0
\$\begingroup\$

I have a system error pop-up on an HMI of one of my machines. Since it only had a number, I tried to reach the manufacturer for the error code description. They said that this error usually is the result of a bad analog input (in my case, one of the thermocouples). I have regular J-types, which are connected directly to PLC. Here's the thing: they all read exactly what you'd expect, but the manufacturer's techs tell me that it could read OK and still be bad, which I don't understand.

Suppose "+" wire's insulation is damaged and there's a short to the ground, then there would be no reading. And if "-" wire is shorted to the shield, and, let's say, it picks up some EMI, but then, it would either get filtered inside the PLC or it would effect the reading.

So, my question is: How can a thermocouple give a good reading and still be bad?

\$\endgroup\$
4
  • 1
    \$\begingroup\$ Shorts in a cable can look exactly like a thermocouple. A thermocouple is made of dissimilar metals shorted together. \$\endgroup\$
    – Andy aka
    Dec 23 '20 at 15:25
  • \$\begingroup\$ And give the reading of exactly correct temperature? I don't think so \$\endgroup\$
    – Viktor
    Dec 23 '20 at 15:28
  • 1
    \$\begingroup\$ @Viktor It'll give a reading of a different location's temperature, wherever the short is (assuming the short is between + and -; otherwise the metals involved probably won't be the right two to make the same type of thermocouple junction! You probably don't have an iron or constantan shield after all.) \$\endgroup\$
    – Hearth
    Dec 23 '20 at 16:39
  • \$\begingroup\$ @Hearth That made zero sense to me) \$\endgroup\$
    – Viktor
    Dec 23 '20 at 16:45
1
\$\begingroup\$

The quick way to test this is to simply disconnect the suspect thermocouple and substitute a similar thermocouple. You don't have to install the substitute in the machine. Simply see if the error goes away.

If you do not have a similar thermocouple, make one by using a piece of thermocouple wire. Quick and simple.

You could also swap the wires of the suspect thermocouple with another within the machine. Then see if the error moves also.

\$\endgroup\$
1
  • \$\begingroup\$ The machine is already fixed, adn it had nothing to do with TC. I am just trying to understand if it's a legal statement, that a TC can give good reading and still cause issues with PLC. In case I have problems like that inthe future. Because right now it seems like the support techs just wanted to get rid of me with a standard "No time to explaine, just shove tomatoes up your ass". \$\endgroup\$
    – Viktor
    Dec 23 '20 at 15:53
0
\$\begingroup\$

legality has nothing to do with it - unless there's specific laws about thermocouples in your country. With any repair problem - check the simplest things first. Wiring is a common fault. Thermocouples being made of wire would also be subject to breakage and shorts. As the others have mentioned, the thermocouple can be broken and give a reading - it just might not be valid, but they are easy to test - heat up the end, temperature reading changes - there's a fair chance it is working, then look elsewhere. Were the support techs trying to 'blow you off'? We may never know.

\$\endgroup\$
1
  • \$\begingroup\$ Why do people never read the question before posting an answer. I realize that thermocouple wire can be broken, just like any other thing in this pointless universe. I thought I was very clear, that tc gives good reading (it means that it shows correct temperature and reacts to change). And despite that, people were trying to convince, that it could still be bad. I was only inquiring to the mechanism, by which it could be physically possible. \$\endgroup\$
    – Viktor
    Dec 25 '20 at 13:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.