Conformity - November 2008 - (Page 30) loop voltage relative to the injected current. The reason is the high pass behavior of the induced loop voltage and the field coupling. Thus, the generator model needs to be sufficiently accurate up to a few GHz. To better visualize the frequency range of validity, a comparison of the spectra was performed. The data shown in Figure 6 indicate a usable frequency range of up to about 3 GHz. Greater precision modeling of the components around the relay structure could provide a better match of the simulated waveform to the measured data in the frequency range of 0.6 to 1.4 GHz. Applications Application #1: Current in a Wire Attached to a Hand Held Device If the device under test is small (e.g. a PDA), then any attached wires will change the currents on the device significantly. As a first step, the current in a cable attached to such a device has been modeled. Due to the high density of components in the device, and as no internal voltages are asked for (see next application for internal voltages), it is possible to model the device as a single metallic block. The wire has been suspended above the ground by a 2 cm thick dielectric (Styrofoam). Attempts to use a 0.5 mm thick dielectric, as required by the standard, led to numerical difficulties that could not be resolved in time for this article. It is subject to intensive investigations. Figure 7 shows the simulation model of the ESD generator with a small handheld device and a cable attached to it. To reduce the domain size of the simulation model, a short ground strap was used. This changes the tail current of the ESD generator wave form, but experience shows that the tail current has little influence on disturbances. Figure 4: Simulated and measured ESD discharge current, for the first 10 ns Figure 3: Details of the pulse forming elements The current was measured in the middle of the first segment using an F-2000 current probe, and is compared to the simulation in Figure 8. The frequency response of the current probe was compensated by de-convolving the measured time domain data. Application #2: Voltage Within the Hand Held Device A significantly more difficult problem is the prediction of the current within a small device. This requires modeling the significant elements of the device. A further complication is posed by our desire to compare with measurement results. Measuring voltages inside a device while performing ESD tests is a very challenging task. It requires very good suppression of any common mode currents, and optimal shielding of the oscilloscope and all connecting cables. We decided to use the voltage between a part of the shield and the PCB as a test case. An off-the-shelf small hand held device was used for the measurements and full wave simulation. Figure 5: Measured and simulated induced loop voltage 30 Conformity november 2008 Figure 9 shows the simulation model of the hand held device. The top plastic half of the hand held device contains a LCD
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.