Connect with us

How do you discharge a capacitor safely.

Discussion in 'Electronic Repair' started by [email protected], Mar 6, 2007.

Scroll to continue with content
  1. Guest

    Thanks for looking.

    How does one discharge a capacitor when working on a PCB, so that one
    does not get electrocuted? Are there any other components that
    maintain a charge after turning off power source. Thank you. please
    answer to .
  2. James Sweet

    James Sweet Guest

    Usually just short it with a screwdriver. CRTs can store a charge too.
  3. hr(bob)

    hr(bob) Guest

    You may get a big spark from a large capacitor, don't jump when it
    happens, you can do as much damage to yourself from ripping a hand on
    nearby sheet metal as from the actual electrical charge. I know from
    experience over manyyears and a few seconds of carelessness.

    H. R. (Bob) Hofmann
  4. the safe way to discharge capacitor are through resistor, use the
    resistor in minimum 10 watts with resistor 10megaohms for AC capaitor
    from 110v to 440v

    for DC from 5 to 220v capacitor us 5 watts and 150kilo-ohms above 220v
    DC use 10mega ohms resistor value,
    use two long lead with insulation and one end ground to chase or PCB
    ground, and other lead to discharge the capacitor.
    take more care when discharging more than 110v capacitor which will
    produced high sparking when discharging.
  5. Arfa Daily

    Arfa Daily Guest

    I wouldn't argue with using a resistor to discharge caps, and also, on
    occasion, just whacking a screwdriver across them - although this can leave
    a bit of a metal deposit 'splat' on the PCB. However, the suggested 10 meg
    10 watt resistor, or even the 150k, seems rather high to me. I think that
    you might struggle to even find 10meg at 10 watts.

    I have a metal film resistor of 2k at probably 3 or 4 watts, that has lived
    for years on the bench screw-magnet, and serves to discharge any caps that I
    ever come across that remain charged due to fault conditions. That includes
    up to 500 volts in some PA amps. This resistor has never even got warm
    during a discharge. If you work out the math, yes, at first the resistor may
    be massively overloaded from a high voltage cap, but it is for such a short
    period of time that the resistor never has a chance to get hot, which is
    what its power rating is all about. A higher value resistor than this (
    though not as high as 150k even ) will do a gentler job of it, with less of
    an initial spark, but will take correspondingly longer to do it.

    Note also that many high voltage cap circuits have bleeder resistors, or
    voltage sharing resistors across them when they are stacked, and these
    should discharge the caps for you. You should always check that they are
    doing their job though, as they do have a tendency to go open. You can also
    discharge caps safely with an older analogue multimeter, which has a much
    lower input resistance than its modern digital counterpart. This has the
    added advantage that you can see the voltage dropping, but is somewhat
    slower than using a resistor of just a few k.

    In general, the following circuitry should discharge the caps for you, and
    caps remaining charged will usually indicate a fault condition such as a
    switch mode power supply not starting up, and leaving the main primary
    filter cap charged to peak line volts, or a filter resistor between two caps
    open circuit, leaving the first one in line, charged.

  6. Guest

    Just how much wattage do you think you are going to get out of a
  7. See the capacitor FAQ at the site below.

    --- sam | Sci.Electronics.Repair FAQ:
    Repair | Main Table of Contents:
    +Lasers | Sam's Laser FAQ:
    | Mirror Sites:

    Important: Anything sent to the email address in the message header above is
    ignored unless my full name AND either lasers or electronics is included in the
    subject line. Or, you can contact me via the Feedback Form in the FAQs.
  8. This is generally frowned upon both due to the pits it will produce in
    the screwdriver, the noise, and potential damage to some types of
    capacitors. See the info at the site below.

    --- sam | Sci.Electronics.Repair FAQ:
    Repair | Main Table of Contents:
    +Lasers | Sam's Laser FAQ:
    | Mirror Sites:

    Important: Anything sent to the email address in the message header above is
    ignored unless my full name AND either lasers or electronics is included in the
    subject line. Or, you can contact me via the Feedback Form in the FAQs.
  9. Ron(UK)

    Ron(UK) Guest

    Rightly or wrongly, I use a 240volt BC pygmy lamp with two leads fitted
    with prods soldered to the pads on the bulb. Works for me.

  10. Guest

    Guest Guest

    xcellent choice, lamp provides indication and nice slope for current rate.
  11. According to my calcs, the maximum power needed for a 10M0 resistor is
    0.01936 watts.
  12. Guest

    Actually I'd like to get my hands on some 240V incandescent bulbs just
    for that purpose.

    I've seen some sets use as high as 940uF for the primary cap (2X470).
    This is a considerable amount of power. I used to have a 75W 130V
    (long life) bulb for the purpose until it got used. I actually put it
    in the manifest as parts to be ordered but it is not going to happen.

    When I see a 75 watt bulb take over a second to discharge it, what
    goes through my mind is, just how long would my body take to discharge
    it ?

    Now I see that they make the PC so that the rectifiers and jumpers can
    be used for 120 or 240, in the US, Canada and Japan at least, they
    double the line voltage. For Europe it appears they just rearrainge
    the jumpers so it is a simple full wave bridge. They just wind up
    using higher voltage lower current devices.

    As it is, I have to discharge each filter in the doubler seperately,
    Going across the whole 320 DC would burn it out. I think a 240V bulb
    would handle it, it is not 320 VDC for long.

    Actually the bulb is the best choice, but a lazy choice. It is
    available and cheap. I am totally against using a screwdriver to short
    out the caps, it no doubt exceeds the filters' ripple current
    capacity, evidenced by the metal deposits from the spark as noted by
    one poster.

    Ideally we could have a constant current source with a low wattage
    bulb to give a visual indication. A regular lightbulb is less
    desirable because it's resistance is low when cold, thus there is a
    surge. If that filament burns out right away when you connect it, you
    might erroneously think the cap(s) were discharged when they were

    Put your solder wick on the drain output of an SMPS IC that still has
    320VDC (or even 160) on it and you'll wish you wore gloves (and maybe
    goggles too !).

    Certain Mits RPTVs and maybe some Hitachis come to mind. There are
    plenty of direct view sets also but I generally don't work on them,
    except for 32" and up. But I have seen where the main cap failed
    because there was no load on it. All bulged, but it wasn't a power
    surge. See some places the power is "dirty". It is essentially 120 VAC
    but there are spikes and all kind of shit, it is not a perfect sine
    wave to say the least. In some of these sets there is no bleedoff if
    the SMPS doesn't start. Therefore the spikes will take the rectified
    voltage quite higher than one would suspect.

    They generally don't use high speed rectifiers for the AC line, which
    means they have a certain effective series inductance. Well that is an
    impedance, and even though a one meg resistor might keep the voltage
    from climbing, it doesn't even have that. Also, if the rectifiers
    outperform their specs, you might have a bit more.

    Do not be surprised if a diod outperforms it's specs. I had a case not
    too long ago, another tech had a set blowing the top damper diode. It
    was a recall, but like an 8 month recall. Well, somehow at the time
    the top damper was replaced with a 400V diode, like a 243636 (MI35 I
    think) but the application clearly calls for a 1500V device.

    I didn't do it, I have no idea how it happened but the 400V diode
    worked for months. Eventually the set came to me. Once this was
    figured out the set and customer have been happy together ever since.

    That's why, I suspect, that sometimes I don't get many answers. I am
    the dogshooter. And folks I got some big ones. But I can see as one
    reads my post they think "Oh yeah, he needs to......" and in the next
    few lines you see I already did that.

    Then at the end of the post you might be thinking "Yup, this dude is
    really fucked". That's my job, fucked. The newest part of my job of
    fucked is those Sonys, the DX1 type chassis, the XBR400s and like
    that. I have had to fix enough screwups on those sets that I am now
    the alpha and omega of the DX1 at the shop. And I said it. Right after
    the words came out of my mouth I got this jumping off a bridge
    sensation, and rightly so. But in the long run, when I don't have to
    follow someone else I'll learn alot more about those sets.

    I would really like to get some 240V bulbs, any ideas for a source ?
    My 120V bulb disappeared shortly after the light in the hall leading
    to a bathroom burnt out, and now I am thinking, if I had a 240V bulb,
    that wouldn't happen.

  13. James Sweet

    James Sweet Guest

    How about a couple of 130V bulbs in series?

    If you want 240V bulbs they're easy enough to get, I've got some
    contacts in the UK I get oddball lightbulbs from and these days you can
    even get them with ES bases that will fit in a standard US socket.
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day