MMM
Results 1 to 20 of 20

Thread: AM2 Windsor IHs removal

  1. #1
    Xtreme Member
    Join Date
    Nov 2006
    Location
    Vienna
    Posts
    176

    AM2 Windsor IHs removal

    hey guys,

    i wanted to remove the IHS of my 6000+, but i didn't manage to do it. The silicon is cut, but the IHS wont get off ...



    i found this picture. Did anyone of you have some experience removing IHS on windsor CPUs? dont want to damage my 6000+ ...

    cheers

  2. #2
    Xtreme Enthusiast
    Join Date
    Jan 2007
    Location
    Stephens City,VA
    Posts
    730
    its soldered you, its not gonna come off without the core
    Asus Crosshair III
    Phenom X2 550@X4
    Cooler Master V10
    2x2gb G-SKILL DDR3
    ASUS HD 4890 Top
    Antec 850w
    WD 500gb
    WD 80gb

  3. #3
    Xtreme Cruncher
    Join Date
    Oct 2004
    Location
    Germany, near Stuttgart
    Posts
    1,076
    of course it is possible to remove the ihs without ripping the core.

    first you have to cut the silicone. then you must heat up the ihs (lighter, soldering iron) to melt the solder.

    try to pull the ihs up... dont use force, it has to pop without much force!
    E2160 @ 360x9, max 390x9 watercooled
    GA-P35-DS3
    Cellshock D9GMH @ 545mhz cl5-5-5-5-15
    Palit 7300gt 500/1000 @ 780/1600 watercooled

  4. #4
    IT Engineer in the making
    Join Date
    Jun 2003
    Location
    Former Kingdom of Bavaria
    Posts
    2,094
    Have a look at the tutorial for S775 CPUs.
    Quote from one of our professors:
    "Reality is hiding in the imaginary part."

  5. #5
    Xtreme Enthusiast
    Join Date
    Jun 2003
    Posts
    763

    hmmm

    Umm, I wouold mention, if the core really is soldered, its going to be getting a REALLY good heat conduction contact with the IHS. Much better than with plain old goop under the hood. So it really woouldn't help much going neked vs staying covered.
    GA-MA790GP-D4SH, 965BE, 3.8Ghz
    Ultra120E, 4x2G Gskill 1066
    2x 150G VRaptor raid, WD640AALS
    8800GT 735/940/1685, LG-BluRay

  6. #6
    Xtreme Cruncher
    Join Date
    Apr 2006
    Location
    Tampa
    Posts
    1,056
    Quote Originally Posted by uwackme View Post
    Umm, I wouold mention, if the core really is soldered, its going to be getting a REALLY good heat conduction contact with the IHS. Much better than with plain old goop under the hood. So it really woouldn't help much going neked vs staying covered.
    I agree, I wouldn't go through all that trouble and possibly killing your cpu. Just lap it if you want better temps.
    Workstation :
    Q6600 @ 3.2ghz/Scythe Infinity (Screw mount mod)
    GA-P35-DS3L (F7)
    4GB G.Skill @ 400mhz 4-4-4-12
    BFG Nvidia 7600gt
    WD 250GB SATAII 16MB (OS/Programs)
    Seagate 250GB SATAII 16MB (Storage)
    NEC DVD+/-RW
    Antec NeoHE 550W
    X-Fi Xtreme Music
    Razer Copperhead
    MS Natural Ergonomic 4000 (Best KB ever!)
    Windows 2008 Enterprise Server x64

    Laptop:
    Dell Latitude D610 - Pentium M 750 1.86Ghz / 1GB RAM

  7. #7
    Love and Peace!
    Join Date
    Dec 2004
    Location
    hiding somewhere!
    Posts
    3,675
    Quote Originally Posted by uwackme View Post
    Umm, I wouold mention, if the core really is soldered, its going to be getting a REALLY good heat conduction contact with the IHS. Much better than with plain old goop under the hood. So it really woouldn't help much going neked vs staying covered.
    not necessarily. the solder layer could be too thick, or the ihs could be slightly crooked, or not flat on the bottom. there are way too many variables to be able to say "it's soldered, therefore it must have good thermal contact"
    Got a fan over those memory sticks? No? Well get to it before you kill them

  8. #8
    IT Engineer in the making
    Join Date
    Jun 2003
    Location
    Former Kingdom of Bavaria
    Posts
    2,094
    Quote Originally Posted by ozzimark View Post
    not necessarily. the solder layer could be too thick, or the ihs could be slightly crooked, or not flat on the bottom. there are way too many variables to be able to say "it's soldered, therefore it must have good thermal contact"
    I fully agree with that statement. In theory, the solder should give you a sufficient thermal junction, but as ozzimark already said, in practice there are way more variables that count.
    Quote from one of our professors:
    "Reality is hiding in the imaginary part."

  9. #9
    Xtreme Cruncher
    Join Date
    Apr 2006
    Location
    Tampa
    Posts
    1,056
    Yeah but still is it worth removing the IHS and possibly ruining your perfectly working CPU?
    Workstation :
    Q6600 @ 3.2ghz/Scythe Infinity (Screw mount mod)
    GA-P35-DS3L (F7)
    4GB G.Skill @ 400mhz 4-4-4-12
    BFG Nvidia 7600gt
    WD 250GB SATAII 16MB (OS/Programs)
    Seagate 250GB SATAII 16MB (Storage)
    NEC DVD+/-RW
    Antec NeoHE 550W
    X-Fi Xtreme Music
    Razer Copperhead
    MS Natural Ergonomic 4000 (Best KB ever!)
    Windows 2008 Enterprise Server x64

    Laptop:
    Dell Latitude D610 - Pentium M 750 1.86Ghz / 1GB RAM

  10. #10
    IT Engineer in the making
    Join Date
    Jun 2003
    Location
    Former Kingdom of Bavaria
    Posts
    2,094
    AFAIK, we never forced anyone to remove the IHS. It's completely up to you to decide whether it's worth it or not.

    The question was: Is it possible and if yes, how. We told the threadstarter how and that he could benefit from better temps afterwards. There will never be a gurantee, though. And I think I don't need to mention that there will always be the risk of damaging your CPU or even your motherboard (indirectly).

    BUT, where are we here? Is this XS or am I missing something?
    Quote from one of our professors:
    "Reality is hiding in the imaginary part."

  11. #11
    I am Xtreme
    Join Date
    Apr 2005
    Location
    Upstate, NY
    Posts
    5,425
    Brisbane 3600+'s aren't soldered, but windsors are.. odd.
    Core i3-550 Clarkdale @ 4.2GHz, 1.36v (Corsair A50 HS/F) LinX Stable
    MSI H55-GD65 Motherboard
    G.Skill 4GBRL DDR3-1600 @ 1755, CL9, 1.55v
    Sapphire Radeon 5750 1GB
    Samsung F4 320GB - WD Green 1TB
    Xigmatek Utgard Case - Corsair VX550

  12. #12
    Xtreme Enthusiast
    Join Date
    Jun 2003
    Posts
    763

    hmmm

    That's the real question. Are the Windsor's soldered?

    My chip is a real gem, I would love to go neked but not if it is soldered. Not worth the bother.

    Even if you get the lid off, there is now solder on the core surface, you need to deal with "leveling" or removing it, and this isn't childs play.

    As to the other points, sure.... hell even the core could itself be installed on the substrate crooked... and there isn't any way to fix that.

    Best solution to crooked anything is alter your HSF mounting mechanism to handle any crookedness. Conform to the actual level of the IHS or core, not some "virtual" flat ideal chip. Spring loaded hold downs, etc do just that, take up the slack for any non-level manufacturing defect/tolerance on the chip itself.

    Ill hold off even finding out until more people delve into Windsors and determine if they are soldered.

    Best solution would be the Micro-torch from RadioShack, just heat the center of the IHS after cutting the silicon COMPLETELY for sure. The core/IHS should seperate with almost NO pressure once the solder melts.

    Weird why AMD would bother, not like they have a rash of millions of IHS pullers out their. Sounds more likely they would do it to improve cooling.
    GA-MA790GP-D4SH, 965BE, 3.8Ghz
    Ultra120E, 4x2G Gskill 1066
    2x 150G VRaptor raid, WD640AALS
    8800GT 735/940/1685, LG-BluRay

  13. #13
    Banned
    Join Date
    Jul 2004
    Location
    Scotland
    Posts
    926
    I have that cpu, they claim if it says 2005 on it, its soldered.

    I read the solder melts at around 100C which is nothing and it can be removed by your nail from core as its soft.

    I am thinking put the sponge on pins that came with CPU to stop bent pins, get some kinda toolmakers clamp/vice, pop complete CPU in a Oven at around 50C for 10mins to get warm uniformly like running in a PC on load, then take out oven and heat with something like a shielded blowtorch or similar like paint remover gun again with heatshield as not too hot, and as the IHS gets hot and solder melts the weight of rest of CPU will let it fall onto worktop.

    Obv cut your silicon 1st but I left that put as obv, test it with a thin piece of thread to make sure its got a gap all round and thread wont damage smd's.

  14. #14
    Xtreme Enthusiast
    Join Date
    Mar 2005
    Location
    Australia
    Posts
    512
    Thats interesting since there are 939 opterons made recently and they aren't soldered (mines a 0704 and it only had grey goop, and i've seen much more recent ones with goop as well). . The yellow square on the bottom of the IHS in the OPs post suggests thermal epoxy rather then soldering to me :S
    Opteron 170 @ 3.0Ghz on air, Opteron 144 @ 3.1Ghz on air

  15. #15
    Banned
    Join Date
    Jul 2004
    Location
    Scotland
    Posts
    926
    Does your IHS say 2001 or 2005 on it ?

  16. #16
    Xtreme Member
    Join Date
    Nov 2006
    Location
    Vienna
    Posts
    176
    I think I'll try it this was today:

    http://www.forumdeluxx.de/forum/show...48&postcount=1

  17. #17
    Xtreme Guru
    Join Date
    May 2007
    Location
    Ace Deuce, Michigan
    Posts
    3,955
    Quote Originally Posted by ozzimark View Post
    not necessarily. the solder layer could be too thick, or the ihs could be slightly crooked, or not flat on the bottom. there are way too many variables to be able to say "it's soldered, therefore it must have good thermal contact"
    True, but if the solder contact is good, then all you need to do is lap the ihs and you'll be good to go. It won't be the same as removing the ihs (as the heat builds up under the ihs and with it off the core gets a bit of airflow, which I find helps with the temps a tiny bit), but it'll still give you a nice temp drop, especially if you lap your ihs too
    Quote Originally Posted by Hans de Vries View Post

    JF-AMD posting: IPC increases!!!!!!! How many times did I tell you!!!

    terrace215 post: IPC decreases, The more I post the more it decreases.
    terrace215 post: IPC decreases, The more I post the more it decreases.
    terrace215 post: IPC decreases, The more I post the more it decreases.
    .....}
    until (interrupt by Movieman)


    Regards, Hans

  18. #18
    Xtreme Member
    Join Date
    Nov 2006
    Location
    Vienna
    Posts
    176
    Tried to remove it, but it didn't work out. The IHs was very hot, i almost burned my fingers ... but it didnt want to come off.

  19. #19
    IT Engineer in the making
    Join Date
    Jun 2003
    Location
    Former Kingdom of Bavaria
    Posts
    2,094
    Then it was either not hot enough (you know that for you, probably 60°C feel freakin' hot, while the solder will at least need ~85°C to melt) or the silicon wasn't fully cut through.
    Quote from one of our professors:
    "Reality is hiding in the imaginary part."

  20. #20
    Xtreme Guru
    Join Date
    May 2007
    Location
    Ace Deuce, Michigan
    Posts
    3,955
    I agree with celemine, you didn't heat it up enough. What you should do is make a brace for the cpu, then just heat it quickly to 85-90c~ (if you do it slowly, there's the chance the solder that holds the core in will melt too) and let the ihs pop off. You should have to pry it off, that'll destroy the chip
    Quote Originally Posted by Hans de Vries View Post

    JF-AMD posting: IPC increases!!!!!!! How many times did I tell you!!!

    terrace215 post: IPC decreases, The more I post the more it decreases.
    terrace215 post: IPC decreases, The more I post the more it decreases.
    terrace215 post: IPC decreases, The more I post the more it decreases.
    .....}
    until (interrupt by Movieman)


    Regards, Hans

Bookmarks

Bookmarks

Posting Permissions

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