Risks of testing maybe dead components? (my bro gave me a carcass for Xmas)

It's a carcass because the PSU blew.

Mobo is ASUS P4P800S, two sticks of 512m PC3200, the CPU is 3Ghz Northwood I think (I took off the heat sink & cleaned it all up, googled the number, reinstalled with fresh paste, but forgot to jot down the number)

Connecting a good psu does nothing (I've just noticed there's no speaker on the board tho, so if it's beeping I don't know.)


However, I have at hand a fully functional and very similar Dell computer, it has a 3ghz northwood, probably the same kind of ram & etc.

Well I have a green light to use it to test parts, but I really don't want to blow it up. Say I put the maybe dead cpu in it see if it fires up, can this break the mobo? What about putting the good cpu in the maybe blown mobo? What about putting maybe dead ram in the good computer, etc etc? I did a quick search and came to the conclusion there's not much at risk here, but if there's something I should know about...


Really want to make this work since it would be hell of an upgrade next to the 1.5ghz Palomino w/256m ram I'm currently using.


On a hunch, I'm guessing the CPU popped along with the PSU.

The memory should be okay to put in the test bed. Be sure to run something like memtest86+ for a good 20-30 hours. If the memory passes everything over that time frame then you can call it good.

There's little chance the processor would kill a good motherboard if it is bad.

Check the motherboard for obvious faults, such as bulged or leaking capacitors. If there are any, then you can replace all the caps on the board and hope for the best.

Ah thanks, yes six caps right next to the CPU socket are slightly bulged, but then the Gigabyte 7VTXE+ I'm currently using is quite much worse with no apparent ill effects. The one on the more recent board look quite easy to replace tho.


And I'm wondering what if I put a known good cpu in this potentially bad mobo, any risks here?


Edit: without being an extensive test, the Dell boots up fine with each ram stick and with the processor from the carcass. Guess the mobo is shot...

Yes, the motherboard is likely damaged.

Go to Badcaps.net for a kit (if you can solder) or get information on having them install a set of replacement capacitors on your board.

Either way, the price is likely less than you would pay for a new board and you'll be doing yourself and the world a favor by repairing this one.

Welp, I replaced the caps and it's still dead, checked the fets also, but they seem alright. Oh well. So I've been looking to grab some other compatible mobo, looks like these are indeed still kinda expensive.


Getting pretty far off topic for cpu forum tho.

*Drives thread off-topic, through a pond, over bushes and toward the edge of a cliff.*

Strange. Does it still produce vcore? You don't actually have to put a CPU in there for it to do so, I don't think.. At least you don't on a P4C800. You can just test from one of the nearby inductors.

No beeps or anything, either?

No beeps, there's a "pop" from the onboard sound thing when i turn it on tho.

Vcore is ~0.07v (!)

OK full disclosure: I was lazy and only changed the row of caps right by the cpu socket, as they were the only ones bloated.

The three big-ish caps near the inductors all measure ~12.5v, similar to what the PSU I just used to test gives, as read out of a molex plug.

On all three of the fets closest to the cpu, drain reads ~0.07, gate & source read ~0.01.

Measurements taken grounded off the PSU case btw (board + psu on a desk)

Well, you seem well-versed enough to know how to change caps with the correct values, so I shan't worry about that. You should change all of the electrolytics, though.

Just changing those should be fine for now... Unless you've about tripled the bulk capacitance, it should be OK with whatever caps you've chosen.

Only thing I can suggest is putting the CPU in now, and trying it again.. Maybe yours doesn't work without the CPU because the CMOS has been cleared at some point so it has no VID value stored to use. If it still gives nearly 0v with a CPU in it... I don't know what else to suggest.

Yeah I got some Nishicon caps of same value but slightly higher voltage.


Oh, wow! W/cpu installed the middle inductor goes up to 0.19v, the two others go to 0.21!

*dramatic look*


Checking out the 3 caps & fets further away from the cpu, they all seem to react the same, but with the board running and cpu installed, the one in the middle gives next to no amps compared to the others. I'm guessing there's something wrong with the driver:

http://www.datasheetarchive.com/pdf/600047.pdf

Tried adding an extra cap to the one in the middle, on the other side of the board, but that didn't help.

Also, the cpu gets hot despite low vcore.



In any case I should be receiving a P4P800E and a ZM600-HP PSU sometime next week.



Edit: it's occurred to me my readings could be quite wrong since I'm using an ancient avm, 30Kohm/v. Still, seeing how these processors suck 80+w, should be able to power a needle right?

Strange how the CPU gets hot with no volts... There could be a short, but I don't see how that could happen on a motherboard... It doesn't make any sense. Are you reading from the output of the inductor? The leg that goes to that row of caps you've replaced..

As for your AVM.. I have no idea