• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

HSF Lapping Service

M

MightyDiomed

Gawd
Joined
Apr 15, 2008
Messages
907
Hey guys, I've got an S1823 HSF, that I would like lapped due to the heat pipes not having full contact with the processor. However, I have no idea how to lap it...and hoenstly don't want to bother.

Are there any reputable boutique builders that offer such a service among other things? -.^
 
I know Jab-tech does and mabey Petra's. Lapping is a pain in the ass, it's worth the extra few bucks to have someone else do it for you, but I hear that they don't do the greatest job in the world.
 
I can do it for you. Lapped my S1283. $20 + shipping costs to and from.

For what it's worth have you taken a straight edge to the base across the heat pipes? When I lapped mine I noticed the heatpipes were already flat with the mounting base. Here are some pics:

Sharpie hash marks across pipes and base prior to lapping:


After about 20-30 passes with 320 grit:


After 20-30 passes with 800 grit:


After 20-30 passes with 1500 grit:



As you can see I didn't go for mirror finish, but that is not the goal of lapping anyways. I also wanted to be sure not to take off too much material as those heatpipes have very thin walls as it is. The pipes and mounting base are perfectly level to one another. I maybe noticed a 2-3c drop in temps after a day or two of the MX-2 setting up. It is also a good idea to have the CPU lapped as well. Here is my Q6600:



Let me know if you are interested.
 
Looks nice!

My heat pipes are pretty flat, however the silver bars in between them...not so much, heh.

I'll PM ya.
 
Eh, I've heard lapping a heatsink with the heatpipes touching the cpu can acctually hurt performance. Im not sure why.
Also, MX-2 dosen't break in. Ive used it quite a bit and have never had temps drop from the first time that I apply the paste.

Very nice job, though. Lapping is such a pain.

Edit, I was wrong about jab-tech. Its crazypcs and I think that they only lap heatsinks that you buy from them.
 
BababooeyHTJ said:
Eh, I've heard lapping a heatsink with the heatpipes touching the cpu can acctually hurt performance. Im not sure why.
Also, MX-2 dosen't break in. Ive used it quite a bit and have never had temps drop from the first time that I apply the paste.

Very nice job, though. Lapping is such a pain.

Edit, I was wrong about jab-tech. Its crazypcs and I think that they only lap heatsinks that you buy from them.
Click to expand...

From the umpteen times I've applied MX-2 I've seen a slight drop in temps after a day or two almost every time. Maybe it's just the paste spreading itself out properly with heat cycles, I don't know. I know it doesn't cure like AS5, although I noticed that if you leave a glob of MX-2 out on a piece of paper and a glob of AS5 next to it the MX-2 will be almost hard in a couple days while the AS5 is still pliable.
 
Sorry, I know tere are different methods of application. Could have something to do with it. Your right about the MX-2 it does dry up very fast. Ive never had an application for more than a couple of months. How does it hold up?

I just used AS5 on my heatsink. Ive only used it a couple of times before. How long does it take to cure fully?
 
Geeze, the Xigmatek S1283 needs to be lapped too? I just ordered a Dark Knight-S1283 after posting my thread about having to lap the TRUE.
 
BababooeyHTJ said:
Sorry, I know tere are different methods of application. Could have something to do with it. Your right about the MX-2 it does dry up very fast. Ive never had an application for more than a couple of months. How does it hold up?

I just used AS5 on my heatsink. Ive only used it a couple of times before. How long does it take to cure fully?
Click to expand...

My application method varies from heatsink to heatsink. But I generally see better temps after a day with MX-2. AS5 takes on average about a week of heat cycles to cure completely. Meaning Power on, load, power off. Do that for a week and it'll be cured.

IsLNdbOi said:
Geeze, the Xigmatek S1283 needs to be lapped too? I just ordered a Dark Knight-S1283 after posting my thread about having to lap the TRUE.
Click to expand...

Maybe it does, maybe it doesn't. Mine was pretty flat from the factory, but I lap everything anyways. Leaves no room for improvement as I like the best results possible the first time around.
 
IsLNdbOi said:
Geeze, the Xigmatek S1283 needs to be lapped too? I just ordered a Dark Knight-S1283 after posting my thread about having to lap the TRUE.
Click to expand...

No. People are just obsessive. 2-3C difference is not very useful in practical terms.
 
I worked metal all my life just about, best way to lap is a figuer eight..
Get some 1200 paper to get the mirror finish but make sure you have an even finish before you move up..

(2 sheets) 200
(1 sheet) 320 400 600 800 1200

Low cost use
200 400 800

You can use so soapy water with wet/dry paper gets clogged up...
 
skyline889 said:
You don't need a mirror finish when you lap but it doesn't look like you lapped the processor evenly. The surface should be uniform like this guy's Q6600:

http://forums.pcper.com/showthread.php?t=441410
Click to expand...

If you are referring to my processor it is flat. I've just remounted several different heatsinks, and remounted heatsinks numerous times to it over the course of a year and the finish has taken a beating. I assure you, it is flat. I see no point in touching up the finish since flat is the goal, not looks.
 
Crazy PC did my first one but I had to redo it. I got another one unlapped and after lapping the heatsink and CPU I got ~8C difference in temps. Well worth it. Both were TRUE heatsinks. I haven't seen to many Xigmateks that were bad but it couldn't hurt. Lapping to the point where everything is flat is a pain in the a$$.
 
Hrmmm... It looks to me like the base of that cooler will NEVER be totally flat. There are "grooves" at the edges of the heatpipes.

Stuff like that makes for nice hotspots.
 
Well, by the time it's made it through the heatspreader the hot spots will only be 1 or 2c hotter than the rest. Now, if you were sticking that straight on the die, there'd be some minor issues :p
 
I've seen drops in upwards of 10c from lapping both processor and heatsink. It all depends on the initial flatness of the mating surfaces. As for the gaps between the pipes and mounting base that is fixed with the proper application of TIM. You need to fill in those gaps first and use a credit card to squeegy off the excess TIM, then apply a thin line of TIM down the two inside aluminum mounting base strips in the middle. See pics:





Sure. to some people it's a waste of time. But for me it's a matter of knowing it's not in the equation from the beginning. Kind of like people who tune cars that spend all day searching for 5 horsepower. 5HP might not mean shit to you, but to them it could be that extra .001 seconds that wins the race.
 
HydroBudz said:
If you are referring to my processor it is flat. I've just remounted several different heatsinks, and remounted heatsinks numerous times to it over the course of a year and the finish has taken a beating. I assure you, it is flat. I see no point in touching up the finish since flat is the goal, not looks.
Click to expand...
Why is there nickel still showing on the edges of the four sides?
 
Because, that is how the HSF is made. Can't get rid of it, thats the only thing I hate about it.

You gotta blow dust off in there whenever you remove thermal paste to blast it out!
 
HydroBudz said:
I've seen drops in upwards of 10c from lapping both processor and heatsink. It all depends on the initial flatness of the mating surfaces. As for the gaps between the pipes and mounting base that is fixed with the proper application of TIM. You need to fill in those gaps first and use a credit card to squeegy off the excess TIM, then apply a thin line of TIM down the two inside aluminum mounting base strips in the middle. See pics:





Sure. to some people it's a waste of time. But for me it's a matter of knowing it's not in the equation from the beginning. Kind of like people who tune cars that spend all day searching for 5 horsepower. 5HP might not mean shit to you, but to them it could be that extra .001 seconds that wins the race.
Click to expand...


You do know that TIM doesn't transfer heat near as well as metal does..... sure filling in the gaps might help a bit.. but there will still be hotspots.
 
The point isn't that TIM transfers better than metal it's that TIM transfers heat much much better than air.
 
MightyDiomed said:
Because, that is how the HSF is made. Can't get rid of it, thats the only thing I hate about it.

You gotta blow dust off in there whenever you remove thermal paste to blast it out!
Click to expand...
Uhh, if this was in response to my post, I'm not talking about the beveled edge of the IHS. Obviously you can't lap that, nor would it make a difference if you did but I'm talking about the top face of the IHS, there's nickel plating showing on all four edges. A good lap job will have a completely flat face edge to edge. The difference between his and an ideal lapped surface is illustrated when you compare his Q6600 to the example I referenced.
 
Trying to get it so that last millimeter on the edges is flat can take a long time (due to the nature of Intel's workmanship) and won't produce any detectable difference. As long as the rest is flat and without noticeable roughness, then it's all good.

As for that particular heatsink base, you could try soldering in the grooves, then lap it down. If you're truly picky :p
 
Yeah but you'd have to use silver solder if you were that picky since typical solder is lead and lead doesn't conduct as well as copper. :D
 
I'd been considering lapping as a beneficial thing for this Q6600 G0 I have, and I'm using the stock Intel HSF till I make a decision on either the Arctic Freezer Pro 7 or the CoolIT Domino watercooler. After reading that post linked above where the guy at PCPer.com got pretty much a ~10C difference just from the lapping alone... hell, where is a hardware store, I need some sandpaper!!! :D

So this weekend if I can get to the Lowes down the road a bit I'll grab some sandpaper and bring this thing into a bit more flat style and see what happens. I'm sure it'll help, so...

~10C difference at idle, that's nothing to dismiss, definitely. For spending an hour or two doing some sanding, I can't imagine anyone not doing this, seriously.

Thanks for that link, skyline. ;)
 
When i first tried for 3.8ghz on my rig in sig, temps at prime small fft load were 81-81-72-72, not good, lapped my TRUE black and my cpu, passed 8 hours prime stable at 3.8ghz, temps of 68-68-62-62, which were the same temps id got on an earlier prime run at 1.4 vcore for 3.6ghz (im using just over 1.5 vcore for my current 3.8ghz clock)
 
flak-spammer said:
Yeah but you'd have to use silver solder if you were that picky since typical solder is lead and lead doesn't conduct as well as copper. :D
Click to expand...

Plus you would have to have a soldering rig that would put out enough heat to melt the solder while being able to overcome the maximum heat dissipation of the heatsink.

And "typical" solder isn't just lead.... it is lead and tin. And you can get quite a few different variations of how much lead and how much tin is in the solder.
 
I wonder how much they tested the application method of two separate lines, seems like a recipe for an air pocket if you ask me *scratches head*
 
cyclone3d said:
Plus you would have to have a soldering rig that would put out enough heat to melt the solder while being able to overcome the maximum heat dissipation of the heatsink.

And "typical" solder isn't just lead.... it is lead and tin. And you can get quite a few different variations of how much lead and how much tin is in the solder.
Click to expand...

True I was just generalizing about solder since lots of kinds solder also have flux in them. But yeah you would have to have something that could out power the passive dissipation but not get the heatsink so hot that it melts fin connections.
 
There's no nickle left on the top of my IHS except for the beveled edges. I don't know what you are seeing, but that's not nickle. Maybe a poor camera angle, something to do with the flash.

As far as my application method, here is a pic of contact patch when I removed the cooler for lapping:



As you can see there is no air pockets or anything abnormal. Perfect contact. Although I did use a tad too much on that application. :p
 
Sweeet. ;)

"Here's a nice little article about that application method."

Oh yeah, forgot I had read that article, been too long :p
 
You must log in or register to reply here.
Back
Top