Author Topic: Why do ATX power supplies use a transformer after the chopper?  (Read 4223 times)

0 Members and 1 Guest are viewing this topic.

Offline WhalesTopic starter

  • Super Contributor
  • ***
  • Posts: 1899
  • Country: au
    • Halestrom
As I understand wound transformers are useful in DC-DC power supplies as they isolate the outputs, but any safety advantage here is completely irrelevant as all ATX power supplies are designed to bridge output GND and mains earth together.

Surely it would be cheaper/simpler to use a wound inductor (single wire) rather than a wound transformer (at least two wires)?   
« Last Edit: March 29, 2015, 02:23:12 am by Whales »
 

Offline c4757p

  • Super Contributor
  • ***
  • Posts: 7799
  • Country: us
  • adieu
Re: Why do ATX power supplies use a transformer after the chopper?
« Reply #1 on: March 29, 2015, 02:38:13 am »
any safety advantage here is completely irrelevant as all ATX power supplies are designed to bridge output GND and mains earth together.

:wtf:

Mains earth and mains neutral aren't the same thing...
No longer active here - try the IRC channel if you just can't be without me :)
 

Online wraper

  • Supporter
  • ****
  • Posts: 16792
  • Country: lv
Re: Why do ATX power supplies use a transformer after the chopper?
« Reply #2 on: March 29, 2015, 02:56:05 am »
LOL   :-DD |O  :palm:  :scared:
 

Online wraper

  • Supporter
  • ****
  • Posts: 16792
  • Country: lv
Re: Why do ATX power supplies use a transformer after the chopper?
« Reply #3 on: March 29, 2015, 03:20:25 am »
I'd like to see what will happen when pass transistor in such buck converter decides to become short.
 

Offline WhalesTopic starter

  • Super Contributor
  • ***
  • Posts: 1899
  • Country: au
    • Halestrom
Re: Why do ATX power supplies use a transformer after the chopper?
« Reply #4 on: March 29, 2015, 04:24:36 am »
I'd like to see what will happen when pass transistor in such buck converter decides to become short.

Ah, I've only been thinking of external failure modes.  For example: connecting the output to active or neutral.  In this situation whether or not the power supply is isolated internally makes no difference as the shared ground will still form a large current path and destroy the device/start a fire.  A 'fully' isolated power supply would not exhibit this.

any safety advantage here is completely irrelevant as all ATX power supplies are designed to bridge output GND and mains earth together.

:wtf:

Mains earth and mains neutral aren't the same thing...
"Output GND" as in the secondary side GND, not mains neutral.


Also using a transformer increases efficiency since on one side of the transformer the voltage is around 400VDC and on the other side of the transformer it is +/-12VDC or so so rather than operating at a small duty cycle due to the large voltage ratio, a more equal duty cycle and transformer is used.
Thanks evb.  I'm curious as to where a 50% duty cycle avoids losses -- is it to do with differences in harmonics?  Transistor switching loss either way would be equal as far as I can see.
« Last Edit: March 29, 2015, 04:30:33 am by Whales »
 

Offline rs20

  • Super Contributor
  • ***
  • Posts: 2317
  • Country: au
Re: Why do ATX power supplies use a transformer after the chopper?
« Reply #5 on: March 29, 2015, 04:38:29 am »
"Output GND" as in the secondary side GND, not mains neutral.

You're proposing a single-wire indicator, so what on earth do you mean by "the secondary side"? Please draw a diagram to explain what you're proposing, you'll end up with at most a diode between mains earth and mains neutral = not good.
 

Offline WhalesTopic starter

  • Super Contributor
  • ***
  • Posts: 1899
  • Country: au
    • Halestrom
Re: Why do ATX power supplies use a transformer after the chopper?
« Reply #6 on: March 29, 2015, 04:50:12 am »
"Output GND" as in the secondary side GND, not mains neutral.

You're proposing a single-wire indicator, so what on earth do you mean by "the secondary side"? Please draw a diagram to explain what you're proposing, you'll end up with at most a diode between mains earth and mains neutral = not good.
I'm aware of that danger now.  I thought c4757b may have misunderstood me.

Before anyone else gets worried: if I had known the answer I would not have asked the question.  No I'm not working on any mains-side circuity. 

There's nothing wrong in not knowing enough about safety.  There is a lot wrong however in pretending that you do if you don't.  I'm being open about what I didn't understand: that's why I asked the question here in the beginners forum :).  The worst thing anyone can do is give people a sense that they should not ask safety questions, especially if they seem dumb to others.
« Last Edit: March 29, 2015, 04:54:24 am by Whales »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf