Lapped core 2 duo IHS pics

Quote:

---

Originally Posted by xoqolatl

Only thing that can go wrong when lapping is when you forget to clean the copper and nickel dust

---

No, you can easily do more harm than good. Anyone who has ever been a part of a machine shop knows how terrible a hand sanding job is for making a flat surface, your eye flat out cannot see the mistakes you're making.

This isn't an "xtreme" mod, and half of the people who do it are doing it "xtremely" stupid. The last thing in the world you want to do is polish it with any metal polish, unless you like degraded performance, not that anyone ever reports or notices the worse performance, everybody's too busy posting pictures of shiney CPU's.

Could somebody put pics here lapping the IHS?!

I think will be better...

thanks

Quote:

---

Originally Posted by falqon

No, you can easily do more harm than good. Anyone who has ever been a part of a machine shop knows how terrible a hand sanding job is for making a flat surface, your eye flat out cannot see the mistakes you're making.

---

Well, sorry but I trust FCG more than you.

http://mekk.homeip.net/cpgallery/dis...e.php?pos=-324
http://mekk.homeip.net/cpgallery/dis...e.php?pos=-323
http://mekk.homeip.net/cpgallery/dis...e.php?pos=-325
http://mekk.homeip.net/cpgallery/dis...e.php?pos=-331

i attempted this with my old athlon 64 that im using and i screwed up big time, i didnt watch where my fingers where pressing and snapped off 4 pins on the processor, took me 4 hours to replace those in a way that would get the processor working again, fortunatly im safe for right now until my new computer is built but i would have rather tried this ahead of time before i do it on the good core 2 duo. I dont know if i goofed up making it even, it seems my temps have actually gone up 1-2c after lapping this thing..

heres some pics, no particular order.
http://www.lazyforums.net/uploadfiles/100_1695.jpg
http://www.lazyforums.net/uploadfiles/100_1696.jpg
http://www.lazyforums.net/uploadfiles/100_1697.jpg
http://www.lazyforums.net/uploadfiles/100_1698.jpg
http://www.lazyforums.net/uploadfiles/100_1699.jpg
http://www.lazyforums.net/uploadfiles/100_1700.jpg

note that these were takin during the process, afterwards it has a mirror finsh that is all copper but i forgot to take pics

Quote:

---

Originally Posted by xoqolatl

Well, sorry but I trust FCG more than you.

---

I can't read it through the firewall, but http://www.overclockers.com/ has an excellent article on the uselessnes of lapping even cheap heat sinks, maybe if enough of us request he'll apply the methodology to IHS's.

hmm, can't direct link to your argument?

@ Spawne32:
1. You must wait some time for AS5 to bake - AS5 instructions say "200 hours and several thermal cycles" - till you can see real effect.
2. Conroe has no pins therefore you can't bend them :) but you did good to try it first on a cheaper CPU.
Good luck lapping your Conroe!

am I the only one who can't seem to get their hs or cpu to look close to an y of these?

Quote:

---

Originally Posted by Halvie

am I the only one who can't seem to get their hs or cpu to look close to an y of these?

---

How do you proceed for your lapping? :)

OK. . .I now know why lapping is so important. Theres no way my FX-60 IHS was flat. Take a looky:

This is the station in front of the hi def TV watching CSI (Grey's Anatomy was a rerun). The pooch (my gf's Shih-Tsu) is parked at my feet.
http://i41.photobucket.com/albums/e2...rk-Log-009.jpg

After one sheet of 320 grit. You can still see the silk screening on the IHS:
http://i41.photobucket.com/albums/e2...rk-Log-012.jpg

After the 2nd sheet of 320 grit. That silk screening is still there albeit more faint this time.
http://i41.photobucket.com/albums/e2...rk-Log-013.jpg

After 3rd sheet of 320. Now we're getting somewhere!
http://i41.photobucket.com/albums/e2...rk-Log-015.jpg

After one sheet of 400 grit to polish it a bit.
http://i41.photobucket.com/albums/e2...rk-Log-016.jpg

After 500 grit. This was the finest grit I had laying about. Gotta get some 1000, 1200, and 1500 tomorrow @ Pep Boys, a local automotive store.
http://i41.photobucket.com/albums/e2...rk-Log-024.jpg
Here's the FX after some 800 grit:
http://i41.photobucket.com/albums/e2...rk-Log-025.jpg

Another 800 grit shot:
http://i41.photobucket.com/albums/e2...rk-Log-026.jpg

Now we're making some progress after the 2nd sheet of 1000 grit.
http://i41.photobucket.com/albums/e2...rk-Log-027.jpg

Here's the finished product:
http://i41.photobucket.com/albums/e2...rk-Log-028.jpg

I can see that the corners are no longer flat. Maybe I went too fast.
http://i41.photobucket.com/albums/e2...rk-Log-029.jpg

http://i41.photobucket.com/albums/e2...rk-Log-031.jpg

Lapping those really helps that much?

lol thats a much better job then what i did on mine philly, i think when i get my c2d im gona drive it up to your house and let you get to work on it :p :rofl:

Quote:

---

Lapping those really helps that much?

---

It surely helps if your IHS is not flat. Mine was like sea, but not as cool. Rather a tropical one.

Quote:

---

Originally Posted by crazyea

hmm, can't direct link to your argument?

---

Good grief, it's on the front page,

http://www.overclockers.com/articles1402/

Notice how every single pic of a processor being lapped has the edges sanded down first? That should at least be one of your first clues that it's not right.

hmmm, I see that he is speaking of waterblocks, which are traditionally more flat (surfaced better) than heatsinks, which are usually a lot flatter (surfaced better) than these cpus. Anyhow, I will quote the article you forwarded me too....

Quote:

---

What then constitutes a surface of sufficient quality?

A sufficiently flat surface is one that is not a significant impediment to heat transfer. Any number I give you must be in some way arbitrary. A flatter surface will transfer heat better no matter how slight the improvement. And, as I have made mention repeatedly before, it is impossible to make a perfect surface. The pertinent question is when to stop.

---

Basically if your cpu has a better finish than your hsf you are waisting your time, and vice versa. It is better to have them both at about the same flatness/finish.

Quote:

---

Originally Posted by falqon

Notice how every single pic of a processor being lapped has the edges sanded down first? That should at least be one of your first clues that it's not right.

---

That is entirely the point as to why we do it.

It is because the cpu is so concave that the outer edges are being sanded first. They make first contact. Once that is evened out the rest becomes sanded as well. The wen sand until the whole surface makes contact at once (ie, it is flatter than before)

Quote:

---

Originally Posted by crazyea

That is entirely the point as to why we do it.

It is because the cpu is so concave that the outer edges are being sanded first. They make first contact. Once that is evened out the rest becomes sanded as well. The wen sand until the whole surface makes contact at once (ie, it is flatter than before)

---

He means post-lapping, usually the corners are rounded post hand-lapping and I believe this has to do with the loose particulate building up towards the corners during the procedure. Whether or not falqon's crusade against hand-lapping is founded, IDK. I have never seen worse temps after my own hand-lapping, and usually if the base in question is concave or convex from the factory, drops in temps that are outside the realm of the margin of error (different ambient temps/TIM application/case cover on or off/fan config) are seen. I prefer popping lids to lapping the noob shields though. Too bad those knuckleheads are gluing the IHS to their cores now :slapass:

Nice and shiny heatspreaders guys, please post improvement of temps as well.
It might induce me to break out some sandingpaper. :D

Quote:

---

Originally Posted by Zeus

Nice and shiny heatspreaders guys, please post improvement of temps as well.
It might induce me to break out some sandingpaper. :D

---

I saw a 8c improvement in idle temps (as low as 23c at boot up with ambient at 22c) and 6c improvement loaded (Orthos prime) after lapping my FX-60. I spent a few hours lapping my new E6300 and will post pics in a while.

Quote:

---

Originally Posted by NonDecepticon

He means post-lapping

---

If he meant "post", why did he say "first"?

Very nice guys :D I'll post a pic of mine.. although its not nearly as good as some of yours :D

These look amazing guys :D

I just wish I had the balls to lap myself, I always think I'll screw it up and end up with worse temps than pre-lap. I need more self confidence!

I lap again mine 540J with 500grid, its not so good looking, but now my temps are... more than 10C DOWN!! WoW! The cap was not curve! Before 4000Mhz 1.5V was ~67C, now are 52-53C! The room temp is just with 1-2C down. Insane!

I think there's no argument against lapping a concave IHS until it's flat. This most definitely is worth it, depending on _how_ concave. So if he's arguing against that, he's simply wrong.

However, I do think it's pointless to lap the IHS until it's mirror-shiny so if that's his point I agree. Not really going to help anything. I would just go until it's flat. In the case of a concave IHS, it's fairly easy to tell - you'll generally hit copper on edges first. After that, when whole thing is copper you can quit. Doesn't matter if there are scratches or it's not mirror finished.