Thermocouples - which part has the different materials?

Thread Starter

cspwcspw

Joined Nov 8, 2016
78
HI - I have some K-type (chromel-alumel) thermocouples, in the picture below. The theory tells me about the junction of the two different types of metal, and shows the two different types of material running back to, say, a voltmeter or an amplifier / AtoD converter.
In this picture, are the chromel-alumel wires running all the way inside the shielding braid, or is the whole sensor in the inch or so at the end, and everything else is ordinary copper leads. Or put another way, if I cut this thing to only be a few inches long, would I be cutting through the chromel-alumel, or could I shorten the leads without changing the probe?

Thanks for any insight. Peter

Thermocouples.png
 

kubeek

Joined Sep 20, 2005
5,794
The wires are all the way to the connector chromel and alumel. You can cut the probe to shorten it, or you can extend it with K-type thermocouple wire. On J type you can solder the wires together (either to make a connection or to make the thermocouple point itself), not sure if solder sticks to K type.

If you change to copper at some point, then you need to measure the tempeature of that junction and use it to provide cold junction compenastion.
 

Thread Starter

cspwcspw

Joined Nov 8, 2016
78
Thanks kubeek - I'm playing, so no need to cut or solder anything. I simply wanted to understand better what was going on, and your reply is very helpful.
Peter
 

Reloadron

Joined Jan 15, 2015
7,501
Your lower image which appears to be a Type T has a small drawing showing the reference junctions where the thermocouple wire meets copper wire. You may wish to do a little reading on CJC "Cold Junction Compensation) and Seebeck Effect when working with thermocouples. This is a fairly good read on the subject and illustrates what can happen when trying to measure a thermocouple output using a voltmeter.

Ron
 

Thread Starter

cspwcspw

Joined Nov 8, 2016
78
Apologies. I did not mean to mislead. I cut and pasted the circuit diagram / drawing out of an AAC article about thermocouples, it doesn't necessarily belong to these images which are the "sales pitches" for the stuff I got. The vendor advertises both as type-K , and the Wikipedia article https://en.wikipedia.org/wiki/Thermocouple says there is at least one coloring scheme (why, oh why, do we need three competing color schemes) that also says the Red/Blue tags identify this as a type K.
 

Reloadron

Joined Jan 15, 2015
7,501
Red and Blue when the wires are insulated have always been representative of a Type T thermocouple. When an outer insulator houses both wires the color is Blue for Type T. Overall they go like this chart shows, including overbraid. Being located in the US I have always used the ANSI standardization for thermocouples, including the color schemes. When buying most thermocouples here in the US they come according to ANSI standardization for the most part.

For the most part the most common thermocouples I have seen used were K, J, T, and S. When selecting a thermocouple most of the decision is based on application to include temperature range, uncertainty limits and the atmosphere.

Ron
 

Thread Starter

cspwcspw

Joined Nov 8, 2016
78
Hi Ron - I think you are correct, and that is a nice chart, thanks. As I mentioned, I am playing around and both my kinds of thermocouple start at about the same temperature (so it seems that the room temp compensation is working), but a simple experiment with some ice water and some boiling water show that the one you mention with the red-blue leads has a steeper voltage gradient, so it "over-reads" my hot water, indicating about 115 degrees Celsius, whereas the other one reads about 95 degrees celsius. (It under-reads on the cold side too!) So that would seem more or less consistent with thar chart: the T type thermocouple has a steeper voltage-change gradient per degree. And the MAX6675 chip and A/D converter assumes K-type, so it explains the over-under readings I am getting.

Thanks again for the input.
 

Reloadron

Joined Jan 15, 2015
7,501
Thermocouples become a world unto themselves. :) Looking back at the link I provided earlier it shows the allowable error limits for the different types with both Standard and Special Limits wire. If you have a need for the Thermocouple Tables they are a downloadable pdf file covering the mV outputs for both degrees C and F. They download in a nice little zip folder. Additionally there is also a little temperature converter for converting C to F and visa verse. The temp converter downloads as a little executable file just open the zip folder and click the executable to install it. Runs on about any Windows system.

Ron.
 

ian field

Joined Oct 27, 2012
6,536
HI - I have some K-type (chromel-alumel) thermocouples, in the picture below. The theory tells me about the junction of the two different types of metal, and shows the two different types of material running back to, say, a voltmeter or an amplifier / AtoD converter.
In this picture, are the chromel-alumel wires running all the way inside the shielding braid, or is the whole sensor in the inch or so at the end, and everything else is ordinary copper leads. Or put another way, if I cut this thing to only be a few inches long, would I be cutting through the chromel-alumel, or could I shorten the leads without changing the probe?

Thanks for any insight. Peter

View attachment 140577
Everyday Practical Electronics recently devoted a Circuit surgery column to that very topic, unfortunately my newsagent has been getting flaky and I'm making the magazines I've got last - maybe not all that recent, but in the past few months.
 
the pics that you posted look more like thermisters. fork terminals would never be used fr thermocouples.

Cold junction compensation has a limited range and a short across the instruments inputs will read room temperature. A lot of instruments (thermometers) would read say 250 C for the lowest temperature. T is good for room temperature measurements and cryogenics. J and K are the most common. R and S have a much lower output. Atmosphere and useable temp are early on creiteria.

You have to watch the range of the instrumentation too.

You can always do the "ice bath" thing with boiling water and a voltmeter and use the tables. One junction at 32 and the other at boiling water. read the voltage and look at the tables. Old school

You need the temperature of the isothermal block that connects to the instrumentation. You then do a reverse lookup to get the mV reading of the isothermal block. then add or subtract (forget which) and look up that voltage in the tables to get temperature.

With thermocouples, Red is negative.

The wire used to extend the thermocouples is called extension wire, not copper. It's made of a lower cost material.
 

Reloadron

Joined Jan 15, 2015
7,501
the pics that you posted look more like thermisters. fork terminals would never be used fr thermocouples.
Keep an open mind with this one but they make barrier strips and actual crimp on terminal lugs for thermocouple alloy. I frequently bought them from Marlin Manufacturing here in Ohio, The lugs come in assorted sizes and have small lettering stamped on them for thermocouple type and polarity. Keep in mind I am only the messenger. :) I have also bought large multipin connectors, like large Cannon plugs on orders with male and female thermocouple pins installed. Omega Engineering also manufacturers a wide range of Barrier Strips, Thermocouple Spade Lugs and Terminal Lugs. I used their stuff also. While not as common more traditional connectors there are plenty used out there. I liked Marlin Manufacturing as they would load a custom 19 inch rack mount panel with connectors. The below is the last project I designed for clave control. I brought in 24 Type K thermocouples used for monitoring and control functions.

C2 Cab Inside.jpg

Lower right corner shows the 24 TC lines coming in and the 24 little Hockey Puck temperature transmitters, manufactured by MINCO and are fully programmable. Damn I loved those little signal conditioners, :)

Ron
 

ebeowulf17

Joined Aug 12, 2014
3,307
fork terminals would never be used fr thermocouples.
*** EDIT: Ignore this post. I was looking at the wrong picture. ***

You've got it backwards.

In the pictured thermocouples, the fork terminal is the sensor end - you mount it to whatever surface you're reading the temperature from. The other ends are bare wire which you can terminate however you want to get into a thermocouple amp.

Count the ends:
  • 5 fork terminals
  • 5 little lengths of heat shrink with a fat wire on one side and two skinny wires on the other side
  • 10 bare wires
*** EDIT: Ignore this post. I was looking at the wrong picture. ***
 
Last edited:

ebeowulf17

Joined Aug 12, 2014
3,307
I think he was right, what you were talking about are ring terminals. He was talking about the fork terminals in the picture below.
Dooooh! :eek: Oh, the shame.:oops:

Please disregard my previous post. I should've double checked what I was looking at before writing that.

Sorry everybody! :rolleyes:
 
Top