1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Electronics Power Factor Correction in power supplies

Discussion in 'Modding' started by g0th, 18 Apr 2005.

  1. g0th

    g0th What's a Dremel?

    Joined:
    12 Apr 2005
    Posts:
    319
    Likes Received:
    1
    I've got a dismantled FSP250-60PNA here, anyway there is something that i haven't seen before in it, a great big heavy (bigger than the main transformer) inductor, with laminated iron core.

    It's in series with the neutral, before the rectifier, and has a a capacitor in parallel with it. It's also shorted out by one pole of the selector switch when the PSU is set for 115V.

    My questions:

    It is for PFC, right? Why haven't i seen one in a PSU before? (Passive) PFC networks are pretty common in modern power supplies, i think.

    Also, why isn't it in circuit for 115V use?
     
  2. chalk_mark

    chalk_mark What's a Dremel?

    Joined:
    8 Aug 2004
    Posts:
    117
    Likes Received:
    0
    It is providing a series drop of about 115 VAC when running on 240 VAC. This steps down the input voltage for dual mains use. Some power supplies use the input filter capacitors, running them in series or parallel for the same result.
     
  3. g0th

    g0th What's a Dremel?

    Joined:
    12 Apr 2005
    Posts:
    319
    Likes Received:
    1
    I thought the way most PSU's did that was by reconfiguring the rectifier as a doubler in 115VAC mode, so the DC rails on the main caps are always ~300V or whatever it is.

    This method you describe sounds a bit nasty, it would screw up the power factor something awful.

    So your saying the AC input to the rectifier is always about 115V RMS?
     
  4. SteveyG

    SteveyG Electromodder

    Joined:
    23 Nov 2002
    Posts:
    3,049
    Likes Received:
    8
    It certainly doesn't do that. The diodes are reconfigued as a voltage doubler when the voltage switch is changed to 110V.

    I've seen several power supplies with no power factor correction at all in them, which may be why you haven't seen the big inductor in there. Some use active power factor correction, which is usually seen as a pcb attached to the heatsinks inside the PSU case. I think generally the whole front end of the switching power supply is altered to accomodate the active power factor correction with the PFC circuitry running at 80 to 200kHz.
     
  5. star882

    star882 What's a Dremel?

    Joined:
    19 Mar 2003
    Posts:
    925
    Likes Received:
    1
    I have several computer power supplies where the active PFC is built onto the main board. With the one that is currently on my desk (only 200w but quite heavy), there are 2 large MOSFETs on a large heatsink on the primary side. One is the normal flyback driver, while the other one is for PFC. A single specialized microcontroller on a primary control board controls the primary side circuitry.
    The secondary side is anything but simple with 3 power MOSFETs and a dedicated controller board covered in SMD components. Along with some ASICs to control the MOSFETs and regulate the voltage, there is an Ashlee Simpson microcontroller to drive the I2C (or SMBus?) control bus and handle some tasks like temperature monitoring, system control, and fault monitoring.
    I have not yet figured out how to switch on the PSU beyond the 5vsb line, but I'll build an I2C bus adapter and see if I could get anything out of it...
    The PSU was for an old, small server (Sun), and I'll see if I could find some information on it. (What I do know, though, is that because the low-level control is handled by ASICs, it's unlikely for an invalid command to cause sudden violent destruction.)
     

Share This Page