ATX PSU to Ham PSU Project log

Discussion in 'Homebrew and Kit Projects' started by KF5JAK, Apr 8, 2012.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-rl
ad: K3QNTad-1
ad: l-gcopper
ad: L-rfparts
ad: Subscribe
  1. KF5JAK

    KF5JAK Ham Member QRZ Page

    I'm working on converting a few ATX Computer power supply units for powering my FT-857d. Because I've only been able to find one or two articles detail this conversion, I'd thought I'd take it upon my self to document the details of my project as I can along so as others may learn from my mistakes and more efficiently diagnose their own problems with this. Stay tuned, more details and pictures in about an hour or two! So far it's three psu's with the 5+v bus wired in series, at 20 amps. Changing soon.
  2. W9GB

    W9GB Ham Member QRZ Page

    The ATX12V is a published specification (size of case, voltage tolerances, connectors used)
    since 2000. It is updated on a regular basis. PSDG2.01.pdf

    The St. Louis Switcher was one of The first articles (May 2002, QST) in amateur radio literature.

    SEVERAL follow-on versions published and found throughout the Internet --
    USE your search engine with correct search words, and you will find them.

    CT2ILQ version

    George Heron version

    Earlier 2008 QRZ question, same desire.

    SparkFun Electronics offers the ATX bench supply kit.

    Last edited: Apr 8, 2012
  3. N0SYA

    N0SYA Ham Member QRZ Page

    Let us know what its noise signature is if any.
  4. KF5JAK

    KF5JAK Ham Member QRZ Page

    0.jpg 1.jpg 2.jpg 3.jpg 4.jpg 5.jpg
    From what you can tell from the pictures, there are four 240w atx psu's. Two are connected in parallel to create ~5.2v at ~40 amps, then wired in series with the other two to come out at ~14.8 volts at ~40 amps. When the radio is plugged in (turned off), voltage reading drops to about ~14.2v. When powered up, drops to 14.1v. During transmit and RF power at 90 watts, 13.3 to 13.5. Haven't got to hook it up to a full size antenna and properly test it, but received noise seemed clean. I had a problem (which I read about before hand) with the negative leads being tied to ground, and that the trace needed be cut to avoid short circuiting. It's pretty nasty wiring right now, but eventually I'll have some power poles or terminals on there instead. Also, I'm using a surge protector power strip to turn it on and off, makes it much easier. Once I get it out and going and do proper test, I'll post again. Thanks everyone!

    EDIT: Forgot to mention, after getting everything connected, I adjusted each variable pot controlling the voltage regulator slightly back and forth to even out power distribution. so each is averaging about ~4.9v.
  5. 4L4AGU

    4L4AGU Ham Member QRZ Page

    Well there are two ways of properly remaking ATX power supply for ham radio use (12V output).

    First of all, you need to check whenever existing 12V line can provide required current? If not, you need to re-wind/re-wire transformer, which can be pain in the back, since transformer is glued securely, there's no way to unglue it without changing ferrite characteristics and so on, so if you have such supply, my advice will be not to even try to mod it.

    If your power supply delivers enough current on 12V line, than you need to re-wire feedback circuit so it now takes control of 12V line, not of 5V line. You may also wish to remove now non-necessary parts, such as group stabilization choke and 5V, -12V, -5V etc. caps and diodes. The trick is, that ATX power supply, which is capable of at least 25A at 12V line, costs itself almost same, as dedicated 12V 25A psu :)

    Personally, I use PSU's like this

    They cost around $11-$15 in China retail shops. I'm adding additional LC circuit at output to cut some RF noise.
  6. KC9VRY

    KC9VRY Ham Member QRZ Page

  7. WA9SVD

    WA9SVD Ham Member QRZ Page

    Actually, that's pretty poor voltage regulation, to go from 14.3 Volts (slightly higher than a normal, regulated linear supply) to 13.3 Volts under load. A properly designed and built linear supply will drop at most 0.2 Volts from 0-20 Ampere load.
    Those also don't appear to be standard "ATX" computer supplies, so the original premise of the thread may be misleading. Many common "ATX" supplies don't have an adjustment that allows the "12 Volt" rail to be adjusted, and the ability to achieve 14.3 Volts (presumably no-load) is an anomaly in and of itself.
Thread Status:
Not open for further replies.

Share This Page

ad: Tetsuo-1