thermocouple noise SCXI 1102 & 1300 terminal block

[DSRTM]

Hi guys, I've searched the previous posts for this question, and did find similar issues but not the exact hardware I'm using.

I've got a SCXI 1101 chassis, the SCXI 1102 card and the 1300 TB measuring a type J TC using a standard Daq assistant in a while loop in LV 8.6.  I've grounded the negative side of the TC connection to chassis ground as recommended in the 1300 instructions.  

I have a part in a press that i'm monitoring temperatures on.  I'm sure the press has it's own ground (although I haven't actually seen a stud on a frame etc).  We were getting noisy temperature signals past roughly 400F., so suspecting a ground loop, I isolated the test part from the press with pretty good results.  It's not perfect however, as I am getting some intermittent noise when the part is jogged in and out of the press - I'm suspecting the isolation isn't perfect so we're making contact to the frame when the part moves 

My question is this - I noticed that some other recommendations in this forum (*for other equipment*) were to remove the ground within the terminal block as the ground on the frame was interfering wiht it.  Can I do this with a floating (non ground referenced) TC with my hardware?

I cannot guarantee that I can get the negative only lead on the TC to touch the panel, while the positive would be not touching - is there anything I can add to this cirucuit to eliminate the effect of the ground loop?

 maybe I'm off base here - but it sure looks like a ground loop problem.  It's tough to explain to coworkers why our Fluke handheld TC reader is reading stable while the $$ NI isn't when i'm not 100% sure what's causing the problem.  I'm assuming the Fluke's TC's are floating and the ground from the press is the only ground??

Thanks

Darren

[Doug_L._Bear]

I would completely agree with your theory of ground loops.  And would say that because your thermocouple is attached to a grounded source, it is in fact not floating and should not be grounded as per the methods in the 1300 manual.

[DSRTM]

It appears to have fixed the problem.  I tried logging without the grounding to the SCXI 1300 and the readings are stable and are matching the old Fluke.

Thanks for the response.

Darren

---

Doug L. Bear wrote:
I would completely agree with your theory of ground loops.  And would say that because your thermocouple is attached to a grounded source, it is in fact not floating and should not be grounded as per the methods in the 1300 manual.

---