Forum Discussion

Altera_Forum's avatar
Altera_Forum
Icon for Honored Contributor rankHonored Contributor
12 years ago

I think my FPGA is dead, I need help : (

I made a FPGA board with a Cyclone III and a EPCS64 over a year ago and always work well, I give it to someone for a couple of days and didn't use it until now and when I try to configure it says:

Info: Started Programmer operation at Tue Jun 18 13:03:56 2013

Info: Device 1 silicon ID is 0x16

Info: Erasing ASP configuration device(s)

Info: Programming device 1

Info: Performing verification on device 1

Error: Verification failed for device number 1

Error: Operation failed

Info: Ended Programmer operation at Tue Jun 18 13:05:24 2013

I configure another board so it is not the program or the usb blaster, then I check the board (continuity between tracks) and discoverd that pin 1 of my jtag connector (DCLK pin) is in gnd contact always, I check and check and check (even remove the EPCS64 from the board) and realize than the FPGA is the problem

When I disconnect this pin from the board everything is fine in the board, but this pin in the FPGA remains in GND contact, I check this pin in my other boards and none have GND contact.

This means that I can´t configure my FPGA anymore?

What could have happened?

It is reparable?

Hope you can help me, sorry for my bad English

2 Replies

  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    12 years ago

    Hi,

    well if the short to GND is identified to be the FPGA Pin, I assume it is dead, indeed... While I never tested this, it might be possible the FPGA can still be configured by JTAG (but seems to be unlikely IMHO).

    This might have happened by plugging or unplugging the USB-Blaster in case the voltage differentials beetween your's USB Blaster "GND" and the board's GND are not equalized by GND being the first pin connected. Thus the USB Blaster should either be connected only with board not being powered or your I/F should include the protection diodes as shown in the handbook, limiting the voltage swing to -0,5 and VCCIO+0,5...

    I have had some prototype Cards with identical effects, as the FPGA was a TQFP144 I saved some for lab testing useage by replaceing the FPGA - the EPCS seem to be either more robust or (more likely) are "protected" by the FPGA... As my Card has an additonal Adapter between the USB Blaster Header and the programming connector I added an additional GND line to have the USB Blaster "grounded" to the FPGA's GND, with this modification I had no additional defects :-D

    Finally it is repairable by exchanging most likely the FPGA..
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    12 years ago

    Hi

    Thanks for reply, my FPGA is dead and although I didn't see what happened

    I suspect it was something like what you mentioned, I will consider use the protection diodes next time u.u