EEVblog #419 - Thermocouple Tutorial

Everything you need to know about how Thermocouples work.
K type thermocouples, the Seebeck effect, the Seebeck coefficient, and cold junction compensation.
Along with some practical measurements with a multimeter to demonstrate the effect.

Seebeck effect on a single conductor:
NIST Tables:

EEVblog Main Web Site:
EEVblog Amazon Store:
Donations:
Projects:
Electronics Info Wiki:

[johnsobota]

Hi Dave - very helpful. BTW you misunderstood the NIST table. Everything in the whole table is referenced to 0 degrees C. The 0-10 columns just give you the last digit of the temp in degrees C. For example if you want to know the mV value corresponding to 1005 degrees, go to the 1000 row and look under column 5.

[sallowsandy]

Amazing, I simply became captivated when you performed your Seebeck experiment with the looped wire. These kinds of things are really what sets apart "just a tutorial" from "an explanation"

[EEVblog]

The Seebeck effect does apply to a single wire, and that is precisely why thermocouples work. A thermocouple is two different metals with two different seebeck effects. It's just that a single wire seebeck effect is not practical, but it does actually exist. See the link in the video description.

[tzampini]

An interesting thing about thermocouples that many may not realize is that although they generate a relatively small voltage, they can generate a high current if their load resistance is low, and the temperature differential is high. The current can be more than half an amp, maybe up to an amp or more. An example of this is a furnace that uses a thermocouple to monitor the gas pilot flame. In this application, it must generate enough current to hold a solenoid open to allow gas flow.

[tomrobottom]

Hi  The bit about the compensation table is wrong.  The scale along the bottom is just to provide the 1 Deg. steps. 

The vertical scale goes up in 10 deg. steps so you go the the 10 deg step that is lower than the temperature that your cold junction is at and the use the horizontal table for the 1deg finer steps.  OK ?

[revealingfactsall]

Brillant! This video and the tear down of the fluke before it are just excellent. I love how this video tied in the mechanics (iso thermal block) of the tear down. Really love this style or combination of videos. I've got a EE degree and been in the buisiness (mostly software side of things) for around 18 years and your knowledge of the subject is expert and your ability to present it in easy to follow format is brilliant. Topics I've not seen since my college days are coming back to me

[EEVblog]

The seebeck effect does happen on a single conductor, it's just not called a thermocouple in this case. A thermocouple by definition uses two dissimilar junctions.

[EEVblog]

In theory, yes, it should cancel out, as it's effectively a thermocouple with two identical metals with identical seebeck coefficients. In practice, it can produce a small dynamic effect like you saw. You may have noticed it actually started going back down the longer the iron was on there.

[abhijeetveerkar]

Great Tutorial. Very good explained. The great thing about you dude is you opened up your instrument for viewers to increase our knowledge. That's the great Seebeck Effect we got :-) . Thanku !

[EE_fun]

I've read about this effect in a physics book, and it says that of course you will get a voltage across the wire if you are using just one metal, but it will be an inner field, which you can't measure from outside with a probe.
So if you are using just one wire and getting a voltage reading, it will be due to other materials (metals) in your measuring circuit. Thats why you have to have the other metal with, of course, a different seebeck coefficient, to not zero out the voltage drop.

[peterbarns]

No doubt, Mr. Jones is god. I could sit 24/7 and watch this man but I've got my own miracles to pull. Thanks for the videos, Mr. Jones.

[karlg.f.]

In the new videos about thermocouples with multimeters, you mentioned this old one - great and informative as usual. However I think I found a flaw at around 12:01 Minutes when you explain the NIST K-Type table.. I think it’s not cold on the X-axis and hot on the Y-axis. It’s a linear list from -270 - 0 °C in one degree steps. If you want to know the value for -13°C, you go the „-10“ line in the „-3“ column and get the value of -0.508 for -13°C.

[PyroRob]

Have watched a lot of videos on TC's and the people making a big deal about the bead on the business end of the TC wire. They go to great extents to make sure there are those same two types of wires through the entire length from the tip back to the meter. All you really need to care about is making only one transition.

When I worked for a US defense contractor in the 80's, I used to make thermocouples for the various temp tests from TC wire on spools of wire. I would cut off 10' of wire, strip and twist one end, and give it a blob of solder, then run the wire back to the patch panel, which was all standard nickle plated bus bars. No need for all the fancy plugs, extension cables, etc. Can use a 16 AWG extension cord if you need to. Besides, what you connect that TC to will not be made of the dissimilar metals.

[UberAlphaSirus]

Dave, thanks, I have pondered ocer this cold junction comp for about 3 months now, I didn't have the time to figure it all out myself. You cleared it all up in 30 minutes. These type of dare I say it "educational" videos is where you excel, clear concise and maybe the odd mistake but hey, they happen and people should take it on the chin. Thanks again

[EEVblog]

Yes, it's all related the thermoelectric effect. There is the "seebeck effect and the "peltier effect", sometimes joined and called the "peltier-seebeck effect".

[UAtubeviewer]

Thanks a lot for your post. This is the most professional explanation about the work of thermocouple I could find on YouTube.. Great job!

[waswestkan]

 Wow more involved than I expect rd this to be, thanks for taking the time to teach some of us, Dave

[ThatElectronicsFool]

Funny that I keep coming across thermocouple stuff right now that it's relevant to my interests... I've been looking at making a reflow oven and am collecting parts at the moment.

[chizzt]

Worth mentioning that the Curie Kink at about 300c was used to regulate early Weller soldering irons..

[neilbradshaw]

At the begining of the video I think the demonstration of the seebeck effect with its explanations could be wrong. The soldering iron was placed near centre of the wire and the only reason for the voltage going up was because the soldering iron was not dead centre. I bet if the iron was place 30% of the way along the wire where the conducted heat would not really heat the banana jack the voltage output would be higher than in your example.