Shorter is better, but long may be good enough
0.0
-0.2
Insertion Loss, dB
-0.4
50 Ohms, tan(δ) = 0.02
-0.6
-0.8
30 Ohms, tan(δ) =0
-1.0 30 Ohms, tan(δ) = 0.02
-1.2
-1.4
-1.6
-1.8
Simulated with Agilent ADS
-2.0
0 2 4 6 8 10 12 14 16 18 20
freq, GHz
Current Distributions
signal
return
1 1
δ= = 2µ f in GHz
σπµ0µr f f
in copper
@ 1000 MHz
Conductor Loss
⎛ RL ⎞
α = - 4.34⎜⎝ + G LZ 0⎟⎠ dB/length
Z0
Conductor Dielectric
α in dB/in loss loss
f in GHz
w is perimeter in mils OD = 10 mils
Z0 in Ohms w = 3 x 10 = 30 mils
-22 f
α conductor ≈ dB/in @ 10 GHz, 0.1 inch long
Z0 w S21 ~ -0.02 dB/inch x 0.1 inch x 3 = ~-0.01 dB
⎛ RL ⎞
α = - 4.34⎜⎝ + G LZ 0⎟⎠ dB/length S 21 ~ −4.34 x
Rdc
~−
Rdc
Z0 50 10
Conductor Dielectric
loss loss
-0.2
-0.3
-0.4
-0.5
-0.6
-0.7
-0.8
-0.9 Simulated with Agilent ADS
-1.0
0 2 4 6 8 10 12 14 16 18 20
freq, GHz
Power Integrity
power
To
regulator Cdecoupling
ground
• Goal: keep the voltage across the power pins constant, even with
current surges
• Strategy: minimize the impedance of the power distribution
• At high frequency, Z = R + iωL
• L is the loop inductance of the power and ground return path
What is inductance?
Inductance Principles -1
Current
section of wire
length, d
Right hand rule
# of field lines ~ current in the wire determines
direction
Inductance Principles -2
Loop Inductance
La
L meter Lab I
Lloop = La + Lb − 2Lab
1. Short lengths
2. Wide conductors
3. Closely spaced return path
4. Multiple power-return
conductors in parallel
• Examples:
9 Z0 = 50 Ohms, Len = 3 mm Lloop = 6.8 x 50 x 3 = 1 nH
9 Z0 = 50 Ohms, Len = 1.5 mm Lloop = 6.8 x 50 x 3 = 0.5 nH
9 Z0 = 20 Ohms, Len = 3 mm Lloop = 6.8 x 20 x 3 = 0.4 nH
9 Z0 = 50 Ohms, Len = 0.1 mm Lloop = 170 x 50 x 0.1 = 0.8 nH
9 Z0 = 70 Ohms, Len = 0.15 mm Lloop = 170 70 x 0.15 = 1.8 nH
• Signal integrity
9 Insertion loss
9 Bandwidth
9 Characteristic impedance
9 Time delay
9 Dielectric loss
9 Conductor loss
• Power integrity
9 Loop inductance
• Other:
9 Return loss
9 Differential impedance
9 Cross talk
9 Ground bounce
The End
Thanks for
listening!
www.BeTheSignal.com
Online Lectures
Feature Articles
PCD&M Monthly Signal Integrity
Column: “No Myths Allowed”
Master Class Workshops
Resources
Live classes through GigaTest Labs
Published by Prentice Hall, 2004