I don't think it's possible for the loadcell to fault in this way, but we are
still checking. The measuring amplifier has already been swapped out on
board, which made no difference. Also, the test system here is using the
swapped out amplifier with no problems at all sofar. So the fault is very
likely in the robostix, either hardware or software.
If you want to take a look at the software we're running, I just attached
"our" software to a reply to Dave on the list. Note though that that version
is not the _exact_ same version as is running, because I don't have access to
that at the moment. It only differs in reading 3 PG inputs as well though.
Due to the fact the problem occurred for the first time during installation
and our current problems with getting access to the ship, we haven't been
able yet to do all tests I would like to do. So I don't know if just
resetting the robostix also removes the error condition. Taking away power
(from all components) and reapplying does remove it every time.
thanks for the help!
On Wednesday 13 June 2007 18:53, jason@... wrote:
> Greetings Esger,
> > I'm convinced it is not a software problem as the robostix program is 1.
> > very simple, and 2. is still running and reporting the other inputs
> > correctly when the brake is stuck at zero.
> Well, I'd have to say, either your brake hardware is malfunctioning, or it
> is probably a software issue.
> If it's hardware, it's most likely not the robostick, but the signal that
> is driving it. If it's software, I'd have to know more about how you wrote
> the program to help. Is it interrupt driven? Are you using counters to
> walk across the ADC channels? Polled mode? Simple fixed routines called
> in sequence?
> My guess it's probably software, but your hardware amp could also be
> going bad. If it's software, you might try just resetting the robostix or
> gumstix/robostix group, and see if that fixes it. If it doesn't, then it's
> most likely hardware drving the robostix.
> Can the gumstix reset the robostix? If so, have it try that, and see if
> the problem persists...
> Good luck!