Resistance is Stainless

An Eccentric Anomaly: Ed Davies's Blog

A couple of weeks ago Paul at the End Of The Road was wiring up some little bus bars using aluminium and M5 stainless bolts. I commented:

Sorry if this sucks eggs but you do know stainless is a lot less electrically conductive that copper or aluminium, don’t you? Cu and Al resistivities are 16.8 and 28.2 nΩ·m respectively whereas stainless is 690 (generic stainless according to Wikipedia) or 770 (314 according to the second link below) so some 30 to 40 times greater.

I use stainless bolts and washers (A2) on my toy LiFePO₄ batteries but I make sure the tinned-copper lug on the lead is in direct contact with the battery terminal with the washer and bolt only providing mechanical connection and a secondary electrical path.

I suspect that if you put enough current through your bus bar to need that much aluminium the bolts will cause a small but noticeable volt drop and might even run a bit warm.

So I decided to measure this to make sure I wasn't saying something silly there. I set up my bench power supply to source about its maximum current, 5 amps, and “short-circuited” it through an M5 zinc plated bolt:

I took care to put the meter crocs directly on the bolt to measure just its voltage drop, not that of the connections:

This gave a voltage drop of 1.5 mV (millivolts) along the bolt. Not a lot.

Then I tried the same trick with the full length of a slightly shorter stainless (A2) bolt:

Now the voltage drop was 7.9 or 8 mV (it varied). Noticeably more than before but nothing like as much as I had expected for the difference.

However, the Wikipedia article referenced above gives the resistivity of carbon steel as 143 nΩ·m (i.e., approaching 10 times that of copper) so geometrically nicely between copper and stainless and matching these results quite well (143 nΩ·m for 1.5 mV drop, 700 nΩ·m for 8 mV drop).

Even 8 mV doesn't sound like a lot. Consider, though, Paul's relatively modest 940 W photovoltaic array. If used on his 24 V batteries in the new generator shed (not what he's using them for at the moment) they'd give a current of about 32 A so this same setup would give a voltage drop of 51.2 mV and a power dissipation of 1.64 W. That's a tiny fraction of the power the PV is generating and it's not going to even get more than a bit warm but it's still a loss you'd rather avoid having an unnecessary number of in your circuit.

Looking at Paul's pictures, he's not using quite as much of the bolt as a conductor as I am in this experiment. Still …

In general, with steel bolts, and particularly with stainless ones, it's well worthwhile making sure you've got a good copper-to-copper (or aluminium) connection rather than relying on current through the bolts, nuts, washers, etc.