What frequency to test capacitance and resistance with LCR meter

[waggs]

When testing caps and resistors (especially low value ones) I find that my DMM is not accurate.  So that lead me to buy an LCR meter - specifically the DE-5000 off ebay.  Because capacitance and resistance change with frequency, what setting do you guys use for testing component values?  I'm building a standard OD pedal like the klon centaur so I would think the 1kHz frequency would be most appropriate, but I'm here to learn.  Thank you.

[Kevin Mitchell]

1st off, welcome 
2nd, you're going beyond what most (or all) builders do! We usually don't test our parts unless there's something wrong. Even in that instance we're more likely to just replace it rather than do any analysis unless we're trying to do something like gain matching.

Consider the whole spectrum of typical raw electric guitar frequencies. But also these parts may be used in part of a circuit where those number are irrelevant - like clocks and LFOs.

Hopefully someone can entertain your question more appropriately as I personally don't test parts unless I have to. And in that case I either do it with my cheap multimeter or simple circuits on breadboard without consideration of the frequency. Perhaps because I haven't had to yet.

[antonis]

Welcome. also..

Reactive items (like capacitors and industors) are used in pedal circuits taking into account their impedance (capacitive or inductive reactance) which is a slightly more complex quantity than resistance, but let it be for the moment..

Roughly speaking, a capacitor of 100nF value say, is used for its "behavior" (impedance) at some frequencies of interest so the only thing you may have to do is to verify its value with a capacitance meter..

P.S.1
Resistors DON'T change their value with frequency simply because stompboxes don't embrace "hyperphysics" principles..

P.S.2
There are some techiques for measuring low value resistors (less than 1 Ohm) but these are beyond the scope/utility of DIY pedal circuits..

[GibsonGM]

I think this is a common path we go down when we get started - we want to measure everything, and learn how all these complex things affect the outcome of the effect we're building or designing.  At audio signal frequencies, as was said - if it ain't broke, it's not very important.  ESR of capacitors, their impedances and so on, really won't change much tone-wise as long as the component is good and within the standard tolerances they're made under.   These things come into play more at RF (radio frequencies), well above anywhere we go!  Luckily, our work is VERY MUCH less demanding!

I think about as complex that we need to get is matching things like FETs for phasers and the like.  Then it DOES make a difference.  As an experiment, playing with this stuff IS fun, though, and helps you understand the inner workings of these parts better.   

[Rob Strand]

d so on, really won't change much tone-wise as long as the component is good and within the standard tolerances they're made under.

Agreed.   There no real need to use an LCR meter to build a pedal.   You might need one for reverse engineering a pedal since the SMD cap values aren't readable.

As far as cap meters goes.   The behaviour and accuracy of DMM cap meters are all over the place - they are often very sensitive to parallel resistances.   The DE-5000 is a pretty good LCR from what I hear.    Typically 1kHz will suit smaller values like 1nF to 1uF.     When you get below 1nF you might consider 10kHz.    Measuring polyester caps at 10kHz can produce a different measurement than at 1kHz that's because the capacitance is not constant with frequency.   It's not a measurement problem it's the capacitor characteristic and the meter is correctly showing the cap's behaviour.   Luckily 100pF caps tend to behave similarly at 1kHz and 10kHz so making a measurement at 10kHz is more accurate from the meter's perspective and still represents the value at 1kHz.

For caps at 100kHz, you really need to understand all the finer aspects of caps and measurement errors.  If not you will get readings which have errors somewhat larger than the LC meter's specs and you might be better of testing at 10kHz.

[PRR]

> capacitance and resistance change with frequency

As said: No. An ordinary 1k resistor will stay 995-1050 Ohms DC to way over 1 MegaHertz. A 100nFd cap will be right close to 100nF to nearly a MHz. Yes, the cap impedance drops but that's what we bought.

For R and C you may as well set the dial on 1KHz, because that's near the center of the Audio Band and there's no real difference even to 50kHz. (OK, Rob squints better than I do. Or is reading hi-Q tuned circuits, a popular pastime in some workshops.)

For iron core inductors it is different. Inductance (not impedance) may drop from 30Hz to 300Hz as eddy-currents force the flux out of the center of the core. Above 10kHz it is common for an audio transformer to be about the same Henries with or withut the iron core.

[amptramp]

I agree with the 1 KHz test frequency and it looks like you have a choice of 100, 120, 1000, 10,000 and 100,000 Hz.  The 1 KHz test is matched to audio frequencies.  I have an AideTek 4070 LCR meter and I use it in every build to verify that I have the correct parts - some manufacturers use a red colour code that looks like orange or vice versa.  Don't let the naysayers distract you - I have seen troubleshooting issues that were nearly impossible to understand caused by wrong value parts.

[waggs]

This forum is awesome!  Thank you everyone!