TPD1E0B04DPYEVM: Using the ESD Eval Board to Confirm Protection Performance

Part Number: TPD1E0B04DPYEVM

We recently purchased a TPD1E0B04DPYEVM to test the performance of the ESD diode before placing it in our design. We followed the test procedure as stated in the user manual except for a few deviations:

1. We generated the ESD strike while the tip of our ESD gun was on the trace between the SMA connectors, not directly on the center pin of J1. The tip of our ESD gun is too large to place directly onto the center pin.

2. We placed two 30 dB attenuators in line between J2 and the oscilloscope. This gives us a 1000:1 voltage division to protect our oscilloscope.

The clamping voltages we recorded were different than what the TPD1E0B04DPY datasheet claimed. We recorded a clamping voltage spike of +226 V on a +8kV strike while the datasheet claims that it should be clamped to about +102 V. We also recorded a clamping voltage spike of -234 V on a -8kV strike while the datasheet claims that it should be clamped to about -106 V.

I can provide pictures of our test set up and of the waveforms that are being generated if it helps.  

Thank you!

3 Replies

  • Hello,
    Yes please send a picture of your set up and resulting waveform.
    Also 1000:1 ratio seems too high. The resolution seems like it would be very poor. However I do not know if that would cause it to be so different
    Also just as a check are you sure the device is soldered down on the pad?
  • In reply to Cameron Phillips:

    Hi Cameron,

    I've attached a schematic of the setup as well as pictures of our test setup. A few questions:

    1. We have J3 and the ground port of the ESD gun connected to earth ground. Should J3 be connected to earth ground or directly connected to the ground of the ESD gun?

    2. The scope we are using only has a bandwidth of 350 MHz. Will this make a large impact in the clamping voltage we are recording? We have a 1 GHz oscilloscope available, but I am not sure about a 2 GHz oscilloscope. Would you recommend switching to the 1 GHz bandwidth scope?

    3. If the 1000:1 voltage ratio seems too high, would you recommend using 40 dB of attenuation before the voltage signal gets to the oscillscope? This would make it 100:1. 

    I'll post a picture of the resulting waveform tomorrow as the main ESD tester is out today (he has all of the oscilloscope data). I don't believe there is a connection issue, but I'll double check.

  • In reply to Erik Dekelbaum:

    Did you get the waveforms?
    Also A higher sampling frequency would help. However, what is the goal overall with the exercise? Are you trying to see if the device will protect a downstream device? Also it is typically used that the 30ns voltage during an ESD event is the clamping voltage not the first peak at 1-2ns.