• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

lapped my IHS on a Q6600 (pics and results)

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Dont freak or anything but I but a grain of rice sized blob on and then use my finger working in circles to evenly spread it out to all the edges prefectly, leaving an almost opaque (able to see through) thickness. I then do the same for my heatsink and place a small dot of extra paste directly in the center of the cpu. Then I put it all together.
 
Well guys I lapped the IFX-14 today so it's nice and smooth + shiny with up to 2500 grit sandpaper. Checked with a plastic card if it's flat as can be.

Unfortunately the joy stops there. Temps did not improve at all. It's back to 81/81/85/85C under load for all cores respectively. This is a QX6700 at 3.33GHz 1.38V Vcore (pencil mod to make it stick).

I used's Zalman's STG-1 thermal grease. I reckon it's too liquad like though reviews seem to love it.

Idle temps for the cores are also at around 50C so there's something wrong somewhere, but I'm too tired (had to rebuild the machine a few times because it wouldn't start) to figure it out.

Others get -10C improvements.

I also checked the QX6700's IHS if it's flat and it's about identical to the lapped IFX-14, save for the shiny part.

Maybe there's not enough pressure between the IFX-14 and QX6700 IHS? Is the penny mod worth it? Seems firm, not loose.
 
@GF: you can try adding a penny or a washer, but I don't think it'll do much for you. 1.38 V is a lot really for air to cool in my opinion.
 
lil_falco said:
Dont freak or anything but I but a grain of rice sized blob on and then use my finger working in circles to evenly spread it out to all the edges prefectly, leaving an almost opaque (able to see through) thickness. I then do the same for my heatsink and place a small dot of extra paste directly in the center of the cpu. Then I put it all together.


... and then you've got too much TIM applied.

Why, oh why, does no one follow the instructions from AS? This isn't voodoo, it's science. Just follow the instructions.

I freaked out. :)
 
aaronjb said:
... and then you've got too much TIM applied.

Why, oh why, does no one follow the instructions from AS? This isn't voodoo, it's science. Just follow the instructions.

I freaked out. :)

I second this. Less is more. Try smaller and smaller amounts. It can't hurt.
 
You will know if you've applied too much goop when you see something like this after you detached the heatsink.











-
 
graysky said:
@gf: you can try adding a penny or a washer, but I don't think it'll do much for you. 1.38 V is a lot really for air to cool in my opinion.

I've added a photo now showing the lapped core, this was up to 2500 grit.

It really makes me wonder just how yourself, reinstated over several forums, and others, are getting only up to 60C-65C (mostly in the upper 50's) on each core under Prime.

I don't see how I'm getting around 80-85C and you guys get anywhere up to 30C less regardless whether it's a B3 or G0 stepping, even at higher MHz.

I had the IFX-14 on without any fans once when just checking if the PC would eventually boot-up and it's getting warm so there must be good contact.

Even if the IFX-14 is only as good as the 120U Extreme, which you guys seem to be using and getting excellent temps, then it's still a mystery why there's a delta of 30C in our results.
 

Attachments

  • DSC00451.JPG
    DSC00451.JPG
    140.3 KB · Views: 403
JamesXP said:
You might have a bad mount or a crazy hot chip?

Well hot chip, hmmm, only as hot as they're supposed to be given it's a B3 quad core, 1.38V Vcore and 3.33GHz clock speed. Even if I run it at stock 2.66GHz the idle temps are just about 50C instead of around 40C at most or 35C what other get on theirs.

Never does the IFX-14 heatsink feel hot which if the CPU was really that hot, it would cause. It's warm at best, mild.
 
JamesXP said:
If the IFX isn't hot theres a bad mount.



Are you using thermal paste? if so which one?

I have to agree with JamesXP.

If the cpu is so hot you would think that the heatsink would be getting hot as well.

Shouldn't the heat pipes at the bottom be nearly as hot as the cpu?
 
JamesXP said:
If the IFX isn't hot theres a bad mount.



Are you using thermal paste? if so which one?

Before I lapped it was I used AS5.

After lapping I used Zalman's STG-1 and painted the IHS all over.

I'm thinking whether the S775 retension bracket is somehow slightly above the IHS causing the former to make first contact with the IFX-14 thus leaving a gap however small between it and the IHS if you follow me.

I need to make some photos of me checking both the IFX-14's base an IHS is flat just for proof my eyes are not playing up.
 
dudleycpa said:
I have to agree with JamesXP.

Shouldn't the heat pipes at the bottom be nearly as hot as the cpu?

I touched them a few times as to try and get a reliable indication and the heatpipes are not hot, merely warm. I can't burn my finger unless I slip and touch the northbridge which is most surely hot.

Seems it's a bad mount. I have a feeling the S775 top bracket is responsible, even if it's only sticking out at most 1mm hmmm. I won't know before I take it apart again on Saturday depending if my P5K arrives tommorow.
 
My overclocked B3 is also at 3.3/1.38v and my load temps are about as high as yours. Under prime95 the temps will go over 80C but everything remains stable. I ran for 48 hours just to be sure (24 hours just prime95, and 24 hours also using ATi tool to heat my GPU/test power consumption).

My idle temps right now are 50-55c between the cores. This is using coretemp; speedfan reports temps ~15C lower which I consider inaccurate.

Max thermal spec for a B3 is 62C TCase which should be about ~77C from the TJunction sensors, so 80C is not really "out of the park" in terms of temps, especially in the artificial scenario of loading all 4 cores using prime.

I'm pretty sure my heatsink is installed just fine, and it is still an improvement over the temps I was getting with the stock heatsink, which I'm 100% sure was installed correctly. I used AS5 and followed the instructions to the T.

I look forward to lapping my CPU soon to see if it makes any difference.
 
GotNoRice said:
My overclocked B3 is also at 3.3/1.38v and my load temps are about as high as yours. Under prime95 the temps will go over 80C but everything remains stable. I ran for 48 hours just to be sure (24 hours just prime95, and 24 hours also using ATi tool to heat my GPU/test power consumption).

My idle temps right now are 50-55c between the cores. This is using coretemp; speedfan reports temps ~15C lower which I consider inaccurate.

Max thermal spec for a B3 is 62C TCase which should be about ~77C from the TJunction sensors, so 80C is not really "out of the park" in terms of temps, especially in the artificial scenario of loading all 4 cores using prime.

I'm pretty sure my heatsink is installed just fine, and it is still an improvement over the temps I was getting with the stock heatsink, which I'm 100% sure was installed correctly. I used AS5 and followed the instructions to the T.

I look forward to lapping my CPU soon to see if it makes any difference.

Keep us posted.
 
GotNoRice said:
Max thermal spec for a B3 is 62C TCase which should be about ~77C from the TJunction sensors, so 80C is not really "out of the park" in terms of temps, especially in the artificial scenario of loading all 4 cores using prime.

Is this true? I haven't heard anyone else say this yet. So does this mean you can run a G0 Q6600 up to ~86*, and a G0 x3210 up to ~100*?

I had assumed you wanted to keep your temps (as read by TAT, CoreTemp) below the thermal specification.

I understand that the TCase and the TJunction are 2 different areas of the chip. I guess I just assumed that the Thermal Specification was related to where the sensors are actually located.

And you're saying TCase, and TJunction run with about a ~15* delta?

Can anyone back-up or deny GotNoRice's claim? I'm very curious. :shrug:
 
GFORCE100 said:
I don't see how I'm getting around 80-85C and you guys get anywhere up to 30C less regardless whether it's a B3 or G0 stepping, even at higher MHz.

Your problem may be two fold: 1.38 V is quite to bit of heat since the heat increases with the square of the vcore. My temps are @ 1.2625V (1.232 V in windows) which is about 150 mV less than your system -- 150 mV is a lot of juice. Here is a copy/paste out of this thread I wrote about minimizing vcore:

Here is a more detailed analysis of two difference vcore settings and the temps they produce on a Q6600 @ 9x266=2.4 GHz as well as @ 9x333=3.0 GHz. The two voltages I used were 1.112 V and 1.232 V (both of these are the load voltage, the actual BIOS settings were 1.1375V and 1.2625V respectively). 2x orthos ran for 30 minutes and the temperatures were averaged over the last 10 minutes of those runs (well after they stabilized). Room temps was 75-76 °F. Notice that the difference in voltage is ONLY 0.120 V or 120 mV, but this seemingly small difference brought the load temps up by an average of 6-7 °C per core!

Code:
Run1 (9x266 @ 1.112 V), Average temps (°C): 51,52,50,50
Run2 (9x266 @ 1.232 V), Average temps (°C): 57,58,57,57
Differences (°C): +6, +6, +7, +7

Now if I add a faster FSB, they increased further:

Code:
Run3 (9x333 @ 1.232 V), Average temps (°C): 61,61,60,60
Differences from lowest voltage (°C): +10, +9, +10, +10
Differences from same voltage (°C): +4, +3, +3, +3

You're a full 10 % faster on your o/c as well as being 150 mV over mine. Did you say you lapped the chip also? If not, I would highly recommend that you do. I dropped off about 10 °C by lapping my Q6600 as well as the base of my ultra-120 x.
 
jason4207 said:
Is this true? I haven't heard anyone else say this yet. So does this mean you can run a G0 Q6600 up to ~86*, and a G0 x3210 up to ~100*?

Yeah. The x3210 is probably trying to account for rackmount enclosures, etc where temperatures could get hotter. Technically a processor should be able to sustain any temperature below the actual TJMax of the processor, at which point throttling should occur. As long as everything is stable there shouldn’t be *too* much to worry about.

I had assumed you wanted to keep your temps (as read by TAT, CoreTemp) below the thermal specification.

I understand that the TCase and the TJunction are 2 different areas of the chip. I guess I just assumed that the Thermal Specification was related to where the sensors are actually located.

And you're saying TCase, and TJunction run with about a ~15* delta?

TJunction sensors are located on each core. Quad chips don’t actually even have a TCase sensor anymore, but intel still outlines the procedure for measuring TCase temperature; it involves cutting a groove into the IHS and placing an external temperature sensor in the approx center of the IHS. See: http://www.overclockers.com/articles1312/8052.jpg

The difference between TCase sensor measurments and TJunction measurements are typically about 15C under load; it’s simply hotter on the cores than on the IHS.
 
jason4207 said:
Is this true? I haven't heard anyone else say this yet. So does this mean you can run a G0 Q6600 up to ~86*, and a G0 x3210 up to ~100*?

I had assumed you wanted to keep your temps (as read by TAT, CoreTemp) below the thermal specification.

I understand that the TCase and the TJunction are 2 different areas of the chip. I guess I just assumed that the Thermal Specification was related to where the sensors are actually located.

And you're saying TCase, and TJunction run with about a ~15* delta?

Can anyone back-up or deny GotNoRice's claim? I'm very curious. :shrug:


Read this thread, it will make it clear what the differences are between TCase and the TJunction etc
 
Back