Clamshell Packages:

"Who Said Air Gaps Don’t Shield Against ESD Events?"

by: Steve Fowler, Fowler Associates, Inc

VPI Mirrex thought the issue needed investigating

In Ted Dangelmayer’s book, ESD Program Management (2^nd Edition) , he states that rigid packages that provide an adequate air gap are more effective for ESD and physical protection than "stressed" flexible packages. Ted has it right. Actually, Dan Anderson first said it so appropriately when he stated, "first, let there be no spark." This was very profound in its simplicity as well as its complexity. Dissipative clamshell packages do not support a spark at most reasonable levels. The energy that is eventually discharged in any ESD event to its surface is reduced to such a level to make it virtually impotent.

This is not to say that ESD shielding bags have no benefit. It is to say that with adequate air gaps the injected energy from external ESD events is sufficiently low to disallow most ESD problems. This means clear 3- dimensional thermoformable packages may be used with very positive success rates for all types of ESD packaging. At the request of VPI Mirrex, Fowler Associates began a study of the ESD susceptibility of productspackaged in clear, static dissipative clamshell packages. This study compared clamshell packages and static shielding bags using actual working boards from Alcatel as well as static event detectors such as the 2N4351 MOSFET ( 150 volt HBM sensitive).

Conclusions:
Our tests show conclusively that dissipative clamshell packages with at least an 1/8^th inch air gap to all ESD sensitive devices performed as well as and sometimes better than static shielding bags especially after the bags had been "handled" in a normal manner - up to ESD events of 20 kV (HBM). In addition, a modifed ESD S11.31 test was performed which injected 10kV instead of the normal 1 kV to subject the packages to higher levels of injected energy for more realistic comparisons. This test also showed conclusively that "air gap shielding" works.

Test Procedures

1.	Packaged functional boards were each subjected to 10 each positive and negative 20 kV HBM discharges directly above the most sensitive areas of the circuits. The boards were then put through a functional test to determine if they had been damaged. There were no failures of functional boards at these high ESD levels.
2.	2N4351 MOSFET IC’s were placed in the clamshell packages and subjected them to 10 each positive and negative 5, 10, 15 & 20 kV HBM discharges directly above the IC’s. The IC’;s were characterized after each discharge series. There were no failures of IC’s at these high ESD levels.
3.	2N4351’s were placed in Clamshell packages and subjected to increasing ESD events until the device failed. This was done in steps with new IC’s used at each step to eliminate any "wounding" of the components in each ESD step. With 1/8th inch or greater air gaps we did not see failures of the IC’s until we reached 35 kV.
4.	2N4351’s were placed in static shielding (metal-in) bags and subjected to increasing ESD events until the device failed. This was done in steps with new IC’s used at each step to eliminate any "wounding" of the components in each ESD step. With an 1/8th inch or greater air gaps we did not see failures of the IC’s until we reached 35 kV.
5.	A modified ESD S11.31 Static Shielding test was performed on both the static shielding bags and the clamshell packages. Both packages had injected energies of less than 0.3 nJoules.

Modified S11.31 Test Set Up

 

 

Figure 1. Human Body Model Test Set Up & 2N4351 MOSFET’s
10 "zaps" over each Printed Circuit Board and MOSFET at each polarity

Human Body Model Functional Tests
100 pF, 1500 Ohms

10 Pulses of Each Polarity for Each Voltage Level

Electrical Characterization of 2N4253 die TO-72 packages
VGS(th) @ TA = 25 OC , VDS = 10 V , ID = 10 mA, (Spec. Limit is 1 to 5 V dc)
IGSS @ TA = 25 OC , VGS = 15 V , VDS = 0 V (Spec. Limit = +/- 10 pA dc)

Functionality Tests	5 kV	10 kV	15 kV	20 kV
	pass/fail	pass/fail	pass/fail	pass/fail
Alcatel Board	--	--	--	pass
Alcatel Board	--	--	--	pass
Alcatel Board	--	--	--	pass
2N4351 MOSFET	pass	pass	pass	pass
2N4351 MOSFET	pass	pass	pass	pass

Human Body Model "Taken to Failure" Tests

10 Pulses of Each Polarity for Each Voltage Level
3 MOSFET’ s Tested at each level for each package type

2N4351 MOSFET’s	20 kV	25 kV	30 kV	35 kV
	pass/fail	pass/fail	pass/fail	pass/fail
Clamshell Package	3-pass	3-pass	3-pass	1-pass, 2-fail
Shielding Bag	3-pass	3-pass	3-pass	1-pass, 2-fail

ESD S11.31 Tests @ 1kV

	#1	#2	#3	#4
	nJ	nJ	nJ	nJ
Clamshell Package	0.22	0.28	0.20	0.10
Shielding Bag	0.19	0.25	0.27	0.19

References

1. D. Hines and W. Duncan, "Shrink Film Packaging Evaluation," EOS/ESD Symposium, EOS 12, Sept. 92, pp. 5A.1.1-4.
2. B. Hoffman, A. Hughes, J. Redden and D. Hines, "Expanded Evaluation Of Static Protective Shrink Wrap Film, " EOS/ESD Symposium, EOS 13, Sept. 93, pp. 173-176.
3. S. Fowler, "Pin-Hole ESD Susceptibility Tests on UNIT PAKS ", Test for Motorola, March, 1994.
4. S. Fowler, "ESD Susceptibility Tests on AirBox Packaged IC's," Test for Motorola, June, 1994.
5. S. Fowler, "Evaluation of Dissipative Shrink Films," Test for Motorola, Feb., 1995.
6. A. Warnecke and C. Deisch, "Evaluating Semi-Rigid Plastic Trays, " "EOS/ESD Technology ", April-May,1992, pp. 6-14.
7. S. Fowler, "ESD Susceptibility Tests on Tape & Reel Air BOX Packages" Test for Motorola, June, 1995.