Preliminary Application Note 2

CALIBRATING TLP SYSTEMS

This note is to provide information on the calibration method we have found to be the most useful for our TLP measurement system. During the development of our model 4002 TLP test system we devised this concept as the simplest method of verifying the accuracy of our measurement system.

This calibration method uses an open, and a short, which are readily available at most laboratories. We also use a 40 to 60 ohm resistor that only needs to be low inductance. A carbon or metal film resistor of 1/2-watt size performs well for this test. It should be measured to about 0.1 % which is easily accomplished these days.

The calibration resistor should be measured before and after a TLP resistance verification test to be certain that the resistance does not change from the test pulses. The DC measured value can be compared to the slope of the line that is plotted after that resistor is tested. Most film resistors will retain a constant resistance for the test pulse amplitudes that most TLP systems produce. If the resistor is damaged during testing, resistor testing can be limited to lower amplitudes or a more robust film resistor can be selected. Carbon composition resistors withstand pulses very well; but their stability is not as good as film resistors. The Model 4002 TLP test system includes an assortment of robust and stable resistors for this verification test.

If the slope of the I/V plot for the resistor verification is sufficiently different from the measured DC value, it can be factored into the values measured on DUT tests.

This test can also identify non-linear slopes that are in the TLP test system. If a non-linearity is found when verifying the resistance of a TLP tester, the resistor should be tested or a different construction resistor should be tested to be certain that the non-linearity is actually in the TLP system.

Once the short (usually a short length of #16 to #20 bare, tinned, or insulated copper wire) is tested, the I/V plot will clearly indicate the internal resistance of the TLP test system. Any non-linearity at high current levels that the TLP test system may have can also be immediately identified. This simple testing can immediately identify any non-linear voltage or current responses. If a TLP test system is found to have non-linear voltage or current responses, the system can then be redesigned to remove the non-linear voltage or current elements.

When the internal resistance of a TLP system is identified by measuring a "short", the resistance inherent in the measuring system itself can then be subtracted from the values measured for each DUT.

While these simple tests are very easy to accomplish, they are rarely done. Many homemade TLP systems, that have some of the problems described above are used as is without being improved. The users learn how to work around their TLP tester limitations, spending an inordinate amount of time in that effort. This time could be put to better use in designing ESD protection with data that can be relied upon for accuracy.

The DIP socket supplied with the Barth Model 4002 TLP system provides short and open pins for this verification test. The short test on wafer can be made by placing the needle tips on the same test pad or two pads that are connected with a low resistance shunt. Lifting one needle off the pad can provide the open circuit.

Return to Designer's Corner.