Bad Thermocouple Input Module?

[crossingnevada]

I am trying to decide if the flex I/O Thermocouple module we have is bad.

I hooked a thermocouple to the module that we have in the main PLC 1756-L63. I configured it as a TYPE K and got a reading of 62.??. It reads within a couple of degrees of the Fluke thermometer that I had it attached to.

I plug it into the Flex I/O thermocouple module (Cat. No. 1794-IT8) and configure it as a type K and I get back 692+/-. Also, if I heat the sensor up and let it cool back down it won’t come back to the same reading, it may come back at 800+/- or some other reading.

Any thoughts folks?

 

[OkiePC]

Hello, and welcome to the forum. I suggest you zip your RSLogix5000 file (if allowed by your company) for best results.

I am not familiar with that combination, but have used Flex analogs and SLC analogs both offering times ten scaling on the card. Also, make sure the CJC connection is valid. These are vague guesses, the experts can give you more details and correct my WAGs.

Ideally, you could test it with a thermocouple simlulator if available, this would prove/disprove the input hardware/software.

Paul

 

[crossingnevada]

Thanks

Thanks Okie I will give it a try.

 

[danw]

1) A quick test of a thermocouple input:

If you short a thermocouple input with a jumper or a paperclip, the reading should be the temperature of the terminal block, probably a little warmer that the surrounding air temperature.

The indicated value is the cold junction compensation measurement, the supposed temperature of the terminal block.

2) One has to consider whether the reading is F° or C°

3) It appears that the problem occurs when you have the thermocouple placed in its operational location.

You could be seeing a ground loop problem, especially if the thermocouple tip is grounded. Differences in ground potential, a ground loop, appear as 'offset' or noise. The ground loop situation is more prevalent with single ended input cards.

First work around is an ungrounded thermocouple.

 

[Ken Roach]

Is the "module that you have in the main 1756-L63" a 1756 series Thermocouple module ?

FLEX and POINT and 1756 are all going to have different selectable data formats. A 1734-IT2I POINT I/O thermocouple module, for example, reports as an Integer datatype with units of 1/10 degrees, so "695" means "69.5 degrees F".

 

[OkiePC]

And, in case in hasn't been mentioned, verify that all your thermocouple extension wire are of the same type as the thermocouple. I prefer to only use terminals rated for the metal, I will twist the metal together and tape them tight if necessary to avoid going through "normal" terminal blocks or relays which can alter your result if indeed it is "off" by 7 degrees.

I think (working from memory) that 752 ish is a value that might indicate max scale error for a type K...but I ain't looking it up today

 

[Kick-It-Up]

Is the "module that you have in the main 1756-L63" a 1756 series Thermocouple module ?

FLEX and POINT and 1756 are all going to have different selectable data formats. A 1734-IT2I POINT I/O thermocouple module, for example, reports as an Integer datatype with units of 1/10 degrees, so "695" means "69.5 degrees F".

1794-IT8 also is in 1/10 degrees

 

[crossingnevada]

Thanks again..

Thank you again for all of your help gentlemen. I have gone through the information provided and I believe that it was a combination of a couple of things; 1.) The unit reports in 10s and 2.) the tip of the thermocouple may have been grounded. Addressed both of them and things are looking better.

-crossingnevada

 

[Bit_Bucket_07]

Thank you again for all of your help gentlemen. I have gone through the information provided and I believe that it was a combination of a couple of things; 1.) The unit reports in 10s and 2.) the tip of the thermocouple may have been grounded. Addressed both of them and things are looking better.

-crossingnevada

Yep... Sounds like you've resolved your problem. I would also like to mention that a go/no-go test for a TC is inadequate. Especially in high temp/high corrosion environments, still working TCs can become inaccurate over time and render non-linear temperature readings over their designed temperature range.