Spikes in J-type thermocouple signals

I

Thread Starter

imran shabbir

I have used J-type thermocouples in a steel industry for monitoring and controlling the temperature of heat treatment furnaces. In all the thermocouple signals, I am getting spikes of about 2 to 3 seconds. Is there any kind of filter available to solve this problem. imran shabbir System Engineer
 
Funny you mentioned that; We had a similar problem and a thermocouple signal conditioner was a viable solution. We used the Phoenix Contact brand but Weidmuller, and Entrelec should have something as well, check out their websites.
 
J

John G. Boland

Imran, you have:

<< J-type thermocouples in a steel industry for monitoring and controlling
the temperature of heat treatment furnaces. In all the thermocouple signals,
I am getting spikes of about 2 to 3 seconds. >>

Are you using isolated signal conditioning modules and shielded t/c lead wire? Are your shields floating on one end? Sometimes the performance is better if the shields are grounded at the instrument end, rather than at the t/c end, theory notwithstanding. It's worth a try.

If it is really all signals and the spikes are synchronized, could the instrument be on the fritz?

Have you recorded the common mode (both leadwires to instrument ground) and
differential mode voltages during the "spikes"?

John G. Boland, president
Strateg!c Method$ Corporation
 
In resonse to J. Roops query:

If "spikes" have a pattern related to number, duration, amplitude, and/or interval timing, check if receiving instrument is equipped with
T/C open-circuit detection.

Phil Corso, PE
Trip-A-Larm Corp
 
J

Johnson Lukose

One installation here had this problem with J-type thermocouples connected to a T/C card of Modicon PLC. When we read the documentation for the T/C card it explained that the card injected a 1mA strobe to the T/Cs to check for open circuit or short circuit. The spike could be detected on the multimeter showing a dancing signal. Just a thought.

thanks.
 
Top