On 26/08/2014 16:05, Mike S wrote:
After some more research, I think I've answered some of my own
questions -
Tellurium copper is used for binding posts, not because it has any
special thermal or EMF mojo, but because it machines much better than
pure copper. And, I suppose, because it sounds like it's extra special.
The Seebeck coefficients (uV/C, relative to Cu) of some relevant
materials:
Cu 0.0
Ag .2
Au .5
Yellow brass 1.5
Phosphor bronze 2.0
63/37 solder 3.0
Sn 3.1
Stainless steel 3.1
Beryllium copper 5.0
Fe -12.3
Ni 22.3
Te -49.25
Based on the extreme Seebeck coefficient of pure tellurium vs. copper,
I'd expect that there might be some coefficient between Cu and CuTe
(0.5% Te), but I could find no reference. The relatively large number
for CuBe is interesting, since that's a common material for banana
plug springs, where one might expect the greatest temperature
differential to occur in such a connection (between the thermal masses
of the binding post/jack and the bulk of the banana plug). Heat has to
flow a considerable distance through the springs, very much more than
when it flows through a surface plating.
The Pomona (Fluke) EM5295-48-0# uses CuBe (gold plated) for the spring
contacts. It seems there might be an improvement to be had by using
the older style pin plugs, where a solid pin was partially sliced into
4 sections which were then spread apart a bit to create tension. That
could eliminate relatively large thermocouples at a thermal gradient,
and might also be expected to have less thermal resistance, allowing
the connection to settle quicker.
But maybe not - I'm still not clear on how plated conductors behave in
this situation. For a high impedance voltage measurement where almost
no current flows, the gold plating may carry the signal, so there is
no real thermocouple (or more correctly, it's entirely contained
within the connector). But if that's the case, why fool around with
special copper connectors when common brass ones would be
easier/cheaper? For current or resistance, the signal would also flow
through the base metal, so does this have an effect (especially for
tinned copper test leads, where there may be a larger temperature
difference between the ends???
Nickle is avoided as a contact material largely because it is subject
to fretting corrosion. Tests done by AMP
(http://www.te.com/documentation/whitepapers/pdf/p154-74.pdf) show
that a Ni to Ni contact can increase from 8 mOhm to 5 Ohms (sic!) in a
short time due to this, while Ag and Au plated contacts exhibit
negligible changes.
Cu (with Be for better machining) seems to be used as the base
material for jacks/plugs to get thermal EMF cancellation to the wiring
on both sides (i.e. use copper everywhere except where there is a
minimal thermal gradient, like platings).
Beryllium copper is a springy material, Tellurium is the material added
to aid machining without adding too much seebeck coefficient.
I remember someone on this list a long time ago saying that NPL used van
damme star quad cable and bought bulk quantities of spade lugs that they
strip all the coating off before crimping and clean before each use.
_______________________________________________
volt-nuts mailing list -- volt-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/volt-nuts
and follow the instructions there.
