Wanted to first say thanks for the previous responses on transformers, by upping the frequency I think I found a small-ish part that might work.
I'm trying to do some semi-accurate current measurements and I have a shunt resistor with 4-wire connection. Shunt is ground referenced (low-side), current is ~2-4 A.
On the PCB would I route the sense wires as a differential pair with the (-) trace terminating at the op-amp's GND? Op-amp is 2-3 inches away from sense resistor. I worry about noise coupling into the pair, I guess I could also surround it by grounded-guard traces/islands.
You measure the voltage across a four terminal resistor with a differential amplifier or an instrumentation amplifier - current sensing resistors are usually very low impedance parts, so a differential amplifer is usually plenty good enough.
The voltage on the low side sense trace isn't the same as the ground at the op amp, and connecting it to the ground at the op amp would turn your four terminal resistor into a three terminal resistor.
What you have to do is to subtract the DC voltage on the low side sense trace from the DC voltage appearing on the high side sense trance, and refer it to the gound voltage at the place where you use/ measure/digitise the voltage drop across the sensing resistance.
Any voltage difference between the two trances will be picked up in this operation, so the two traces should be routed close together. If they were loose wires you would twist them to minimise the area included in the loop around the sense resistor, the sense eads and the op amp input stage. You can pull a similar trick with printed circuit traces by crossing them over at regular intervals. Routing them over solid ground plane helps a lot; grounded guard traces on either side improves matters a bit more but it is rarely worth the trouble.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.