Jump to content
Search In
  • More options...
Find results that contain...
Find results in...

Maes

Members
  • Content count

    17317
  • Joined

  • Last visited

About Maes

  • Rank
    Here's an old post I made on the subject,

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

Single Status Update

See all updates by Maes

  1. My Athlon 64 3200+ temperature was raising steadily, even with an ambient temperature of 18-19 C both the core and average CPU temp were as high as 42 C idle (too much for this type of CPU).

    So I opened the case, cleaned the CPU fan carefully (it tends to become quite grimy in a matter of months), vacuumed the cooler's fins, tried again...and there was a slight temp. drop: 40 C for CPU, 41 C for core.

    Still, I could do better. So I took out the cooler block, washed it with soap and water to remove the last bits of grime that vacuuming couldn't remove, greased up the fan's bearing with general purpose brown lithium grease, and then noticed that the CPU paste was dried and cracked, with a lot of empty zones.

    So I just applied some of the same brown lithum grease on the CPU, put it all together, fired it up and....it was a tad quieter, and CPU temp: 35 C, Core temp: 36 C. The same exact values I had with an "arctic silver-alike" product. Talk about cheat & ghetto... Grease: t3h ultim4t3 ov3rcl0x0r toolz!

    1. Show previous comments  1 more
    2. Maes

      Maes

      Well, with the original paste (that is, when I got the box in 2005) I remember that CPU temp was 32 C, but since there were many BIOS upgrades ever since that altered the temp reading and the Cool'n quiet functionality was erratic under different mobos (I used the same CPU in a different mobo for a while) I can't really tell.

      From what I can tell now, it performs just as well as a generic "silver paste" product, while being MUCH cheaper (like 1.5 Eur for 100 ml).

      I had been thinking that as long as one keeps the film thickness really minimal, the thermal conductivity of the paste itself matters little, at least within one order of magnitude, but I need to review some equations in a thermal design textbook I have before claiming something like that. In the meantime, I see that it works pretty well.

      Anyway, I'll see how well it holds up when the hot season really kicks in...

    3. AirRaid

      AirRaid

      Oh, you're right about the thickness issue. The thermal paste, whatever you use, is only there to completely seal any cracks/inperfections in the contact patch between cpu core and heatsink. Ideally the heatsink would be directly 'welded' to the cpu core, leaving no possible air gaps.

    4. Maes

      Maes

      My latest experiment seems to confirm that: because the temperature eventually rose again to 39-40, I went out and bought some Arctic Silver ceramique (no particular preference, it's all I could find). I'm still breaking it in but now the temp is at 36-37 when idle, which was what the grease gave me at first.

      The grease worked well for a few days, but then its performance declined. Nothing catastrophic, mind you, it's not like the CPU went up in smoke, it was just sub-optimal.

      However the bottom line is that you can use pretty much any kind of paste if the need should arise (at least mineral oil based pastes, not watery stuff), and running a CPU on grease for a few days won't hurt it.

      Perhaps I should try applying it again with the "arctic silver" method, e.g. slightly damping the heatsink slightly, and then applying a small ball of grease on the CPU without spreading it, and pressing the heatsink on top of it.

      Anyway...if you ever wondered whether you could use lithium/white/brown grease, vaseline, wax etc. as CPU paste...yes, you can. Your CPU won't go up in smoke or explode, and will even work as well as dedicated paste for a few days, but it won't last as long.

×