Thermocouple wire

[realolman]

There is thermocouple wire that you can twist the two conductors together and it will form a thermocouple that can be hooked to an instrument, controller, or whatever.

Then there is also the need to extend the thermocouple signal to a remote location using thermocouple extension wires.

Are the two the same thing?

... and if not could someone point me to a source for thermocouple extension wire for a type J thermocouple?

Thanks

 

[winnie]

In a pinch they are the same thing, but if you have time to buy the right stuff, there are differences.

With thermocouples, you have to take care that there are no junctions of dissimilar metals anywhere except at the measuring point and at the measuring instrument. Any other junctions of dissimilar metal will act as an additional thermocouple in series with your intentional one, and all of the thermal EMFs will add up in series. Thus thermocouple extension wire must be made of the same alloys as the thermocouple itself. This means that the extension wire and all plugs and connectors are specific to the particular thermocouple type.

I'd suggest going to the Omega Engineering site and just reading up on the differences between thermocouple wire, thermocouple extension wire, etc. http://www.omega.com they sell lots of the stuff, but they also have lots of very useful 'handbook' information about how to use thermocouples.

-Jon

 

[GeorgeB]

There is thermocouple wire that you can twist the two conductors together and it will form a thermocouple that can be hooked to an instrument, controller, or whatever.

Then there is also the need to extend the thermocouple signal to a remote location using thermocouple extension wires.

Are the two the same thing?

... and if not could someone point me to a source for thermocouple extension wire for a type J thermocouple?

Thanks

+1 on Jon's response; I believe that Omega's guides and info will answer your questions. But to say what you will find in better detail there ...

Twisting together SORTA works, spot welding is better;
Specifications are higher on Tc than Tc extension, but I made many thermocouples from extension GRADE wire;

I've forgotten alloys, but we used either J or K ... red and white ... and silver soldered the twisted ends together, then ran a bunch up in our environmental chamber and confirmed they matched before using otherwise. Only bad solder joints ever failed.

Omega is good, but sometimes expensive. I believe that you will find Belden and Alpha have extension cable, but it has been years since I looked.

 

[LarryFine]

There is thermocouple wire that you can twist the two conductors together and it will form a thermocouple that can be hooked to an instrument, controller, or whatever.

You should know that the thermocouple is the joint of two dissimilar metals, and not the lead wires. An extension is an extension, but you don't want the extension causing misreadings, as Jon said. I've never heard of home-made thermocouples, though.

 

[winnie]

Larry,

You are exactly right, the 'couple' is the junction of the two metals. If the extension leads are not made of exactly the same metals, then at every metal transition you get another junction, so you are forced to keep the same two metals throughout the system, or you need to somehow compensate (or determine that it is acceptable to ignore) for the other junctions.

Because the extension wire is made out of the same metals, if you have a non-critical application, you can simply make a junction from the extension wire. Just twisting the wires together is enough if you don't need reliability.

Apparently 'extension' grade wire has different specifications and tolerances than 'thermocouple' grade wire, and it is pretty clear than homebrew couples will have poorer tolerances than carefully manufactured couples. But it was pretty common to just use a twisted wire thermocouple, a ice reference bath, and a multimeter to make measurements in my undergraduate chem labs.

-Jon

 

[K8MHZ]

You should know that the thermocouple is the joint of two dissimilar metals, and not the lead wires. An extension is an extension, but you don't want the extension causing misreadings, as Jon said. I've never heard of home-made thermocouples, though.

I worked in a research facility that made their own thermocouples. It took special welding apparatus and a weld tech with impeccable skills. Agreed, it's not common, but it is done.

 

[ELA]

First job out of school we made our own using a spot welder all the time.
It was not hard to learn the technique and it saved a lot of cash.

 

[fjrivera]

If accuracy is not an issue, you can twist the ends of thermocuple cable to form a junction. Also, you can make the joint using spot welding. More accurate readings can be achieved by calibrating the readings from the thermocouple in the transducer thus minimizing junction error. Many temperatures transducers and controllers has built-in cold junction error compensator.

... and if not could someone point me to a source for thermocouple extension wire for a type J thermocouple?

For this maybe, it's easier to install a transducer near where the thermocouple is installed and from the transducer just run the signal cable to the remote location. Also you can use a long thermocouble cable but as for any cable as longer the cable higher the losses. I haven't heard of thermocouple extension cables.

 

[LarryFine]

You are exactly right, the 'couple' is the junction of the two metals. If the extension leads are not made of exactly the same metals, then at every metal transition you get another junction, so you are forced to keep the same two metals throughout the system, or you need to somehow compensate (or determine that it is acceptable to ignore) for the other junctions.

Well, there must be at least one other dissimilar junction where 'normal' wires are joined, or terminations are made.

 

[StephenSDH]

Any other junctions of dissimilar metal will act as an additional thermocouple in series with your intentional one, and all of the thermal EMFs will add up in series.

This is and isn't true. If you go from thermocouple cable to a dissimilar metal and then back to thermocouple cable, it actually cancels out and the signal is unaffected. It does add, but it is the same as +1 + -1 = 0. If possible don't add junctions.

 

[LarryFine]

It does add, but it is the same as +1 + -1 = 0. If possible don't add junctions.

Unless you series them with the right polarities.

Actually, I think they'll still generate electricity as long as you don't heat the "backwards" junctions.

 

[PhaseShift]

Interesting thread. We deal with thermocouples all the time.

Does anyone care to give a basic theoretical explanation of how a thermocouple works, and how using dissimilar extension wire creates more junctions causing false readings.

 

[GeorgeB]

If you go from thermocouple cable to a dissimilar metal and then back to thermocouple cable, it actually cancels out and the signal is unaffected. It does add, but it is the same as +1 + -1 = 0. If possible don't add junctions.

There being no effect only applies if both dissimilar junctions are at the same temperature; one splice near the furnace at 250, one in the control room at 70 ... error. Now if they are always a stable temperature, the error may often be calibrated out, but the field interconnection is rarely in that situation.

 

[gar]

100205-0833 EST

PhaseShift:

Every node (connection point) of dissimilar materials can be represented as a voltage source (a battery).

Consider an iron-constantan thermocouple. At one end of this wire pair twist the two wires together to form a junction. You have created a battery of variable voltage proportional to temperature. At the other end do the same thing.

If both ends are at the same temperature the result is two batteries (really should be called cells) in series opposition with equal voltages and no current flows in the closed path.

Next open one of the wires and insert a voltmeter made of the same material as in the wire opened. No new junctions created. The voltage read on the meter is proportional to the difference in temperature of the two end nodes (junctions). Put one of the ends in a bath of ice water, 0 deg C. If the meter is calibrated for centigrade, then the display will read the temperature in centigrade of the other end of the thermocouple wire.

If the scaling of the meter is in F, and 0 is zero volts (you are using your millivoltmeter as the readout) (about 55 microvolts/C or 30.6 microvolts/F), then the reading needs to have 32 deg added to give F readings.

If you understand what I have said so far, then I will continue with what happens when you use dissimilar wire for an extension.

.

 

[StephenSDH]

There being no effect only applies if both dissimilar junctions are at the same temperature; one splice near the furnace at 250, one in the control room at 70 ... error. Now if they are always a stable temperature, the error may often be calibrated out, but the field interconnection is rarely in that situation.

Right. I was thinking more a steel terminal block, or a slip ring. But you are right. It cancels only if the temperature is the same at both junctions.

 

[LarryFine]

If you understand what I have said so far, then I will continue with what happens when you use dissimilar wire for an extension.

I do; please do.

 

[Steve Kelly]

T/C wire source

T/C wire source

Omega Engineering is a good source for extension T/C wire. Here's the link
http://www.omega.com/toc_asp/sectionSC.asp?section=A&book=temperature

They also have an excellent technical reference section that may be of use to you in answering your questions. In the link below, you'll find an extensive section on thermocouples to your right. Hope this helps

http://www.omega.com/techref/


Steve

 

[gar]

100205-1610 EST

Larry:

First read the section from Principle of Operation from
http://en.wikipedia.org/wiki/Thermocouple

Principle of operation
Main article: Seebeck effect
In 1821, the German?Estonian physicist Thomas Johann Seebeck discovered that when any conductor is subjected to a thermal gradient, it will generate a voltage. This is now known as the thermoelectric effect or Seebeck effect. Any attempt to measure this voltage necessarily involves connecting another conductor to the "hot" end. This additional conductor will then also experience the temperature gradient, and develop a voltage of its own which will oppose the original. Fortunately, the magnitude of the effect depends on the metal in use. Using a dissimilar metal to complete the circuit creates a circuit in which the two legs generate different voltages, leaving a small difference in voltage available for measurement. That difference increases with temperature, and is between 1 and 70 microvolts per degree Celsius (?V/?C) for standard metal combinations.

It should be noted that there are a lot of minor variations in material composition, uniformity of the material, and the effects of temperature.

You can theoretically view the voltage from a thermocouple as the difference of the voltage generated along one wire (material) compared to the voltage along the other wire as the output voltage of the open circuit thermocouple resulting from the temperature gradient along the two wires. In the real world you can not measure this without introducing more dissimilar junctions. With the two wires being of different materials with a different voltage change from a given temperature change there will be a generated voltage from the temperature difference.

Next, assume the two thermal couple wires are not copper and both have a thermal electric output with copper that is non-zero. At one end the thermocouple wire is connected together to form the measurement junction. The other end is connected together with a copper wire. The copper wire and its two junctions are all at the same temperature. Because there is no temperature gradient along the copper wire there is no voltage generated along the copper wire. Therefore, this is the equivalent of the thermocouple wires being connected together instead of coupling thru the copper wire.

Make the copper wire a long loop. From a temperature measurement perspective the temperature difference measured is that of the probe point temperature to the temperature at the point where the copper loop and the thermocouple wires connect together. This reference temperature is not the end point of the copper loop. I am assuming the end point of the copper loop is the location of the measuring instrument and also the location of the absolute temperature reference.

Thus, one wants to have the uniform homogeneous thermocouple extend to the location where the reference temperature is defined. Basically the location of the measuring instrument.

It would be possible to make the thermocouple to copper termination some place out in a field and locate some kind of absolute temperature measurement element, such as a thermistor, at that copper junction location and bring back its signal to provide the zero reference in the instrument. But why do that instead of bringing the thermocouple wire back to the measuring instrument?

For those asking the question about thermocouple extension wire. The wire composition needs to be the same as the thermocouple wire, but insulation would not have to meet the requirements of the probe and therefore the extension wire could be a lower cost.

.

 

[LarryFine]

I think of gas-appliance pilot thermocouples, which generate microvolts to operate the gas valve.

The copper tube is basically a coaxial cable, but I don't know what the inner conductor is made of.

 

[gar]

100205-2000 EST

Larry:

Looking on the Internet I did not get an easy answer to your question on gas appliance thermocouples.

I did find a reference to a patent. In their claims reference was made to nickel and chromium alloys. Looking in the Handbook of Chemistry and Physics under Thermoelectric Power (a misnomer) I find Ni-chrome has a +25 microvolt value and Nickel has a -19. Thus, appropriate alloys could be fairly good for voltage generation.

A pilot light thermocouple in combination with its holding solenoid is going to be designed for maximum power transfer. This might mean about 0.2 to 0.5 ohms for coil resistance. These resistance values are from one old used thermocouple and one new one. Quite likely the old one failed because the internal resistance got too high. These resistances might be much higher in a flame, and therefore the solenoid coil would be higher by design.

Note for an ideal voltage source and linear resistances that maximum power transfer occurs when the load resistance equals the source resistance.

.