# Thermocouple

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

1. ### DannyGuest

Hi

I am making a thermocouple from Copper/Constantan, i plan to twist the
ends together at the appropriate junctions, and then solder them in
place, a spot welder is a little bit out of my league, would the solder
affect the performance of the thermocouple or not?

Danny

2. ### John WoodgateGuest

I read in sci.electronics.design that Danny <>
Provided it doesn't melt, no. The thermal voltage is not developed at
the junctions, contrary to popular belief, but in the wires themselves.
Even though you may have two junctions, copper-solder and solder-
constantan, they are both at the same temperature, so there is no
spurious voltage.

3. ### James MeyerGuest

This is 100% in line with my research into thermocouples.

Jim

4. ### JeffMGuest

Even though you may have two junctions,
This assumes that the copper-solder junction
produces a voltage which is equal and opposite
to that of the solder-constantan junction
(facts not in evidence).

5. ### John WoodgateGuest

<>) about 'Thermocouple',
The junctions DO NOT produce any voltages, contrary to popular belief.
Thermoelectric voltages are produced by the temperature gradients in the
wires.

6. ### Spehro PefhanyGuest

If there is no temperature difference, then there is no voltage.
Always.

Best regards,
Spehro Pefhany

7. ### John PopelishGuest

Junctions do not produce voltage. Changes in temperature along
conductors is what produces thermocouple voltage. As long as the
solder joint is small enough to be at essentially one temperature, it
produces no error voltage.

8. ### James MeyerGuest

The facts that *are* in evidence state that the voltage generated by a
thermocouple is *not* generated at the junctions. The voltage is generated
along the length of the conductors due to the temperature differences between
the ends of the conductors.

In point of fact, a junction is not required at all for the generation
of a voltage. A conductor with a temperature difference between its ends will
generate a voltage. No junction is required unless you want to make use of a
current flow due to the voltage generated.

A little research before you post will reduce the number of times you
put your foot in your mouth.

Jim

9. ### Luhan MonatGuest

Wow, this is not what I learned in my university physics class! It all
had to do with how the free electrons interacted at the junction. Also,
I have never heard of voltage created by temperature gradients.

Do you have some acedemic reference your position? I would like to
check this out and settle the matter for myself.

10. ### Spehro PefhanyGuest

No, it's the free electrons in the wires, which have a temperature
gradient along them. The distribution of the gradient in a homogenous
wire is not particularly important (law of Magnus). A T/C measures the
difference between those potentials in different metals.

An ideal T/C would have a zero-width junction. Obviously no
temperature delta across that with finite heat flow. Real T/Cs have
almost no delta-T across or even near the alloyed region that is the
junction.
Look up Seebeck (coefficient), as well as Peltier and Thomson (that
Lord Kelvin bloke).

Best regards,
Spehro Pefhany

11. ### John WoodgateGuest

I read in sci.electronics.design that Luhan Monat
No, I don't have any references; I don't have access to a university
library. But I feel sure you will find the information on the Net with a
Google search, with appropriate references.

12. ### John PopelishGuest

Do you have a name for the effect that causes this small gradient at
the junction, so I could look it up?

I have never heard of such a thing. I have used intrinsic junctions
made by welding the two thermocouple wires to some third metal object,
to measure the approximate average of temperature between the two
contact points. The composition of the metal between the two
thermocouple leads appears to have no effect on the temperature
reading.

13. ### Luhan MonatGuest

Yep, its the gradient in each wire plus a small contribution from the
junction itself.

14. ### John PopelishGuest

Here is an introduction to thermoelectricity:
http://www.chem.cornell.edu/fjd3/thermo/intro.html

I see no mention of a voltage at the junction of two dissimilar
materials.

Besides, if such a junction could produce a voltage (that could create
a continuous current) at a single temperature, it would violate the
conservation of energy. It would produce continuous power with no
absorption of energy from anywhere.

15. ### Charles DH WilliamsGuest

There is a so-called 'contact-potential' which is a contribution to
the electrochemical potential but other contributions (from a charge
double-layer at the boundary) cancel it out so there is no net
electrochemical potential difference across a metal-metal
junction at equilibrium.

Strictly speaking, a normal voltmeter measures electrochemical
potential difference, not electrostatic potential difference.

Charles.

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.
Continue to site