thermocouple noise problem / earthing advice required 1

[wolfbag]

Hi,
This is probably something I should know but its a long time since those Measurement Systems lectures in college...

I am using K-type thermocouples in an application where there are high ambient AC voltages. If I use a standalone, battery-powered datalogger which is unearthed the TC signal is fine. If I use a PC-based data acqusition system (old NI435 card) then the signal is swamped with noise. This makes me think it may be an earthing issue. My problem is that I really want to capture several channels of data, and the standalone logger only has one channel, so I have to use the PC-based system.

So should I a)earth the thermocouple itself, a pain in the ass but can be done, or b)disconnect the earth from the PC and/or DAQ board or c)follow the advice of someone who knows what they are doing? I can live with the risk of having an unearthed PC, plus both the PC and DAQ board are old enough that no-one will cry if they get torched.

Thanks for your help.

 

[ilan7]

Think I have the answer - but do not know if you can do it:
Connect the T/C to a shilded compensation wire. The shild must be connected to one point at the measuring device, and that point MUST be connected to a spacial ground electrode and not the same ground of the rest of the system (the AC system).
The other option of disconnecting the ground can work but with the risk of damaging the hole measuring system and the risk of getting electrical shocks.

 

[epoxyman]

Wolfbag

I have a somewhat similar aplication. Is the part with the thermocouple attached conductive? If so, is it possible to ground the part at a location other than at the thermocouple itself? This could ground the tc but maybe get away from the pain of replacing the tc you mentioned.

Regards,
Epoxyman