Honored Contributor
Honored ContributorHi,
I observed similar problems while doing prototyping. Many times "broken" cards could be reused by changing the FPGA (I had a TQFP package), sometimes the EPCS1 needed to be replaced as well. Between perfect operation and "nothing anymore" it worked sometimes, sometimes not... I never really identified the reason but problems stopped by ensuring the GND of the USB-Blaster being the first signal connected to the target PCB with the FPGA. Thus I assume the programming pins might be damaged by overvoltage by static discharge if these and not the GND line is connected first. (The USB-Master was referenced to PC, the FPGA to it's powersupply, somewhat all being capacitive referenced to earth...) This either ends in connecting and disconnecting the USB Blaster only with the target being completely unpowered (disconnected from power line) or (as I did) connect the GND of the Byteblaster permanently with the target (I use an adapter from the USB-Blaster connector to the smaller connector on the target anyway) Taking a look in the datasheet for Cyclone III and IV, they have added protection circuit to the programming pins to keep voltages at DCLK, DATA, ... within devices limits... Maybe this does not solve your problem but helps to scratch other cards... Carlhermann
