I am using cyclone V FPGA. During my testing process, somehow I gave 12V to the IO bank instead of 3.3V. The board is shorted and I found FPGA has gone bad.

Now to prevent such accidents in the future I am configuring my power sequencer such that it will turn off the rail immediately as it sees a higher voltage.

The question is: if the power sequencer shuts down the voltage rail after 500ms after finding the fault, will this damage my device. How much time can an FPGA bear voltages above absolute maximum?

I am attaching a screen shot of fault settings which is not final.enter image description here

  • 3
    \$\begingroup\$ Damage will be instantaneous. \$\endgroup\$ Jun 12 '18 at 8:14
  • 6
    \$\begingroup\$ You should consider that anything more than zero seconds above the absolute maximum can potentially destroy the device. And 500ms is an eternity from an electronic standpoint. \$\endgroup\$
    – dim
    Jun 12 '18 at 8:16
  • \$\begingroup\$ Why wouldn't you just use a TVS diode? That's far faster than any sequencing logic and may buy you enough time for the sequencer to deal with things before the TVS diode gives out due to heat. TVS diodes also rise in voltage the more current they must conduct though so your sequencer is going to have to be a helluva lot faster than 500ms even with a TVS diode to hold the fort temporarily. \$\endgroup\$
    – DKNguyen
    Apr 12 '19 at 3:24
  • \$\begingroup\$ Probably a lot less than 500ms. \$\endgroup\$
    – user253751
    Apr 12 '19 at 4:22

Technically the time is zero, period. The device was designed and tested for the absolute maximum ratings and those cannot be exceeded, not even for a short time.

However, one thing that could be done is placing a zener or other diode protection scheme on the offending voltage rail (by blue wiring if necessary)


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.