HAS any one used a 50 ch DMM for resistance measurement?

[fmeraz]

What i'm trying to do is simple, i'm trying to monitor 50 channels or should i say 50 points of resistance measurement and then datalogging a change but i have a restriction and that is the sampling rate. i can't use a realy since there is a DEBOUNCE issue so i'm stuck right now on this. i have to pretty much scan a 50 channels in 20uS or less and then logg everything that changes. please help thanks.

[Henrik_Volkers]

Without digging into specs ....

You have a 50ch DMM with 50kHz scanrate?

If continious exitation isn't a problem and you don't have to measure with alternating current to get rid of DC offsets:

One (or more) current source(s) , all (a fraction) resistors in series plus one reference, than is a differential voltage reading that can easely be performend with that scanrate.

If you use 3-4 PXI-2501 (25kHzmax FET-muxer) and 3 to 8 channels at 100kHz?  

[Dynamik]

Hi CrukZ,

      I am NOT a "hardware" person, but recently researched this subject and (with a hardware person) built a shorts-test station.  One solution that I/we considered was to use an analog DAQ board.  We'd have put some voltage - say 5V - on a buss, then branched off the buss with 50 individual high-precision resistors - one to each testpoint (one input goes straight to the ?V buss).  The DAQ board does a single-ended voltage measurement at each test-point - the mid-point of a voltage-divider - and use the measured voltage to calculate current and/so resistance.  Two 1.25MHz DAQ boards (triggered together), at 25-channels each, would get the DAQ-rate you're after.

You'd have to play with the test-voltage and resistor choices depending on the current/voltage specs of your device.  I think the closer the series resistor matches expected DUT resistance, the better it is for the calcs, but then really low resistance means higher current so it's "a balancing act".  The more expected resistance varies, the worse it is for the calcs. 

In our case, the tested circuits included diodes and current didN'T change linearly with test-voltage.  However, we successfully emulated a hand-held Fluke DMM @1.2V on the 400 ohm scale.  Done-deal! 

BTW, I won't feel bad if someone describes why this is a terrible idea.

I am NOT a hardware guy person

[fmeraz]

Wow, that is very intresting i think i should look into that one... Thanks!!!