Anyone else have this AM2 problem???

mETRo

[H]ard|Gawd
Joined
Feb 24, 2005
Messages
1,045
I had an AM2 3500+ that I replaced with an X2 3800+ today. My 3500+ would NOT come out of the socket by hand. I literally had to use force and pry it out with a pair of pliers, thus damaging the chip/wafer around the edges. There was nothing else I could do, I messed with it for over an hour.

I believe the new clamping system puts way too much force on the CPU in the socket (most of the AM2 boards I've seen are bowed due to the heatsink clamp). Mine was just stuck. The 3800+ is in there now, but I emailed AMD about it, so we'll see if they tell me to go stick it or help me out.
 
ZIF sockets have used the same clamping system for years and i've never run into this problem. maybe it's your board?
 
Maybe your socket broke. It's ZIF just as always, nothing different.
 
I know all about the ZIF sockets... I'm talking about the new AM2 lock-down tabs for the heatsink. They put an ungodly amount of force on the processor, and I could not for the life of me get the CPU out of the socket after having it in there only 2 months. The socket slides like it should when I unlock or lock the lever.

This is far from the first PC I have built. The CPU went in just fine, it just didn't want to come out.
 
the mounting screws on the old Alpha 8045 and Swiftech socket A HSFs used to clamp down really tight. i never had any issues removing the CPU from those boards or even from the AM2 setup i built recently.
 
The Gigabyte M55plus-S3G needed a crazy amount of force to get the heatsink to clamp to the motherboard. It took so much force that two of the plastic rivets that hold the CPU socket back plate to the back side of the motherboard popped out. The heatsink was then just floating around on the CPU. Glad I noticed before pushing the power button or that new CPU might have been toast.

I had to remove the heatsink and start again. It took 200 lbs of force to pop the rivets back into the motherboard and getting the heatsink clamp to lock into position took far more force than it should have. I've put together plenty of previous generation AMD systems and the force required was nothing compared to AM2. If I ever do a processor swap I hope I don't run into the same problems you did.
 
Sounds like a PICNIC error or an I D 10 T error ;)
 
I thought it was just me when I put together my first AM2 system, the MB was definatley bent in the area of the cpu socket after putting on the heatsink, (which does go on quite easily) but i have since seen 5 or 6 boards that do the same thing !!

very strange ....

but who knows... maybe it is me ;)
 
yeah i've noticed the boards bend with most through-board mounting systems. but there isn't any reason why it would cause the CPU to get stuck in the socket unless something is wrong with the socket.

think about all of the people who do watercooling and phase change setups, i know they clamp those suckers down tight and i know most of them don't have any problems.
 
I have never had a problem with any build except this one. The CPU went in fine, like normal. It just wouldn't come out of the socket unless it was pulled out. I spoke to an AMD rep this mroning and they are going to take care of it for me. The clamping system on the AM2's is kinda strange, it really does require a ton of force to lock it properly (that doesn't mean it's hard to do, it's very simple).

No, this was not a user error and I am not an idiot, I know how to build systems. How many people have actually already swapped out their AM2's so far? And how many people have Biostar motherboards? I wanna get a feel for what the real cause of it is/was.

Oh and none of my other "through the board" mounting systems have ever 'bent' the board before this one. The guy at AMD seemed genuinely concerned and was pretty helpful, so we'll see what they say when they get it.
 
The force required to install an AM2 CPU on a Gigabyte board I built up scared me. :eek:
Glad it wasn't my board!

Now I know it's not just a Gigabyte problem. I installed an AMD OEM heatpipe cooler on a socket 754 system soon after and it was a piece of cake to lock the cam & lever properly.

Solid contact is great for heat transfer but I think AMD might have gotten a little carried away on their heatsink design specs.
 
mETRo said:
I'm talking about the new AM2 lock-down tabs for the heatsink.
the am2 and s939/754 stock heatsinks are identical...

and no, i haven't had that problem.
 
it was INCREDIBLY hard to attach the heatsink to my x2 3800 (Am2)... getting it off wasn't that hard (i think I lucked out, took 2 minutes) and my waterblock just screws in where the plastic mounting around the CPU was - so it's really easy to screw in and out.
 
on my rs485m-m motherboard it was easy to get the heatsink on, on my k9nu-neo it was ridiculous... i dont like putting that much force into something that is so delicate honestly.
 
that's why the heatspreaders are there though.. makes it a lot harder to break the core, and with the backplate, the mobo won't warp like those silly lga775 boards do with the stock heatsink :(
 
(cf)Eclipse said:
the am2 and s939/754 stock heatsinks are identical...

Almost... but not 100%, at least the two that I have here on my desk aren't. They're both factory heatsinks. We'll see what AMD says.


Thanks guys for letting me know I'm not the only one who seems to think this much force on the CPU is un-necessary.
 
well, they do change heatsink and clipping pieces every once in a while.. but i have 3 amd stock heatsinks, two s939 opterons and one AM2 sempron. while they all look slightly different from each other, they all have no problem clipping into s754, s939 and AM2 stock brackets. :D

have you had a problem with removal again, or was it just that once?
 
(cf)Eclipse said:
well, they do change heatsink and clipping pieces every once in a while.. but i have 3 amd stock heatsinks, two s939 opterons and one AM2 sempron. while they all look slightly different from each other, they all have no problem clipping into s754, s939 and AM2 stock brackets. :D

have you had a problem with removal again, or was it just that once?


I haven't tried it a second time as I haven't had the need to. When I upgrade again I'll test it, but I don't want to risk damaging 2 processors if I don't have to.
 
Back
Top