Electronics > Metrology

Fine resistance adjustment for standard resistors

<< < (2/2)

TiN:
Another jewel from zlymex :)

Vgkid:
Thanks for the information.

zlymex:
Thanks everyone for the remarks. I actually done nothing except teardown, photo and copy schematics from their manuals. ;)

Vgkid:
It is nice having everything in one spot.

splin:
So what differences are there in thermal EMFs between series and parallel arrangements? Consider the two examples zlymex shows in (3) Statistical arrays; assuming that each soldered joint has a 4uV/K potential, what might be the maximum total thermal EMF be if there is a temperature gradient of 100mK:

1) across the 20 x 20.01 Ohm parallel resistors in the Fluke 742A-1 (roughly top to bottom in the photo)?

2) across the Fluke 752A's nine 120k, one 119.8k and one 200 ohm resistors in series (say left to right in photo)? Would bottom left to top right be worse?

In series there is the potential for all the EMFs to add up but of course thermal symmetries should allow most to cancel out if the temperature gradients are uniform - especially no turbulent air currents, but parallel still looks to be the better option? Of course it's often, but not always, possible to cancel out thermal EMFs when using such instruments by reversing connections but it would still be better to minimize them.

This also applies to the series/parallel trimmer arrangements though avoiding having a resistor in series with the main resistor altogether would require unrealistically high trimmer values (with poor stability and TCs) for anything above a few hundred ohms - ie. a 1k resistor would require a 10Mohm parallel trimmer for 100ppm trim range. (The trimmer itself could comprise series fixed and variable resistors without adding significant EMF because of the 1k/10M divider).

This also begs the question as to what might be typical temperature gradients encountered within calibration instruments in a lab environment? Would 100mK be much higher than reality for a passive device such as a Hamon divider which doesn't contain a significant heat source? What about when it's placed on or near a powered instrument such as a DVM dissipating 20 or 30W?

[EDIT] Thermal EMFs will probably be small at soldered joints between resistors because the temperature difference across the thin solder layer between the two wires will be small. However thermal EMFs will also arise in the connections within the resistors, between the resistance wire and the lead-out wires. Also variations in the composition of lead-out wires and connecting wires subject to temperature gradients will also generate thermal EMFs, thus resistors connected in series ought to come from the same batch to minimize this problem (but even resistors from the same batch may not necessarily have all lead-outs made from the same reel of wire).

Navigation

[0] Message Index

[*] Previous page