Wiring Thermocouples

I know all the theoretical values of wiring Thermocouple to similar metal extension wire and terminal blocks, but in practice I really don't see any useful difference in the input values.

Granted, I am not worried about high precision, just comparative numbers, but our system has used standard terminal blocks for years (with the appropriate thermocouple extension wire) with no decernable problems. I have also had to use regular wire to extend the thermocouple on occasion with good results. I only have to extend them maybe 30 feet, so maybe I don't degrade the signal significantly.

So the question is, what are the real down sides to just using wire to extend the thermocouple to the input card?

Just had a problem that relates to this. We wired some standard compensating cable to a FLEX Thermocople card and connected a standard thermocouple. We read the data ok and carried on installing on other equipment. Then on one installation i had values around 0 degrees. After checking everything it turned out that i had wired the couples incorrectly, in this case because the temp was higher on this application (ten degrees higher) the temperature reading actually went in the opposite direction to what you should expect. My expert tip check against the specs as to which core should be +ve and which should be -ve. Luckily it only took a day to sort everything out!

The only impact is on accuracy.

Each time you transition from one metal to another you create, in effect, another thermocouple. This will induce a voltage into the cicuit. I don't know enough to explain the details properly. The magnitude opf the error depends on the metals and the relative temperatures, and the significance depends on the application. In some cases the induced error is not significant, in some cases it is.

Omega Engineering has a good explanation in their handbook, and I'm sure Minco and other suppliers do as well.

The law of intermediate metals (see Omega thermocouple references) indicates that you can use a terminal of different material IF the juctions at each side are exactly the same temperature (the same thing has to occur at the insturment anyways). However, best practice is to use a TC block. I wouldn't abandon best practice and skip the TC block just to save 5 bucks.

Alaric said:

The law of intermediate metals (see Omega thermocouple references) indicates that you can use a terminal of different material IF the juctions at each side are exactly the same temperature...

Click to expand...

Reminds me of the time a user complained that there had to be something wrong with a readout device in the Lab. It ended up being one side of a multipin quick connect feedthrough was 130F hotter than the other side that came out of a pressure vessel. It was in error of over 5F from what was expected. Ends up it was assembled by a new technician that mistakenly used copper for both pins for that thermocouple instead of the the correct material for the second pin. Talk about something hard to figure out when he had done the other 12 thermocouples correctly.

The design expectation is that the T/C to copper transistion is where the cold junction compensation will be made.

To use copper wire, rather than thermocouple extension wire, at some point other than the final connection can cause serious errors, because the receiver end (AI input: PLC input, recorder, DCS input, panel meter, whatever) measures the temperature at that connection point and calculates the cold junction compensation.

However, the cold junction point, the T/C to copper transition point, has been made elsewhere in the system. The error will be the difference between the temperature of the receiver connection point and transition point where the copper to T/C was made.

The connection point for your receiver AI can easily be elevated above amb ient temperature in an enclosure or a panel, whereas transistion connection to copper wire could be many degrees different than inside the panel temperatures.

Where ever TC junctions are made, the assumption is that both sides, or both legs, plus and minus, will be kept at the same temperature. Having the same temperature on both legs is known as "isothermal". When kept at the same temperature, the junction created with the junction block is nulled, with the plus side error nulling the minus side error.

The special junction blocks made for thermocouple junctions are really isothermal blocks, built on a chunk of metal that is supposed to maintain both terminal points at the same temperature, and in fact, does a good job of that. The fact that the screw terminals are made of the same alloy as the thermocouple wire is nice, but is NOT critical to reliable and repeatable temperature measurement. Isothermal terminal connections are critical though.

The use of standard terminal blocks will fly if the terminal blocks stay at the same temperature. The use of copper wire, rather than TC extension is fraught with the possibility of error.

Dan