Help with thermocouple type

[BrunoARG]

Hello everyone

I am working in a small personal project which is a poor man's soldering station, or so. I have made the complete diagram, the solder iron and hot air gun connection and sensors positioning, the 1st order (proportional) temperature controller with PWM output which will activate an opto diac, connected to a triac and the heaating element.(yes, won't control the output power via PPM to avoid any complication).

The matter here is.. I bought a thermocouple, but they couldn't tell me what type it was. All I know is that this one is used in stoves, as safety protection, when there is no flame, gas won't flow outside.

So I did some research and found this:


And the thermocouple is this one:


Junction head


Connector


I had to see what was inside, so I decided to get into it...

This is the conenctor. You can see a copper wire insulated with a sort of cheap fiberglass thread. A brass piece connected to it, an insulator ring and the external "copper" tube.


This is the junction or metals welding. They look the SAME color but the end of that filament was actually part of the "copper" core.


And I made the following tests:

Took a fridge magnet, and realised that (as thought) the outside "copper" tube was not magnetic, but the silver-looking head did.

The metal that looks silver and is welded to the head is NOT magnetic, but the head is.

So, there are two posibilities:
K type --> Nichrome // Ni-Al

J type --> Iron // Constatan (Cu-Ni).

By the color of the core ("copper") it must be Ni-Cu (because nichrome looks silver).
By the color of the head, being magnetic, it must be IRON, and it should be because it's the other metal used in the thermocouple.

I am pretty sure it's a J-type thermocouple, then. But how about the external tube? The core is Ni-Cu and the outside is something with copper (or just copper, doesn't matter), it should be the same material as the second metal (iron) or it would generate another seebeck potencial in the joint (at the end of the head).

I don't understand this thermocouple, can you help me please?

 

[barry]

Are you sure this is a thermcouple and not a thermal switch? It's pretty unlikely they would use a thermocouple to simply sense the presence or absence of flame.

 

[BrunoARG]

Yes, pretty sure.

They put it right next to the flame, and when it gets hot, the small voltage allow to open a valve. Even I put it in my stove and measured about 13mV with a multimeter

 

[Warpspeed]

Yes thermocouples are used as direct flame sensors.

Usually the thermocouple is connected DIRECTLY to a gas solenoid valve held open by the pilot flame. While the voltage is very low, there is sufficient current to hold open the gas solenoid valve.

I do not know what type of thermocouple materials are used, but you may be able to work it out from thermocouple tables of voltage versus temperature, especially if you have narrowed down the range of possible types.
Here are the thermocouple voltages for J type in Fahrenheit:
https://www.pyromation.com/Downloads/Data/emfj_f.pdf.

Try putting it into boiling water and measure the voltage.
The voltage will be the difference between ambient temperature and boiling temperature.

For instance 212F minus (70F day) = 142 F
142F = 3.175 mV

In a direct flame, or at soldering temperature you will obviously get much more than that. But if you see around 3mV its probably a J type.

For K type :
https://www.pyromation.com/Downloads/Data/emfk_f.pdf

Boiling water test, 142F = 2.483 mV.
Only slightly different, but different enough to tell for sure.

 

[FvM]

The thermo couple in question is obviously designed to source a high current for a solenoid and not to achieve a particular measurement accuracy. It must not necessarily use a standard material pair.

 

[BrunoARG]

Thank you guys for your replies.

Yes, I measured it under boiling water and it was around 3mV (couldn't measure accurately as it was a cheap multimeter).
At the flame I measured upto 13mV... don't know the temperature but maybe around 400ºC. It should have been 11mV, it was 13 but who knows the temperature? at a point the thermocouple head was red hot.

But it could be that it's some cheap kind of thermocouple that is not any standard one, as FvM said, it only needs to generate a voltage when hot, something that almost any different metal junction does.

I think I will buy a multimeter K type thermocouple, won't take risk using something I don't know what really is.