Twist and solder k type thermocouple wires

[Faringdon]

Hi,
Dont  have access to the welder at the mo'
What temperature is it ok for, if i twist together the thermocouple wires, and then solder that twist?. I cant solder them if i dont twist the wires together since they wont take the solder, as the  vicious flux is at the main lab, not where i am now.

[jpanhalt]

If you are sure you can solder K-type, go ahead.  Brazing might be more effective.  You cannot rely on a twist alone.

With TIG or hydrogen torch, I can weld a TC in 10 seconds or less.  How many to you need?  Try "evaluating" a TIG welder before buying, assuming the dealer doesn't already know you.

[Stray Electron]

    I don't think that you can solder thermocouple wires together since the solder alloy will interfere with electrical response of the two materials used in the thermocouple.    I saw something in a old electronics magazine once where they welded a thermocouple joint together using a homemade carbon arc welder for welding together thermocouple joints. They used two carbon center electrode out of carbon zinc batteries for the electrodes and powered the circuit from a large 12 car battery and wired together with heavy gauge welding cable.  I don't remember if they used the center electrodes from a D sized battery or from one of the larger cylindrical batteries. 

[magic]

I don't think that you can solder thermocouple wires together since the solder alloy will interfere with electrical response of the two materials used in the thermocouple.

"The thermocouple" is the pair of wires running from the blob at the end to the plug, not the metal blob itself. When the ends of uncontaminated wires are at the same temperature as the blob, it doesn't matter what the blob is contaminated with.

There are bigger practical problems: the junction melts at 180°C and it's difficult to solder these metals in the first place.

I once repaired a K type thermocouple by soldering, using dilute phosphoric acid as flux. It passed the boiling water test closely enough.

[jonpaul]

Will oxidize and give false readings.

Easy to buy ready made from Omega, Fischer, Thermo-Electric, Fluke $10..30 with connectors

Jon

[jpanhalt]

As Magic stated, the "Law of Intermediate Metals" (https://www.omega.co.uk/literature/transactions/volume1/infared1.html ) applies.  Oxides may affect conduction or far more likely cause an open circuit with K-type.  Neither should give a wrong reading with a decent circuit. 

[Faringdon]

Thanks, the idea of the twisting, was to  ensure the wire ends are physically together, because as you kindly describe, soldering them is a little difficult.
It seemed to work OK though....and if there was oxidation, then its not likely to be able to insulate the wires from each other significantly, since the wires are twisted tightly together anyway.

[jpanhalt]

Good luck.  I tried tightly twisted K-type, and while it did work on occasion, if I touched it or wiggled it, I got an OC (open circuit) error with the TC amplifier (MAX 31856) I was using.  That amplifier is a bit problematic in that it gives over-voltage and OC errors probably when others don't.  Welding the TC's completely eliminated the OC error.  Adding solder that doesn't actually wet the wires is no better than adding epoxy. 

Your comment about needing to twist before soldering made me suspect at least one of the wires is not wetting/tinning.  I would tin both wires, then twist together and reheat with a little flux/solder.

Be careful about noise.

[jonpaul]

Hi: These chromel–alumel  metal alloys will NOT make reliable low R contact over time/temperature/vibration with either twisting OR soldering!

 YOU CANNOT SOLDER THEM !

Twisting will seem to work a while and eventually go intermittent or give false readings.

Suggest that you BUY READY MADE TC  or get a proper TC welder, Not that hard to DIY the welder.

Jon

PS:

We used and made 100s of Type K TC  in 1980s..1990s for QC/debugging of T1 and othet telecom rack terminal and network equipment.

Used Omega #32..36 Kynar type K, Omega connectors.

 welding 100s of pcs at a friend's lab. Took perhaps 1..2 hrs.

[Faringdon]

Code: [Select]

No regulatory will accept a PCB fuse as it is not reliable and will not break fire safe.Thanks, and in the face of it, you are of course correct.

However, we do not need them to work "over time"..but just for the immediate test that we are about to do......we will attach them to the FET....in 10 mins the FET will get up to steady state temperature.....and the soldered/twisted thermocouple wires will last ok for 10 mins...then,, during the following 2 hours of the soak test, any sudden changes in the FET temperature  would of course, indicate that the thermocouple had come to grief.......so we would at least know if it happened.