Close

Failure analysis (ohm's law rehabilitated)

A project log for (re)Mangetizing neodymium magnets

My attempts at replicating what Polymagnet are doing

deepsoicDeepSOIC 11/24/2016 at 23:062 Comments

As I wrote in previous log, the current in the first attempt was way lower than I expected. I want to investigate, what actually happens, so that I can predict the outcome next time.

I've added a current shunt resistor (0.015 ohm, should not affect the circuit). And captured both current and voltage traces of the pulse.

Turns out my quick estimate of 80 A was way off. The actual peak current was 150 A, which is only two times lower than 300 A I wanted.

That is of course assuming the shunt resistor is accurate at currents this high. Which I'm not really sure about, but I hope that's the best resistor for the purpose, on account of being small, and its resistive layer being very very thin.

Is this waveform as one would expect? Not sure yet, maybe it's time to add some SPICE to the meal =)

UPDATE:

Wow, that is almost spot on!

I didn't know the ESR of capacitor, so I just gave it a guess to be 0.2 ohms. The simulation waveforms are very similar to the ones I measured.

So, THERE'S NOTHING WRONG WITH OHM'S LAW. "The problem was between driver seat and steering wheel".

UPDATE2:

I tried to acquire the pulse with larger capacitor, but current sense resistor exploded. Oops.

Does it still obey Ohm's law? =)

Discussions

Jarrett wrote 11/24/2016 at 23:12 point

You've got a poorly tuned resonant LC tank :P

I'd expect a little bit of bounce, and non-linear discharge curves like that.

  Are you sure? yes | no

DeepSOIC wrote 11/24/2016 at 23:20 point

Yep, it's an LC tank, apparently close to being critically damped, which is actually great!

  Are you sure? yes | no