Why Drain Time Is Important to ESD

Drain time is an important parameter in ESD control. Ignoring it can be almost as detrimental to your product as having no ESD-control program at all.

For example, the electrical field from a charged static-protective bag could damage an ESD-sensitive (ESDS) item, according to the Field-Induced Model (FIM). In this case, most of the charge must drain onto a grounded operator’s hand while he or she is manipulating the bag before adding or removing an item. Similarly, a conductive tote box must drain its charge onto a static-dissipative benchtop before an ESDS item can be added or removed.

A short drain time is also needed for conductive or static-dissipative shoe soles to discharge to a conductive floor and minimize the skin voltage of personnel. This prevents ESD damage, according to the Human Body Model (HBM). In contrast, the drain time from the lead or contact of a charged ESDS item to whatever surface it touches must be long enough to slow or “soften” the discharge to prevent damage, according to the Charged Device Model (CDM).

For judging the acceptability of ESD-control products such as bags, tote boxes, benchtops and floors, the drain time must be:

Defined.

Specified as a required maximum or minimum.

Measured either directly or indirectly.

For the purpose of defining drain time, a charged bag, tote box, person or ESDS item can be visualized as one plate of a capacitor draining through a resistor. Then:

ln (V_t / V_o) = -t / RC

where: V_t is the voltage at time t in seconds

V_o is the initial voltage (at t = o)

R is the resistance in ohms

C is the capacitance in farads

A suggested figure of merit for t is the RC time constant, so that V_t / V_o = 0.37 and drainage is 63% complete. But in this discussion, 90% drainage is assumed necessary to drop V_t to a safe level. ¹ Therefore:

drain time = t = [-ln(0.1)] RC = (2.303) RC

This definition of t, like any assumption, will not fit all situations. Perhaps V_o = 5,000 V, and a residual V_t of 500 V remains damaging. Or V_o = 300 V and the time to reach V_t = 30 V is needlessly long for an ESDS item sensitive to 100 V. Any generalized requirement is merely a guideline.

Given this definition of t, what maximum or minimum is required for various operations? The maximum time of 2 s for bags or tote boxes, as shown in Table 1, is imagined as preceding transfer of an ESDS item into or out of the container.

In contrast, the much shorter maximum of 0.1 s is proposed for suppressing surges of voltage on a walking person. The minimum time of 0.1 s for draining the lead of a device, such as a dual-in-line package (DIP), is estimated based on testing in which V_o was 500 V to 1,000 V. ²

Now comes the difficult problem of measuring or estimating drain time by a standardized test. The Electronic Industries Association (EIA) 541 static decay test monitors the electrical field, not the charge itself, so a conductive layer within a material will collapse the field by capacitive coupling or voltage suppression in a misleadingly brief time. ¹

Meanwhile, if the same material has a nonconductive surface, the charge (surplus or deficit of electrons) will not drain at all. This hidden charge, or cryptocharge, might cause ESD damage. ³

Consequently, the estimation of the drain time by the EIA 541 decay test should be confined to materials without a buried conductive layer, such as conventional antistatic polyethylene. But even then, the arbitrary conditions of the test will seldom match reality. For example, the charge on an antistatic bag may drain onto an operator’s fingers gently touching the bag (relatively high resistance) or grasping it tightly (lower resistance).

In another example, a charged conductive tote box resting firmly on a hard static-dissipative benchtop will drain much faster than a warped box making only point contact. In the equation t = (2.303) RC, R is much higher for the warped than the unwarped box. C also can vary widely, even during the draining process, if the box tilts toward the surface while making contact. As the tilt angle decreases, the dielectric layer (air gap) of the box-air-ground capacitor becomes thinner and C rises.

Considering all the imponderables in any direct measurement of decay or drain time, t per se might as well be ignored in favor of a guideline R as in the last column of Table 1. That is, the required t is implicit in R.

A suggested standard method of measuring R is the flooring resistance test of NFPA 99 or ASTM F 150 with the exception that the electrodes (5 lb, 2.5″ dia) are placed 2″ apart (between edges), rather than 3 ft apart, and readings are taken at 100 V, rather than 500 V, to represent the final stages of drainage.

The 2″ spacing simulates a surface resistivity (W /sq) determination in which the word “surface” refers to electrode placement rather than the locus of current flow. This close spacing allows conveniently small samples to be tested. Tests showed relatively little effect when the spacing varied between 1″ and 10″.

Table 2 gives data for the proposed method using two NFPA 99 electrodes. Also, the resistance was measured between an NFPA 99 electrode and a probe of copper wire, representing a DIP lead, placed 2″ from the edge of the electrode. ²

Observations drawn from Tables 1 and 2 are:

Resistance readings were often lower at 100 V than at 10 V, and voltage dependence of R complicated a calculation of t. This is one reason why guideline resistance values can be only approximate.

Copper probe readings were sometimes erratic, since R might be in an actual discharge from a device lead and a calculation of t would be uncertain.

The copper probe identified small regions of differing conductivity in Material 5. But the NFPA 99 electrodes, with their relatively large area, gave an averaging effect and more reproducible readings, suitable for a guideline.

The ratio of copper probe resistance to standard NFPA 99 electrode resistance varied from approximately 10 to 10³. A factor of 10² was chosen for calculating the guideline of 10⁸ W , minimum, in the last column of Table 1.

The guideline resistance of 10¹¹ W , maximum, for drainage of bags was met by all the materials.

The guideline resistance of 10¹⁰ W , maximum, for drainage of tote boxes was met by Materials 8, 9 and 10 for benchtops.

The guideline resistance of 10⁸ W , maximum, for drainage through shoe soles was met by Material 5. Material 3, which did not meet the guideline, was intended merely as a backup for reducing, not eliminating, charges on personnel.

The guideline resistance of 10⁸ W , maximum, for CDM safety was met by all the materials except 5, 6 and 10. Test results with MOSFETs sensitive to 100 V support this guideline. ²

The effect of relative humidity (RH) on R for Materials 1, 2, 3, 4 and 7 must be considered. For example, Material 4 becomes nonconductive (surface resistivity greater than 10¹² W ) when the RH falls below approximately 8%.

In conclusion, a standard test is suggested for measuring surface resistance as an indirect estimate of drain time. Order-of-magnitude maximum or minimum resistance values are proposed for preventing damage from ESD events modeled by the FIM, HBM and CDM.

References

1. Baumgartner, G., “Electrostatic Decay Measurement Theory and Applications,” ESD Association and IEEE EOS/ESD Symposium Proceedings, Las Vegas, NV, 1995, pp. 262-272.

2. Kolyer, J. M., and Watson, D. E., “CDM and Work Surface Selection, Part II,” EE-Evaluation Engineering, October 1991, pp. 110-117.

3. Kolyer, J. M., and Watson, D. E., “Hidden Charges on ESD-Protective Packaging,” EE-Evaluation Engineering, September 1992, pp. 94-100.

About the Author

John M. Kolyer joined Rockwell International in 1973 and today is a Senior Engineering Specialist in ESD control and nonmetallic materials sciences. He received a Ph.D. in chemistry from the University of Pennsylvania in 1960. Dr. Kolyer also has authored many papers and articles and a book on ESD control. Rockwell International, Autonetics and Missile Systems Division, 3370 Miraloma Ave., Anaheim, CA 92803,

(714) 762-6144.

Table 1

Table 2

Copyright 1996 Nelson Publishing Inc.

August 1996