[Fixed] Double action power on sequence after PSU upgrade

[Levent]

Hi,

 

I recently upgraded the PSU in my desktop to a Seasonic Focus Plus 650W (SSR-650FX) from a VS450.

Ever since the upgrade when I click the power on button:

1. Clicking the power button, a relay clicks.
2. Device does nothing for 2 seconds.
3. Device turns on.

Issue I am having with this is that, when device loses power, it becomes unresponsive to WoL packets. This was not the case with VS450, and now I have to power on the device every time a power loss happens since WoL doesnt work.

 

Any ideas?

 

B450 Aorus M

4x8G RAM

R5 3600

GTX970

2T 670P.

 

 

 

EDIT:

 

THIS PROBLEM WAS FIXED BY DISABLING "POWER LOADING" IN BIOS!!!!

Edited April 19, 2022 by Levent
FIXED!

[--SID--]

Seems like the motherboard has to train the RAM every boot. Is the CMOS battery empty?

[Levent]

18 minutes ago, --SID-- said:

Seems like the motherboard has to train the RAM every boot. Is the CMOS battery empty?

Not an issue with the RAM, swapping the previous power supply instantly resolves the issue. Cmos battery is fully charged and keeps the saved values. I also tested with another kit when I upgraded the PSU, same thing.

 

I updated the post.

Edited March 13, 2022 by Levent

[Biohazard777]

1 hour ago, Levent said:

Issue I am having with this is that, when device loses power, it becomes unresponsive to WoL packets. This was not the case with VS450, and now I have to power on the device every time a power loss happens since WoL doesnt work.

 

Any ideas?

Try disabling ErP in BIOS.

[Levent]

14 hours ago, Biohazard777 said:

Try disabling ErP in BIOS.

That fixed the WoL issue. I guess I cant do much about the PSU turning on an off before actually computer starting?

[Biohazard777]

On 3/13/2022 at 10:03 PM, Levent said:

• Clicking the power button, a relay clicks.
• Device does nothing for 2 seconds.
• Device turns on.

12 hours ago, Levent said:

I guess I cant do much about the PSU turning on an off before actually computer starting?

Hm... power good signal (PWR_OK) should be set to high (5V) by the PSU no sooner than 100ms and no higher than 500ms, 2000ms is way outside the spec.

spec:

Spoiler

https://www.intel.com/content/dam/www/public/us/en/documents/guides/power-supply-design-guide-june.pdf

 

It could be it really needs those 2s to stabilize the outputs... which ain't good, and I don't see a way to fix it (without replacing the PSU ofc.)

You could do what the fire hazard cheap PSUs do... wire the one of the 5V outputs to PWR_OK. Joking, don't do that except for testing purposes, because in case something goes wrong with the PSU that signal is withdrawn, and if it is hard wired to 5V that ain't gonna happen.
Anyhow I am no expert when it comes to PSUs, perhaps we can summon @jonnyGURUto help you out with this issues, could be something else I am missing.

[Guest]

No mechanical relay is fast enough to support T1 timings.  Thing is, the inrush current relay should only click when power is initially applied to the PSU.  Once the bulk cap is energized, that relay should remain closed.  There's something else afoot here.

 

[Levent]

4 hours ago, jonnyGURU said:

No mechanical relay is fast enough to support T1 timings.  Thing is, the inrush current relay should only click when power is initially applied to the PSU.  Once the bulk cap is energized, that relay should remain closed.  There's something else afoot here.

 

Here it is in action, if its anything to go by. Sorry for shitty video and editing.

It is me smashing the DEL key at the end by the way.