Load life stability of enameled copper wire used for TC adjustment?

[aronake]

I well know method of adjusting TC is to add some copper wire with high TC.

Have anyone done experiments, seen any information or have views on what "load life stability" these coper wires may have? the trade off is to reduce TC with 1 ppm / K, but introducing X ppm / year in drift this may not be a good idea, but would depend on application.

There have been cases where i have added 1 ohm to a 10K resistor to get it compensated. So even if the copper wire resistance would move 1% total change would be very small, but may add to the load life stability of the base resistor.

When I have done these compensators I have covered them in a thin layer of epoxy to reduce future air impact. Does that sound like a good or a bad idea? If they have some load life stability issues, and other ideas how to best treat them?

[dietert1]

Copper forms oxide layers that become a protector after some time. Remember that copper is/was used on the roof. One could do an accelerated aging test to see how resistance is affected by the formation of oxide. It might depend on wire thickness (total amount of copper used).
Using transformer wire coated for isolation should be a good method. Don't know whether epoxy can harm the coating.

Regards, DIeter

[aronake]

Copper forms oxide layers that become a protector after some time. Remember that copper is/was used on the roof. One could do an accelerated aging test to see how resistance is affected by the formation of oxide. It might depend on wire thickness (total amount of copper used).
Using transformer wire coated for isolation should be a good method. Don't know whether epoxy can harm the coating.

Regards, DIeter

This is very much in line with my thinking. But at the same time, quite decent resistors can have quite some load life drift, which should still be much less than a good old copper wire in air (save for oxidation layer once that is formed and the coating).

[WattsThat]

There is negligible dissipation in the 1 ohm resistor as it is the ratio of the two resistors. If you have 100mW in the 10k, you’ll have 10uW in the 1 ohm. Do you think that will drift due to heating?

[donlisms]

Furthermore, a drift of 1% has an effective drift of .01 / 10k = 1ppm.

Just by intuition, it seems unlikely to have a drift of 1%.