Content uploaded by Jawad Yousaf
Author content
All content in this area was uploaded by Jawad Yousaf on Feb 15, 2018
Content may be subject to copyright.
Effect of ESD Generator Ground Strap Configuration
on ESD Waveform
Jawad Yousaf, Jaeyoung Shin, Rao Leqian, Wansoo
Nah*
Department of Electrical and Computer Engineering,
Sungkyunkwan University, Suwon, Republic of Korea
*wsnah@skku.edu
Jinsung Youn**, Daehee Lee, Chanseok Hwang
Design Technology Team, Memory Division
Samsung Electronics Co, Ltd
Hwaseong, Republic of Korea
**jinsung.youn@samsung.com
Abstract—In this study, an analysis of the effect of the
electrostatic discharge (ESD) gun grounding strap on the
generated ESD stress waveform is presented. The reference ESD
waveform is measured using the standard calibration setup of the
IEC 61000-4-2 standard for the different (round, short and long)
configurations of the grounding strap. The post processing
analysis shows that increasing the length of the simulator ground
strap to 3m reduces the variations in the ESD w aveform,
particularly ringing after the first peak, in comparison with
standard reference waveform characteristics.
Keywords—Electrostatic Discharge (ESD) Gun, ESD standard
waveform, IEC 61000-4-2
I.
I
NTRODUCTION
System level ESD testing of equipment under test (EUT)
requires the generation of the standard stressing waveform
using an ESD simulator or ESD gun. An ESD simulator is used
for mimicking a typical human-metal charged scenario as per
the standard requirements [1], [2]. The main function of the
ESD gun is the zapping of the ESD signal of different levels on
the EUT to check its immunity against the stressed waveform.
The IEC 61000-4-2 standard describes the procedure for the
calibration of the injected waveform by the ESD gun. The ESD
generator waveform is verified for the different test levels as
per the defined detailed setup in [1].
The generator characteristics are dependent on the type of
generator, experimental setup including the discharge current
return cable, and the environmental conditions [1], [3]. The
ESD generator is equipped with a 2m long grounding cable
which provides the return discharging path to the gun at the
time of the injection. In the standard setup, the discharging
cable is connected to a specific point on the metallic board
containing the standard 2Ω ESD current target. It forms a loop
which determines the waveform ringing after the first peak and
variations in the second and higher peaks of the generated ESD
waveform. The characteristics (peak value of current, current at
30 ns, and 60 ns timing and rising time) of the discharged
current waveforms must be within the specified limits of the
standard [1].
The grounding return path plays a vital role for the
successful verification of the ESD stressing waveform as per
the standard requirements. The grounding cable length,
position, formed loop, and its connection position to the
grounding plane, all have an impact on the waveform
characteristics [1], [3]. The grounding cable of the generator
has to be positioned in a well-defined manner to assure the
generation of the well-specified shape of discharged ESD
waveform [3].
The standard ESD generator has a 2m long discharge gun
return cable. However, the length of the cable can be increased
to 3m for the testing of either table top or floor-standing tall
equipment [1], where the 2m discharge cable is insufficient in
length. Also in actual ESD testing, the connection point of the
return cable is the electrical earth ground for the system level
ESD testing.
In this study, variations in the standard ESD waveform are
analyzed for different length (2m and 3m), loop configurations
(standard loop and modified loop) and ground connection
points of the discharge return cable. The comparison of
measured waveforms for the stated configurations with the
standard IEC 61000-4-2 setup is demonstrated. The details of
the adopted measurement strategy, measurement setup, and
analyzed result are given in the following sections.
II.
MEASUREMENT STRATEGY
This section describes the details of the measurement
strategy and measurement setup. Figure 1 shows the
measurement setup for different cases of the analysis.
Fig. 1 ESD waveform measurement configurations.
978-1-5386-3912-2/17/$31.00 ©2017 IEEE
120 121
2017 Asia-Pacic International Symposium on Electromagnetic Compatibility (APEMC), June 20-23, 2017, Seoul, Korea 2017 Asia-Pacic International Symposium on Electromagnetic Compatibility (APEMC), June 20-23, 2017, Seoul, Korea
978-1-5386-3912-2/17/$31.00 ©2017 IEEE
Fig. 6 Measured waveform results for the round 2m loop configuration of Case
C (Figure 3 (b)).
Fig. 7 Measured waveform results for the long 3m loop configuration of Case
D (Figure 3 (c)).
IV. A
NALYSIS
This section presents the comparison of the waveform
characteristics for all cases with the ideal waveform
parameters. Figure 8 depicts the comparison of the ESD
stressing waveforms for the 4kV level for all four cases. The
details of the waveform parameters of these cases are given in
Table 1. It can be observed from Table 1, that for all the
modified grounding configuration cases, the rise time, peak
current value, and current at 30 ns are within the ideal
threshold values. However, deviations occur for the current
values at 60 ns for the cases where the gun discharge return
cable is directly connected to the earth ground. However for
the waveform characteristics up to 30 ns, the best results are
achieved for the longer ground loop configurations i.e., Case
D where the ringing in the waveform after the 1
st
peak is also
minimum as compared to other cases. The small variation at
60 ns for Case D could be reduced by grounding the cable at
point P on the load resistor mounted board as like in Case A.
The increase in the length of the return grounding loop
enhances the inductance of the ground which produces less
deviations from the ideal values. The reduction in the ringing
effect of the waveform after the first peak in Case D is a
noticeable observation; as it is quite difficult to reduce this
waveform ringing in all commercial ESD generators [4]. The
ideal ESD waveform as defined by [1], does not have this
waveform ringing. Therefore for a more ideal generation of
the ESD waveform, especially using commercial ESD
generators, a 3m long ground strap can be used.
Fig. 8 Comparison of waveforms for different grounding configurations for the
+4 kV discharge voltage.
TABLE I. C
OMPARISON OF
4
K
V
W
AVEFORMS PARAMETERS FOR
DIFFERENT GROUDING CONFIGURTATIONS
.
Rising Time (ns)
Peak
I
(A)
30ns
I
(A)
60 sn
I
(A)
Ideal
0.8
(0.6 ~ 1)
15
(12.8 ~ 17.2)
8
(5.6 ~ 10.4)
4
(2.8 ~ 5.2)
Case A
0.7
14.6
9.3
4.9
Case B
0.8
14.6
9.0
5.3
Case C
0.7
14.8
9.3
5.5
Case D
0.7
15.0
6.6
5.3
V. C
ONCLUSION
This study has described the effect of changes in the ground
loop configurations of the ESD gun discharge return cable on
waveform characteristics. The analysis of experimental results
shows that the ESD waveform characteristics are highly
dependent on the length, loop, and ground point of the return
cable. The increase in the length and grounding area of
discharge return cable results in lesser ringing of discharge
waveform after the first peak and less deviation of the
waveform parameters from the ideal values. Future study can
be done to evaluate the 3m long cable effect for the ESD guns
of different manufactures.
A
CKNOWLEDGEMENT
This work was financially supported by Semiconductor
Industry Collaborative Project between Sungkyunkwan
University and Samsung Electronics Co. Ltd
R
EFERENCES
[1] Electromagnetic Compatibility (EMC) - Part 4-2: Testing and
measurement techniques – Electrostatic discharge immunity test, IEC
Std. 61000-4-2, 2001.
[2] Bo Pu et al., "Modeling and prediction of electromagnetic inmunity for
intergrated circuits", Journal of electromagetic engineering and science,
vol.13, no.1, March, 2013, pp54-61
[3] Caniggia and F. Maradei, "Circuit and Numerical Modeling of
Electrostatic Discharge Generators," in IEEE Transactions on Industry
Applications, vol. 42, no. 6, pp. 1350-1357, Nov.-dec. 2006.
The ESD signal is injected on the standard 4GHz 2Ω load
target using a NoiseKen TC-815 gun, in the contact discharge
mode. The ESD current can be measured across the 2Ω load,
which is connected to the LeCroy WP7300 3GHz oscilloscope
through a 26dB attenuator for protection of the oscilloscope.
The ground of the oscilloscope is connected to earth ground
for the common reference ground of the measurement setup.
The ESD waveform is measured for four different
configurations of the grounding return cable of the generator.
Case A is the IEC 61000-4-2 standard configuration in which
the 2m long return cable is pulled up right backward and then
connected to the connection point (P) on the ground plane of
load resistor mounting board. Figure 2 depicts the actual
measurement setup of Case A.
Fig. 2 Standard ESD waveform measurement setup (Case A: 2m standard
grounding loop configuration).
Fig. 3 Modified grounding strap configurations, (a) Case B: 2m short loop, (b)
Case C: 2m round loop, (c) Case D: 3m long loop.
In Case B, the grounding strap is pulled up right back in
ordinary way to form a short loop of ground strap and then is
directly connected to the earth ground at point Q as can be
noted in Figure 1 and Figure 3 (a). Case C is similar to the
standard Case A, but here the grounding strap is attached to
point Q on the earth ground for the formation of longer loop as
compared to standard setup (Figure 3 (b)). The length of the
grounding strap is kept to be 2m long for all three cases A, B
and C. In the last configuration (Case D) as shown in Figure 1
and Figure 3 (c), the length of grounding strap is increased to
3 meter and is connected to the earth grounding point Q,
pulled up right back as in Case A. The load resistor mounting
board is also connected to the earth ground plane at the same
point Q. The return cable is connected to the electrical earth
ground point Q as in actual system level ESD testing and the
ESD generator is connected to the earth ground. Also we want
to investigate the effect of the formed short loop (in Case B)
and longer ground loop (in Cases C & D) as compared to the
standard loop of Case A on the peak current, waveform
ringing after the first peak, current at 30 ns and 60 ns, and the
rise time specification of the reference waveform.
III. M
EASUREMENT RESUTLS
The ESD waveform generator waveforms are measured for
all fours levels of standard voltages i.e. 2kV, 4kV, 6kV, and
8kV. The waveform characteristics, i.e., first peak current, rise
time, current at 30 ns and 60 ns are compared with the ideal
generator waveform characteristics in [1].
Figure 4 shows the measured results for the standard (Case
A) grounding configuration. It can be noted from Figure 4 that
the measured waveform characteristics for all voltage levels are
within the defined limits of IEC standard [1]. As shown in
Figures 5, 6 & 7, the change in the grounding point from the
target board to the earth ground in Case B (2m short ground
loop), Case C (2m long ground loop), and Case D (3m long
ground loop) takes the current at 60 ns out of the defined
threshold limit of the IEC standard.
Fig. 4 Measured waveform results for the standard 2m loop configuration of
Case A (Figure 2).
Fig. 5 Measured waveform results for the short 2m loop configuration of Case
B (Figure 3 (a)).
122 123
2017 Asia-Pacic International Symposium on Electromagnetic Compatibility (APEMC), June 20-23, 2017, Seoul, Korea 2017 Asia-Pacic International Symposium on Electromagnetic Compatibility (APEMC), June 20-23, 2017, Seoul, Korea
Fig. 6 Measured waveform results for the round 2m loop configuration of Case
C (Figure 3 (b)).
Fig. 7 Measured waveform results for the long 3m loop configuration of Case
D (Figure 3 (c)).
IV. A
NALYSIS
This section presents the comparison of the waveform
characteristics for all cases with the ideal waveform
parameters. Figure 8 depicts the comparison of the ESD
stressing waveforms for the 4kV level for all four cases. The
details of the waveform parameters of these cases are given in
Table 1. It can be observed from Table 1, that for all the
modified grounding configuration cases, the rise time, peak
current value, and current at 30 ns are within the ideal
threshold values. However, deviations occur for the current
values at 60 ns for the cases where the gun discharge return
cable is directly connected to the earth ground. However for
the waveform characteristics up to 30 ns, the best results are
achieved for the longer ground loop configurations i.e., Case
D where the ringing in the waveform after the 1
st
peak is also
minimum as compared to other cases. The small variation at
60 ns for Case D could be reduced by grounding the cable at
point P on the load resistor mounted board as like in Case A.
The increase in the length of the return grounding loop
enhances the inductance of the ground which produces less
deviations from the ideal values. The reduction in the ringing
effect of the waveform after the first peak in Case D is a
noticeable observation; as it is quite difficult to reduce this
waveform ringing in all commercial ESD generators [4]. The
ideal ESD waveform as defined by [1], does not have this
waveform ringing. Therefore for a more ideal generation of
the ESD waveform, especially using commercial ESD
generators, a 3m long ground strap can be used.
Fig. 8 Comparison of waveforms for different grounding configurations for the
+4 kV discharge voltage.
TABLE I. C
OMPARISON OF
4
K
V
W
AVEFORMS PARAMETERS FOR
DIFFERENT GROUDING CONFIGURTATIONS
.
Rising Time (ns)
Peak
I
(A)
30ns
I
(A)
60 sn
I
(A)
Ideal
0.8
(0.6 ~ 1)
15
(12.8 ~ 17.2)
8
(5.6 ~ 10.4)
4
(2.8 ~ 5.2)
Case A
0.7
14.6
9.3
4.9
Case B
0.8
14.6
9.0
5.3
Case C
0.7
14.8
9.3
5.5
Case D
0.7
15.0
6.6
5.3
V. C
ONCLUSION
This study has described the effect of changes in the ground
loop configurations of the ESD gun discharge return cable on
waveform characteristics. The analysis of experimental results
shows that the ESD waveform characteristics are highly
dependent on the length, loop, and ground point of the return
cable. The increase in the length and grounding area of
discharge return cable results in lesser ringing of discharge
waveform after the first peak and less deviation of the
waveform parameters from the ideal values. Future study can
be done to evaluate the 3m long cable effect for the ESD guns
of different manufactures.
A
CKNOWLEDGEMENT
This work was financially supported by Semiconductor
Industry Collaborative Project between Sungkyunkwan
University and Samsung Electronics Co. Ltd
R
EFERENCES
[1] Electromagnetic Compatibility (EMC) - Part 4-2: Testing and
measurement techniques – Electrostatic discharge immunity test, IEC
Std. 61000-4-2, 2001.
[2] Bo Pu et al., "Modeling and prediction of electromagnetic inmunity for
intergrated circuits", Journal of electromagetic engineering and science,
vol.13, no.1, March, 2013, pp54-61
[3] Caniggia and F. Maradei, "Circuit and Numerical Modeling of
Electrostatic Discharge Generators," in IEEE Transactions on Industry
Applications, vol. 42, no. 6, pp. 1350-1357, Nov.-dec. 2006.
The ESD signal is injected on the standard 4GHz 2Ω load
target using a NoiseKen TC-815 gun, in the contact discharge
mode. The ESD current can be measured across the 2Ω load,
which is connected to the LeCroy WP7300 3GHz oscilloscope
through a 26dB attenuator for protection of the oscilloscope.
The ground of the oscilloscope is connected to earth ground
for the common reference ground of the measurement setup.
The ESD waveform is measured for four different
configurations of the grounding return cable of the generator.
Case A is the IEC 61000-4-2 standard configuration in which
the 2m long return cable is pulled up right backward and then
connected to the connection point (P) on the ground plane of
load resistor mounting board. Figure 2 depicts the actual
measurement setup of Case A.
Fig. 2 Standard ESD waveform measurement setup (Case A: 2m standard
grounding loop configuration).
Fig. 3 Modified grounding strap configurations, (a) Case B: 2m short loop, (b)
Case C: 2m round loop, (c) Case D: 3m long loop.
In Case B, the grounding strap is pulled up right back in
ordinary way to form a short loop of ground strap and then is
directly connected to the earth ground at point Q as can be
noted in Figure 1 and Figure 3 (a). Case C is similar to the
standard Case A, but here the grounding strap is attached to
point Q on the earth ground for the formation of longer loop as
compared to standard setup (Figure 3 (b)). The length of the
grounding strap is kept to be 2m long for all three cases A, B
and C. In the last configuration (Case D) as shown in Figure 1
and Figure 3 (c), the length of grounding strap is increased to
3 meter and is connected to the earth grounding point Q,
pulled up right back as in Case A. The load resistor mounting
board is also connected to the earth ground plane at the same
point Q. The return cable is connected to the electrical earth
ground point Q as in actual system level ESD testing and the
ESD generator is connected to the earth ground. Also we want
to investigate the effect of the formed short loop (in Case B)
and longer ground loop (in Cases C & D) as compared to the
standard loop of Case A on the peak current, waveform
ringing after the first peak, current at 30 ns and 60 ns, and the
rise time specification of the reference waveform.
III. M
EASUREMENT RESUTLS
The ESD waveform generator waveforms are measured for
all fours levels of standard voltages i.e. 2kV, 4kV, 6kV, and
8kV. The waveform characteristics, i.e., first peak current, rise
time, current at 30 ns and 60 ns are compared with the ideal
generator waveform characteristics in [1].
Figure 4 shows the measured results for the standard (Case
A) grounding configuration. It can be noted from Figure 4 that
the measured waveform characteristics for all voltage levels are
within the defined limits of IEC standard [1]. As shown in
Figures 5, 6 & 7, the change in the grounding point from the
target board to the earth ground in Case B (2m short ground
loop), Case C (2m long ground loop), and Case D (3m long
ground loop) takes the current at 60 ns out of the defined
threshold limit of the IEC standard.
Fig. 4 Measured waveform results for the standard 2m loop configuration of
Case A (Figure 2).
Fig. 5 Measured waveform results for the short 2m loop configuration of Case
B (Figure 3 (a)).
122 123
2017 Asia-Pacic International Symposium on Electromagnetic Compatibility (APEMC), June 20-23, 2017, Seoul, Korea 2017 Asia-Pacic International Symposium on Electromagnetic Compatibility (APEMC), June 20-23, 2017, Seoul, Korea
