conductivity signal discontinuity

[vdolnik]

 I collect my conductivity detector data with NI 9239 / cDAQ 9171 / Signal Express. Recently I experienced discontinuities in my signal (see the picture attached), i.e., signal started going down to "zero resistance". The detector electrodes are connected to electrolyte at + 3 kV, the circuitry is galvanically separated from the signal, so there should not be a problem. Anyway, I tried another detector, another NI 9239, with only partial success if any. Restarting the computer with Signal Express on it typically brought the problem back after some period of proper operation. Can anyone identify the most probable cause of these discontinuities? Dirty connectors do not seem to be the problem.

[rsmayer]

When you use the other 9239 module, what do you mean by "with only partial success if any"? Are you experiencing similar behavior or are you getting completely different results?

Also just to clarify, all of your measurements are accurate immediately after restarting your computer? How quickly does the error come back?

-Ryan

[AnalogKid2DigitalMan]

Could the electrolyte be breaking down causing an arc between the electrodes?

-AK2DM

[vdolnik]

Thank you for your suggestion. There could be electrolysis on the electrodes for a different reason but manifests as bubble formation and an infinite resistance, i.e., signal 9-10 V. It is annoying but, fortunately, I know how to solve this problem. "Zero resistance" discontinuity is something different. 

[vdolnik]

The discontinuity problem occurs irregularly, so its temporary absence does not mean it permanently disappeared. (Yesterday there were not any discontinuities, I am not sure about today.) The rest of the collected data is not affected. The export of data in Excel, however, provides only the initial part till the (first) discontinuity.