PCB Meter Shunts

[German_EE]

Does anybody have experience or data regarding high current shunts made using FR4 PCB material? I need to make a 50A and a 100A shunt and I'm looking for alternatives to wirewound devices.

[dom0]

You probably know that, but cooper has a terrible tempco... like, 2000 times higher than good precision resistors, if I remember correctly. Another question would be PCB variance, the 35 µm cooper thickness has quite some of it according to people who measured that.

[SeanB]

You are stuck with either wirewound devices or with cutting a shunt out of some stainless steel plate. Copper board and a local temperature sensor then apply compensation scaling will work, but will lose accuracy at the high end and add an extra time constant.

[German_EE]

Stainless steel plate is an idea I had not thought of, thank you for that.

This is not a precision application, the meter will probably be calibrated in 5A increments so variation over temperature will not be a problem. I'm just worried about wirewound devices going open circuit and blowing up the meter movement.

[dom0]

Oh we're talking about electromechanical meters? I've seen many shunts for those apparently made of simple copper bus bar. I guess thermal drift isn't much of an issue with +-2.5 % or higher tolerance meters anyways.

[daqq]

See:

http://electronicdesign.com/power/battery-board-measures-current-without-shunt-resistor
http://www.embedded.com/design/automotive-design/4010022/Hall-effect-sensor-ASIC-integration-shrinks-current-transducers-Part-1-of-2-

[tautech]

I have made small current shunts using several strands of Constantan and ceramic bodied screw terminal blocks. Final "trimming" can be done with a crease from sidecutters and/or a dividing trimmer across your meter.

Googleing shows there is a better alloy for shunts: Manganin.
http://en.wikipedia.org/wiki/Manganin

[German_EE]

Success!!  Three pieces of 15mm wide clock spring each 31mm from clamp to clamp. One piece can be used to make a 50A shunt and two pieces can be used for a 100A device, adjust distance between clamps for calibration. They glow cherry red under full load but this is not a problem.

A fun way to spend a Saturday morning.

[TimFox]

Do they get hot enough to light a cigarette?

[German_EE]

Probably, I have not smoked for over thirty years. As I said the portions of spring glow cherry red under full load so they are probably operating at the limit.

[mzzj]

Success!!  Three pieces of 15mm wide clock spring each 31mm from clamp to clamp. One piece can be used to make a 50A shunt and two pieces can be used for a 100A device, adjust distance between clamps for calibration. They glow cherry red under full load but this is not a problem.

A fun way to spend a Saturday morning.

And you think that this will be more reliable than wirewound resistor? 

I would use* something like this if I want to be cheap http://www.ebay.com/itm/Analog-Current-Meter-Divider-DC-100A-75mV-Shunt-Resistor-for-Amp-Meter-Ammeter-/161366204100?pt=LH_DefaultDomain_3&hash=item25922cdec4

* I have bunch of 50 to 600 amp shunts like this that I have kept from junkyard equipment:
http://www.thermovolt.se/index.php/products/standard-60-mv-shunts/without-socket/20-150-a-60-mv-0-5