Are all capacitors equal

Discussion in 'The Projects Forum' started by WrinkledCheese, May 30, 2010.

  1. WrinkledCheese

    Thread Starter New Member

    May 30, 2010
    Quick answer no, that is why they have different "ratings" and types.

    I tried figuring out why I'm getting the behaviour on Wikipedia and other places but I don't really know anything about circuitry. I've built a circuit following a diagram. I can find problems in circuits IE cold solders, and more recently blown capacitors. So I have recently come into a new project. Fixing a 22" widescreen monitor. The original owner said they heard a "pop" that they thought was a message coming in via facebook. I immediately thought it was a blown capacitor but I don't know anything. So they gave it to me.

    Anyway, enough background. I took the thing apart and checked for obvious signs of damaged circuitry. I found a capacitor that had one cold solder lookin solder. There is also a white stuff that seems to be on all the capacitors that was severely cracked. The guy at "The Source", think Radio Shack, said it was some sort of paste. So I took to Google and figured out a way to test the capacitors with a multimeter. Using the ohmmeter I would see a gradual increase until it overloads whatever setting it's on. For the capacitors of the exact same rating and voltage, in the circuit - 220 μf 25v 105 deg. C - it takes about 2-3 seconds to overload 200kohm setting. I bought a "variety pack" of capacitors at the source because they don't sell singles anymore. I looked through the packs till I found one of the same rating, 220μf 25v 105 deg. C. After I found one, I put it back into the circuit, using the PBC board markings to make sure I put positive and negative in the right holes. I turned it on and I got nothing. I took it back apart and tested the capacitor. This is where things were weird for me. It takes about 10-15 seconds to overload 200k ohm setting on my multimeter.

    The capacitor is the same height as the original but it is much fatter. I would say about 2x the circumferance. OR about 1.3 to 1.5 the radius. [EDIT]It looks that way but it's circumferance is about 20% larger on the new one now that I've measured it.[/EDIT] Does this have an effect on the overall circuit? Would a soldering iron heating up the capacitor have an effect on the capacitors ability to function?

    I would say yes if it reaches 105 deg. C but like I said, I know nothing. I'm a programmer by trade... more of a learn as you go type hobbyist. Maybe I should use a heat-sink when connecting the capacitor. I checked all the other capacitors, they all behave the same, overloading quickly.

    IMAG0065.JPG is the circuit board with a green - and + where the - and + terminals should go.

    IMAG0067.JPG is the two capacitors. the black one(top) is the one that will not show any resistance when hooked up to the miltimeter, the brown one(bottom) is the new one that takes too long to charge. You can see the "white stuff" I was talking about on the black one.
    Last edited: May 30, 2010
  2. bertus


    Apr 5, 2008
  3. WrinkledCheese

    Thread Starter New Member

    May 30, 2010
    Based on that site, I think the original is an aluminium capacitor and the replacement is also and aluminium capacitor. The only thing I can think of based on my very limited knowledge is that the new capacitor takes longer to correct it's leakage. I did notice that as the resistance goes up it does start to go up faster, but still takes 20+ seconds to overload 200k ohm as compared to 2-3 seconds for the old one...well identical ones that are still working.

    This is my process for testing them:
    1) short the terminals to remove charge.
    2) attach negative lead to negative multimeter lead.
    3) attach positive lead to positive multimeter lead.
    4) count how long it takes to overload.
  4. thatoneguy


    Feb 19, 2009
    The capacitor that failed could have the "ESR Plague", where an internal resistance has bulit up.

    Symptoms include bulged tops of electrolytics, leaking fluids, or ripped coverings.

    Replace capacitor with same value (Electroyltic, preferably low ESR) and it should work.

    Testing caps by timing the charge rate is only somewhat effective, since other than shorts or open states, capacitor issues such as ESR are only detected with an AC ohmmeter or LCR meter that shows the Dissipation Factor.

    Also, you should use a resistor, about 1k or more to discharge the capacitor, shorting the leads together can be harmful in the long run.

    Examine the entire board for any caps that are bulged or looking different than the rest, and search for "ESR" isssues on this forum, which will have a lot of information for you.
  5. WrinkledCheese

    Thread Starter New Member

    May 30, 2010
    I Googled how to test a capacitor with a multimeter. I found that you can get results, not the best, with an ohmmeter. So I tested the capacitor I thought to be faulty and it would not show any resistance change and stayed overloaded.

    I replaced it with one of the exact same specs(220 μf 25v 105 °C) It didn't work.

    I tested all the capacitors and they all showed rising resistance up to 200k in a matter of seconds. The capacitor I bought from "The Source" formerly "Radio Shack" takes upwards of 20 seconds to overload 200k ohm on my multimeter.

    I did notice there are 2 other capacitors on the board that do have the bulged cap that I read is an indication of failure. Another 220μf 25v and a 1000μf 25v. They are in close proximity to the one that failed and on the same circuit. These however behave the exact same as others on the board with the same ratings.

    Like you said, I don't have the proper equipment so it's possible they are blown, my testing just doesn't show that.

    There is a commercial electronics place around here but they are open 9-5 Mon-Fri so I have to hit them on Monday. I will also start using 1k resistors to discharge/short them.

    "The Source" packaging doesn't mention ESR ratings. I'm assuming the commercial shop will have exactly what I'm looking for. I will also pick up a proper capacitor tester. I may want to invest in a better multimeter as mine is a $10 "Radio Shack" digital multimeter.

    Thank you all for your help and links to information. I've read quite a bit about "ESR" but nothing mentioned what it stood for. From what I gathered it was an acronym for the "leaking" behaviour's actual name. I will Google it to find out.

    I seemed to be somewhat correct. It seems ESR(Equivalent Series Resistance) is the natural resistance of the materials used to create a capacitor. Nothing to do with the leaking from what I can tell. I haven't a clue though.

    Thank you all again!
    Last edited: May 30, 2010
  6. studiot

    AAC Fanatic!

    Nov 9, 2007
    It is very unlikely that the only fault was the failure of a single capacitor.
    It is even more unlikely that there would be no response at all from the circuitry with that capacitor simply missing. The unit would likely work, but not properly.

    So I conclude that your unit has bigger ills that this one capacitor failure.
  7. WrinkledCheese

    Thread Starter New Member

    May 30, 2010
    I suck at explaining myself...

    Good to know!!! I figured once one blew the rest would be protected via in-circuit resistors etc. I know NOTHING lol

    I wouldn't even try turning it on with anything missing. What I meant to say was I put the new capacitor in and still had no output on the monitor, although when I moved my mouse when the computer was in "power save mode" the LED changed from orange to green, which didn't happen before replacing the capacitor I tested as faulted.

    I figured as much ;) There are two more capacitors with "mushroomed" caps that I'm going to replace. I hope that's the end of it.

    If you look at the circuitry picture in my first post, you can see two more "mushroomed" capacitors. One immediately left of the green -/+ symbols where the capacitor I removed belongs and one to the "north-east" about an inch.

    Thanks!!! Some good info in this post. I should probably pick up a circuitry book and read it from cover to cover.

    If it ain't broke. Fix it until it is. LMFAO this one never gets old.
  8. WrinkledCheese

    Thread Starter New Member

    May 30, 2010
    I went to the local industrial electronics store and they charged me $2.63 for the 3 capacitors I THINK I need. I priced their cheepest multimeter with capacitance built in and it was $270 + tax. They said they're aimed at industrial applications and I told him I do electronics as a hobby so that's a little out of my price range and I will have to stick to my $10 Radio Shack multimeter for now.

    The only brand they carried was Fluke, which is the industry best he said. I will see if I can find one for a cheaper price somewhere else. I'm looking for something in the $45 range. Nope, not $50...$45 lol

    I found this one, it doesn't have the capacitance symbol but it does say "cap".

    If anyone can find one on e-bay for a reasonable price that would be awesome, or let me know if this one will do the job. I'd like to keep it under $100.

    Just found another...just $4.99 out of my price range.

    Unfortunately the one at princess auto only goes up to 20 microfarads(uf) and starts at 0. The packaging says from 0-20mF, and on the device itself it said 2nF to 2uF then 20uF. Damn!!!

    So I just googled, and got a wikipedia page:
    it says there is a difference between mF and μF. The packaging says 0-20mF(200,000μF) but the device itself only shows 20μF. Their slogan is "no sale is final until you're satisfied" so I might as well pick it up, give it a shot and if it works, awesome, if not, return it. Claim the packaging is misleading.

    The manual says 20μF :/ ah well...back to the drawing board.

    I found this project that is cheap and looks like fun. Says it has a range to 500uF

    I also found this one, it's not a multimeter but a capacitance meter only. $15 not bad. The pictures show it going up to 20mF but the auction says it goes up to 20,000uF which is 2mF.
    Last edited: May 31, 2010
  9. thatoneguy


    Feb 19, 2009
    Fluke meters are about the best you can get, rugged, accurate, and it takes a lot to kill one.

    You could get by with a cheaper "All-in-one" Multimeter that tests capacitors/inductors/transistors, plus voltage/current/diode/resistance measurements for about $45 on eBay from a no-name chinese place.

    The accuracy is about 10%, but it's better than not knowing at all.