computer data acquistion problem

[mark]

Hello all<p>I have a thermoconductivity detector (basically a resistance bridge as far as I know) that puts out approx 4 mV with a bridge current of about 100 mA. I'm trying to read the signal into a computer through a USB data acquistion module (DI-700 from DataQ). When I connect the signal leads to the module I register a signal of 0.6 V instead of 4mV and the bridge current drops by half to about 50 mA. I thought maybe the signal was too low, so I hooked up a x10 and x100 amplifier using an op-amp, but the signal I read on the acquistion card just goes up to 6V and 18V (the limit of the PS on the op-amp). When I connect my voltmeter to the output of the op-amp without the data acquistion card, I see the correct voltage and the bridge current stays at 100 mA. The weird thing is that I don't even need the + part of the data acquisition card to be connected for the signal to jump: if I connect the analog gnd connection only, the reading on the voltmeter jumps.<p>So I was thinking it might be a ground problem. To bring the signal from the detector I'm just using a pair of wires. The detector output has a connection for a cable ground to use a shielded cable, but the data aquisition module doesn't have a cable ground to attach to.<p>If you have any suggestions, please let me know
Thanks,
Mark<p>PS The data acquisition module works fine if I feed it 90 mV from a 9V battery through a voltage divider.

[russlk]

I suspect the resistance bridge has an internal ground, so when you connect the output to ground, it shorts part of the bridge.

[mark]

Russ, thanks for the advice. I ended up using two channels in differential mode and it works much better now.