Page 1 of 3 123 LastLast
Results 1 to 10 of 22

Thread: presslab's powersuply..

  1. #1
    Variable Bitrate
    Join Date
    Jul 2001
    Location
    The Netherlands
    Posts
    307

    Post presslab's powersuply..

    Arby or presslab,

    as you may know I keep getting very high ratings on my +5 and +12V outputs (+8 and +26) to be precise. I tried giving the psu some load to stabalize these voltages, but that didn't work.. Now arby said on his site :" If one voltage is right, but another wrong, then you didn't wind the transformer properly. "

    is this true? I thought the dc site from the original psu regulates the voltages? I was thinking that when I wind the transformer wrong the DC site would blowup or something, but everything seems to go ok (except for the IRF's that get so damn hot, even with a very large heatsink....)

    Marsjell

    [ 01-07-2002: Message edited by: marsjell ]
    If at first you don't succeed.........
    destroy all evidence you tried

    for info on t6369c LCD or Presslab's powersupply check:
    http://www.namms.tk <=updated!!

  2. #2
    Newbie
    Join Date
    Jul 2001
    Posts
    4

    Post

    Yeah, I think you wound the transformer wrong, the reason you get the correct voltages on the -5volts is because that if obtained by regulating the -12volts (or in your case -26) The reason the -12 is -26 is because this doesn't use a regulator, but rather uses the same windings as the +12volts does. It is also possibly that you wound the primary incorrently by my calculations, if you were to wind 2turns on the primary, rather that 3, the outputs would be about 8.3volts on the 5volts, and about 19.9 on the 12volts, so, from that calculation, it may seem that either you have incorrectly wound the primary and the 12volt windings, or my calculations were incorrect..

    The secondary part of the circuit does not regulate the power, it just converts it to DC (using mospecs), and filters it with capacitors, and inductors

  3. #3
    Newbie
    Join Date
    Jun 2001
    Location
    Sebastopol, CA
    Posts
    55

    Post

    Originally posted by marsjell:
    <STRONG>Arby or presslab,

    as you may know I keep getting very high ratings on my +5 and +12V outputs (+8 and +26) to be precise. I tried giving the psu some load to stabalize these voltages, but that didn't work.. Now arby said on his site :" If one voltage is right, but another wrong, then you didn't wind the transformer properly. "

    is this true? I thought the dc site from the original psu regulates the voltages? I was thinking that when I wind the transformer wrong the DC site would blowup or something, but everything seems to go ok (except for the IRF's that get so damn hot, even with a very large heatsink....)

    Marsjell

    [ 01-07-2002: Message edited by: marsjell ]</STRONG>
    Hi Marsjell,

    It does seem like it may be wound incorrectly then, or possibly connected wrong. If you hook the negative side of the +5V winding to the +3.3V winding positive, this would give you +8.3V (about) if your power supply was not designed to work like this. Do you have the original transformer? Did you take it apart and note the number of turns and their connections? If you don't remember the number of turns it might be easier to start with a new PSU, with the old transformer and primary circuit. I also had to rewind my transformer a couple times to get it just right. The IRF540s will get hot, but with no load they should be below 120 deg F (hot to the touch, but doesn't burn). They have a maximum operating case temperature of about 300 deg F. That's pretty toasty! I had a fan in my PSU.

    PressLab

  4. #4
    Newbie
    Join Date
    Jul 2001
    Posts
    4

    Post

    Yeah, I didn't think of actually connecting the transformer wrong... But I don't think the supply is ATX.... is it marsjell???

  5. #5
    Maximum Bitrate
    Join Date
    Nov 1999
    Location
    Sydney NSW Australia
    Posts
    581

    Post

    PressLab a quick question ...
    How can we reduce the heat dispertion of the mosfets? I dont mean by putting a fan on it or something, i mean is there anyways of making those transistors not to get that hot using the circutry?

    Thanks

    [ 01-07-2002: Message edited by: Fosgate ]
    Fosgate

    System Comp V3 - In progress.
    Low power MB with C7 CPU, DC-DC PSU, car ECU link, USB TV, GPS, 7" TFT, Wireless, Voice.

  6. #6
    Newbie
    Join Date
    Jun 2001
    Location
    Sebastopol, CA
    Posts
    55

    Post

    Originally posted by Fosgate:
    <STRONG>PressLab a quick question ...
    How can we reduce the heat dispertion of the mosfets? I dont mean by putting a fan on it or something, i mean is there anyways of making those transistors not to get that hot using the circutry?

    Thanks

    [ 01-07-2002: Message edited by: Fosgate ]</STRONG>
    Yeah, use a MOSFET with a lower Rds(on). A good example would be the IRL1404. It's a bit more expensive though. Most MOSFETs are designed to withstand very high temperatures, but a lower Rds(on) will make the supply more efficient.

    PressLab

  7. #7
    Variable Bitrate
    Join Date
    Jul 2001
    Location
    The Netherlands
    Posts
    307

    Post

    presslab,

    I followed your advice and started on a new psu, this time a AT 'cause I didn't have another ATX lying around. Now I finished winding the transformer I noticed that there is absolutly no optocoupler on it, there are however 2 small transformers.

    Do you have a suggestion where to connect the voltage feedback? The +5V output?

    thanks again
    If at first you don't succeed.........
    destroy all evidence you tried

    for info on t6369c LCD or Presslab's powersupply check:
    http://www.namms.tk <=updated!!

  8. #8
    Maximum Bitrate
    Join Date
    Nov 1999
    Location
    Sydney NSW Australia
    Posts
    581

    Post

    Originally posted by PressLab:
    <STRONG>

    Yeah, use a MOSFET with a lower Rds(on). A good example would be the IRL1404. It's a bit more expensive though. Most MOSFETs are designed to withstand very high temperatures, but a lower Rds(on) will make the supply more efficient.

    PressLab</STRONG>
    Yeah thanks PressLab... I'll look into it.
    Fosgate

    System Comp V3 - In progress.
    Low power MB with C7 CPU, DC-DC PSU, car ECU link, USB TV, GPS, 7" TFT, Wireless, Voice.

  9. #9
    Variable Bitrate
    Join Date
    Jul 2001
    Location
    The Netherlands
    Posts
    307

    Post

    hm just tried the rewound AT PS (220V input)but now I have really high values.. somewhere around 31V and 26V maybe I've been stupid, arby wrote that you should have somewhere between 40-50 windings on the prim. side, I had 36.

    My windings where :

    prim side:
    36 windings single wire

    sec side:
    3 windings 4 wires
    3 windings 4 wires (wound in oposit(?) direction)
    these windings connect directly to the ground on the "output-side"

    4 windings 2 wires
    4 windings 2 wires

    do these windings mean I have to rewind in another way?

    please help, as I am totally stuck...

    thanks, Marsjell

    [ 01-10-2002: Message edited by: marsjell ]
    If at first you don't succeed.........
    destroy all evidence you tried

    for info on t6369c LCD or Presslab's powersupply check:
    http://www.namms.tk <=updated!!

  10. #10
    Newbie
    Join Date
    Jul 2001
    Posts
    4

    Post

    No, you would wind it the same way as described... If the PS didn't have an optocoupler, where did you connect the opto leads???

Page 1 of 3 123 LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •