Results 1 to 10 of 10
  1. #1
    cOnTrOl fReAk Guest

    question about lapping?

    How do I go about lapping my SK-6 heatsink. Just switched to a Alpha Pal 8045 with a nothing 80mm fan on it and I'm about 1-2 degrees higher than my SK-6 was with a Delta 30 CFM. I am going to try a better fan on the 8045, but while it's out I thought that this would be a time to try and lap the SK-6.
    What grit of wet/dry sandpaper should I use ? Do I lap it wet or dry? About how long will it take to do a reallly good job. I have plenty of time while it's just sitting on my desk. Plus we're snowed in here in NC today.

  2. #2
    Joined
    Dec 2000
    Posts
    1,656
    Most of the SK6's I've seen have a pretty good bottom. Lapping it will get you nothing but sore arms. But if you want to spend 2 hours for maybe a .5C improvement go right ahead.
    "mmmmmm..........beer" Homer J. Simpson
    "mmmmmm...........Hofbräu Dunkel" Tek


  3. #3
    cOnTrOl fReAk Guest
    If that's the case then I think that I'll pass. I proablly spend my time lookng for a better fan for my 8045. I am stilll curious as to how I go about lapping though.

  4. #4
    Joined
    Aug 2001
    Location
    Boring place where you cross the river,Dorset,England
    Posts
    1,109
    It probably wont get you much of a temp' drop, but all the SK6's i've seen,(mine included ), have shite awful finishes on the bottom!.IMO its not worth lapping to a high polish, but its almost 'a must!' to lap to a 'matt' finish
    Abit Nf7S, T~bredB@2.3gig
    512mb 3200Twinmoss
    Radion9700
    S/B,5.1
    40gig~IBM~GPX60
    H2o cooled...

  5. #5
    Joined
    Jul 2001
    Posts
    2,115
    There are many little different ways to lap. Just make sure you use a large flat block with the sand paper so you don't make it into a smooth, but not flat surface. I agree with the matte finish plus ASII is designed to fill in any small imperfections.
    Athlon XP 2400+- Giga-Byte GA-7DXR F10a BIOS Alpha 8045 w/ L1A blowing ||| Enlight 7237 w/ 430W Antec True Power 2 Intake 7V L1A's in front 1 Exhaust L1A in back ||| 512MB Crucial DDR Registered ECC 16-8-8-2-6-2-2 ||| Sony DRU-10 ||| Lite-On 106 ||| Samsung SP1614N ||| Leadtek Ti4400 (removed fans & use Zalman FB123 w/92mm on quiet) IRQ 10 ||| Linksys LNE100TX IRQ 7 PCI#5 ||| Turtle Beach Santa Cruz IRQ 5 ||| WinXP Pro

  6. #6
    Joined
    May 2001
    Location
    Newnan GA, USA
    Posts
    5,628
    get a pice of glass bigger than than a sheet of sand paper 9x12 is good the thicker the better (I use a pice of 1/4" thickest I could find) clean it well to remove any dirt that could cause a bump in the paper set the glass on a stack of news paper to absorb all the mess u are about to make preferably on a surface that is cleanable (no bare wood)

    use wet sand paper with water the paper will last better and wont "fill" as easily thsi also keeps u from breathing all the dust and keeps the mess confined to the area that gets wet and it lubricates the whole proces making it easier

    I like to use soapy water it makes the paper stick to the glass better and keeps the paper cleaner, wet the back of the paper and stick it to the glass

    the first stage is to make a flat surface the rest is removing the scratches from that leveling step

    the selection of the first grit is a catch 22 if u use a rughf grit (80-180) it flatten faster but take longer to remove the damage from the rughf paper

    if u use a fine grit to start (200-600) it will take a long time to flatten but less time to clean up

    so what u start with depends on how bad the bottom is (and what u have/can get localy also) I started my sk6 with 200 then 400-600-800-1000-1500-2000

    try to pass the paper under the heattsink in one direstion (north south)try to keep it flat on the paper if u put a lot of pressure on one side u will put a bevel in the bottom it will also want to grab to help with all of this hold low on the sink and let the weight do most of the work

    pick it up and look often, there will be a few "low" spots that show up (they will not be sanded) when u get to the bottom of these "lows" u are done with the first step

    u should now have a perfect flat surface with a bunch of scratches in one direction move up to the next highest paper u have (400) and sand in the oppisite direction (east west) when all the north south scratches are gone u are done with that step and move to the next (600) and go north south untill all the est west scrathes are gone and so on and so forth till u get to as far as u want to go

    then clean off all that junk off the fins and bottom with soap and hot water and then remove all your finger prints from the bottom with alcohol (or they will be corosion in the morning and u will have to hit it with 2000 again) and never touch the bottom again

    I do not know how they treated the sk6 but the original surface will not tarnish the lapped surface will tarnish where ever u touch it (thsi is not a problem as the dime sixed part we are worried about will be sealed from the air by asll soon and will never tarnish


    soem say that u should not go to i mirror finish that it will accualy hurt, "u need grooves for the as to squeeze out of"

    I dont buy it copper is a much better heat conductor than arctic silver u want as much direct contact between the sink and core, well mirroed heat sink need next to no compound "almost transparent" to fill the enebitable voids and u will have no squeez problem

    after all this U may get a few C


    there is more than one way to skin a cat, If I missed something please do tell
    Last edited by RavenTai; 01-04-2002 at 02:10 AM.

  7. #7
    Joined
    Aug 2001
    Location
    Mumbai, India
    Posts
    422
    I checked my SK6 with a steel ruler and found that the base was definately concave and it had circular milling lines on the base. I lapped it to get rid of this concavity and not for a mirror finish effect. It took me 4 hours to get it flat and I ended up with a mirror finish. But the difference to the temperature is a BIG ZERO. The temperatures readings have remained the same.

    I wonder if the circular lines on the base would have had any permanent effect on the CPU's core.

    I picked up a lot of information on lapping before I started the process. I did things "by the book" but there has been not even one iddly piddly degree in temperature difference.

  8. #8
    Joined
    Aug 2001
    Location
    Boring place where you cross the river,Dorset,England
    Posts
    1,109
    I read a article somewhere once(one of the overclocking sites i think), with this dude who worked with cooling industrial diamonds!, apparently they did a test, polishing did not create as good a contact as a matt finish!!, i know it does'nt sound right,and they were diamonds, but its gospel
    Abit Nf7S, T~bredB@2.3gig
    512mb 3200Twinmoss
    Radion9700
    S/B,5.1
    40gig~IBM~GPX60
    H2o cooled...

  9. #9
    Joined
    Aug 2001
    Location
    Boring place where you cross the river,Dorset,England
    Posts
    1,109
    IMPORTANT!!!, when you have finished lapping a copper H/S, you MUST leave it for at least 24hours before applying Artic Silver!!, A/S ic CORROSIVE to copper unless it has had time for the surface to Oxidise!, oxidising occurs naturaly when it comes into contact with..,duh!, Oxigine/Air
    Abit Nf7S, T~bredB@2.3gig
    512mb 3200Twinmoss
    Radion9700
    S/B,5.1
    40gig~IBM~GPX60
    H2o cooled...

  10. #10
    Joined
    Aug 2001
    Location
    Mumbai, India
    Posts
    422
    MadDogMe

    NOW you mention this !!! I have lapped the SK6 3 times and each time I have immediately applied AS2 and attached it to my CPU.

    But being located in this ultra humid tropical peninsula city I suppose the waiting period would be shorter here.
    Last edited by subman; 01-05-2002 at 11:03 AM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •