Asus MOBO P5S800-VM/vintage

[hifihifi]

When starting this MOBO the 3.3v and 5v comes up but there is a high current drain, and the ATX powersupply gets in protection mode.
This is before the cpu power is starting.

I see no bad caps, I remove the northbridge, but the problem still exist.
Any ideas or schema's are welcome.

Greetz

Ivo

 

[Sceadwian]

It takes a LOT of current to trip an ATX supply off, if your motherboard is drawing enough to do that on it's own there's a VERY good chance there's no hope of doing anything with it. Whatever is wrong with it will be blatantly obvious on a detailed visual inspections as something is gonna be toasty.

 

[Reloadron]

When the PSU receives a PS_ON command from the motherboard it should start. If all the voltages come up within tolerance within about 500 mS (about 1/2 second) the PSU should send a PWR_OK signal to the motherboard. You only mention the 3.3 and 5 volt rails, what is the 12 volt rail doing during this time period? All of the voltages should come up at the same time. If you have not done so already disconnect things like CD/DVD drives, floppy (if there is one) and anything non essential. This link includes exactly what happens from the moment you press the computer's front panel push button to start the system. If the load is so excessive the PSU will not come online there is a heck of a load out there. Also, I would be curious what the 12 volt rail(s) are doing at power up.

What is the make and model of the PSU?

Ron

 

[hifihifi]

this is not a ATX supply problem but a MOBO problem
The 3,3V is powering up with a surge but looses voltage fast to 2,5V

Anyone a schema or suggestings?

 

[Sceadwian]

Dead motherboard, replace it, it's not practical to repair something like that, typical modern motherboards are 5 layer PCBoards not something you want to try to rework. If you're sure it's not the ATX supply that's the problem and the voltage is actually dropping to 2.5 volts as you say then the motherboard is most definitely fried.

 

[hifihifi]

Soldering this is no problem, I have special equiment for that.

I just want to study this.

 

[Reloadron]

The problem you have is a multilayer motherboard that apparently has what closely amounts to a dead short on it somewhere. A visual inspection shows nothing apparent or obvious as the problem. You have disconnected all unnecessary peripherals from the PSU so you know the problem is on the motherboard. You are working with a known good PSU that is not just folding over to to a PSU problem. The motherboard could have a failed cap but all the caps look fine visually. That really doesn't mean much as caps can fail and look just fine. There is no easy way to isolate components or sections of the board.

Actually the Asus P5S800 was a very good board years ago. I built a system using it, handed it down to my daughter and today my granddaughter uses that thing.

I guess you could disconnect the PSU, isolate the motherboard completely and take some DC resistance readings on the motherboard main PSU connector. I doubt it will tell you much. You are sure that nothing like a loose standoff became lodged under the board. This system was working fine and then one day just failed correct?

Ron

 

[hifihifi]

Indead, one day OK, other day dead.
This is a electronics question and not a IT question.

 

[Reloadron]

I realize it is an electronics question. The problem is that a motherboard was designed and built as an expendable component. When they fail, they generally get replaced. Sometimes you get lucky and the problem is apparent such as cooked or charred component and sometimes you don't. The fact that the PSU powers up and immediately shuts down within about 1/2 second says it is doing exactly what it should be doing. That being nothing but a symptom that says it is being overloaded. The actual root cause could be any number of components. The loading could be on any of the PSU rails. Since the board can't be powered up there is no logical way to troubleshoot it. You can poke and prod but that is as good as it gets. You can start by removing every component and testing the ones easily identified. That being an endless process. I don't have any ideas.

Ron

 

[hifihifi]

I have measured, and if the ossciloscoop works as a extra ground, the motherboard know stays on.
But the 3,3 is bad. If it reached 3,3V it falls again to 2,5V and 0,5 sec later again and again...
It is a regulating or power component problem, but without schema it is hard and for example I check a component with caption A016, but never find this component what it is. It's a mosfet or normal transistor or something, but its important to know witch to determine if that component is broken or not, whithout soldering.

If someone has a scherma I would be pleased...

 

[Exo]

The 3.3V comes straight out of the power supply, it's not regulated by a onboard regulator on the motherboard.