Overclock.net banner

Socket burn mobo + CPU - Please Help

1K views 23 replies 7 participants last post by  TerenceJ 
#1 ·
Edit: Pics

2iaeno4.jpg

wqvm9.jpg


Alright so this is a bit of a lengthy story, so here it goes...

I have a Dell PC with an i5-750 CPU that I wanted to overlock so I decided to go ahead and buy an evga p55 ftw from someone who sold it to me for a pretty darn good price (130$). It got here yesterday and I spent last night hooking everything up and when I finally turned it on it would stay on for just a split second and then shut off and kept looping like that endlessly. I called it a night and figured I'd tourbleshoot in the morning. Read up a bunch of stuff online and decided to try some ideas out. Unhooked everything except for one stick of ram and the processor / heatsink. Reconnected all the wires and cables, resat that heatsink, and finally just very lightly untightened the heatsink to avoid too much pressure and flexing on the board as someone suggested (using a Corsair A70). It seemed to help, the board would now stay on for a good five seconds before turning off and looping, and even better it was showing some debugg LED codes. Unfortunately neither the manual, nor the website, or the tech support guys has any idea what 62 and 68 stand for.

Then my dad figured it might be an incompatibility issue with the RAM, seeing as i'm using some random RAM from that Dell PC which is made by...Samsung. So I figured he's probably right and I'd buy some RAM come Monday. In the mean time I have no PC so I decided to go ahead and hook everything back up to my Dell, and in doing so I notcied something terribly wrong. The socket on the Motherboard is fried on the top left corner and so is my chip. Also, when I called the evga tech support they don't even seem to have the board listed as being registerd which is giving me a really bad feeling in the pit of my stomach despite the seller telling me he had it registered. I'm still waiting on his e-mail response.

So I have a couple questions...firstly being how in the hell does that even happen. I made sure to clear the CMOS everytime before trying to power the computer on. The pins looked perfect on the motherbaord before I ever tried anything with it. If this is a second hand board it most definitely worked in the past no? This isn't the first time I've installed a cpu / heatsink and as far as I can tell I did everything right, didn't rush things and made sure to do a good job. Now the mobo is using a foxconn socket and not a lotes, but again it worked for the first owner, and apparently that socket issue has been fixed / was only a problem for people overvlocking to 5+ Ghz.

My other question is, would Intel potentially RMA a new chip for me? I don't even know if this one is working, but I'm scared to pop it into the H55 mobo on my Dell and fubar that too. I doubt Dell would cover anything seeing as they have a very strict policy on their warranties. And finally, if the seller did in fact lie to me, do I have any kind of recourse? I payed for the board through paypal with a CC, I would have bought it through amazon but he was a US seller and I live in Canada and they don't allow those kinds of transactions. Would evga potentially help me if the board was never registered?

Any advice would be greatly appreciated guys...I'm feeling pretty ****y right about now.
 
See less See more
2
#11 ·
btw don't be too hard on yourself. This is how things go when you start this hobby. You're out a few hundred bucks (maybe), but you've learned a bit. Many of the savviest folks on this forum - much, MUCH more knowledgeable than me - have a fried component or two in their past. As for me, I still have no idea what I did to make the capacitor on that Enermax psu pop...but I'm a lot more careful about raising volts and fsb on a new rig now. And any money you may have lost will be regained, with interest, when you get a $200 chip to run much much faster than the most expensive $1000+ chip on the market.
coolsmiley.png
 
#14 ·
Quote:
Originally Posted by seward;13419703
btw don't be too hard on yourself. This is how things go when you start this hobby. You're out a few hundred bucks (maybe), but you've learned a bit. Many of the savviest folks on this forum - much, MUCH more knowledgeable than me - have a fried component or two in their past. As for me, I still have no idea what I did to make the capacitor on that Enermax psu pop...but I'm a lot more careful about raising volts and fsb on a new rig now. And any money you may have lost will be regained, with interest, when you get a $200 chip to run much much faster than the most expensive $1000+ chip on the market.
coolsmiley.png
Well the thing is, I'm not quite sure if I've learned anything because I don't think I did anything wrong, aside from buying second hand and using a chip that isn't on warranty. The thing is out of everything that could possibly happen, this is the last thing I would have expected. I didn't even think it was possible to fry a chip just because of "bad luck" without even ever attempting to OC.

Also, I'm still hopeful for a reply from the guy I bought the board from, he seemed like a very stand up guy.
 
#16 ·
Quote:
Originally Posted by myst88;13419788
Well the thing is, I'm not quite sure if I've learned anything because I don't think I did anything wrong, aside from buying second hand and using a chip that isn't on warranty. The thing is out of everything that could possibly happen, this is the last thing I would have expected. I didn't even think it was possible to fry a chip just because of "bad luck" without even ever attempting to OC.

Also, I'm still hopeful for a reply from the guy I bought the board from, he seemed like a very stand up guy.
The problem is the socket design. Maybe it's too small, so that can happen. 1366 sockets did not have that issue, probably because it is bigger and it passed more testing. Not really sure, but you are not the only one with that intel issue.

Next time, you should be looking for information before buying anything.
 
#17 ·
Oh I doubt it's stable, no need to run prime 95, I wouldn't want to risk completely destroying it at this point. Also the funny thing is I did do a lot of research before buying the motherboard, I spent a good week on and off. I knew about the faulty foxconn sockets, but everything I read was about people doing extreme overclocks at 5 + Ghz and that apparently there was absolutely no issue otherwise. I was only planning on taking the chip to max 4.0 Ghz so I figured I was fine, sadly I didn't even get the opportunity to attempt it.
 
#18 ·
Quote:


Originally Posted by myst88
View Post

Oh I doubt it's stable, no need to run prime 95, I wouldn't want to risk completely destroying it at this point. Also the funny thing is I did do a lot of research before buying the motherboard, I spent a good week on and off. I knew about the faulty foxconn sockets, but everything I read was about people doing extreme overclocks at 5 + Ghz and that apparently there was absolutely no issue otherwise. I was only planning on taking the chip to max 4.0 Ghz so I figured I was fine, sadly I didn't even get the opportunity to attempt it.

Have you tried cleaning the CPU contact points with isopropanyl?
 
#19 ·
If the burn marks are so bad Isopropyl will not get it off. Perhaps try some a little fine steel wool pad.
 
#21 ·
Quote:


Originally Posted by myst88
View Post

No, I didn't know you could "clean" them, are there any guides you guys are aware of. Seems like it might be worth a shot.

You know what isopropanyl is?

Take a piece of paper, soake it in isopropanyl and start rubbing over the contact points..
 
#23 ·
Quote:
Originally Posted by otaku_ex;13429628
maybe you can rma the chip to dell if it is clean enough. You can use benzine too. Since it's working, or at least it boots normally, I don't think they will be able to know you attempted to overclock the chip.
But he didn't even attempt it, it died immediately on first boot...
 
#24 ·
When are you reading that its downclocking? Are the games not running stable?

You know when a CPU is not under load, it might have the CIE and intel speedstep enabled so it downclocks. If you prime 95 for a second and watch the core speed via CPU-Z and watch the temp with Coretemp or real temp you'll prob see the core speed kick up to what its supposed to be.

I doubt it would damage it anymore, as soon as the program detects a problem of an unstable processor it will stop the test.
wink.gif
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top