Author Topic: ATX power supply rebuild  (Read 3737 times)

0 Members and 1 Guest are viewing this topic.

Offline icecoldTopic starter

  • Newbie
  • Posts: 4
  • Country: at
ATX power supply rebuild
« on: December 27, 2016, 12:21:37 pm »
Hello,

I rebuilt an ATX power supply to regulate the 12V line and made it adjustable. After disabling the under-voltage protection it can go down to 3V now, but below that it starts to freak out (hissing noises and awful waveforms across the transformer). I think this is because of pulse-skipping.

Is it possible to make it stable below 3V?

Its very similar to this one: http://320volt.com/wp-content/uploads/2011/05/patriot-switching-power-supply-sg6105d-cef2n60-ee19-ee16-ei33-f12c20c-s20c40c.png

thanks :)
 

Offline Seekonk

  • Super Contributor
  • ***
  • Posts: 1938
  • Country: us
Re: ATX power supply rebuild
« Reply #1 on: December 27, 2016, 01:20:47 pm »
I don't see much reason for it.  The real problem is that the output is opto isolated from the line power.  You can't get below the 1.7V the led needs to turn on without another external supply.
 

Offline icecoldTopic starter

  • Newbie
  • Posts: 4
  • Country: at
Re: ATX power supply rebuild
« Reply #2 on: December 27, 2016, 01:43:15 pm »
The opto is used for the 5V standby supply which works fine even if the main supply is unstable. I altered the feedback loop on pin 17 of the SG6105D to only regulate 12V and made it adjustable which works fine between 3V - 15V on the 12V rail. It must be something around T3 or its drive circuit which makes it unstable if pulses are skipped.
 

Offline icecoldTopic starter

  • Newbie
  • Posts: 4
  • Country: at
Re: ATX power supply rebuild
« Reply #3 on: December 28, 2016, 01:13:36 pm »
It actually was a similar problem. The input of the error amplifier (SG6105D pin 17) is compared to a 2.5V reference which makes the output unable to go below this value. At the extreme setting of direct connection of the output to the input of the error amplifier the psu gets unstable (maybe because it bridges the filters).
 

Offline danadak

  • Super Contributor
  • ***
  • Posts: 1875
  • Country: us
  • Reactor Operator SSN-583, Retired EE
Re: ATX power supply rebuild
« Reply #4 on: December 28, 2016, 01:19:31 pm »
You could always add an LDO but that wastes power, very inefficient.
But then if you "synched" the LDO ref leg to your control over the ATX
Vout might be able to minimize power loss while still keeping solution
in regulation. Just a thought.


Regards, Dana.
Love Cypress PSOC, ATTiny, Bit Slice, OpAmps, Oscilloscopes, and Analog Gurus like Pease, Miller, Widlar, Dobkin, obsessed with being an engineer
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf