NI Home > Community > NI Discussion Forums

LabVIEW

Showing results for 
Search instead for 
Do you mean 
Reply
Member
metrisys
Posts: 18
0 Kudos

How to make thermocouple from thermocouple wire?

Hi there,
I'm going to measure a temperature up to 150 C from 7 thermocouples with CompactDAQ and 2 NI 9211 Modules. The  length of each  thermocouples has to be approx. 5 meters. 
Because of the thermocouple lenght, i have to use “Thermocouple Wire” to make my own thermocouples.
Unfortunately, I couldn’t manage to find how such information.
How to connect the conductors of that “Thermocouple Wire”?
Knight of NI
johnsold
Posts: 10,294

Re: How to make thermocouple from thermocouple wire?

The best way is to weld the wires together to form the junction. Several companies make spot welders for that purpose (Hot Spot).

Depending on the the environment and the amount of error you can tolerate, you may be able to get by with just twisting the wires tightly together or soldering the junction. Some thermocouple wires do not adhere well to solder and the solder adds two or more metals to the junctions. Clamping the junction with screws can also work if the massive junction created by the clamps does not interfere thermally.

Lynn
Member
MeCoOp
Posts: 198
0 Kudos

Re: How to make thermocouple from thermocouple wire?

Not sure if I fully understand your question but did you check out omega.com....they probably have something to suite your application.

 

LabVIEW 2012 - Windows 7

CLAD
Active Participant
LabViewGuruWannabe
Posts: 221
0 Kudos

Re: How to make thermocouple from thermocouple wire?

Take a look at the information from Omega. They have a great amount of technical information on thermocouples, including wire and connectors.

In short, to do it right, you'll have the thermocouple ends with a welded bead or attached to a probe. Any connections between the tip and the NI hardware can be accomplished with screw-together connectors.

In a pinch, you can take thermocouple wire and twist the two wires together at the end for the sensing tip. Notice, you'll get a much smoother line for your temperature data if the tip has a probe or at least attached to a solid metal object than you would if it were just twisted together at the tip and in open air.

B-)
Trusted Enthusiast
AnalogKid2DigitalMan
Posts: 3,184
0 Kudos

Re: How to make thermocouple from thermocouple wire?

You do not neccessarily have to make your own T/C's from T/C wire. You can buy the thermocouples off the shelf , then use the appropriate matching connectors and thermocuple extension wire to exten dyour lead length. This way, if a T/C fails you simply unplug the bad unit and plug in a new one.

Omega's site has a lot of useful information. Make sure you have the polarities correct, use the proper wire and connectors that match the T/C you use.

~~~~~~~~~~~~~~~~~~~~~~~~~~
"It’s the questions that drive us.”
~~~~~~~~~~~~~~~~~~~~~~~~~~
Knight of NI
altenbach
Posts: 27,813
0 Kudos

Re: How to make thermocouple from thermocouple wire?

I agree omega is probably the place to look.

Omega has nice connectors for each thermocouple type, so you can easily make "extension cords" if needed to be used with off-the shelf thermocouples. Since each of the connector terminals must match the metal of each wire, make sure you get the connectors for the correct type.

If you want to make your own thermocouple junction, one easy way is to spot weld the two wires together.


LabVIEW Champion . Do more with less code and in less time .

Active Participant
Conseils
Posts: 585
0 Kudos

Re: How to make thermocouple from thermocouple wire?

[ Edited ]
Just twist the two metal wires together and bung it into a gas welding torch flame......or similar....

Watch out the wires get hot !!! of course, ideally you need to make a single round bead of a consistent size, it takes practice.... I am not even sure I can get it first time now, its been a good many years since I last made one.

We don't do this any more for a number of reasons. The first of which is safety (hot naked flames and stuff....) and also that modern thermocouple manufacturers go to great trouble to reduce impurities in the junction (higher batch to batch precision and so on). There are also issues over tracability and so on.

Anyway you don't need to have a long thermocouple, you can use a short thermocouple and a 'compensating' cable. This is cable which simply contains the same materials as the thermocouple. There are special connectors available to enable you to put cable mounting plugs and sockets on the cable as you need.

Now you can have a tough, long cable and a short, expensive, easily replaced sensor.



Message Edited by Conseils on 12-17-2007 08:36 PM
Active Participant
Gabi1
Posts: 634
0 Kudos

Re: How to make thermocouple from thermocouple wire?

let me add my 2c:

if you already have termocouples, with a junction, and only want to extend the wire all the way to your DAQ instrument, depending on your desired accuracy you have several choices:

1) simply connect the 2 cold side (measurement side) to simple electrical wire (shielded is a little bit better). with this, assuming the room temperature is uniform, you might expect a negative systematic error of a few degrees, on top of the room noise, typically for this kind of connections of 1Deg, for K types.

2) connect to as you call: thermocouple wire. much better quality, you have to be sure that the cold side connect to a wire made of identical material, fitted exactly for each side of the thermocouple. for that, it is common to use special type of non-symmetric connectors, also made of same material for each electrode. as everybody told you, Omega is the best. be aware: you will not avoid systematic error, rather reduce significantly the noise.

3) on top of 2, add a thermo bridge compensator (also Omega), for simulating a cold spot at 0 Deg Celsius

4) instead of 2, connect the thermo bridge to a differential amplifier. (again, yes Omega). i like this best, as you can have a full range of 10V instead of the typical few mV. noise is kept to a minumum because of the proximity of the amplifier to the source (shorter wires), and because the measuring device is totally decoupled. significant reduction of ground loops and general ground problems. of course, the amplified signal can now run in a standard BNC for 10 of meters...

good luck!

 

-----------------------------------------------------------------------------------------------------
... And here's where I keep assorted lengths of wires...
Member
Tohatsu
Posts: 60
0 Kudos

Re: How to make thermocouple from thermocouple wire?

A few stuff to think about, as long as the wires touch eachother in the end (hot spot), you can practically do whatever you want, adding a third or fourth metal will not matter, so the "impurities" talked about earlier is nothing to worry about.
 
5 meters is not very long, and if you have TC wire, simply strip 7-8mm off the end, and twist them together. If you can, do apply, as suggested before, some heat and braze/solder it together. Remember that the amount of metal applied will change the response time of the TC, the greater the mass, the slower the system gets, but TC is usually very fast anyway so...
 
Then you just strip the wires at the other end and connect it to your data unit.
 
Now, let's say the TC breaks at the end, what do you do then? well, simply cut off that 7-8mm of stripped wire, strip another fresh 7-8mm and twist those togheter...no need to buy a new TC or something like that...the entire wire will work as a thermocoulpe.
 
Also, I am no familiar with your hardware, but cold junction compensation is something you might want to look into. A cold spot is where tou connect your TC wires to another metal, and that will cause for an error. Having a temp. sensor at that spot and compensating your readings with that value, will increase your accuracy.