I am using a cap discharge welder to weld Stainless steel to a mild steel galv plate. I know this should not work but I have managed to get the parts to "stick" enough so I can bowl feed them to a final assembly. My question is how do I measure the resistance between the electrodes so I can accurately track the differences between the parts. I am getting differing results when welding because the parts are different in their density so the resistance will differ between the electrodes. I can us a scope for the voltage but not sure a multi meter will be accurate enough to give significant results.

  • \$\begingroup\$ Are you seeking a type of quality control ? Able to discern between different parts or difference in the weld. Are you wanting to know resistance before welding? Perhaps you are looking for Welding Signature Analysis ? \$\endgroup\$
    – Marla
    Jul 30, 2016 at 21:19
  • \$\begingroup\$ In some cases a constant-current welder is needed. If you can't get a good filet weld then slow down and wiggle the rods. Are some of the welds oxidizing? \$\endgroup\$
    – user105652
    Jul 30, 2016 at 21:24
  • \$\begingroup\$ If you clean the pieces well at the location where they make contact, then clamp them together with approximately the same force as your spot welder would apply, then you should be able to use an Ohm meter to measure the resistance between the pieces. The resistance will change of course if a spot weld is performed due to heating effect. \$\endgroup\$
    – zeffur
    Jul 30, 2016 at 21:47
  • \$\begingroup\$ I wouldn't attempt to use an ohmmeter, I'd use good old E=IR. Add an ammeter shunt so you can measure the amps of flow, and it's trivial to measure voltage at the electrodes. \$\endgroup\$ Jul 30, 2016 at 22:37

1 Answer 1


If you have a power supply with constant current mode that will work down to a short (most linear ones will, the 50A switching ones I have kick out at about 1-2V so would be more problematic) simply pass that current through the part and measure the resulting voltage drop with a multimeter.

Suppose you have 5A available, most inexpensive meters can resolve 100uV, so you can 'see' 20u ohm resolution, not bad! Using just the ohms function, a resolution of 0.1 ohm would be more typical, with lead and contact resistance adding more errors.

You want to use 4 wires to do this, the 'outer' ones supply the 'force' current and the inner ones to the meter sense the voltage. If done properly, only the outer ones carry any significant current and contact resistance hardly matters. This called a Kelvin connection and is the best way to measure extremely low resistances.

One minor tweak in your situation is to swap the 5A polarity and measure the voltage both ways since the different metals at each contact may cause a bit of thermocouple voltage.

Of course this is for the part without the welder connected. If you want to measure the welding current get a current sensor such as those made be LEM and connect the welding lead through it (also apply suitable power). You can probably set your scope to display resistance directly during the weld pulse if it can show V/I using two channels.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.