P4SD - case LEN

[MXM]

Hello everybody I already had written a whole post on this thread and when i clicked submit i lost it all

Anyways, this board ive been working on since yesterday came with also, no signals, no video display, no POST, nada...

Some caps have been already replaced, some havent, mainly due to the fact that IMO they look Ok -but im not even 50% reliable when it come to caps-

After the caps replacement I still got nothing, then I began doing manual resoldering of the whole board, but at some point -unfinished- I decided to give it a full blood bath -which is how i like calling the hot air proc.- and let it cool off all night long, then i came back this morning, i cleaned it a little with ISO cause it had some soldering grease (is it conductive, btw?) and then i tested, results:

Unsuccesful, but present, POST...

At the middle, or so, of the memory check (before the memcheck began it had already displayed some keyb error, with plugged keyb) the display turned into little coloured squares, no particular pattern on the color changing and all about the same size, and i was wondering if someone has spotted this type of failure, and also would like to read any comment or idea on how to fix this board...

I havent plugged it yet after the test cause Im planning on finished the manual resoldering...

I am aware that the replacements I use are nothing but pure crappy brands but thats all ive got, my bad

Some pics of the board:

General overview...


Detailed pics...
















Any type of info, link, docs, tips, is appreciated.

Thanks.

[MXM]

Re: P4SD - case LEN

UPDATE: Im done halfway with the resoldering, finished the bottom side, after I tested and results were:

Unsuccessful but present POST, this time it got stuck faster than last time during the mem check, and (if it may have some meaning) the little squares appeared again this time they were all on the purple violet tones. After this I turned it off and tried to boot up again and got no POST. I havent tested it again as Im up for the rest of the resoldering.

Any tip or idea on where may this fail be at? Id truly appreciate it.

Thanks.

[MXM]

Re: P4SD - case LEN

UPDATE: Im done with resoldering the board both sides, I'd be a liar if I told you I resoldered every single point but I did as much as possible....

I havent got good results, after the resoldering I tested and got nothing, then I decided to apply some heat to it and let it cool off all night long, I did and tested this morning with same no positive results, no POST, nada...

I was wondering if someone could help me out on this...

Any info, link, manual, doc, suggestion is accepted and very wel appreciated.

Thanks.

[Leopard]

Re: P4SD - case LEN

I cannot see any pictures in the first post. They would help a lot in this case.

[MXM]

Re: P4SD - case LEN

dude i dont know what happened, they were there i promise.
i am reposting em since i cant edit (can i?) that message.

one second

[MXM]

Re: P4SD - case LEN

Pics from first post:



















Thanks

[Scenic]

Re: P4SD - case LEN

general purpose caps on a mainboard? (2nd pic NCC SMG)

also.. what the hell happened on pic 3 ?!
above the yellow bios reset button and the green LED. looks like something was spilled onto that board o.O
and what's with those bumps on the FET to the right of the oscon?

[MXM]

Re: P4SD - case LEN

ok lemme check and i'll tell you, im gonna take pics of it, since its been already resoldered.

I added some soldering grease in some random areas while resoldering just to have it easier.... about this i got two questions:
is that grease conductive?
may i wash the boards with plain water and cheap soap cause if you saw grease is prolly cause the ISO didnt get entirely rid of it...

about the caps..... I know, i do not have better (yet) I am hoping next month (next tues) to have more budget and be able to buy more and better. my bad.

the bumps is prolly the combination of bad quality soldering roll and my bad soldering technique.

Thanks for the reply. I appreciate.

[MXM]

Re: P4SD - case LEN

pics from it after the resoldering.... the areas i didnt touch were basically those hard to reach, in between the DIMM slots and such.... everything else was resoldered. Sorry about the bad technique, is me plus bad the rush and the rest, my bad.

The board still hasnt shown no signs of living after the full resodlering and heat applied. last thing it displayed was the fail POST with little squares all over...













Thanks for replying.

[370forlife]

Re: P4SD - case LEN

Umm. What caps did you use to replace them?

[MXM]

Re: P4SD - case LEN

really cheap and bad ones dude.... mostly wendell and ltec... i just dont have the choice... but i can say ive recapped many others that come out working Ok.

[c_hegge]

Re: P4SD - case LEN

really cheap and bad ones dude.... mostly wendell and ltec... i just dont have the choice... but i can say ive recapped many others that come out working Ok.

Shame on you! You can get good caps like rubycon from badcaps.net

https://www.badcaps.net/pages.php?vid=22

[MXM]

Re: P4SD - case LEN

dude i am actually lookin for better quality caps... just got many stuff around...

UPDATE: I decided Id apply some heavy heat to the board, so I did and the results were:

First it booted and gave POST, actually pretty good, got excited and everything... so i mounted it to its base and set the processor as it should, it worked fine for at least 20 mins, in that time I put it under memory tests, memtest+ had at least passed one time all its tests, and so i let it be tested for more and is when it failed, it got simply froze.

So I uninstalled the processor and tried another one, it didnt boot, i tried another processor (a third one) and in this ocassion it showed the 3red flash error, which according to the manual is "Processor Not Seated/Installed" error, so I thought try another, I tried another proc (4th) and it booted and in the middle of the POST memory checks it failed, the screen went to little squares (as I noticed on other fails on other boards) of different colors changing no particular pattern -at least not that i noticed- and since then it has not shown any type of signal. I went back to the proc that gave me the positivi initial results and got nothing again....

Any idea is very welcome. The board looks as in the pics only difference is the excesive heat I applied to it -since worse is nothing I took my chances-

i did let it cool off for a long long time before testing.

Peace.

[MXM]

Re: P4SD - case LEN

UPDATE: After my lunch break i came back and installed the proc that gave me good results at the beggining, now the best im getting is the 5 red flash, which according to the manual is "Memory Error" please give me the benefit of the doubt, is not the DIMMs since i got plenty to test, some times the POST will show memorry errors too and sometimes is just the 5red flashes, some other times i still get Nothing...ive tried different combinations of memory slots, i keep gething those errors, no particular patern...and yeah mostly after few errors I go back to nothing on screen, when POST ends succesfully, or at least the memcheck part, i normally shut it off to try new tests, after the heat-cool off I brushed the DIMM slots with ISO.... I was wondering if someone had an idea on what could I do to try to fix this one?? the caps close to the mem slots were changed, wel the 2x 1500uf/6.3v were, the original were TEAPO cheapo and i replaced em for worse LTec, but dont have other, I have other brand but they are 1800uf/6.3v if someone recommends me to recap for those I would. The other is a 3300uf/6.3v but is a Rubycon and looks Ok so i have not thought on replacing that one.... I dont have suspects on any surface element but if someone has something please advise... please.


Thanks

[seanc]

Re: P4SD - case LEN

Have you checked under the northbridge for any failed SMT caps?

[Scenic]

Re: P4SD - case LEN

this kinda sounds like problems with the BGA of the northbridge.. (pixelized video, no cpu error, ram error, or nothing at all..)

edit: post #6 .. i see a ripped off surface mount cap to the right of the northbridge heatsink

[seanc]

Re: P4SD - case LEN

Agreed, C37

[c_hegge]

Re: P4SD - case LEN

Good spotting. Never noticed that until now

[MXM]

Re: P4SD - case LEN

guys ur amazing how can u spot that from a pic and i cant spot that in real stuff !!!
kool i still am tryin to find it out

i am resoldering the area asap and will update.

THANK YOU ALL.

[MXM]

Re: P4SD - case LEN

UPDATE: Hey.... This is what the board looks as I type:

im sorry took me longer but i accidentally sucked C66 and had to go n get it from a dead board











after resoldering i tested for continuity and came out Ok.

Then i thought id give it a try, i did and it booted so i set it up for testing on memtest and it begun working out not so ok, i noticed the wall time wasnt counting as it should -like for a split sec the seconds got stuck and then suddenly advance one or two seconds quickly-.... and the cpu fan was at it max speed... and after a little time it went dead; No more display (black screen) and thats it. Since it hasnt booted again.

Any idea is welcome.

Thanks.