Extending a type K thermocouple....

[KI6LNG]

Ok Hams,

I'm working on a hot tub project. I am replacing the analog thermostats (no longer working) with a PID Industrial temperature controller and a Solid State relay to mate into the existing tub wiring. The thermocouple is type k and about 36" long. I'd like to mount the control unit about 25' from the tub for convenience and will need to extend the thermocouple. I know you are "supposed" to use a type k extension wire, but I was wondering what impact regular old speaker wire (to save some dough) would have on the readings of the thermocouple if any? I know some of you are probably very familiar with these units and it's a bit off topic, but I think a fellow ham is the right place to start with this one...

Thanks!

Rob
KI6LNG

[K8ERV]

My co. used J couples, but this may apply. You may be able to buy "extension" K which is cheaper. I think it is the same but not carefully tested or guaranteed for accuracy.

In theory, you can use any wire if all of the junctions are at the same temp. I think (but not certain) that only the adjacent junctions need to qualify. You might bench test the system. I would put each adjacent pair thermally together (wrap in metal foil?). I think if you then heat that pair you will not introduce an error.

Maybe someone else has more experience with this.

TOM K8ERV Montrose Colo

[AE7RZ]

It's all about resistance. The sensor converts resistance to display degrees. If you can keep the resistance exact you'll be OK. I suspect you will not be able to extend 25'!

[KI6LNG]

It's all about resistance. The sensor converts resistance to display degrees. If you can keep the resistance exact you'll be OK. I suspect you will not be able to extend 25'!

Some of the actual extension wire on ebay is 50' long so I'm hoping this is not the case!

[K8ERV]

Check WIKI for T-C info. These generate a voltage ALONG the wire, not at the shorted junction. I didn't know that. But couples do NOT work by a change of resistance. Some sensors do, but not couples.

Of course as in any system the wire resistance can be important if it carries much current. Most accurate temp meters have a high Z input so the resistance is not important. I don't know what your load is.

TOM K8ERV Montrose Colo

[KO6WB]

Extending a type K therocouple requires the use of type K wire. The reason is how thermocouples work. They use different metals to produce a potential that can be measured. When you use wire that is a different metal then the wire in the thermocouple wire, you produce a new thermocouple and that interfers with the actual reading. This can be quite substantial. The easiest way to make a thermocouple is to twist the two different wires in a thermocouple together. Bingo you have a thermocouple. In type K thermocouples there is a red and yellow wire, in type J it's red and white.
Good luck with your install and use the proper wire. No reason to go cheap with it.
73
Gary

[WA6TKD]

The problem will be an error introduced to the measurement if there is any difference in temperature between the 'cold' reference temperature compensation performed at the controller input and the temperature of the junction at the cold end of the thermocouple and copper extension wire. If both these points are near the same ambient temperature then the error with using copper extension will be minimum, but if not the error could be significant.

Omega engineer has lots of technical information on the theory and use of thermocouples:

http://www.omega.com/thermocouples.html

[AF6LJ]

Extending a type K therocouple requires the use of type K wire. The reason is how thermocouples work. They use different metals to produce a potential that can be measured. When you use wire that is a different metal then the wire in the thermocouple wire you produce a new thermocouple and that interfers with the actual reading. This can be quite substantial. The easiest way to make a thermocouple is to twist the two different wires in a thermocouple together. Bingo you have a thermocouple. In type K thermocouples there is a red and yellow wire, in type J it's red and white.
Good luck with your install and use the proper wire. No reason to go cheap with it.
73
Gary

In a word;
Exactly.


You would be better off extending the controls for the controller.

[KI6LNG]

The problem will be an error introduced to the measurement if there is any difference in temperature between the 'cold' reference temperature compensation performed at the controller input and the temperature of the junction at the cold end of the thermocouple and copper extension wire. If both these points are near the same ambient temperature then the error with using copper extension will be minimum, but if not the error could be significant.

Omega engineer has lots of technical information on the theory and use of thermocouples:

http://www.omega.com/thermocouples.html

So if the temperature at the connection point on the control box and then where the copper wire meets the thermocouple wire are roughly the same temp then the copper wire should effectively be "invisible" to the control unit?

In a word;
Exactly.


You would be better off extending the controls for the controller.

Do you agree with the above?

[KO6WB]

You're not listening!!!!! Do not try to go cheap with a temp control. The end result will be unsatisfactory and could cost you much more. If you can relocate the controller closer to the thermocouple and run the wiring to it, that would be a better choice. I work with thermocouples every day. Mostly type K and we have some type J. We also have RTD's which are resistive units and with them you can use copper wire. Some of our runs from the point where the thermocouple is and to the point where the controller is are hundreds of feet.
73
Gary