FPGA hangs

[Masoud Naderi]

Hi all,
I have two boards with the exactly same fpga (spartanIIE) and same
code inside them. One of the boards hangs 2~10 minute after power on.
I want to find out the reason. Is it due to power problem on one of
the boards? grouding or ...?
please let me know your ideas.
regards.

 

[Peter Alfke]

First:
check that both boards really operate with the same input signals. Swap the
boards and check where the failure moves.
Next:
Vary ambient temperature and also (separately) Vcc, and note their impact on
the behavior.

These two simple test should give you some clues.
"Minutes" almost always points to a temperature effect, unless you have
extremely long counters in the chip.
It could also be caused by hanging undriven pins that very slowly drift High
or Low.
Hope this helps.
Peter Alfke

From: naderimisc@yahoo.com (Masoud Naderi)
Organization: http://groups.google.com
Newsgroups: comp.arch.fpga
Date: 5 Mar 2004 14:17:52 -0800
Subject: FPGA hangs

Hi all,
I have two boards with the exactly same fpga (spartanIIE) and same
code inside them. One of the boards hangs 2~10 minute after power on.
I want to find out the reason. Is it due to power problem on one of
the boards? grouding or ...?
please let me know your ideas.
regards.

 

[Masoud Naderi]

Dear peter,
Thank you for your reply. I have some questions:

First:
check that both boards really operate with the same input signals. Swap the
boards and check where the failure moves.
The enviroment, input signals, temperature are exactly the same for

both boards.

Next:
Vary ambient temperature and also (separately) Vcc, and note their impact on
the behavior.

Power on these boards comes from a good DC/DC converter from DATEL.
There is no power variation on the boards.

It could also be caused by hanging undriven pins that very slowly drift High...

I have some undriven and UNUSED pin on the fpga (XC2S150E). No signal
handled by these pins. Is it important? Should I connect these pins to
ground or Vcc ?

regards.
M. Naderi

 

[Kevin Neilson]

"Masoud Naderi" <naderimisc@yahoo.com> wrote in message
news:2ba3bbea.0403061113.7404aca5@posting.google.com...

Dear peter,
Thank you for your reply. I have some questions:
First:
check that both boards really operate with the same input signals. Swap
the
boards and check where the failure moves.
The enviroment, input signals, temperature are exactly the same for
both boards.

It sure sounds like a temperature problem. Variations in process could

cause one part to fail at one temperature and the other to operate fine. I
once had a customer that reported that my design would work for several
minutes and then start to fail. He blew some coolant on it and it started
working again. He had to end up beefing up the fan system. Xilinx parts
tend to use a little more power than expected.

I also had a customer with a similar symptom and the cause was insufficient
decoupling capacitance, so you might check that as well. Those caps seem
like they're not doing much, but they are important and you should have one
for eac VCC pin.

-Kevin

 

[rickman]

Masoud Naderi wrote:

Dear peter,
Thank you for your reply. I have some questions:
First:
check that both boards really operate with the same input signals. Swap the
boards and check where the failure moves.
The enviroment, input signals, temperature are exactly the same for
both boards.

Next:
Vary ambient temperature and also (separately) Vcc, and note their impact on
the behavior.

Power on these boards comes from a good DC/DC converter from DATEL.
There is no power variation on the boards.

It could also be caused by hanging undriven pins that very slowly drift High...

I have some undriven and UNUSED pin on the fpga (XC2S150E). No signal
handled by these pins. Is it important? Should I connect these pins to
ground or Vcc ?

A floating input that is not used is an unlikely source or problems.
But you can make sure by configuring your device to drive these pins to
a given value, even if it is just a pullup.

--

Rick "rickman" Collins

rick.collins@XYarius.com
Ignore the reply address. To email me use the above address with the XY
removed.

Arius - A Signal Processing Solutions Company
Specializing in DSP and FPGA design URL http://www.arius.com
4 King Ave 301-682-7772 Voice
Frederick, MD 21701-3110 301-682-7666 FAX