Amd 965 Be Overclock questions

[caddi daddi]

that 620 is a little on the small side, i use a 920 on a 965 at 4.2 and it does well.
if you have a hard time finding arctic silver 5 radio shack has it.

 

[trents]

It might be sold under the Radio Shack name, however, and they will charge you 250% of what you would pay for it from NewEgg.

 

[xCaDillaCx937x]

It might be sold under the Radio Shack name, however, and they will charge you 250% of what you would pay for it from NewEgg.

yea i see they had it for 12.99 lol newegg like 7$.

i went ahead and set everything back to auto (stock speeds/volts). redid the paste this time i tried to spread a thin layer across the cpu. im running p95 50c so far has been my max for about 1/2hr run time. this seems high for stock speeds. i dont know, i was looking into getting a new case. maybe a new cooler. or maybe my cpu is just runs hot? i remember i think my temps were around the same with the stock cooler.

 

[caddi daddi]

for me it's a 6 hour trip to the computer store, if i need it it's there.

 

[trents]

yea i see they had it for 12.99 lol newegg like 7$.

i went ahead and set everything back to auto (stock speeds/volts). redid the paste this time i tried to spread a thin layer across the cpu. im running p95 50c so far has been my max for about 1/2hr run time. this seems high for stock speeds. i dont know, i was looking into getting a new case. maybe a new cooler. or maybe my cpu is just runs hot? i remember i think my temps were around the same with the stock cooler.

If you tried to spread the paste evenly, you already did it wrong. The right way is to put a pea-sized blob in the middle of the CPU and not try to spread it. Just clamp the block down evenly and let the pressure spread it. That way you don't introduce air bubbles into the paste layer.

Ebay is the best place to get TIM. Sometimes you can get a great price on a large tube or distressed inventory prices.

 

[xCaDillaCx937x]

If you tried to spread the paste evenly, you already did it wrong. The right way is to put a pea-sized blob in the middle of the CPU and not try to spread it. Just clamp the block down evenly and let the pressure spread it. That way you don't introduce air bubbles into the paste layer.

yea i usually do the thin line or the pea sized method but last time i did the pea sized it had the footprint that left the cpu bare in some spots. same with the line so i tried the spreading with a card. i always tighten like im changing a tire sort of. top left screw 1 turn , then bottom right , then top right and so forth till it becomes finger tight.

 

[keny]

Have you examined the block face with a ruler/flat edge to make sure its not concave in any way and the same with the CPU. Trents has said the only way to apply tim successfully and if it is leaving the CPU bare in places then you're not putting enough on, but I have ran coolers with just the centre circle part of the CPU lid covered and it has worked almost as good as a full coverage mount, so something seems suspicious as to how you're temps are so high.

 

[xCaDillaCx937x]

Have you examined the block face with a ruler/flat edge to make sure its not concave in any way and the same with the CPU. Trents has said the only way to apply tim successfully and if it is leaving the CPU bare in places then you're not putting enough on, but I have ran coolers with just the centre circle part of the CPU lid covered and it has worked almost as good as a full coverage mount, so something seems suspicious as to how you're temps are so high.

if your talking about the bottom copper plate side , i looked at it but didnt notice anything unusual. when i get my new case i will redo it all and check the plate.

 

[RGone]

The mounting screws or at least the hold-down ears on the Antec 620 would never hold up to the pressure I put on my water blocks. They just don't have enough beef honestly.

You could certainly put the pea sized blob of heat sink compound on the cpu and then position quiet accurately the block itself and knowing the back of my mobo has a spacer to keep the board from bending, i apply pressure by hand to the block (real pressure) and then remove and look at the contact pattern. If not good contact patch then you have found A problem. IF the contact now looks good then you can clean and REapply the pea sized blob on cpu and make sure you hold down on cpu the ENTIRE time you tigthten the hold-down screws and you will have more clamping pressure than just sitting the block up there and tightening the hold-down screws.

There are many things that an individual can do on his own that you will find work for you. You being the operative word here. When trying to 'teach', there is a prescribed method that most originally adhere to.

Like when I begin to use a new cpu on a block I have had for years, I know what the contact patch should look like, so I press it down by hand really hard. Then I look at the patch showing on the new unknown cpu. If okay, I clean the block and add just a half-sized pea blob of heat sink compound to the cpu and mount away. I don't get air most likely due the compound I use and it is n0t AS5 since I hate it for being capacitive at least. But if you learn to do it 'by the book', over time you can learn the working shortcuts.

I used to mount a cpu at least 7 to 10 times a week. I did not have time for a 200 hour burn-in of AS5. I learned a trick around burn-in times. Over the course of mounting so many cpus over a 4 year period, I found my own 'working' methods and had plenty of temp baselines to check out my mounting procedure from each cpu change.

Everybody hates mounting cpus. I do too. But having to do so and so much, got me better suited to do the job and do it where the mounting was not an issue.
RGone...