Guru
Joined: 25/02/2006
Location: AustraliaPosts: 495
| Posted: 12:15am 29 Jun 2011 |
While I don't have a MM, I've had to deal with multiplexed ADC inputs over many years.
A common problem (you may not have encountered this (yet) but you could probably test and confirm the behaviour) - if you try to read more than one analog input (quickly), the first input will influence the second.
To test/prove - take two or three trimpots and set the wiper of each to a seperate analog input. Set the inputs to "similar" voltages and run the program to display each AD input. NOW adjust the inputs of the first AD your software reads. The further you get from the voltage of the second input, the more you influence the second input!
I found you can minimise this effect by selecting the channel, delaying a few milliseconds and then reading the value. (Not all software gives you this level of control).
The problem (as I understand it) is the input capacitance of the A/D converter itself, and the effective resistance of the input source. Low impeadance drives can charge or discharge the A/D input more quickly. Tracks the input better and reduces jitter a LOT.
Its not always practical to do, but you should try using a low-impeadance source (opamp would probably do) to buffer your input and see if it helps.