P5KPL-AM/PS won't power on

[pfrcom]

Seeing this P5KPL-AM/PS is built with Chemicon(?) polymers and Panasonic electrolytics, I'm surprised that it won't power on with a known good power supply

Its Nikos P75N02LDG mosfets seemed to be shorted between Drain & Source, so I thought it'd be worth swapping in some AP 90T03GH which don't show the same behaviour - in comparable positions on a scrap P5K SE/EPU

Data sheets attached for both mosfets

Having removed the P75N02LDG, I find the board still won't power on, so I'm wondering whether to continue, or is this another example of intractable Asus Sudden Death Syndrome

What do the electronics gurus here think ?

[momaka]

Re: P5KPL-AM/PS won't power on

Having removed the P75N02LDG, I find the board still won't power on, so I'm wondering whether to continue, or is this another example of intractable Asus Sudden Death Syndrome

Won't power ON at all (i.e. PSU ticks and fans just twitch) or the motherboard turns ON but doesn't POST?

If the latter, it could be ASUS ASDS. But check voltages on all major motherboard rails first anyways. These are usually (but no limited to): CPU Vcore, NB/chipset, RAM Vcc, RAM Vtt, and possibly 3.3V stand-by and SB.

Its Nikos P75N02LDG mosfets seemed to be shorted between Drain & Source

Is this out of circuit or in circuit?
If out of circuit, make sure to discharge the Gate to Source before taking measurements. I made this mistake a few times and nearly thought the MOSFET was shorted.

[pfrcom]

Re: P5KPL-AM/PS won't power on

Won't power ON at all (i.e. PSU ticks and fans just twitch) or the motherboard turns ON but doesn't POST?

Doesn't power on at all, don't see fan twitch, can't hear PSU

Now I've removed those Nikos P75N02LDG, their solder pads exhibit the same behaviour, namely shorted between Drain & Source (get audible beep on DMM continuity test)

Is this out of circuit or in circuit?

That was in circuit

[momaka]

Re: P5KPL-AM/PS won't power on

Doesn't power on at all, don't see fan twitch, can't hear PSU

This is not ASUS ASDS then. ASUS ASDS is more of when board powers ON and fans spin but you get nothing on the screen (no POST). So this may actually be repairable.

Now I've removed those Nikos P75N02LDG, their solder pads exhibit the same behaviour, namely shorted between Drain & Source (get audible beep on DMM continuity test)

So the MOSFETs test shorted between S and D out of the board or just the pads on the motherboard?
...
Well, in any case, take a good high quality picture (as close to as possible to 2000x2000 or whatever the forum limits are) of the entire board and circle where these MOSFETs are.

Most motherboards have an "always-on" 2.6V or 3.3V power supply rail that gets powered from the 5VSB. Sometimes it is implemented with just a single MOSFET in a linear design, other times it's done with a buck converter. If this rail is dead, the motherboard will not power on.

Also, don't use diode/continuity test exclusively. That is, if you find that two points have continuity, then switch back to the lowest resistance scale on your multimeter and measure the resistance. If it's not 2 Ohms or less (or whatever is the lowest your multimeter can measure), then it may not necessarily be a short. Reason I mention this here is because most mutlimeters will "beep" when there is less than 50 Ohms resistance (though it does vary a lot between different multimeters).
For all the information that we have so far here, you could be measuring across the NB or SB. These normally do exibit low resistance between their supply/core rail and ground.

That was in circuit

When in doubt, always check MOSFETs out of circuit.
I bet the old ones are fine. If not, we have to find how they are connected on the board, what's controlling them, and what's killing them. It's a multi-step process, but not complicated at all, and you don't really need any fancy tools or skills. Just a multimeter and some time to map out what's connected to what.

[pfrcom]

Re: P5KPL-AM/PS won't power on

take a good high quality picture (as close to as possible to 2000x2000 or whatever the forum limits are) of the entire board and circle where these MOSFETs are

Also, don't use diode/continuity test exclusively. That is, if you find that two points have continuity, then switch back to the lowest resistance scale on your multimeter and measure the resistance.

Pics attached - circling unnecessary, seeing MOSFETs have been removed

Interesting point - without a processor fitted, the board will power on

Also without processor, no continuity between the pads where Drain & Source connections go - just a quick beep from my Fluke 77, which in my limited experience denotes a working circuit

However, when processor was fitted, the board would no longer power on, and get a continuous beep from the DMM between those same pads - the measurement was 9.9 ohms

[momaka]

Re: P5KPL-AM/PS won't power on

Also without processor, no continuity between the pads where Drain & Source connections go - just a quick beep from my Fluke 77, which in my limited experience denotes a working circuit

However, when processor was fitted, the board would no longer power on, and get a continuous beep from the DMM between those same pads - the measurement was 9.9 ohms

That's normal. Those MOSFETs are in the CPU VRM - some between 12V and V_core, the others between V_core and ground. The ones that appear shorted with the CPU in there are more than likely between V_core and ground. So the continuity you see is the low resistance of the CPU (most CPUs are like that).

Check for short circuit between 12V rail on the CPU connector and V_core (V_core is connected to the Drain pads on those MOSFETs you removed). There may be a problem there. Also check the CPU socket thorouhly for bent pins.

[pfrcom]

Re: P5KPL-AM/PS won't power on

check the CPU socket thorouhly for bent pins.

Socket looks fine to me - pic attached

[pfrcom]

Re: P5KPL-AM/PS won't power on

Check for short circuit between 12V rail on the CPU connector and V_core (V_core is connected to the Drain pads on those MOSFETs you removed)

With Fluke 77 in audible continuity mode, gives a quick beep when one probe is touching a Drain pad and the other is touching either pin in the 12V connector where yellow wires connect from the power supply

@momaka Are you willing to do some serious hand-holding, or should I just scrap the board ?

[momaka]

Re: P5KPL-AM/PS won't power on

Sorry for the long delay in response. Yes, I don't mind "hand-holding" through the process.

This is my standard procedure, but if you have any questions, feel free to ask:
https://www.badcaps.net/forum/showpo...5&postcount=13
When you're done, post what results you get.

[Christian_moh]

Re: P5KPL-AM/PS won't power on

Same problem here

[Christian_moh]

Re: P5KPL-AM/PS won't power on

But it powers on for a second and turn off itself.
I have checked those mosfets niko p75n0dlg and they are short.I need to find a similiar mosfet.What would you guys recommend.any help would be much appreciated.

[momaka]

Re: P5KPL-AM/PS won't power on

But it powers on for a second and turn off itself.
I have checked those mosfets niko p75n0dlg and they are short.I need to find a similiar mosfet.What would you guys recommend.any help would be much appreciated.

Read post #5 and #6 again.

If the "short circuit" disappears when the CPU is removed, then your MOSFETs are likely OK. CPU, RAM, Northbridge, Southbrdige... all of these have fairly low resistance to ground on their main voltage supply rails. Most multimeters will beep / show continuity when you try to measure across these, so DO NOT use continuity to test MOSFETs. Instead, measure resistance on the lowest scale possible (if using a manual range multimeter.) A true short-circuit will show the same resistance (or no more than 0.1 Ohms away) as when you take your multimeter probes and touch them together. Depending on how good (or not) your multimeter is, it will likely show 2 Ohms or less when the probes are touching together. So a real short-circuit will show up as that same exact resistance. If it's more than that even by just 1 Ohm, this may not necessarily be a short-circuit. For example, it's pretty normal for many CPUs to show less than 5 Ohms between V_core and ground.