Announcement

Collapse
No announcement yet.

AC on DC side of PSU?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    AC on DC side of PSU?

    Hello. The other day I was testing some led's that I intended to use on a new (hopefully) fashionable lid for my computer case. I tested one of them using a resistor and a power supply. But I hadn't even attached the led when I noticed something odd.

    When my powersupply is switched off (but plugged into the wall socket), there's an AC current running between me and the ground wire of the power supply - and between me and the 12v wire I meant to use as well in fact.
    I tried this with two different power supplies I had laying about (an antec 380w and a modular 620w corsair). The current varies appearently, but it's somewhere between 54 and 110 volts, and as voltage goes down so do herz count. I've seen anything from 25 to 7 hz. Between the two wires, there's less than 0.1v but still enough to detect a 50hz frequency.
    When I switch the unit on the voltage becomes half of whatever it was before, and the herz rating changes to 26-27.
    There's even enough current to have the diode light up enough that I can tell the colors apart (3 diodes in one housing sharing a common ground) when I use myself as 'plus' and the ground from the psu as ground.

    Why? Why do I detect ac current on my power supply?

    When the unit is turned on I'm measuring very close to zero voltage ac between ground and +12v, but I still have a 50hz reading on the multimeter. Switching from raw dc reading (of 12.08v on the antec) to ac+dc I notice the voltage reading changing, but the herz is still @ 50.

    I'm just a simple enthusiast. I would've always claimed there's no ac on the dc side of a power supply - but appearently that's not so.

    Why?


    background:
    System's in denmark. Means 230v @ 50hz from the wall socket.
    multimeter is an mx53 (http://www.aemc.com/products/html/mi...bname=products)
    Power supply is turned on by shorting green and a ground on the atx plug using wire from another atx plug that I had cut to pieces a while ago.


    I already asked toms hardware some days ago, but the only answer I got was to ask an electrician (which I had done, but to no avail) or try posting here.

    #2
    Re: AC on DC side of PSU?

    Addon - in case it's important, I might want to mention that there is no earth in the socket. It's not the newest house.

    Comment


      #3
      Re: AC on DC side of PSU?

      Originally posted by neiro
      Addon - in case it's important, I might want to mention that there is no earth in the socket. It's not the newest house.
      not safe.

      hence why we are getting the house we just bought upgraded.
      sigpic

      (Insert witty quote here)

      Comment


        #4
        Re: AC on DC side of PSU?

        I think this is leakage current through Y filter caps in psu's input filter. Y caps are connected between line and ground and neutral and ground. If you would have your house grounded, this leaking current would go into the ground. The current is very small and I think it's nothing to worry about (except your not grounded house ).

        Comment


          #5
          Re: AC on DC side of PSU?

          Get an alligator clip and connect your computer chassis to something grounded, like your house heating radiator for example, I bet you the AC voltage will be gone then
          "The one who says it cannot be done should never interrupt the one who is doing it."

          Comment


            #6
            Re: AC on DC side of PSU?

            what mur said.
            i see this a lot in modern unearthed junk like tv's and set-top boxes.

            i almost got thrown off a ladder once by an unearthed satellite receiver that was putting about 380v onto the earth side of the LNB cable via the tuner-can.

            Comment


              #7
              Re: AC on DC side of PSU?

              I took my multimeter with me to work and tested a normal hp psu from one of the hundreds of computers here.
              First I tested with no ground wire, and it was outputting 46 volts at a fairly high 38hz. Then I changed to a grounded outlet, and it completely vanished!

              So you guys are right. It's the missing grounding. Maybe this is something I should do something about. If it can make a led glow, it can also break stuff.

              Comment


                #8
                Re: AC on DC side of PSU?

                I know this isn't computer-related, but grounding related. I was in the process of bench testing a 1000-watt metal halide parking lot light assembly without a ground attached. Although I didn't touch it, my DMM was reading over 300 volts to ground. I double-checked all the connections and found nothing wrong. After a ground wire to the frame, the voltage issue went away. The ballast in it was a common core-and-coil. Fluorescent fixtures, especially those with electronic ballasts (a SMPS!) can be dangerous if not grounded properly.
                Stupidity should be a crime, especially for drivers. I have NO patience for them.

                Comment

                Working...
                X