I have a circuit that provides power (1.8VD) to an FPGA and some ADCs. I have seen in the past where damage (ESD, EOS, etc) occurs to one of the ADCs. This causes a short from 1.8VD to GND internal to the ADC, which can be measured. The FPGA and PROM device are connected to this 1.8VD supply. Therefore, when 1.8VD is shorted to GND, the FPGA doesn't startup correctly.

I have seen a few of these boards when powered on, do not measure 1.8V but measure around 0.1V. However, when I measure the resistance from 1.8VD to GND, it appears nominal and no short is measured.

Is it possible that the ADC isn't fully shorted internally but still can hold this voltage rail down? Any thoughts, comments or insight would be helpful. I also suspect it is possible that they ADC is shorted internally to some other signal or plane other than GND.

  • \$\begingroup\$ What is nominal, in this case? \$\endgroup\$ – user253751 Feb 12 '19 at 0:50

You are experiencing a phenomenon where ESD voltage spikes enter your IC chip and cause one of two things to happen:

  1. Stray PNPN junctions on the silicon can latch up like an SCR and can cause massive current to flow where it is not intended and thus cause fusing and melt damage of the silicon.
  2. The voltage spike being high can also cause insulation layers on the chip to be punched through and create direct circuit shorts. These shorts can cause circuit malfunction or even outright silicon damage.

Note it is all too common that ESD damage to ICs cannot be measured with a typical ohms measurement with a DMM because the meter applies just a low voltage to the test connections. However when the chip attempts to power up the supply will get high enough to exceed some threshold for the silicon chip to startup. Once damaged circuitry attempts to operate it can cause the voltage rail clamping that you see. You may want to use an oscilloscope to capture the initial rise of your 1.8V supply and you may very well see it rise up to a little over 1.0V -> 1.2V before it then gets pulled quickly down to near ground.

I suspect that your 1.8V supply has current limiting on it. In this case that is a good thing because without it you would likely be seeing burnout smoke holes in the top of these chips.

|improve this answer|||||
  • \$\begingroup\$ Thank you Michael for your detailed response. This is very helpful and will help me to continue my investigation. \$\endgroup\$ – Tyler Wilson Feb 12 '19 at 16:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.