David Hawkins wrote:
> Could you test one thing;
>
> 1. back-to-back reads of the same channel, with no channel change
> in between
>
> versus
>
> 2. sequential reads of channels.
>
> If you find that 1 is stable, and 2 is not, then I suspect your
> channels are not being driven by low-enough impedance.
[snip]
> If you probe the ADC inputs, you should be able to see
> glitching.
>
> Of course, this is only one of many things that can be wrong,
> but no one has suggested it, so I figured it was worth
> making you aware of this possibility.
Some good ideas there. I had only recently noticed the note about the input
impedance restriction of the circuit which drives the A/D inputs. I don't know
if the engineer who made the sensor interface board was aware of this. The
interface board was a copy of an earlier version of the product which used a
microcontroller and a different A/D converter.
At any rate, perserverence (or was it just plain stubborness) paid off in the
end. I found the cause of the problem around 5pm on Friday.
One line of Port B was used to turn a 1kHz signal source on and off. With the
new kernel module, that line operates at 1kHz all the time. What I eventually
discovered is the line was being turned off by the application during
initialization. This affected the generation of the 1kHz signals. Scoping the
lines I noticed that the waveform was flipping upside down every so often. It
explained why the A/D voltages were varying from what they should be reading
to stable negative voltages to readings in between.
The lines that were writing to Port B in the application were harder to spot
as they used a macro defined in a header file. It also took a while to find
since I wasn't sure if I was dealing with a software issue or a hardware
(possibly timing related) issue.
Thanks to those that offered thoughts, ideas, and questions. I knew it was
probably some stupid little thing I wasn't seeing and it turned out I was
right. I have been faced with this type of problem in the past where some
little hard-to-find thing was causing a problem. Fortunately, its been quite a
while since I last faced a similar situation.
--
Cheers!
Kevin.
http://www.ve3syb.ca/ |"What are we going to do today, Borg?"
Owner of Elecraft K2 #2172 |"Same thing we always do, Pinkutus:
| Try to assimilate the world!"
#include <disclaimer/favourite> | -Pinkutus & the Borg
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/ts-7000/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/ts-7000/join
(Yahoo! ID required)
<*> To change settings via email:
<*> To unsubscribe from this group, send an email to:
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|