PSU pin mod and motherboard power up

[ajrettke]

I'm aware of the pin mod that can be done to PSU's to get them to turn on:
http://www.overclock.net/faqs/96712-how-jump-start-power-supply-psu.html

But if this is done while the PSU is connected to the motherboard, will it, A: damage the motherboard? and B: Will it turn the board on (i.e. wakeup, start posting etc)?

The reason I ask is I'm modding an NES and putting in a mini-ITX etc. I want to use the original power button, which when pushed to the "on" position it completes the circuit and stays closed (doesn't momentarily close then open the circuit like the reset button).

Would it work if I just spliced the two wires into the PSU that turns the board on? That way when the circuit is closed the PSU turns on...

If this won't work or if someone knows an easier way let me know!

[Psycogeec]

i might suggest doing the grounding with the proper resister instead.
from what i understand which is little . ..
the curcuit on the board does not hard ground it like that, it is a logic Low, one of them chip things sending out a "low" signal down that line. this signal is also releaced when something is going horribly wrong , stopping the flow of power to the motherboard.

on high tech electronics sites , as opposed to the DIY sites, i have seen them use a resister of the appropriate size instead.

I donno that is something to think about, if something goes horribly wrong, there will still be a continual major power flow?
and what is it that shuts OFF the power when there is other stuff horribly wrong, like a major cooling problem?

another thing peops seem to be leaving out when they mention that a PSU "Blows up" when it is off, is it is not "off" untill the power switch or the wall plug is removed, we bypass remembring that there is a 5V standby line that stays powered. but that is not on topic for this thread.

[ajrettke]

Hmmm I do see what you're saying about safety...
I think I may just connect it like normal and do a quick click to turn it on, then click it again to turn it off...still will use the original switch...just won't be depressed while it's on. Not ideal but definitly safer.

[Psycogeec]

oh, i thought you were having to run a non-momentary. for some reason.
yes, if its just the bump to turn it on then the chips can still turn it off.

but a bump wont turn it off again, if the chip is still connected holding it low.

how about moding the switch itself instead? put a cheap momentary behind the same button item, then wire it normal?

[ajrettke]

the reset on the NES is momentary, but power is toggle open/close. I wanted the look of the depressed/closed and red light on (like old school NES)....but I can just cycle it quick and it will look momentary to the mobo and boot it up.

I'd have to take apart the switch and move the contact further back so it only connects when it's depressed all the way (and when it moves back forward slightly it opens up again. If I ruin it I need to buy almost a new NES!

[Bobnova]

The PSU wants a continues signal to stay on, just wire the green wire to ground via the normal nintendo switch (and a resistor, if you like). It'll work perfectly.

[ajrettke]

Right, but will that wakeup the motherboard and have it start to post? Or do I need to trigger the state select from the power sw?

[Psycogeec]

i think as soon as you ground it, the power supply will turn on. (as we know)
as soon as you unground it without the chip setting it low, the power will go right off again.
BUT i dont even know if you can force the computer to turn on that way anymore to begin with.
i would have to get down there with my Paper Clip :-) and check that out.

there is another idea that came to mind, use a Capcitor , which will be uncharged when you make the inital connection, but then charge up as it sat there for a while, basically releacing it again, IF the chip set low, and everything magically worked (turned on actually and locked) , then the thing could still shut itself off.
major Kludge, the capacitor size would have to be tuned, and there needs to be quite a Long Low before the computer STAYS on, not just a quick blip, because the power has to stabelise first.
but its one more crasy idea to get a minor isolation going of the button after turn on.

[Bobnova]

I will test tomorrow and see what happens when i jumpstart my PSU at work.
It'll be about 11 hours, from now.

[ajrettke]

Thanks guys....just be careful! I don't want you guys blowing up some computers!

[Bobnova]

I'm a bit late, but my ECS G31 mobo will not post when paperclipped. The PSU turns on and the fans spin and such, but the mobo does nothing.

[Jmtyra]

That's for testing that out bob...I was following along with this experiment, and was rather curious as to how/if that would work or not.

Glad nothing bad happened during your 'testing'. :D

[Bobnova]

There's a reason i was testing on a $30 mobo with a $10 cpu :D

Course my ddr2-1200-4-4-4-4 ram was in there, now that i think about it. I'm glad nothing bad happened too!