I have alot of ATX powersupplyies, i'm planning to use those for some of my experiments... where the first will be a bench supply.

I will probably only need +3.3v +5v +12v.

From what i read an old powersupply which has more ampeeres on the 5v side needs at least 1A of minimum current.... so i will prolly need 10ohm 10W dummy load... but i will test first if it shuts down.

Anyway the question is about the POWER_GOOD / P.G. cable and the POWER_ON / PS-ON ...

  1. The POWER_GOOD outputs HIGH wehen the powersupply's voltages are stabilized.
  2. To turn the powersupply on i need to attach the POWER_ON to the ground.

Can i do something like the following to turn it on when everything is stable?


simulate this circuit – Schematic created using CircuitLab

Or is the POWER_GOOD only active after i ground the POWER_ON?

Another question that is less important...

There is also the +5v standby voltage. Those 5v are always on even if the power_on is not tied to the ground. Is that stable?? I could attach an attiny or whatever microcontrolller to turn on and off the powersupply, maybe wirelessly..and also a display that shows the various voltages. Most of the powersupplies i have say that the max output of the +5 VSB (standby voltage) is more than one amp.

  • \$\begingroup\$ you can test the POWER_GOOD pin on the PS ( check if it's high ) without grounding the POWER_ON \$\endgroup\$ – Chebhou Feb 20 '16 at 13:12
  • \$\begingroup\$ From my experience you need dummy loads on some of the rails to get the voltage regulation to work on the other ones. It varies from supply to power supply. It is worse with the crappy, cheap ones. \$\endgroup\$ – Eric Urban Feb 20 '16 at 13:16

The described circuit should not work as the sequencing is backwards.

From this Wiki page:

The ATX specification defines the Power-Good signal as a +5 volt (V) signal generated in the power supply when it has passed its internal self-tests and the outputs have stabilized. This normally takes between 0.1 and 0.5 seconds after the power supply is switched on. The signal is then sent to the motherboard, where it is received by the processor timer chip that controls the reset line to the processor.

So basically, the sequence would be as follows:

  1. Ground the PS_ON signal to enable the power supply - the PSU sits in standby mode supplying only +5VSB until this signal is pulled low.
  2. Hold all circuitry in a reset state while the POWER_GOOD signal is low.
  3. Once POWER_GOOD asserts (goes to +5V), release your circuitry (e.g. any MCUs, etc.) from reset.

This extra period is specified to allow the power supply rails to stabilise and reach the correct voltage before you start running any processors.

The +5VSB rail is supplied specifically for the reason you suggest. It is there to power things like the RTC (so as no to drain the battery) and also any control circuitry which switches on the power supply. The PSU will specify how much current you can draw, but it is usually in the order of 500mA or so. Powering an ATTiny from this would not be an issue at all.

| improve this answer | |
  • \$\begingroup\$ basically it's simpler to just ground the power_on and use the power_good as reference... doing something like this.built-to-spec.com/blog/wp-content/uploads/2011/12/back.jpg \$\endgroup\$ – cocco Feb 21 '16 at 6:30
  • \$\begingroup\$ @cocco That is one option, yes. If you are using an ATTiny or whatever MCU, you could connect the power good signal to the MCU reset pin as well. \$\endgroup\$ – Tom Carpenter Feb 21 '16 at 6:31
  • \$\begingroup\$ but to stop the power from outgoing i would need mosfets... lots of them or relays... to feed my "experiment" with stable power ..i would just get a stable attiny doing so. \$\endgroup\$ – cocco Feb 21 '16 at 6:34
  • \$\begingroup\$ not to talk about the voltage drops... caused by the mosfets \$\endgroup\$ – cocco Feb 21 '16 at 6:35
  • \$\begingroup\$ @cocco yes that's true. And to be honest for most applications 500ms while the power ramps up isn't really going to be noticeable. \$\endgroup\$ – Tom Carpenter Feb 21 '16 at 6:36

As you suspected, you're putting the horse before the cart. POWER_GOOD indicates that the outputs are stable after POWER_ON has been asserted. On a real motherboard, when you press the power button, it asserts PWR_ON and then waits for PWR_GOOD before proceeding. That circuitry functions because it's being powered by the 5V stand-by output from the power supply.

The 5V stand-by is on whenever the power supply is receiving AC power, regardless of the state of PWR_ON and is stable as long as AC power is supplied. There is no electrical warning that power is dropping, which you probably don't need anyway.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.