lapped my q6600 (pics and temp results)

Discussion in 'Intel Processors' started by graysky, May 6, 2007.

  1. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    Well, after lapping my HS, I've had this nagging little voice in my head telling me to do the same from the CPU. I did the job with 800 grit sandpaper. Initially, I told myself I'd just buff what's there right now just to see if it's level. After about 30 laps in one direction and 30 in the other direction I discovered I had quite a concave IHS. So I just kept at it. Two 9x11 pieces of 800 grit later paper later I was left with a darn flat layer of copper looking back at me. I finished the job and put a mild shine on it with a sheet of 1000 grit I got from the local auto parts store just for the f*ck of it.

    Here are a few pics and the temp. results I got from lapping both my CPU and HS. I would recommend that anyone wanting the best $20 decrease in temps should consider lapping both the CPU and HS.

    Hardware details: Q6600 @ 9x333 and vcore of 1.2625V in the BIOS, P5B Deluxe (vdroop modded) cooled w/ an Ultra-120 Extreme (lapped) with Scythe/s-flex SFF21F 1600RPM fan, in a P182 case:

    Temp results:
    [​IMG]
    Each temp. point represents an average of data collected over approx. 1 h time period during the 2nd pass of a 2-pass x264 encode of a 720x480 DVD source using a high quality video profile. Data points were logged by Speedfan every 3-4 seconds over this time period. The average CPU usage was >99 % on all 4 cores throughout the experiments. Also room temp was between 20-22 °C.

    After about 5 minutes of lapping in each direction with 800 grit. You can see how the nickel plating has come off around the edges first which shows you just how concave this thing really was:
    [​IMG]

    After more lapping most of the nickel plating has been removed expect in the really low areas (the camera flash fired so close to the chip makes all the scratches show up much more so than they do under normal light):
    [​IMG]

    Switched to 1000 grit, here's the result:
    [​IMG]

    Another angle shows the nice dull reflection, still very so slightly concave at the extreme edges, but good enough for me:
    [​IMG]

    I would recommend that anyone wanting the best $20 decrease in temps should consider lapping both the CPU and HS.
     
  2. furyagain

    furyagain Limp Gawd

    Messages:
    297
    Joined:
    Feb 24, 2007
    i willnever lap my cpu..
    never

    i will never be able to sell it to anyone else if i do that to it..:(
     
  3. grimpy

    grimpy n00b

    Messages:
    46
    Joined:
    Apr 7, 2006
    ofcourse you can sell it just at a reduced price as you wont have any warranty with it.
     
  4. danny_discus

    danny_discus Gawd

    Messages:
    609
    Joined:
    Dec 13, 2006
    Nice work, this is real results anyone can use and benefit from. Especially enjoy the comparisons.
     
  5. Mansize_tissue

    Mansize_tissue 2[H]4U

    Messages:
    2,501
    Joined:
    Jan 1, 2007
    There won't be any warranty anyway, will there, if it's sold on? I thought that the warranty was only applicable to the person who first purchased the CPU.
     
  6. yaric

    yaric [H]Lite

    Messages:
    119
    Joined:
    Aug 9, 2006
    I would buy a used lapped Q6600... as long as they at least guaranteed me it would work on arrival and not just be a DOA.
     
  7. Zok

    Zok Limp Gawd

    Messages:
    193
    Joined:
    Sep 5, 2003
    Nice work! You've inspired me to do the same thing on my future E4400. I need all the temp decrease in a SFF I can get...
     
  8. theseeker

    theseeker 2[H]4U

    Messages:
    2,245
    Joined:
    Apr 27, 2005
    I am trying to bring myself to lapp my QX! Based upon your temps I need too.
     
  9. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    @Zok and Seeker: very cool. Just be sure you rigorously document your stock temps so you can report back to the community. Looking forward to reading each of your posts :)
     
  10. ziddey

    ziddey [H]ardness Supreme

    Messages:
    5,392
    Joined:
    Dec 24, 2002
    lol I was able to sell my water damaged e6400 just fine. lapped was actually a bonus in the sale and not frowned upon at all.

    OP, nice results!!
     
  11. theseeker

    theseeker 2[H]4U

    Messages:
    2,245
    Joined:
    Apr 27, 2005
    In the event that I lapp it, I will keep the forum posted. However, I am getting a bit tired of taking my system apart, but I am strongly considering it. I have very liitle time these days and my wife does not appreciate me working 24/7 and spending my few off hours tweaking the rig. Then again, I just bought her a very nice anniversary gift that would pay for about 7.5 rigs!
     
  12. william_fontaine

    william_fontaine Gawd

    Messages:
    737
    Joined:
    Oct 26, 2006
    wait, how can you work 24/7 and have any off hours at all? LOL

    and wow, I never thought "lapping" would have such a good effect on heat transfer.

    P.S. I still think there ought to be a better word for this than lapping. like sanding, or smoothing, or something.
     
  13. DeFex

    DeFex [H]ard|DCer of the Month - June 2011

    Messages:
    5,248
    Joined:
    Feb 13, 2001
    lapping is a machining operation. it is to make the surface flat, and also to get the desired roughness. "sanding" and "smoothing" are something else.

    the flatter and less rough the CPU and heatsink are, the more metal will be touching, and there will be better heat transfer.
     
  14. SpoogeMonkey

    SpoogeMonkey 2[H]4U

    Messages:
    2,931
    Joined:
    Feb 20, 2005
    While we're on this subject, how many people have popped the lid on the c2d's? Are the cores making better contact with the ihs than the x2's did?
     
  15. stoney_titan

    stoney_titan [H]ard|Gawd

    Messages:
    2,018
    Joined:
    Feb 19, 2007
    Wow.... The temp. decrease you had is insane. Speechless.
     
  16. jws2346

    jws2346 [H]ard|Gawd

    Messages:
    1,768
    Joined:
    Feb 28, 2007
    Although I've never been too keen on lapping my cpu heat spreader. I had some doubt if it was really necessary to lap it. The last cpu heat spreader I lapped was my "fire breathing" PD 805 (along with adding several case fans, a larger cfm fan for my BT, modifying my case again, etc), but after seeing your fantastic results I'll probably get the glass and sandpaper out and lap one of my E6600 cpu heat spreaders(I always lap the HSF) Very impressive post :eek:
     
  17. WhiteZero

    WhiteZero 2[H]4U

    Messages:
    3,638
    Joined:
    Oct 21, 2004
    Wow.. I really had my doubts about the performance gain of lapping... now I have seen the liiiiiight! lol
    Anyway, I think we can assume that, with such good results, the the heatspreadder on the top of thoes CPU's are just crappy to begin with, which is why lapping had such an awesome effect.

    I can only immage how using a pro lapping kit that has upto 2500 grit paper can imporve this.
    :D

    I've never lapped before, so I guess by the time I get a new CPU I'll be looking more into it. But when you lapped that IHS, did you take it off first or what?
     
  18. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    From what I've read, going beyond say 800 grit doesn't buy you anything (shiny surfaces don't mater. FLAT does. The IHS was not removed, just flipped the chip over.
     
  19. jws2346

    jws2346 [H]ard|Gawd

    Messages:
    1,768
    Joined:
    Feb 28, 2007
    I have to agree with graysky, the important thing is the heat spreader is level and flat. Going beyond 800 grit is probably not necessary (I'm anal, I always go to 1500 grit). I read a few Lapping Guides around the "Net" and they all agree that levelness or flatness is the primary objective. I've even heard of people polishing their chips with metal polish to make the heat spreader shine, but I was always afraid I might not clean the heat spreader off good enough or I would be handling the chip too much. No, you don't take the heat spreader off (I think they're soldered on and are hard to get off) you just flip the chip over.
    Don't forget the protective plastic thing to protect the contacts
    Don't forget to tape the paper to a piece of glass or mirror (something very flat)
    Probably the most important thing to me whenever I lap anything is take your time and use circular motions, the process takes time.
    Make sure you clean the finished lapped cpu super clean before you reinstall it. Google cpu lapping and I think you'll find some good technigues.

    Edit: Oh yeah, remember the warranty is history, which may or may not be important to you. Check your heat spreader first to make sure it's in need of lapping because it may make the chip harder to sell in the future.
     
  20. Mansize_tissue

    Mansize_tissue 2[H]4U

    Messages:
    2,501
    Joined:
    Jan 1, 2007
    jws2346, what happens if water gets on the non-IHS side of the CPU; will it be fubar-ed? Is it also okay to get rubbing alcohol on the non-IHS side when cleaning it (grit from the sandpaper might get on there)?
     
  21. jws2346

    jws2346 [H]ard|Gawd

    Messages:
    1,768
    Joined:
    Feb 28, 2007
    Hell, I don't think it would hurt it. I saw a picture of some guy soaking his cpu in a glass of gasoline or something (it scared me a little, I wouldn't recommend it). I've used a toothbrush soaked in rubbing alcohol many times. With the 775 chips (no pins) you don't even have to be careful of bending pins. (I think the 775 chips are pretty tough) Just make sure the chip is dry and clean when you reinstall it.
     
  22. blade52x

    blade52x 2[H]4U

    Messages:
    2,965
    Joined:
    Sep 9, 2006
    How do you clean a lapped C2D?

    I think I read somewhere that alcohol and copper is not a good mix - or is this not true?
     
  23. jws2346

    jws2346 [H]ard|Gawd

    Messages:
    1,768
    Joined:
    Feb 28, 2007
    Oh sh*t, I've been using wood grain alcohol on everything computer wise (except of course my fan bearings, etc) I have been using wood grain alcohol (not drinking alcohol) to clean my lapped cpus, my lapped HSF's, my lapped NB and SB heat sinks, moffet sinks, etc just about anything I wanted clean and sterile so I could replace the TIM (I do love my AS5) I even ordered a lapping kit (just some sandpaper) and the retailer recommended using alcohol and now I read where it's not a good idea, the computer Gods must like me because I've never had any problems. You asked "How do you clean a lapped C2D?" now I have no idea, probably carefully.

    Edit: Oops, I meant to say in my previous posts I used isopropyl alcohol, being "intellect challenged" I thought "wood grain" alcohol and "isopropyl" alcohol were the same things (D'oh)
     
  24. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    Wood alcohol shouldn't be a problem at all. The key is low residue and low boiling.
     
  25. WhiteZero

    WhiteZero 2[H]4U

    Messages:
    3,638
    Joined:
    Oct 21, 2004
    I got a bottle of 91% Isopropyl alcohol I like to use.
     
  26. yaric

    yaric [H]Lite

    Messages:
    119
    Joined:
    Aug 9, 2006
    Going beyond 800 likely doesn't show any difference as the thermal compound should fill everything in and make it have very good contact.

    I'm wondering what would happen if you made them absolutley perfect and polished them both then just pressed them together with quite a bit of force. I wonder which would conduct better. That or the normal lapping with thermal compound.
     
  27. JCNiest5

    JCNiest5 2[H]4U

    Messages:
    3,757
    Joined:
    Apr 25, 2005
    The best combination would be an absolutely perfect polished surface PLUS really good thermal compound.
     
  28. WhiteZero

    WhiteZero 2[H]4U

    Messages:
    3,638
    Joined:
    Oct 21, 2004
    So when you get to the point of having your IHS and your HSF lapped to the point that they stick togeather just by pressing them against eachother (I've seen videos of that)... do you still want to use thermal compound?

    I would like you ALWAYS want to use thermal compound.
     
  29. djgizmo

    djgizmo [H]ard|Gawd

    Messages:
    1,429
    Joined:
    Dec 24, 2006
    If you don't mind me asking, why circular motions? I have been taught to use back and forth motions (say for 10-20 strokes) turn it 90 degrees (1/4 turn to the right) and repeat.

    Would you say circular is more effcient or more exact?

    Thanks!
     
  30. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    Back and forth is correct... circular motions tend to cause areas of uneven pressure which you DON'T want to do. The goal if flat, not shiny. BTW, you can get it pretty shiny doing back and forth laps. Resist the urge lap in circles; it's not a chick, it's a CPU :)
     
  31. ziddey

    ziddey [H]ardness Supreme

    Messages:
    5,392
    Joined:
    Dec 24, 2002
    the helicopter?
     
  32. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    11-May: Updated the data table and added a few more pics.
     
  33. Hornswoggler

    Hornswoggler 2[H]4U

    Messages:
    2,325
    Joined:
    Jun 29, 2004
    Nice work! I really like the data and chart!
     
  34. Majin

    Majin 2[H]4U

    Messages:
    2,477
    Joined:
    Feb 4, 2005
    I know what Lapping means and I understand the reason for doing it.
    What I am not sure on is, the proper way to go about lapping a CPU or HS.
    First, what kind of sand paper should you use? I know moving form 600 grit to a finer 1000+ grit is the way to go, but is there a particular band or type of paper to use?
    Second, how do you go about lapping, do you put the paper on a flat surface and then move the CPU/HS across the paper or do you hold the CPU/HS and use a couple of fingers to press the paper to the CPU/HS?
    Third, wet lapping, obviously not for the CPU but for the HS, what’s the process?
    Forth, what’s the best way to protect the CPU when lapping, all the socket pins and keeping filings off the underside of the CPU.

    As you can see I need some input. Anyone have a guide or link they trust that can point me in the right direction. As a noobie I think I’ll do the HS only for now but I would like to learn.
     
  35. WhiteZero

    WhiteZero 2[H]4U

    Messages:
    3,638
    Joined:
    Oct 21, 2004
    I cant help but wonder what the temps would be if the IHS was lapped and not the HSF.
    o_O
    Wondering if the huge temp difference he got from lapping the IHS was the sum of it and the lapped HSF. Or would you still get the big % decrease from just the IHS lap. I wonder this because his temperature difference with just the HSF lapped wasnt all that great, but the IHS was. One could potentially skip the HSF lapping and just lap the IHS if your lazy I guess.

    But maybe you'll only get that huge temp difference if both are lapped, which would make a little more sence.
     
  36. Mansize_tissue

    Mansize_tissue 2[H]4U

    Messages:
    2,501
    Joined:
    Jan 1, 2007
    It depends on the condition of both to begin with. If the IHS is obviously very concave, or convex, then lapping it would yield large improvements. Conversely, if the HSF was already pretty flat, the improvement gained from lapping it might not be so great. That's how i assess this particular situation, anyway.

    With a Core 2 Duo, it doesn't have pins so there's not really a problem here. Nevertheless, to protect it the plastic cover which came witht he CPU would be ideal.
     
  37. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    Like MT said, all depends on how bad each one is (bad meaning convex or concave). I would recommend you start with the cheapest component: the HS. Get the technique down on it and then move on to the CPU. You'll be happy w/ both lapped anyway. Just my opinion.
     
  38. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    Buy wet/dry paper. I don't think the brand really matters. I used 3M since that's all the auto parts store had. I'd say start w/ 320-400, then 600, then 800 or 1000. The point of that is each time you decrease the particle size, you're more or less filling in the lines from the previous lapping. The 320 or 220 is used to remove the nickel and more importantly FLATEN the item. From there, it isn't about getting the surface flat, it's about getting it smooth.

    That's more or less it. You can google around for "lapping e6600" or "lapping guide q6600" etc. Read a bunch before you try it. What I did was to duct tape a thin strip (maybe about 1/4 of the 9x11 sheet) down to a FLAT piece of glass. The glass was in turn tapped down to the counter top so it wouldn't move. I put a little soapy water (literally 1 drop of dish soap in about 1 liter of water) on the sandpaper, just enough to wet it, not soaked. Then slowly and carefully, move the HS front-to-back. Don't let it bump or jump around. That can actually be a little tough at first. Don't apply much pressure at all. Just let gravity and the sandpaper (and the weight of your hand) do the work. I did about 45-60 seconds of back-and-forth, then I'd rotate the HS 90 degrees and repeat. The tedious part is cleaning off the sand paper between runs. Use water and paper towels. Also change the paper often. Once you use about 60 % of the grit or so you're pretty much wasting your time; use a fresh cutting surface.

    Once you're happy with the flatness, and the nickel layer is gone, switch to a higher grit and repeat. You can check the flatness with a marker (sharpie or whatever). Either draw some dots (a pattern along the edges and in the center) or draw an "X" or whatever on the base of the HS. Then lap 5-10 laps in one direction, rotate 90 degrees and repeat. Flip is over and see how the marker has worn off. If it comes off uniformly, that's a pretty good sign the thing is flat, if not, keep at it. Remember, do NOT try to press harder on the parts that are higher. Let gravity do it for you to get a nice even lapping. That's pretty much all there is to it. The key is slow and steady, minimal pressure, and keeping the sandpaper clean.

    Oh, I've read that you shouldn't go in circles since that tends to put uneven pressure on the HS or CPU. I did both of mine going back-and-forth exclusively.

    Wet lapping is indeed for the CPU, you just use less water and be VERY careful not to get it wet. Blot the sandpaper with a paper towel so only a small amount of liquid is on it.

    I dunno about pins: the q6600 has none. I made very sure my hands were bone dry before touching it. Doing the HS first will help you loads; there's no substitute for hands-on experience with anything. I'm sure you''ll develop a good technique.
     
  39. Majin

    Majin 2[H]4U

    Messages:
    2,477
    Joined:
    Feb 4, 2005
    Thanks alot Graysky.

    I just got a Scythe Ninja Rev B. for my Q6600 and want to get the best results from it, so I want to lap it first before putting on the Arctic Silver.
     
  40. graysky

    graysky Gawd

    Messages:
    620
    Joined:
    May 6, 2007
    I'd encourage you not to do that. Go ahead and put it on before lapping. Leave it for a couple days and record your temps doing something reproducible. I've been using x264.exe via MeGUI to do an encode of a DVD that takes just over an hour w/ the settings I used. You can use orthos or prime95, whatever maxes out all 4 cores. If you end up doing orthos, be sure you run it for the same amount of time as your "experiment" (say 1 h).

    Also, make note of the room temp before, during the after the test so you can have that logged as well. Speedfan will do a great job logging your core temps to a txt file which you can open in any spreadsheet. Look at the numbers when the test is finishes, you can see when the reach a steady state, or average them after 10 minutes of preheating, etc. Keep the data so you know how efficient your lap job has been and so you can share the results with the group.