LGA775 Flimsy CPU Socket Thread

[Sentential]

I honestally do not want this to turn into a witch-hunt. Nor do I blame Asus for my problems. I want to see how many other people are having problems with damaged LGA775 sockets. Im not sure if this is a recurring theme or not.

I belive the reason my first P5WD2 failed was from pressure applied by my Sonic tower. My Abit AS8 does not suffer from any issues yet both Asus boards I have recieved do
____________

EDIT: I found the INQ news story. Judge for yourself:

http://www.theinquirer.net/?article=14865

EDIT2: Asus appears to be using a different type of Foxconn CPU socket than is Abit

[Overclocker550]

avoid all mobos with weak sockets. If you boycott, they will get their act strait

[jtracc20]

my mobo is the P5P800, and i am sure its no foxcon. never had a issue, and i have a thermaltake jungle on it.

[Alchemy1]

I don't have any problems, but I have noticed that my base kind of pushes some of the capacitors that are close to the socket. I am scared I am going to snap one of them one day, but I have good temps now. So, no need to take the HS off again until I get a new chip...HOPEFULY.

[crimedog]

i have two p5wd2-p in my posession, one was an RMA (has a foxconn socket) the other was a refurb from newegg with the normal socket. both have had phase change or dry ice tube strapped to them, both of which put an extremely high amount of pressure on the chip/board. both have been working fine and have taken a lot of chip swapping.

that's all i have to say i don't see any difference in the sockets and the boards continue to work great. i have also bent pins, burnt pins, and abused the boards so i don't think there is a problem with the sockets at the time.

[MikeyLikesItSI]

here is a vote for a P5WD2-P WITH a foxconn socket.

[Sentential]

so i don't think there is a problem with the sockets at the time.
That is entirely possible. I just find it odd that I have had two boards go tits up in two different cases and 4 different CPUs. All showing the same exact failure. I honestally thought it might have been a grounding issue but I have since ruled that out.

Here is what I have seen so far so that you understand why I feel this way.

Board #1 adds Sonic Tower. First try = no post. after a couple tries it boots. Ignorance / weird joo-joo blamed.

Board #1 Changes CPU to 650. Same thing first try no post. Second no post. Nothing there after no post. un-do pressure on sink and just lay it flat againt the chip with no mounting mechanism. Boots fine. Apply pressure, no POST. Remove pressure boot fine. Half way pressure, boots fine. Adds another 25% = no post. Gives up and leaves it be.

Board#1 changes to 660. No post on first try with all pressure. No post with no pressure. No post period. Board sent for RMA and sends 660 back.

Board #2 Celeron D with Sonic tower. Boots first try. Reboot, no post second try. No post 3rd. No post period. Removes pressure. No post. I thnk its a grounding issue. I remove it and use stock cooling. Posts!

Board #2 + now heavily sheilded with plastic to prevent grounding + Sonic Tower. First try No post. Second without Pressure, Posts. Adds all pressure. No post. Lets run for 30 mins @ no post and reboots. Posts. At this point I assume that it must be related to the Sonic tower some how and so i go back to my old XP120.

Board #2 Celly D + XP120. First try post. Clear CMOS. No post. Reboot, No post. Let hang for 30 mins and reboot. Post. Reboot again posts. (now to present)

I hope this makes people understand my logic.

[Overclocker550]

you had less problems when you had your a64. May be time to go back to amd, this time Venice are out, you had a winny last time

[CCUABIDExORxDIE]

Yeah right! Like that's gonna work. I've avoided all nForce 4 chip(sets) because of their unstability, unreliability and hard to overclock, yet nVidia keeps puking them out like a drunk trucker. So I'll never spend a penny on an nForce 4 MoBo.


I have no problems with my P5WD2-Premium and I'm using a heavy a$$ cooler, the CNPS7700cu which is a 900+ grams and that's over two pounds and I thighten those screws till they couldn't be screwed in no more and I don't see FOXCONN stamped anywhere in the socket. :cool:



take into consideration your cooler is low profile. its not as tall as a sonic tower. the xp120 didnt have issues cause its a low profile cooler. take a tower hsf, and there in lies the issue, possibly. thats my 2 cents.

[MikeyLikesItSI]

take into consideration your cooler is low profile. its not as tall as a sonic tower. the xp120 didnt have issues cause its a low profile cooler. take a tower hsf, and there in lies the issue, possibly. thats my 2 cents.

Simply laying the case over on its side would remove the high/low profile as a problem, then its just gravity + retention mechanism at work.

[CCUABIDExORxDIE]

who has room to lay a case flat?? and whose gonna run a system flat. and by now, the damage has been done, or has started to be done.

[MikeyLikesItSI]

who has room to lay a case flat?? and whose gonna run a system flat. and by now, the damage has been done, or has started to be done.


i meant for testing purposes, you can at least lay the case on the side just to rule out the profile part of the equation.

[speed bump]

Water cooling here or else a massively modded stock cooler.

No problems with a DFI 875p-t or MSI 915p

I believe this was maybe a known issue when LGA came out also.

[Sentential]

This is very strange. It is entirely possible that this is not related to the pressure on the sink. Tho one thing else comes to mind. Right where the screw hole is alot of FETs. Perhaps maybe some metal shavings coming from the screwing in and out of the heatsink is shorting something out?

That might explain it as the FETs are not in any sort of way on the AS8 so no metal shavings from the screw hole would contact any solder points

[Ross]

had phase change or dry ice tube strapped to them, both of which put an extremely high amount of pressure on the chip/board.
I 2nd this: I have my evap torqued so tight, I swear the board is *almost* bending...at least the 4 corners of the 1/4" thick backplate are. No problems with the socket and it's been on and off 2-3 times to swap procs. Foxconn socket.

I rarely put a mobo screw in that hole by the 8-pin P4 plug...is that the hole you are talking about? It's a pain to get since I have to put the harness in there before screwing it down (wires run behind mobo). I should also note that I am using the V2000, which is "BTX-style" and the mobo is mounted upside down, so any shavings would fall away from the fets anyway...

[fordf250]

I have had that happen lots with the mount for the silent tower,same as the sonic tower mount? The problem was the plate on the back of the MB touching some of the soldered pins grounding throught the cpu socket, the little foam pad for the plate was not good enought, now I think they come with a nylon piece.Since then I have removed and installed many cpus in my as8 and 875p-t and never had that happen, tighted some of my abortion hsf to the point where they did bend the MB with no problem, even bent a pin 5 times, same pin still no problem.

[Sentential]

I have had that happen lots with the mount for the silent tower,same as the sonic tower mount? The problem was the plate on the back of the MB touching some of the soldered pins grounding throught the cpu socket, the little foam pad for the plate was not good enought, now I think they come with a nylon piece.Since then I have removed and installed many cpus in my as8 and 875p-t and never had that happen, tighted some of my abortion hsf to the point where they did bend the MB with no problem, even bent a pin 5 times, same pin still no problem.
Hmmm Ill take another look at that. Atleast the issue of grounding isnt ruled out anymore.