How to check capacitance?

[spwath]

I put my multimeter in the right mode, and touched the two ends to the pins of a capacitor, and nothing happens on the screen.

 

How are you supposed to check capacitors capacitance?

[CNY RMB]

Hi,

what multimeter are you using and also check that the test leads touch the correct side of the capacitor. I am assuming you are trying to measure an electrolytic cap, so polarity matters. The side with a line is usually the negative side, sometimes a little + or - is also visible on the cap. Just check that they are the right way round. Also definitely ensure the cap is discharged by placing a resistor across the pins first and then measure.

Jakob

 

[spwath]

1 minute ago, CNY RMB said:

Hi,

what multimeter are you using and also check that the test leads touch the correct side of the capacitor. I am assuming you are trying to measure an electrolytic cap, so polarity matters. The side with a line is usually the negative side, sometimes a little + or - is also visible on the cap. Just check that they are the right way round. Also definitely ensure the cap is discharged by placing a resistor across the pins first and then measure.

Jakob

 

IM using some radioshack multimeter

 

If i dont have a resistor, how long should I wait for caps to discharge?

[CNY RMB]

Just now, spwath said:

IM using some radioshack multimeter

 

If i dont have a resistor, how long should I wait for caps to discharge?

Hard to say, it depends on the capacitance. If it is a big cap, it can take weeks to fully discharge. If you have a fairly small cap you can simply short the two pins with some metal. Then, if you are trying to find out whether the cap is the problem in your circuit I would first check that the cap hasn't actually shorted out, but also isn't an open circuit. Also, physical damage may indicate a fault. Fluke has a nice guide on measuring capacitance and common problems with capacitors.

http://en-us.fluke.com/training/training-library/test-tools/digital-multimeters/how-to-measure-capacitance-with-a-digital-multimeter.html

 

Jakob

[GER_T4IGA]

15 minutes ago, spwath said:

IM using some radioshack multimeter

 

If i dont have a resistor, how long should I wait for caps to discharge?

Discharge rates can usually be seen in the manufacturer's product specifications. I have seen some nice U/t diagrams in the past. Idk if it will be to easy to find that if your cpa is from a circuit though

[spwath]

37 minutes ago, CNY RMB said:

Hard to say, it depends on the capacitance. If it is a big cap, it can take weeks to fully discharge. If you have a fairly small cap you can simply short the two pins with some metal. Then, if you are trying to find out whether the cap is the problem in your circuit I would first check that the cap hasn't actually shorted out, but also isn't an open circuit. Also, physical damage may indicate a fault. Fluke has a nice guide on measuring capacitance and common problems with capacitors.

http://en-us.fluke.com/training/training-library/test-tools/digital-multimeters/how-to-measure-capacitance-with-a-digital-multimeter.html

 

Jakob

I have a audio mixer that sounds bad, trying to figure out if bad caps are the problem.

 

is 10uF 5V small?

 

Or 47uF 25v?
 

[CNY RMB]

Those are small caps anything below 500uF (0.5mF) @ around 30V should be fine.They don't really store much energy.

[SpaceNugget]

48 minutes ago, spwath said:

I have a audio mixer that sounds bad, trying to figure out if bad caps are the problem.

 

is 10uF 5V small?

 

Or 47uF 25v?
 

the capacitance can still be in spec when caps go bad, you need to check equivalent series resistance with an... equivalent series resistance (esr) meter. Your radio shack meter isn't up to this job unfortunately.

[mariushm]

Most electrolytic capacitors have polarity (positive and negative) but the large majority of multimeters don't care about how the leads are connected to a capacitor, because the test signal is low enough that even if connected incorrectly, the test pulses won't damage the capacitor.

 

Multimeters measure the capacitance by sending a "stream" of pulses to the capacitor and measuring how quickly the voltage raises on the capacitor. If the voltage raises fast, it means the capacitor has a low capacitance, if it takes a long time for the voltage to reach the expected threshold it means the capacitor is of a large value. The target voltage is often in the area of 1 .. 2v or something like that, very small.

 

Because of this, a lot of multimeters on the market are not able to measure capacitors with high uF values. Some multimeters can't measure above 100uF, others stop at around 1000 uF or at around their digit count (if it's a 1999 digit multimeter, they usually can't measure above 2000uF, any capacitor higher than that will cause the meter to show OL "over limit" or something like that)

 

It's also important to know that the capacitance value of a capacitor is not a good indicator of its quality. A capacitor can often fail and still have its capacitance within that 20% tolerance standard for electrolytic capacitors.

Capacitors have some technical parameters like ESR and ESL which are affected by the operating frequency of the circuit the capacitors they're in. When you test a capacitor with a multimeter, the capacitors are basically put in a circuit which runs at something like 100 Hz or slightly higher.

When the capacitors are in circuits typical in switch mode power supplies (like in computer power supplies, the monitor's power supply, the DC-DC converter that powers the cpu on the motherboard and so on), all these circuits run at high frequencies, usually above 40-60 kHz (because human ears can hear sounds up to around 16-18kHz, power supplies are typically designed to operate at frequencies above double that, so above around 36 kHz). 

Very good quality ATX power supplies nowadays switch at around 125-150 kHz , older designs typically hover at around 60 kHz and really old Pentium 4 300-450w power supplies typically worked at around 35-40 kHz.

The point is, when capacitor is in circuits running at such high frequencies, those properties ESL and ESR can have completely different values compared to the same values when circuit works at those frequencies the multimeter uses to test a capacitor. A bad capacitor can show its bad behavior at those frequencies but look totally fine when measuring with a multimeter.

 

The only accurate way to test a capacitor would be to use a LCR meter with ESR measurement function at 100 kHz (it's important, there are LCR meters with ESR measurement but only at 120 Hz and 1kHz which isn't enough). They're not cheap, they usually cost  from about 80$ and up.. probably more like $110-130 ... 

Here's a couple of proper LCR multimeters with ESR measurement :

 

Uni-T UT612 : https://www.tester.co.uk/uni-t-ut612-lcr-meter

 

(jump at 32:10 if it doesn't do it automatically)

 

 

Mastech MS5308  (long discussion thread on eevBlog about it, with reviews and pictures : http://www.eevblog.com/forum/testgear/mastech-ms5308-lcr-meter-with-esr-measurement-on-discount-at-the-moment/ )

 

 

There are ESR meters on eBay and various websites which are much cheaper and in theory they're not really ESR meters, but do a particular measurement that for all intents and purposes, gives a result that is about 95-99% as accurate as a proper ESR measurement with a more expensive tool, so these $15 .. $50 tools are good enough if you don't want to invest on a proper LCR meter.

 

Commercial expensive ESR meters  (you're better off buying one of the  LCR with ESR functions above, unless you find one worth less than $100)  : https://www.amazon.com/Peak-ESR70-Atlas/dp/B005NIBEYU

 

 

I personally have a microESR which I bought from radiodevices.ru , but I've ordered directly from the designer of the tool, who has an account on Badcaps.net forum.  I think it cost me about $50 about a year ago. It's better to contact him through the website or on the forum I mentioned before ordering.

 

There's also cheaper ESR meters on eBay these days, lots of people copy the same design and just use different cases or lcd displays.

Here's just one example (I haven't tested this tool personally, but I don't see why it wouldn't work) : http://www.ebay.co.uk/itm/Portable-MK328-LCR-ESR-Tester-transistor-inductance-capacitance-resistance-meter-/171927184554?hash=item2807a8a4aa:g:MdoAAOSwk0pVgETb

 

 

 

 

[mariushm]

A sidenote / ps about ceramic capacitors and tantalum capacitors.

 

Everyone should also be aware that  ceramic capacitors have some particular characteristics that make them special compared to electrolytic capacitors. Unless you really know what you're doing, never replace a ceramic capacitor with an electrolytic capacitor (even of the non polar type).

 

For example, ceramic capacitors have voltage bias ... simply put, a 10uF 16v ceramic capacitor won't be 10uF when it's on a 5v circuit, the capacitance would be much lower. That's why you'd often see ceramic capacitors rated for 50v or even higher when the voltages in the circuit could be lower than 3.3v or 5v.

 

See https://www.maximintegrated.com/en/app-notes/index.mvp/id/5527

 

also see

 

 

 

Tantalum capacitors typically have ESR values lower than electrolytic capacitors (and they're more consistent over a wider range of frequencies) but are much more sensitive to over-voltages and they react badly / violently , as in they blow up or go in flames, unlike electrolytics which may just vent a bit . Most engineers recommend not going with the voltage over 70% of tantalum capacitor's voltage rating.. for example you shouldn't use a 16v rated tantalum capacitor in a power supply that's meant to output 12v , you'd better use a 25v or 35v rated tantalum. 

 

Sometimes it's a good idea to replace them just to make the product safer, but one should keep in mind the ESR issue... for example a 100uF 25v tantalum would have a much lower ESR compared to a regular electrolytic capacitor, but a polymer capacitor would have equal or even lower ESR compared to the tantalum capacitor. If an electrolytic is used, maybe you'd keep the capacitance value but go up to 50-63v rating and bigger volume electrolytic capacitor, since the ESR value of a capacitor is most of the time lower as the volume of a capacitor goes up. Still, check the datasheets of capacitors to be sure the ESR is close enough. 

 

 

 

[Silverwolf_7]

You can't check the values of a component if it is still installed in the circuit. The test signal that the multimeter sends out will find a different way around.