LCR meters and how to measure parasitics.

[David_]

Hello to you all.

A while back I bought a MASTECH MS5308 LCR meter, it's a low cost meter but not of the cheapest verity. Its my first "kind of real" LCR meter and when bought I thought that considering the specs vs price it was a very good perches, and I still do.
However it does show sometimes that it is some form of indirect digital measurement(at least I think that that is a viable interpretation of the results) like when I measure some small resistance and I can get the meter to indicate a huge inductance or while a small capacitance can be viewed as a few H coil.
But I tend to trust the displayed D, Q ESR and θ but when bought I thought that given the ranges this meter has I should be able to use it to measure some parasitics, but it would seem as I was wrong.

But I'm not sure if that is correct since I don't know "How does one measure the parasitic inductance of a capacitor/resistor, or the capacitance of a inductor/resistor?"
First I thought that I could hook up a capacitor the the LCR meter and set it to measure inductance and in that way find out the ESL of the capacitor, or measure the capacitance across a toroidal wound inductor... But at least with this meter that is giving me results that makes me question if that at all is how one deduces the parasitics of a component.

How do you measure the capacitance of a inductor and the inductance of a capacitor?

 

[volker@muehlhaus]

How do you measure the capacitance of a inductor and the inductance of a capacitor?

You can only do that indirectly by measuring at different frequencies. For example an inductor with parallel capacitance between the turns will show an increase in L over frequency. From the change in measured effective value, you can then try to extract a nominal DC value and a parasitic element that causes the frequency response.

However, this isn't exact because other effects (e.g. skin effect) will also change the frequency response. Skin effect in conductors causes a drop (!) in inductance when the frequency increases. From just measuring at two frequencies, you can't separate an increase in effective L from parasitic capacitance and a decrease from skin effect.

 

[FvM]

To indentify parasitics respectively to understand the results obtained with a LCR meter at different frequencies, you need an equivalent circuit of the component and a rough estimation of expectable impedance versus frequency.

There are simple cases like an inductor with parallel capacitance, if the capacitance is behaving as being lumped. And complicated cases with distributed parasitics, multiple self resonances etc.

 

[David_]

Okey, I think I get the concept.
I have a for my level of things decent function generator(Rigol DG1022), its a dual channel 10MHz function generator.
My LCR meter only got frequency steps of 100Hz, 120Hz, 1kHz, 10kHz and 100kHz and that does not feel suitable for this kind of test.
I am not clear on how I would set up a "setup" for this kind of test but I am really only interested in measuring ESL of capacitors(ceramic SMD 10nF-100nF & 1µF-10µF, and maybe 10-100µF electrolytic but that's a low priority in comparison to the ceramic. Its capacitors for decoupling primarily) ideally and parallel capacitance of inductors(or any capacitance of inductors though I don't have a specific inductance range in mind but I think it will be mostly toroidal ferrite/powdered iron core inductors... but say 10-1000µH for use in between 10kHz - 1MHz)

Actually I have no idea where to begin, is it possible that you could off the top of your head roughly describe an example of a scenario on how to measure any of the two given parameters?
Just so I have a better idea of how or where to start.

I do have a Analog Discovery which is surprisingly handy, it contains nice software for Network analyzer, Spectrum Analyzer and Oscilloscope as well as a function generator all above is dual channel(14bit ADC/DAC) but might I be required to build some kind of impedance bridge or is this kind of experiment at all possible without real lab grade equipment?

 

[chuckey]

Parasitic lead inductance of ceramic capacitors is often used to make a series resonant circuit at the decoupling frequency in use, i.e. at the IF frequency if its going to be used as an emitter decoupling capacitor. So the first point to underline that the lead length is paramount. I am thinking that if its for PCB mounting, if you got a piece of blank PCB, cut two parallel lines across it and mount your capacitor across one line and a reference inductor across the other, then applying a signal across the outside and measuring the two voltage drops at different frequencies you can get an accurate value.
Frank