lapped my IHS on a Q6600 (pics and results)

[GFORCE100]

Coretemp accuracy is the matter of luck.

Any OC-er friends around the neighbourhood ? Most never use them, maybe you could buy it cheap ?

Nah, I don't know anyone who has one nearby that's gathering dust in his closed. It would surely buckle under the heat load anyway but true, it would also give a comparison between the temps.

Either way, 3GHz at 68C core temp and 3.33GHz at up to 90C core temp, a 22C delta is too large logic would have it.

If only I owned a computer shop with CPU Coolers.

 

[Vagabond102]

Well, the only other thing I can think of, gforce100, is it's possible that your cpu has an IHS seperation issue. I can't be sure, but I've read somewhere that there have been a couple people that have had this issue where the core->IHS was not making proper contact.

The only thing that I can think of is to pop the top and run naked, or another CPU.

I'm really sorry for your heating issues, as I could imagine myself having one on a 300.00 part that I can't return!

 

[JamesXP]

Don't even try popping your Q6600 Top >_>

 

[treatmentx]

I forgot about this thread, lol, will post more pictures when I get home from work. I have it, just havne't posted it.

I'm getting REALLY good temps across the board after lapping... and after reading around, it seems like lapping has made a bigger impact on either A) quads or B) the g0 batches, can't really seem to conclude from either theory, but seems like the numbers are pointing in one of those two directiosn so far.

It does take a bit of work though, I sanded off some pieces of my finger (ALSO have pictures, and will put that up later lol). I think there's a method to the madness, and it's really all about how you mount your sandpaper, how you apply pressure to the chip. Even pressure is the key.

Anyone heard of flash corrosion and the possibility of it? There's a warning on the AS5 page, but I have never heard anyone having problems of it... still, I dish-soaped my water, then sanded as opposed to just water.

 

[graysky]

@GF100: you might drop the $60 on a Ultra-120 Extreme and see how that does out-of-the-box know that you can either: a) return it for a refund if the temps are the same (less a restocking fee perhaps) or b) find the temps are lower and then lap it and be happy.

Just my 2 cents

 

[Vagabond102]

Don't even try popping your Q6600 Top >_>

I wasn't really suggesting it, but I guess I expect a bit of intelligence...

 

[Mycobacteria]

Sorry for the nube questions...

1 Is the proc is made of copper or silver?

2 Is it possible to be too zelous and rub untill all the metal is gone? Has anyone done this? (pics would be good)

3 How thick is the proc and how would we measure it ie how many inches of metal have you guys shaved off? (Thoughts of the lapping the proc till its so thin that the proc colapses with the weight of the giant HSFs we use.)

4 Does lapping the proc void the warranty? Lapping the proc its not stated as a warranty issue because technically I havent over clocked it...

5 If lapping is really that great why hasnt intel lapped it for us? (saves my fingers)

 

[Vagabond102]

Sorry for the nube questions...

1 Is the proc is made of copper or silver?

2 Is it possible to be too zelous and rub untill all the metal is gone? Has anyone done this? (pics would be good)

3 How thick is the proc and how would we measure it ie how many inches of metal have you guys shaved off? (Thoughts of the lapping the proc till its so thin that the proc colapses with the weight of the giant HSFs we use.)

4 Does lapping the proc void the warranty? Lapping the proc its not stated as a warranty issue because technically I havent over clocked it...

5 If lapping is really that great why hasnt intel lapped it for us? (saves my fingers)

Proc? Many different materials

Never seen someone lap the IHS to nothing

I think you're thinking IHS, not proc...the IHS covers the cores of the cpu and is sufficiently thick. If you lap to the point of removing all material or enough for it to become weak, you've done something no one I've seen has done.

Lapping does void the warranty.

Lapping helps shave off up to roughtly 10C.

There are great sticky threads and guides about lapping. It does void the warranty. It would be very difficult, provided decent motor skills, to mess up the lapping process.

 

[treatmentx]

So here it is, to help those who are considering lapping. Pictures as promised.

600grit


1200grit, little specks are just debri.


1200grit, at 1200 it's pretty smooth already, some people stop at 800grit.


2000grit, it begins to shine.


2400grit, I was trying to show the flatness


2400grit, i aimed with as little flash as possible to see flatness.


2400grit, cleaned the debri.


2400grit, sitting in it's home. This is a EVGA 680i A1/T1 board, Apogee GT block.


2400grit right before I wiped, AS5'd and put my waterblock on.


2400grit, the chip's previous home.

 

[treatmentx]

The smoothness on my finger from lapping


A simple overclock, still toying with everything. Check out the temps at idle in 25C ambient

 

[bing]

Next time, save you finger's skin by doing this way.











-

 

[treatmentx]

So here it is, to help those who are considering lapping. Pictures as promised.

600grit


1200grit, little specks are just debri.


1200grit, at 1200 it's pretty smooth already, some people stop at 800grit.


2000grit, it begins to shine.


2400grit, I was trying to show the flatness


2400grit, i aimed with as little flash as possible to see flatness.


2400grit, cleaned the debri.


2400grit, sitting in it's home. This is a EVGA 680i A1/T1 board, Apogee GT block.


2400grit right before I wiped, AS5'd and put my waterblock on.


2400grit, the chip's previous home.

I didn't understand what you meant by tape, like double sided tape? I didn't want to "spin". I was reading about lapping, and the aerospace engineers say figure 8 or "left-right" lift "left-right". So I went with the "left-right-left."

Perhaps you meant something different, but I thought you were talking about spinning and double sided tape so I just did my way =|

 

[bing]

Oh, you got it wrong, I meant those single sided tape is use to stick the sandwitched CPU, the back cover and the styrofoam together by rolling & sticking them.

Imagine a burger, circulate and wrap it's side with a duct tape, it will stick and secure the upper bread and bottom together while isolate the meat & cheese in the middle.

Looking at your pic, I assume you're holding the CPU on their corners and caused your fingers got lapped in the process.

My suggested method is to make a better holding place to hold the cpu while lapping and also to spread the pressing force more evenly rather than on thos 4 sharp corners, and protect the cpu on "wet" sanding.

 

[treatmentx]

oh believe me, lol, I held it in everyway possible... side, corners, center... felt like my fingers were falling off. But now that I re-read what you were saying, I got it lol. Would've been saved my fingers that way lol. And it's all about the even force

 

[bing]

Glad you got it !

I wish I took a pic when I did that !

Instead of styrofoam, I used big very stiff inert rubber block with size of the cpu but with the height around 5-10 cm.

It feels like holding a cigarette's box size of pencil eraser, and the lapping process went very well instead of torturing the finger like yours !











-

 

[Mycobacteria]

I "believe" the results are skewed a 10 deg difference between lapped and normal is ALOT!

Do you guys think the results are a little exgerated?

If you had to chose what would you rather have a new zalman HSF or a lapped proc?

 

[treatmentx]

I "believe" the results are skewed a 10 deg difference between lapped and normal is ALOT!

Do you guys think the results are a little exgerated?

If you had to chose what would you rather have a new zalman HSF or a lapped proc?

lol, I never claimed 10C, I think some people have, but my own internal testing have shown 2-3C difference on idle, 3-6C on load. I have provided a range for "worst improvement" and "best improvement" for the different clocks I do.

 

[orion456]

lol, I never claimed 10C, I think some people have, but my own internal testing have shown 2-3C difference on idle, 3-6C on load. I have provided a range for "worst improvement" and "best improvement" for the different clocks I do.

It looks like around 3-6c on load. My Q6600 B3 after lapping and some burning in is down by 3c. My friend's Xeon has a drop around 8c. But he had a signficant hollow in the center. Now they both have a very thin spread of AS5 all over which I am much happier having.

 

[GFORCE100]

I lapped thw QX6700 up to 2000 grit so now both the IFX-14 and it are lapped and guess what, temps haven't changed.

The only thing that's changed is that the temp variation between the cores is now only 1-2C 90% of the time.

At 3.33GHz default Vcore I'm still getting 84/84/86/86C in Coretemp 0.95 in Prime95 25.14. Let it run for a bit longer and it's 86/86/89/89 so temps are even worse.
At 3.00GHz 1.25V Vcore I'm getting 68/68/65/65C in Coretemp 0.95 in Prime95 25.14. A 20C delta between these for just 333MHz more? That's what peltiers do, not air coolers, just underlines there's a spanner in the works.

I checked if the S775 top retension bracket is sitting higher than the IHS voiding the latter contacting with the IFX-14 but now, there's a 1mm height difference.

At this stage I'm just lost and somewhat annoyed at what the problem is. Just about everything I could have done has been leaving little else.

I would haved expected temps to now be in the 70's at most.

I reckon the IFX-14 isn't performing up to scratch though why remains a mystery.

Something somewhere is not giving up the fight. If only I had some of that paper dentists use where you place between your teeth to check their texture. It would allow checking just how well the IFX-14 contacts with the IHS which now should be almost 100% ideal.

Further to the above. It turns out for whatever reason the IFX-14 isn't contacting the IHS. I tested this with some 3M 200MP 0.6mm adhesive and stuck the pad on. Placing the IFX-14 before screwing it allowed it to stick (this could be felt well) so it was contacting there and then. However after screwing it on and turning power on, CPU temp in the BIOS was 85C!!!

So I know it's not making contact, but don't know exactly why. With STG-1 or AS5 it contacts a lot better but temps are still high as previously indicated above.

I got fed up in the end so switched back to the Titan Amanda TEC cooler.

 

[1138]

Here is my Q6600 G0.......

I used 600, 800, 1000, & 1500 Grit automotive sandpaper.

Before

600 Grit 1

600 Grit 2

600 Grit 3

600 Grit 4

DONE

Installed

My temps didn't change much, if any, after I lapped my Q6600 & HS.