Measuring ultra-low ESR caps

[Harvs]

Has anyone got a simplish way of measuring the ESR of caps in the range of 5-50mOhm range? 

Not just for gross error (i.e. finding a cap that's 10x the ESR it should be), but to measure the ESR of a functional cap.

I've search around and haven't found much.

I'm thinking of just buffering the output of a sig gen with a power mosfet and 100mOhm current shunt, put a couple of amps through them, then use a scope to do a ratio measurement. 

Any easier ideas?

[robrenz]

Use one of the several relatively low cost LCR meters mentioned on this forum with .001 ohm ESR or series resistance resolution.

[kripton2035]

choose your esr meter here :
http://kripton2035.free.fr/esr-repository.html
regards,

[BravoV]

Since you mentioned a simplish way & easier, assuming you got yourself a decent sound card in your computer, and also a dual precision op-amp and good enough precision resistor (I bet you must have this somewhere in your component bins right ?), try Visual Analyzer, its freeware -> http://www.sillanumsoft.org/ZRLC.htm , should be pretty easy & doable even at breadboard.



Hope this helps.

[Harvs]

Thanks guys.  I guess something like the Agilent U1733C would do the trick, at about $400...

But I have to admit I'm still not entirely sure.  I would have expected when measuring 5-10 mOhm ESR you'd need be using a kelvin 4-wire system. 

Even just measuring a 5 mOhm standard resistor without a 4 wire system is pretty hopeless.

I can see a homebrew system being more accurate then any two wire system 

Wouldn't be too hard to build either.

Cheers

[robrenz]

IMO the Agilent is not a contender because it does not use 4 wire connections at all, not even the internal component clips
The Mastech and the DE-5000 both use 4 wire kelvin connections. I compared the  DE-5000 to my calibrated LOM-510A micro ohm meter on a nominally 50mOhm resistor.  LOM-510A reads 58.76mOhm (max error +/- 35uOhm) and the DE-5000 in series resistance mode at 1kHz reads a very slow bobble between 58-59mOhm and it is on 59 most of the time. I always promote LCR's on the forum for inexpensive accurate low ohms and this is why.

[Harvs]

Cool, I didn't realise the DE-5000 did 4 wire measurements until now, just watched the teardown of it and now understand how it's connected. Would probably still need a custom set of leads to do proper 4 wire measurements, but that's pretty easy. Something for the shopping list :-)

[robrenz]

Cool, I didn't realise the DE-5000 did 4 wire measurements until now, just watched the teardown of it and now understand how it's connected. Would probably still need a custom set of leads to do proper 4 wire measurements, but that's pretty easy. Something for the shopping list :-)

It comes with the alligator clip attachment that is 4 wire but the alligator clips themselves are not kelvin. Just cut them off and add your own kelvin clips, the source and sense wire and guard wires are there.  The connection box is very mod friendly as you can open it up and attach your own wires. The optional tweezer attachment is kelvin right up to the tweezer tips.  Just be aware the DE-5000 does not have any input protection.

[Harvs]

Yeah it looks from the video that the alligator clip box just uses a double sided PCB as the contacts that slot into the meter.  This would make it very easy, and quite cheap, to make different custom leads sets.

[Prizmatic]

Since you mentioned a simplish way & easier, assuming you got yourself a decent sound card in your computer, and also a dual precision op-amp and good enough precision resistor (I bet you must have this somewhere in your component bins right ?), try Visual Analyzer, its freeware -> http://www.sillanumsoft.org/ZRLC.htm , should be pretty easy & doable even at breadboard.


Hope this helps.

Very interesting, thanks.

[PA4TIM]

It arrived today.
I have not tested the ESR yet but the pF and uH ranges are very accurate. See the value on the cap, measured with my GR1620.

[robrenz]

Great PA4TIM!  look forward to seeing more in the "DE-5000 review thread."  (yet to be posted).  I will contribute low ohms accuracy testing to it as soon as I get a chance.

[Harvs]

Cool...