Testing electrical systems for electrostatic discharge (ESD) robustness usually involves using IEC 61000-4-2 as a benchmark. This standard defines the stress current waveforms for each voltage level, how to calibrate the ESD pulse source, the test environment for the measurements, and the pass and fail criteria. It also provides guidance on how to perform the tests.
But when performing ESD testing on electrical systems, it’s not clear how much stress the unit under test is actually withstanding. This is especially true for portable products in which there’s no ground connection during the stress condition. Still, it’s possible to measure the actual stress current during the ESD testing of a handheld, battery-powered product. It’s even possible to employ simple calculations to show how additional information is obtainable from the measurements taken.
SYSTEM-LEVEL ESD TESTING PROCEDURES
When dealing with small products, engineers most often perform system-level testing for ESD in a specialized test environment (Fig. 1). For IEC 61000-4-2 testing, this consists of a metal ground plane on the floor, a wooden table, a metal horizontal coupling plane on the table with a 0.94-M connection to the ground plane, and a 0.5-mm insulator on top of the horizontal coupling plane.
Necessarily, such a test environment enables repeatable results. First, place the equipment under test (EUT) on top of the insulating surface, and then apply stress to the EUT in a number of different ways. For instance, apply contact discharge to conducting surfaces, including all of the metal cases and the grounded metal shells of the connectors. Also, deliver air discharge to surrounding insulating surfaces with an emphasis on likely ESD paths, such as seams in the casing of the device as well as all and any vent holes and keypads.
Indirect discharge tests are part of the job, too, particularly on the horizontal and vertical coupling planes to simulate the effects of electromagnetic interference (EMI) caused by ESD events in nearby objects. The difficulty that often arises and complicates matters for the design engineer is that the amount of actual stress applied to the EUT isn’t always obvious during these measurements.
ESD MEASUREMENT IN A PORTABLE PRODUCT
Our example here uses a portable, battery-powered personal digital assistant (PDA). First, deliver stress in contact discharge mode to the metal ground shell of the USB port. Then measure current with a transformer-type current probe specifying a 1-GHz upper bandwidth (the Fischer Custom Communications F-65A being most desirable) and connect to a standard 1-GHz bandwidth oscilloscope. Note that the internal diameter of the probe needs to be large enough to fit around the 12-mm diameter tip of an IEC 61000-4-2 compliant ESD gun.
To simplify descriptions in this particular example, take measurements with reference to a 0.6-meter square ground plane directly on a tabletop as opposed to using the full IEC test setup (Fig. 2). The ground strap of the ESD gun connects to one corner of the ground plane. All measurements are taken at a voltage level of 8 kV. The first measurement is directly to the center of the ground plane. Results of these measurements employ an expanded time scale.
In a second measurement procedure, the test engineer places the PDA face down on the ground plane so the metal shield of its miniature USB connector is easily accessible. The current into the PDA is much lower than the current injection directly into the ground plane (Fig. 3). To further reduce the current, insert a 0.9- cm insulator between the PDA and the ground plane.
The current reduction witnessed isn’t uniform in nature, though. The initial current spike is lower by 10% for the PDA directly on the ground plane and 33% for the PDA on top of the 0.9-cm insulator (Fig. 4). At 20 ns, the stress on the PDA directly on the ground plane is lower by 69% and the stress to the PDA on the insulator is lower by 93%. The reduction in current is the result of the PDA charging up during stress conditions. After each stress procedure completes, we must ground the PDA to return it to an uncharged state before the next measurement.
The schematic in Figure 5 is quite viable for a quantitative understanding of the measurements. The 150-pF capacitor and the 330- resistor are the standard circuit elements for an IEC 61000-4-2 compliant ESD gun. The parasitic capacitance between the gun and the ground plane provides the initial current spike included in the IEC 61000-4-2 current waveform. It exhibits a capacitance value of just a few picofarads, which we can ignore in the quantitative discussions.
The engineer places the current probe around the tip of the discharge gun and initiates a discharge directly to the ground plane, represented by the short seen in the diagram. Integrating the current for the discharge in Figure 3 directly to the ground plane out to a time period covering 300 ns yields a capacitance of 1.21 µC, which is almost exactly the same figure predicted for a 150-pF capacitor charged to 8 kV.
Discharge to the PDA is represented via a parallel plate capacitor, with one plate being part of the PDA and the other being the ground plane. Early in the pulse, the PDA capacitance provides low impedance and the current injected into the PDA is similar to the discharge to ground. As the pulse progresses, charge accumulates in the PDA-to-ground plane capacitor. The potential on the PDA will therefore rise until the potential between the PDA and the ground plane and the voltage across the 150-pF capacitor are equal. At this point, no further current flows, even though the ESD gun’s capacitor hasn’t fully discharged.
Integrating the current for the discharge to the PDA provides the amount of charge transferred from the ESD gun to the PDA. Knowledge of the charge transferred from the ESD gun to the PDA allows for calculating the voltage remaining on the ESD gun’s 150-pF capacitor and, in turn, revealing the voltage on the PDA. Also, being aware of both the voltage and the measured charge on the PDA, it’s now possible to calculate the capacitance between the PDA and the ground plane (see the table).
The measurements indicate that the PDA directly on the ground plane charges to 6253 V, while the PDA on top of the 0.9-cm insulator charges to 7381 V. These figures correspond to capacitances of 41.9 pF and 12.6 pF for the two cases. A capacitance measurement with the PDA directly on the ground plane using a 1-kHz inductance, capacitance, and resistance (LCR) meter yielded a value of 34 pF. This level of agreement is considered quite reasonable given the differences in measurement technique.
The measurements show that the current from an ESD gun is measurable during ESD stress of electrical systems. In this case, we saw how the amount of stress on a portable system such as a PDA or mobile phone can be much less than the full stress available from the ESD gun.
A simple circuit model and calculations based on the measured current can produce useful information, such as the charge delivered to the system being tested, voltage that the system was raised to, and an approximation of the system’s capacitance to ground. The peak current in the initial current spike doesn’t decrease as much as the remainder of the current waveform.
This finding is also consistent with the model in which the PDA’s capacitance to ground provides a low impedance path to ground for the initial current spike. This measurement technique is extendable to air discharge, provided that accidental discharge to the current probe isn’t interpreted as discharge to the device under test.