You need to apply a lot of pressure I think, how much of a gap do you have between the IHS and the PCB?
Jokester
Jokester
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
S939 A64 IHS Removal
- Thread starter Jokester
- Start date
sgoaty said:
Should do I would have thought, was this the one that was hot? Maybe the IHS isn't flat or something. There's not that much TIM on the inside to be honest.
Jokester
sgoaty said:
I would avoid a pad to be honest. Try it with AS5 and put it under something heavy without the sealant and see if it makes good contact.
Jokester
Associate
can i ask, why would you want to put this back on if it only makes the processor warmer...
Soldato
Selling the cpu...
Sending the cpu to RMA...
Protecting the core from damage possibly from a heatsink or what not...
The IHS does'nt make the cpu warmer. With the IHS on, there is a layer between the base of the cooling unit and the core. Sometimes, you can get a good contact with the IHS on the core and the temps are good. Other times, you can get the IHS having a bad contact so the heat isnt transfered properly. If this the case, you are definatly gonig to see the temps decrease as the core had direct contact with the base of the cooling unit. It may only shed off 1 degree or 2 if the IHS has good contact. It will be better to keep the cpu cooler but some might not risk it for 1 or 2 degree's.
Sending the cpu to RMA...
Protecting the core from damage possibly from a heatsink or what not...
The IHS does'nt make the cpu warmer. With the IHS on, there is a layer between the base of the cooling unit and the core. Sometimes, you can get a good contact with the IHS on the core and the temps are good. Other times, you can get the IHS having a bad contact so the heat isnt transfered properly. If this the case, you are definatly gonig to see the temps decrease as the core had direct contact with the base of the cooling unit. It may only shed off 1 degree or 2 if the IHS has good contact. It will be better to keep the cpu cooler but some might not risk it for 1 or 2 degree's.
Soldato
How well is this likely to work with a pelted x2 4200+?
Would condensation be more of a problem with the exposed chip?
Would condensation be more of a problem with the exposed chip?
Associate
all i know is, it is not happening to my X2 4400+ after taking mine off the 148 and then blowing it by accident i am not doing it to the X2, although is it worth sending the opteron in for RMA as it is still in warrenty, and wont they take the IHS oss when you send it to RMA
Are you using a cold plate between the TEC and IHS?
Jokester
Jokester
Soldato
Jokester said:
I will be when I re-assemble it all tomorrow
Cyanide said:
I used to use liquid tape over the top of small chips next to the core but as far as I can tell it not necessary. If you've got a cold plate then it might drop you temps a degree or two or more if you're lucky as the IHS is just more metal in the way between the chip and the cooler.
Jokester
Soldato
Jokester said:
There's a supply of insulating wax tape at work that we use to melt onto 33kv cables... that might do it
What sort of temps would I see without the coldplate?
All in all though, if it's only for the sake of a couple of degrees I'm not really willing to risk my chip.
Yet...
toucam said:
Ive finally got a ninja and am planiing to fit it but will wait till exams are over before doing it in case of probs. Have looked at it and i reckon just file off 1mm from the underside of the mount (the 4 points and the 2 scre holes) that comes with the ninja. Let me know if you get it working ok. Jokester mentioned that the mount may need modification.
I decided to do this on my 4000+ SD to try it before I started hacking away at my dual core CPUs (I chose the SD because it was the hottest of all my machines - it is in a scabby old case with virtually no ventilation or air movement).
Before the procedure, temps were about 36c idle and 42c load on stock.
I took great care when cutting the seal but still managed to knock one of the little chips to the side when my knife slipped. I seriously thought it was done for but I teased it back into position and it seemed to sit pretty snugly so I decided to give it a go.
Heatsink is an SI120 so the mount needed modification (file down the feet and screw guides on the bottom). I did this and got the CPU mounted, applied AS5 and clamped on the heatsink.
I got everthing back together and plugged it all in and switched on, fearing the worst.
To my joy she booted up no problem (I set everything to stock for the procedure). Unfortunately the temps were even higher than before (46c and 51c)
I surmised that this was due to insufficient contact between heatsink and CPU so I hauled it all apart and filed down the heatsink mount a little more and screwed it all back in.
I then proceeded to clean the AS5 from the heatsink and CPU. To my horror, I caught the chip with the cloth and knocked it on the floor and When I picked it up, I immediately noticed that the little chip I had moved before was now completely gone
.jpg)
It was definitely done for this time so I stopped for a cigarette and had a browse through the OcUK shop, looking at 3700+ chips.
Being an eternal optimist, I decided to chance my luck before clicking the order button so I reassembled everything with a deep sense of foreboding.
Well, to cut a long story short, I switched it on and lo and behold, the little beauty booted up perfectly Idle temp now 32c and 38c load at stock - an improvement of 4 degrees from start.
I don't know what function the little chip I knocked off had but I may find out tomorrow when I try to o/c it
Stan
Before the procedure, temps were about 36c idle and 42c load on stock.
I took great care when cutting the seal but still managed to knock one of the little chips to the side when my knife slipped. I seriously thought it was done for but I teased it back into position and it seemed to sit pretty snugly so I decided to give it a go.
Heatsink is an SI120 so the mount needed modification (file down the feet and screw guides on the bottom). I did this and got the CPU mounted, applied AS5 and clamped on the heatsink.
I got everthing back together and plugged it all in and switched on, fearing the worst.
To my joy she booted up no problem (I set everything to stock for the procedure). Unfortunately the temps were even higher than before (46c and 51c)
I surmised that this was due to insufficient contact between heatsink and CPU so I hauled it all apart and filed down the heatsink mount a little more and screwed it all back in.
I then proceeded to clean the AS5 from the heatsink and CPU. To my horror, I caught the chip with the cloth and knocked it on the floor and When I picked it up, I immediately noticed that the little chip I had moved before was now completely gone
It was definitely done for this time so I stopped for a cigarette and had a browse through the OcUK shop, looking at 3700+ chips.
Being an eternal optimist, I decided to chance my luck before clicking the order button so I reassembled everything with a deep sense of foreboding.
Well, to cut a long story short, I switched it on and lo and behold, the little beauty booted up perfectly
Idle temp now 32c and 38c load at stock - an improvement of 4 degrees from start.I don't know what function the little chip I knocked off had but I may find out tomorrow when I try to o/c it
Stan
Soldato
Cache chip?
Phnom_Penh said:
The cache is on the core, not sure what they do, but if you lose one sometimes they work, sometimes they don't.
Jokester
