Connect with us


Discussion in 'Electronic Design' started by Danny, Nov 26, 2003.

Scroll to continue with content
  1. Danny

    Danny Guest


    I am making a thermocouple from 28swg copper and constantan wire, i plan
    to twist the ends together and solder the joints, will this solder have
    any affect on the characteristic of the thermocouple? I only chose
    soldering because a spot welder is a little bit out of my league.

  2. No. Be sure to get the right alloy if you expect type T output.

    Best regards,
    Spehro Pefhany
  3. Mac

    Mac Guest

    You could probably fuse the wires with a car battery. I once turned a
    piece of 9-strand 12 gauge cable into single strand cable by accidentally
    shorting it, very briefly, across a car battery's terminals. Come to think
    of it, you could use a large capacity slow-blow fuse to limit the total
    energy. Just a thought.

    Of course, these kinds of hit-and-miss techniques are not for everyone.
    But the kind of person who makes his own thermocouple seems a good

  4. Roger_Nickel

    Roger_Nickel Guest

    Ordinary solder will not work. If you must solder, use silver solder.
    You will end up with two thermocouples (copper-easyflo,
    easyflo-constantin) but even though technically dodgy it works O.K at
    low temperatures. Spot welding is definitely best , maybe you can
    improvise. I once worked with a thermocouple welder which used a
    microswitch to switch power to the primary of a low output voltage
    transformer. IIRC it had about a dozen turns of no. 8 copper wire on the
    secondary. The microswitch was tripped by a weight falling down an
    adjustable inclined rail and holding the switch closed as it passed.
    Simple and effective, necessity is the mother of invention.
  5. Attempting to spot weld round wires together with an improvised welder
    is a good way of making a sensor both amateurish and unreliable. Inert
    gas shielded welding is the "right" way, BTW.

    Best regards,
    Spehro Pefhany

  6. I've heard that chemical impurities in the junction affect the accuracy of
    the thermocouple. (I know nothing about it, myself.)

    Does anyone have a quantitative sense of the extent of the effect? Like,
    how much worse is a hand-soldered thermocouple from a professionally-welded
  7. No, not if they are confined to the *junction*.

    Best regards,
    Spehro Pefhany

  8. Which fits what John Woodgate said, too.

    So why not always just solder? Is it a matter of dealing with extreme
    temperatures (past the melting point of the solder)?
  9. That's reassuring. ;-)
    (except for a few application areas, mostly T/Cs *are* used above the
    melting point of solder or at cryogenic temperatures)

    * Many (most, actually) T/C alloys are not easily solderable or "wet"
    very poorly
    * Temperature range severely limited (most base metal alloys can go
    to 500-1000°C and precious metal alloys to maybe 1750°C.)
    * Poor mechanical strength of the joint
    * Cost is not really lower (just capital cost of equipment)
    * Lead-based solders are not food-safe or are not chemical resistant

    Best regards,
    Spehro Pefhany
  10. Walter Harley wrote:
    That is what I understand. Any soft solder, by definition, has a
    lower melting point than the wires being joined. A weld is rated for
    as high a temperature as the wires.
  11. I read in that Walter Harley
    Well, somewhat below the actual melting point the tin migrates out of
    the solder and the joint becomes bad.
  12. But it will still work for a while even with the solder melted,
    assuming something like surface tension holds it in place, so the
    exact limits are kind of inexact.

    Similarly, the upper limits for base-metal thermocouples (even when
    manufactured optimally) may be much lower than the melting points of
    the alloys because of service life considerations.

    Best regards,
    Spehro Pefhany
  13. Boris Mohar

    Boris Mohar Guest

    Would arc welding work? The method I have in mind is to use the TC junction as
    one electrode and arc directly to it from another (carbon?) electrode
    ( AC? DC?) until a bead forms.


    Boris Mohar

    Got Knock? - see:
    Viatrack Printed Circuit Designs
    Aurora, Ontario
  14. That's pretty much how it's done, but you need the whole kit with (the
    right) inert gas and so on to do it right. The Home Depot specials
    won't do the trick.

    Best regards,
    Spehro Pefhany
  15. Winston

    Winston Guest

    Like this.

  16. Barry Lennox

    Barry Lennox Guest

    Yes, but only with the correct scale of equipment, otherwise it would
    be a nightmare.

    There's several commercial systems that either dump a cap (usually
    variable for power control; about 35-75 volts) through the junction,
    or use a step-down transformer as spot-welders. A previous employer
    had one for the former, and they work very well.

    For my own use, I use a modified microwave oven transformer with 3
    thick turns on the sec. The primary side is controlled by an SSR,
    variable from 0.05 to 1.0 secs. A copper anvil and small knife bar on
    the sec hold the TC wires together, and splat! it's done. I have made
    perhaps 200+ TCs this way without any problems whatsoever.

    Barry Lennox
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day