11-10-2006 02:56 PM
11-12-2006 01:50 PM
01-21-2007 01:31 PM
Thanks for your help.
I finally traced the channel interference problem to poor settling time on the analog-to-digital converter. I slowed down the convert rate to less than 25kS/s and the problem went away. It was not the linear scales as I had thought.
HP
01-28-2007 12:06 PM
Your comment that you solved your "channel-interference" problem (that is normally called crosstalk) points up a potential problem with all multiplexed data acquisition systems (and some others but these are the worst).
The specified performance of the systems assumes that the source (transducer and cabling) impedance is essentally zero. If this is not the case, the resistence combined with the capcitance of the multiplexer buss will produce a low-pass filter that "slows" the slew of the voltage when the channels are switched. The voltage read will be a mix of the earlier and present channel.
The solution is to put an amplifier ahead of the channel. But this is expensive. You don't get something for nothing... ss