Announcement

Collapse
No announcement yet.

Socket 478 CPU.... Stuck?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    #21
    Re: Socket 478 CPU.... Stuck?

    Stevo, any of these are within 2C of eachother or less.

    Arctic Silver 3 or 5
    Arctic silver ceramique
    Coolermaster Premium
    ThermalRight OEM stuff that comes with thier HSs
    Zalman OEM that comes with thier HSs
    Shin Etsu...If you can find it

    Arctic Cooling MX-1 is the TIM fanboy favorite ATM

    That liquid metal stuff does work good, or so I've read, but it is expensive, hard to apply, next to impossible to remove, is conductive, attacks aluminum and excess will run over the board like tiny balls of mercury. Makes you wanna run out and buy some...Don't it?

    Go to Arctic Silver's home page for application instructions. I like using one BB+ sized dot in the center of the heat spreader, then just set the sink down on it, this avoids trapped air.

    NOTE: AMD heat spreaders are generally flat. Intel heat spreaders are very slightly concave. Most heat sinks are dead flat or near it.
    Bottom line, Intel chips need slightly more TIM than AMD chips.

    The last HS I bought was a ThermalRight SI-128 and it's base was .0035" concave. So I lapped it. I have no idea if you're into DIY projects but if you are, there is a pdf on the left that needs reading. http://www.easypckits.com/

    Comment


      #22
      Re: Socket 478 CPU.... Stuck?

      I used to lap most heatsinks in the the Coppermine/Athlon era, when contact was critical. A figure-8 pattern on a flat piece of fine-grit emory with water will produce a very slightly convex surface, with closest contact near the center of the HS - which usually worked well with a small amount of AS3 TIM.

      I've also lapped Coppermine die to reduce the bulk thermal resistance - It worked well for an P3/800
      which was unstable prior to lapping. This has to be done extremely carefully to avoid broken and chipped die. Practice first on a dead CPU to gain confidence and understand the process. Usually, you can remove a few hundred microns of thickness safely.

      Comment

      Working...
      X