# Thermocouple type-K

I have a type-K thermo-couple. In all the thermocouple, I have seen so far there is just two wires and a joint. But in the one I got, there is third wire which is connected to the joint and is twisted around the ceramic shield of the thermocouple. I do not know what does it do?
Any idea ....Thank you

berkeman
Mentor
Can you post a picture? What is the product model number that you bought? Can you post a datasheet link?

Artyman
Can you post a picture? What is the product model number that you bought? Can you post a datasheet link?
thank you for the response. I do not have a datasheet. I just found it in a very old furnace.

berkeman
Mentor
Don @dlgoff gave a great answer in a recent thermocouple thread. Maybe he can comment on this one...

dlgoff
Gold Member
I'm fairly sure that this extra wire is for grounding via a thermowell (the ceramic wire separaters tells me the thermocouple is intended to be slide down a well.) i.e. the extra wire makes a connection with the inside of the thermowell which gets attached to some grounded equipment. Here's a thermowell that gets threaded into the equipment. They can also be permanently welded into the equipment.

From this Omega technical learning page:
Thermocouple junctions may be grounded or ungrounded. They are often covered with protective metal but may be left exposed to improve response time. Grounding is frequently required to prevent buildup of static charge, which may negatively impact accuracy.

jim hardy, berkeman and Artyman
I'm fairly sure that this extra wire is for grounding via a thermowell (the ceramic wire separaters tells me the thermocouple is intended to be slide down a well.) i.e. the extra wire makes a connection with the inside of the thermowell which gets attached to some grounded equipment. Here's a thermowell that gets threaded into the equipment. They can also be permanently welded into the equipment.

View attachment 210176

From this Omega technical learning page:
Thanks dlgoff. My first guess was the grounding issue, but the confusing part is "why this shield grounding wire is welded to joint?"....

My first guess was the grounding issue, but the confusing part is "why this shield grounding wire is welded to joint?
Variations on the theme exist, but thermocouples come in grounded or ungrounded, and shielded or exposed junction types. On this one the junction is exposed, and equipped with a grounding wire.

If it were a shielded type (rather than being contained within a thermowell like the one @dlgoff described) it would typically be contained within a closed stainless steel or Inconel tube packed with magnesium oxide (to insulate the connection wires), and the junction would either be welded to the tip (grounded type), or offset back a bit to allow headspace for MgO insulation (ungrounded). In a shielded thermocouple, the grounded type has slightly faster response due to the direct connection between junction and sheath. An exposed junction TC has even faster response.

When used in the furnace, did it measure and control the temperature of the air within, or was it positioned directly in the flame as part of a proven ignition safety circuit? An ungrounded thermocouple can't form ground loops, and is generally preferred from a measurement perspective. I've never seen this sort of TC either, and unless it is part of a gas burner safety circuit the reason behind the intentional grounding wire is throwing me.

Artyman, jim hardy, dlgoff and 1 other person
Averagesupernova
Gold Member
They power the gas valve in home appliances and gas log fireplaces.
http://www.homedepot.com/p/Rheem-PR...r-Natural-Gas-Water-Heaters-SP20824/206523600
But not type K. I believe those are type J. I always had a hard time believing that a thermocouple could power a solenoid valve or relay, but it is most certainly the case. Here are some I am familiar with:
-
I recall reading in the encyclopedia many years ago about a candle powered single transistor radio. It used a thermocouple of some sort (most likely called a thermopile) and if memory serves me it developed about a half a volt. I never attempted to build one, but it fascinated me. I have tried googling for it but not much turns up.

... I always had a hard time believing that a thermocouple could power a solenoid valve or relay, but it is most certainly the case. ....
But it doesn't really "power a valve" as we normally think. Recall that with these gas valves, you press the valve in and hold it until the thermocouple gets hot and provides a holding current. At that point, the current can maintain the holding force, but there is no way it could pull it in on its own. It's at a very delicate balancing point, and there is just enough electro-magnetic force to hold it.

Just like a magnet that has attached it self to a metal plate holds hard, but move it a short distance away, and there is very little force. It's a damn clever arrangement.

Averagesupernova
Gold Member
A thermocouple relay certainly does 'pull in'. A valve may not but a relay does.

A thermocouple relay certainly does 'pull in'. A valve may not but a relay does.
OK, I'm sure there are designs that do pull in. I was thinking in terms of the ones I'm familiar with - the thermocouples on the older style North American residential gas water heaters and furnaces - the kind Jim Hardy was referring to. Those only hold, they do not pull in. I recall wondering how my old gas water heater could work, since it didn't need an AC power connection. And that was it - I provided the 'pull in" force when I lit the pilot, it only had to supply the hold-in force with that tiny power from the thermocouple ( µwatts? mV and mA I think).

jim hardy
Gold Member
Dearly Missed
the kind Jim Hardy was referring to. Those only hold, they do not pull in.

Actually on the gas fireplaces i've encountered there's a low millivolt thermocouple, about twelve millivolts for the hold in valve that operates as you say, with a pushbutton for safety

and a thermopile about 500 millivolts for the on-off valve that's controlled by a wall switch....or sometimes by a thermostat(yikes! that's outlawed in a lot of places) ..

https://customer.honeywell.com/resources/techlit/TechLitDocuments/69-0000s/69-1219.pdf

Last edited:
Variations on the theme exist, but thermocouples come in grounded or ungrounded, and shielded or exposed junction types. On this one the junction is exposed, and equipped with a grounding wire.

If it were a shielded type (rather than being contained within a thermowell like the one @dlgoff described) it would typically be contained within a closed stainless steel or Inconel tube packed with magnesium oxide (to insulate the connection wires), and the junction would either be welded to the tip (grounded type), or offset back a bit to allow headspace for MgO insulation (ungrounded). In a shielded thermocouple, the grounded type has slightly faster response due to the direct connection between junction and sheath. An exposed junction TC has even faster response.

When used in the furnace, did it measure and control the temperature of the air within, or was it positioned directly in the flame as part of a proven ignition safety circuit? An ungrounded thermocouple can't form ground loops, and is generally preferred from a measurement perspective. I've never seen this sort of TC either, and unless it is part of a gas burner safety circuit the reason behind the intentional grounding wire is throwing me.
Thanks. The furnace does not work. I should find appropriate thermocouple and temp-controller to warm up the furnace. In the manual of this furnace they just said the typeK TC, thermal controller and heating element. there is well for the thermocouple with ceramic shield. Please find more captured photos for this furnace 's thermocouple.

#### Attachments

• IMG_1269.jpg
30.7 KB · Views: 546
• IMG_1270.jpg
26.2 KB · Views: 526
• IMG_1271.jpg
33.6 KB · Views: 510
• IMG_1272 (2).jpg
38.8 KB · Views: 546
jim hardy
Gold Member
Dearly Missed
The furnace does not work.

You're troubleshooting a gas furnace ? I hope it's for a kiln or something, not heating an occupied building.

Read its resistance both ways with a DMM you should get very few ohms.
then
Connect a meter to the two wires at signal end and heat it gently with a flame - barbecue lighter or something even a candle.. You should see roughly two millivolts for every hundred degreesF.

Do not ever try to fix a gas valve in a heating system - just replace it. .

If this is for heating an occupied building - call your furnace serviceman there's huge liability and PF can't risk being party to amateur repairs. .Nor can I.

old jim

dlgoff, Asymptotic, berkeman and 1 other person
berkeman
Mentor

berkeman
Mentor
Do not ever try to fix a gas valve in a heating system - just replace it. .

If this is for heating an occupied building - call your furnace serviceman there's huge liability and PF can't risk being party to amateur repairs. .Nor can I.

dlgoff
berkeman
Mentor
After a PM conversation with the OP, this thread can be re-opened. It does not look like there is a CO risk in his project.
this tube is an electric tube furnace, not a gas one. Also, this wont use for building with people, this is very small and we use it just for mm size ceramic samples.

Artyman
jim hardy
Gold Member
Dearly Missed
Whew ! Thanks, guys !

berkeman
After a PM conversation with the OP, this thread can be re-opened. It does not look like there is a CO risk in his project.
Thank you for consideration. Yes it is an electric tube furnace for lab work (mm size ceramic fabrication). I hope every one could get it from the right point. I am not a technician.

Thanks. The furnace does not work. I should find appropriate thermocouple and temp-controller to warm up the furnace. In the manual of this furnace they just said the typeK TC, thermal controller and heating element. there is well for the thermocouple with ceramic shield.
... it is an electric tube furnace for lab work (mm size ceramic fabrication). I hope every one could get it from the right point. I am not a technician.

What is the manufacturer and model number of the furnace?

• What temperature is displayed by the temperature controller?
• Does the controller have a lamp/LED/something else? that turns on to indicate when heating output is commanded on?
• What is used for heater power control? A solid-state relay, electromechanical relay, or something else?

There is a good possibility the root cause of this 'does not work' problem is not, or at least not exclusively, a thermocouple issue.

I'm betting those brass (or perhaps copper) terminals were originally parts of the broken ceramic device the thermocouple extension wire runs through. What lies underneath looks like a thermally insulated pass-thru hole to route TC extension wire out of the oven chamber, and to the controller.

Wires from the TC itself, and TC extension wire are in contact with each another under the screw of each brass block. A thermocouple is formed any time two different metals come into contact with one another, and (if the wires weren't in contact with each other) each of these connections would exhibit thermoelectric behavior.

Generally speaking, the possibility of short circuits between wires is greater when thermocouple (+) and (-) and (in this case) ground wires are poorly insulated and in close proximity to one another (shown circled in red). This ought to manifest as a displayed temperature remaining suspiciously lower than expected - the measured temperature signal is an average of the intended thermocouple junction and 'false' thermocouple created where the wires have shorted together.

You're troubleshooting a gas furnace ? I hope it's for a kiln or something, not heating an occupied building. ...

Do not ever try to fix a gas valve in a heating system - just replace it. ...

old jim

Agreed, and I'm glad the thread was re-opened. Although a little off-topic from the OP's application, sharing some safety info is always a good thing I think. So I will share this:

About 30 years ago, I had the thermocouple sensor in my gas furnace valve go bad. I knew enough to troubleshoot it down to that component (thermocouple was good), but the local repair outlet would not sell me the replacement part. I found a "thermocouple tester" in the hardware store, which consisted of this sensor and an indicator light, so I bought the "tester" and was going to use it to replace the bad sensor in the gas valve. If memory serves, there just wasn't any good way to get the valve apart to replace that component, so I think I ended up with a whole new gas valve anyway. Or maybe I was able to fix it, I just don't remember.

But it happened again a few years ago in my newer home on a ~ 15 year old furnace. This time, it was clear that there was absolutely no way you could take that valve apart to replace any component. In one way, I was mad about that - it seems a shame to throw away an entire valve because one little $5 component was defective. But after thinking about it some more, I realized that if that component could be replaced, it could also be by-passed. And if it can be done, some one will do it. So I resigned myself to the idea that safety for all was more important than my being able to do a repair. In industrial settings, where only licensed repair techs will be doing work, I believe these systems may be more modular which means you could replace a component, but also bypass that component. And hopefully a trained tech would not do such a thing... but I wonder? Artyman jim hardy Science Advisor Gold Member Dearly Missed Ahh an electric kiln of sorts. A thermocouple in proximity to sizeable electric heaters - grounding it seems a good idea. I never encountered one like that, mine were in thermowells like the one dlgoff posted. If memory serves, there just wasn't any good way to get the valve apart to replace that component, so I think I ended up with a whole new gas valve anyway. Or maybe I was able to fix it, I just don't remember. Some of them look like a copper tube going into the valve My local hardware store carries them in several lengths https://www.robertshaw.com/ProductDetail.aspx?id=2147492934&cat=2147484248 Replacing the thermocouple is about all you can do to them . dlgoff Agreed, and I'm glad the thread was re-opened. Although a little off-topic from the OP's application, sharing some safety info is always a good thing I think. So I will share this: About 30 years ago, I had the thermocouple sensor in my gas furnace valve go bad. I knew enough to troubleshoot it down to that component (thermocouple was good), but the local repair outlet would not sell me the replacement part. I found a "thermocouple tester" in the hardware store, which consisted of this sensor and an indicator light, so I bought the "tester" and was going to use it to replace the bad sensor in the gas valve. If memory serves, there just wasn't any good way to get the valve apart to replace that component, so I think I ended up with a whole new gas valve anyway. Or maybe I was able to fix it, I just don't remember. But it happened again a few years ago in my newer home on a ~ 15 year old furnace. This time, it was clear that there was absolutely no way you could take that valve apart to replace any component. In one way, I was mad about that - it seems a shame to throw away an entire valve because one little$5 component was defective. But after thinking about it some more, I realized that if that component could be replaced, it could also be by-passed. And if it can be done, some one will do it. So I resigned myself to the idea that safety for all was more important than my being able to do a repair.

In industrial settings, where only licensed repair techs will be doing work, I believe these systems may be more modular which means you could replace a component, but also bypass that component. And hopefully a trained tech would not do such a thing... but I wonder?
Agreed, and I'm glad the thread was re-opened. Although a little off-topic from the OP's application, sharing some safety info is always a good thing I think. So I will share this:

About 30 years ago, I had the thermocouple sensor in my gas furnace valve go bad. I knew enough to troubleshoot it down to that component (thermocouple was good), but the local repair outlet would not sell me the replacement part. I found a "thermocouple tester" in the hardware store, which consisted of this sensor and an indicator light, so I bought the "tester" and was going to use it to replace the bad sensor in the gas valve. If memory serves, there just wasn't any good way to get the valve apart to replace that component, so I think I ended up with a whole new gas valve anyway. Or maybe I was able to fix it, I just don't remember.

But it happened again a few years ago in my newer home on a ~ 15 year old furnace. This time, it was clear that there was absolutely no way you could take that valve apart to replace any component. In one way, I was mad about that - it seems a shame to throw away an entire valve because one little \$5 component was defective. But after thinking about it some more, I realized that if that component could be replaced, it could also be by-passed. And if it can be done, some one will do it. So I resigned myself to the idea that safety for all was more important than my being able to do a repair.

In industrial settings, where only licensed repair techs will be doing work, I believe these systems may be more modular which means you could replace a component, but also bypass that component. And hopefully a trained tech would not do such a thing... but I wonder?
Thank you for sharing the safety notes. But please do not go again in Gas Furnace topic. It is not a gas furnace at all. I like to use the information related to electric tube furnace and do not like to see this topic is closed again! again thank you for sharing your information.