i7 Lapping Mini Guide

Discussion in 'Modding Worklogs' started by Trucrymz, Nov 14, 2009.

  1. HyDrA

    HyDrA Member

    Joined:
    Jul 1, 2001
    Messages:
    3,821
    Location:
    SA
    Very true v81!

    I once drilled the HSF mounting holes on a video card to a larger diameter, and lapped the GPU itself :) It worked great afterwards!
     
  2. Frontl1ne

    Frontl1ne Member

    Joined:
    Mar 10, 2008
    Messages:
    3,886
    How do you lap a giant heatsink? :confused:
     
  3. alexc2005

    alexc2005 Member

    Joined:
    Sep 24, 2007
    Messages:
    2,311
    Location:
    Brisbane
    i liked the guide.

    Much appreciates, sadly there will always be haters out there.

    Pretty sure he isnt forcing his opinions on anyone...

    id find it very hard to not apply direct pressure, ive toyed with the idea of creating a lapping machine but it hasnt got any further than that sadly :p
     
  4. flinchy

    flinchy Member

    Joined:
    Jan 18, 2007
    Messages:
    6,355
    Location:
    Northlakes
    um no

    the inside was lapped down BEFORE the outside, hence the middle was sticking out further

    hence, it is convex.

    not sure if anyone else picked up on this earlier though.


    i also found my CPU performed about 2-3 degrees better when i had a very bad lap on it. bit of roughness was probably good for the TIM so sink into imperfections.
     
  5. Scoobth

    Scoobth Member

    Joined:
    Oct 16, 2005
    Messages:
    178
    Location:
    Adelaide
    Very carefully!

    I lapped my Thermalright Ultra 120, and basiclly you let the heatsinks weight do all the work. you hold the base and slide it back and forwads gently and rotate it 90degrees every 20 (or 30 whatever you prefer) times.

    No really downward pressure from you is needed except to keep it stable and flat.

    When you are on the very coarse paper you may put a hand on top to help keep it stable and one on the bottom and move both arms in unisen to keep it level.

    GL :p
     
  6. BeanerSA

    BeanerSA Member

    Joined:
    Aug 26, 2009
    Messages:
    3,644
    Location:
    Gateway to the Barossa
    I'm sure this has been covered on threads before, but isn't there an assumption here, that the glass is actually flat.
    I would have thought a surface plate would have been a more reliable option, but realise not everybody has access to one.
     
  7. flinchy

    flinchy Member

    Joined:
    Jan 18, 2007
    Messages:
    6,355
    Location:
    Northlakes
    did you cut the webbings of your hand on the sharp fins on the TRUE? i did

    feels bad man.

    it's also not really possible to move it gently... it's a 1kg tall bit of metal hahaha.

    ED: i used a slab of granite (well.. ceaserstone) for mine. flat as.
     
    Last edited: Nov 17, 2009
  8. OP
    OP
    Trucrymz

    Trucrymz (Banned or Deleted)

    Joined:
    Jul 16, 2008
    Messages:
    362
    Another one?? Do you guys not know what convex means? it means the centre is highest point whilst the outer edges is the lowest point. if you bother looking at the picture, you will see that the sides around all four edges have wear. As I said, the 4 was the first to wear, followed by the centre, and then the outer 'valleys'. Yes the middle is bowed in the centre, so on a whole, the heatspreader could be described as was concave with a bowed centre or whatever else you wanna call it, just dont try to tell me its convex when it so clearly is not..

    see here;

    Convex
    [​IMG]


    Not Convex
    [​IMG]


    Convex
    [​IMG]


    Not Convex
    [​IMG]

    Hopefully no one else tries to tell me its 'Convex"
     
    Last edited: Nov 17, 2009
  9. OP
    OP
    Trucrymz

    Trucrymz (Banned or Deleted)

    Joined:
    Jul 16, 2008
    Messages:
    362
    large heatsinks require more steady hands and patience to lapp. Its pretty similar, but its more difficult to not apply downwards pressure, you just want to guide and stabilise the heatsink with your hands and let the HS's weight do the work.

    I would rub using a figure 8 motion, and definitely add water or oil to lubricate the paper. Make sure you turn it 90degress often, (I turn after 10-15 reps for HS's) to minimze wearing down one side more. Also with HS's I do not aim for a really shiny finish, just a flat one, which is normally achieved at around 800-1200 grit. test with a razor.

    Also if you've got a direct touch heat pipe (Xigmatek, OCZ, etc) you have to be extra carefull not to take too much off the base as those heat pipes are pretty thin.

    I'll be lapping my waterblock soon and will post the process here, it might help you a little...
     
    Last edited: Nov 17, 2009
  10. goodguy82

    goodguy82 Member

    Joined:
    Feb 28, 2006
    Messages:
    4,855
    Location:
    Adelaide
    woooo crazy shit :Paranoid: :thumbup:
     
  11. v81

    v81 Member

    Joined:
    Jan 31, 2005
    Messages:
    666
    Location:
    SE Vic
    The discription you gave maded sense ot me.

    Don't worry about the others.
     
  12. Frontl1ne

    Frontl1ne Member

    Joined:
    Mar 10, 2008
    Messages:
    3,886
    oh shit man, no way i'm trying something like that :Paranoid:
     
  13. v81

    v81 Member

    Joined:
    Jan 31, 2005
    Messages:
    666
    Location:
    SE Vic
    Quality glass made useing the float process is a pretty damn good reference for a flat surface.

    Excepting the obvious of course, like like a large thin sheet supported at the edges it should be fine.
     
  14. Decky

    Decky Member

    Joined:
    Sep 4, 2001
    Messages:
    300
    ...and how you are going to stop copper corrosion?
     
  15. OP
    OP
    Trucrymz

    Trucrymz (Banned or Deleted)

    Joined:
    Jul 16, 2008
    Messages:
    362
    Hey thanks, I was starting to think I was nutty :tongue:, I thought I described it pretty clearly, but just cant figure how some are so adamant that it is convex...
     
    Last edited: Nov 17, 2009
  16. flinchy

    flinchy Member

    Joined:
    Jan 18, 2007
    Messages:
    6,355
    Location:
    Northlakes
    you don't know what either are then

    technically speaking it's both convex AND concave (or neither)

    you can't have concave where the middle (which is usually the 'cave' part) is actually convex compared to it's immediate surrounds

    that CPU is (was) convex, there are no two ways about it

    just because the very outer edges were raised higher.


    if it's concave, the middle has to be lower than everything surrounding it.


    however, when talking about lapping, and thermal properties of cooling that CPU, it was convex. (the centre of it was pushing the HSF away from contacting the rest of the CPU)
     
  17. OP
    OP
    Trucrymz

    Trucrymz (Banned or Deleted)

    Joined:
    Jul 16, 2008
    Messages:
    362
    Here we go again, this is starting to be funny...

    Technically, I did not say it is concave (actually I did but I corrected myself) I said;

    "the heatspreader could be described as concave with a bowed centre or whatever else you wanna call it"

    Logically, you are saying exactly the same thing that I am. that IT IS NOT CONVEX

    Basically I lapped my CPU cos I believe a perfectly flat (or close) surface will have better efficiency than, Concave, Convex or ConWhateverYouWantToCallIt type surface. And I think most will agree with me on that...
     
    Last edited: Nov 17, 2009
  18. OP
    OP
    Trucrymz

    Trucrymz (Banned or Deleted)

    Joined:
    Jul 16, 2008
    Messages:
    362
    umm... did you notice your first 3 statements all contradict each other, first its both or neither,

    then you cant have both as you cant have concave with convex n the middle

    then it definitely IS convex

    ??? make up your mind....

    As I said, I think IT DEFINITELY IS.... CONWHATEVERYOUWANTTOCALLIT.
     
    Last edited: Nov 17, 2009
  19. flinchy

    flinchy Member

    Joined:
    Jan 18, 2007
    Messages:
    6,355
    Location:
    Northlakes
    actually, the reason to lap a PCU is NOT to get it flat (will explain in a moment) but to bare the copper, as it has much much greater thermally conductive properties compared to the zinc

    the reason it's not to get it flat?

    can't remember which computer site tested it.. but they lapped their CPU/HSF down to perfection, not a single slight rise in either surface. idle temps were something like 2 degrees above ambient.

    but under load the CPU naturally becomes convex under heating, the whole middle of it bows out. this completely throws off the HSF, and stops cooling altogether. (hence why you still need TIM) (they didn't use TIM for their tests, forgot to mention that...)

    ED: at above.. yeah.. i tend to confuse myself like that sometimes... as my train of thought progresses.
     
  20. HyDrA

    HyDrA Member

    Joined:
    Jul 1, 2001
    Messages:
    3,821
    Location:
    SA
    But in the real world, people use a TIM - doesn't that kind of invalidate that test?

    And real world testing in this case I feel is more reliable, provided one can replicate the conditions eg. temperature, humidity etc.
     

Share This Page

Advertisement: