[emac]

Hi, ran into a problem and was looking for some help.  Over the coming months I was planning on testing out some capacitors to see what I think of them.  I have an Extech mutimeter that my wife had gotten me for Christmas, but when I've tried to test out capacitance, the values range from pretty much dead on to nowhere near close to what the value on the cap is.  So, I wanted to get some input on this.  Am I doing something wrong (1 lead to each side of the cap) or do I need to get a proper capacitance meter?  And if I need to buy something, what is recommended?  Thanks.

[JoshK]

I usually measure caps with my LC meter (inductance/capacitance).  My DMM doesn't seem to give consistent results even with the same cap, so I switched to using the LC meter which does.   I bought the LC meter to trim inductors in speaker crossovers.   It was $50, but well worth the investment in my opinion.  I use it often enough to justify the cost.

I think a good DMM should give you an accurate reading, but mine doesn't.  I am not familiar with your meter.

Putting one lead to each end of the cap is how you measure a cap. 

[Scott F.]

I agree with Josh. To get reasonably accurate measurements you need to move to a dedicated LC/R meter rather than a multimeter.

I use a Wavetek LCR-55 but it looks as if its been discontinued. You might be able to find a used one. If not the Amprobe LCR55A looks to be a decent option (actually a direct knockoff except in red rather than the dark gray of the Wavetek). Its integral transistor checker has come in handy a time or two also.

You are looking at about $150 for the Amprobe which in the grand scheme of things isn't too bad especially if you are a dedicated DIYer and want precision.

[dweekie]

What range of values are you planning to test for the most part?  There are a couple kits and there's the Atlas LCR-40 that comes to mind at the moment that have gotten some great reviews as far as accuracy into small values goes.

[emac]

What range of values are you planning to test for the most part?  There are a couple kits and there's the Atlas LCR-40 that comes to mind at the moment that have gotten some great reviews as far as accuracy into small values goes.

Mostly in the 0.01 - 30uF range for XO applications.  I don't need everything to be exact, just close enough.

[dweekie]

http://www.peakelec.co.uk/acatalog/jz_lcr40.html

When I was searching for the most versatile LCR meter for the best price, that kind of stuck. It's $111 shipped and has an inductance range that dips low enough to be used for inductance values in power supplies or class D output filters, which can't be said for most other LCR meters out there.  I'm not sure if that's within the price you want to spend though, but it does have a nice range in case you need it in the future....

I have a cheap LCR meter that I spent about $40 on that doesn't really work as well as I like at the moment, and I'm still thinking hard about that Atlas right now. 

[markC]

Decent multimeters aren't likely going to give you accurate results down to .01uf. My U. E .I brand, which I use for work, seems to measure quite well from .10 to higher electrolytic values. Cost was about $C 175.

[analognut]

A couple of posters here mentioned that multimeters will give inconsistent results when measuring capacitance. I am very curious about this. Is that because they were trying to measure C at too low of a value for the range the meter was in? I would think that if you were using the correct range on the meter the result would have to be the same each time!

Is it possible the cap needs to be discharged before the measurement is taken? I'm guessing that shorting the leads before measuring would discharge the cap.

I have a very good quality Wavetek multimeter and can't remember when the last time was I used it on a cap, or if I ever did. I must try this and see for myself! 

[markC]

My meter is auto range and gives consistant results. I have some .01 caps in "the parts bin". I'll try measuring them and see what happens.

[Niteshade]

I built a nice "capacitance meter" based on a 555 timer IC way back when. It would produce a specific frequency for the capacitor that was in the circuit. A frequency of so many CPS would reference to a capacitor's value. It was crude, but worked well.