Thermal Paste needed for Slot 1 CPU?


Recommended Posts

I?ve ordered a Pentium 3 550mhz (SL3F7), minus heatsink. I already have a 600mhz (SL3JM) version siting around, that never worked properly, that has attached a large passive heatsink. I had to modify this h/s, cutting off the ends, so that CPU could be seated in the slot (it still only mounted adequately, not 100% securely). Anyway, question is: if I swap that h/s onto the 550, do I need then to use any thermal paste between CPU and sink? There?s a white plastic looking patch on the h/s where it mounts to the CPU.

Link to comment
https://www.neowin.net/forum/topic/847680-thermal-paste-needed-for-slot-1-cpu/
Share on other sites

Most of the people on this board are probably not old enough to remember the P3 :-p

The white plastic thermal pad should be cleaned off and new thermal paste applied between the CPU and the heat sink. I would probably recommend using a fan, as I'm not sure a passive HS is adequate for a P3 550. I had a room with two P3 1 GHz PCs and an Athlon 733 MHz at one point and remember it being quite a hot room.

Sounds like cotton swabs and alcohol is the ideal way to clean the old paste off:

http://www.overclock.net/amd-cpus/335074-b...-paste-off.html

Hope this helps!

Very helpful, cheers! I drew a blank from a local hardware store and chemist, for those cleaners mentioned in that link. I removed the white pad from the h/s using my thumbnail and Auto Glym Engine & Machine Cleaner ?basically a ?powerful water-based biodegradable degreaser?. I?m reluctant to try that on the CPU. I?ll try again for something more appropriate tomorrow, unless you think I can get away with the Auto Glym cleaner?

I?ve got some Arctic Ceramic, that would be okay to use as thermal paste?

Perhaps the 600MHz CPU never worked well because it was using just a passive heat sink.

Incidentally, back in the P3 days, I typically just used contact without any fancy thermal pastes to attach and active heat sink/fan to a slot1 P3.

These days, the thermal paste may cost more than the CPU.

The 600 booted up fine, and I could use various programmes, but FireFox crashed if you visited various web pages. FF had to be used as the browser, so I dumped the 600.

That just sounds to me like it needs a proper heatsink/fan combo. It's designed to be 600/100 but how well does it run at 400MHz (400/66). It should be locked at 6x.

Very helpful, cheers! I drew a blank from a local hardware store and chemist, for those cleaners mentioned in that link. I removed the white pad from the h/s using my thumbnail and Auto Glym Engine & Machine Cleaner ?basically a ?powerful water-based biodegradable degreaser?. I?m reluctant to try that on the CPU. I?ll try again for something more appropriate tomorrow, unless you think I can get away with the Auto Glym cleaner?

Honestly, I would avoid things like that altogether. You want something that is completely pure and that won't leave behind any sort of residue. Ideally something like isopropyl alcohol (but not things like nail-polish remover, because that has additives), although acetone will also work extremely well provided the surfaces are metal. Heat spreaders and sinks usually are, but going from memory I think most of the slot 1 boards expose the processor directly and don't have a heat spreader, and so could potentially be damaged by acetone. Go with isopropyl alcohol, and give the heat sink a going over as well, in case the automotive cleaner left anything behind.

Provided the heat sink and processor physically touch, any HTC compound should be fine. I don't know what that particular Arctic product is, but it'll probably be fine.

Thanks for everyones advice.

That just sounds to me like it needs a proper heatsink/fan combo. It's designed to be 600/100 but how well does it run at 400MHz (400/66). It should be locked at 6x.
But why didn't the whole system crash, if you're suggesting the CPU thermally shut down? It was just FF that crashed and only on certain web pages, always the same pages. I'll see if it runs at 400.

Well the 550 isn?t working either. Just like the 600, it boots fine, and no problem with other programs, and no problem with most sites FireFox visits, apart from a few where ?Doctor Watson for Windows NT? interrupts things with: ?an application error has occurred and an application error log is being generated. FireFox exe.?

It?s not overheating, I aimed a fan at it.

According to this site, these CPU's should work with my P2B rev 1.10:

http://homepage.hispeed.ch/rscheidegger/p2...pgrade_faq.html

I was thinking maybe one last chance and get the 400 version of the Celeron 333, that it?s already running.

Edited by Deisel Weisel

Isopropyl Alcohol works well to clean

Arctic MX-2/3 is great paste.. Its like Arctic silver 5 but without the set in time

Two different processors, but the same problem.... Sounds to me like you're barking up the wrong tree. I would re-think my diagnosis if I were you.

I'm thinking he should run Memtest and test his memory for defects.

Two different processors, but the same problem.... Sounds to me like you're barking up the wrong tree. I would re-think my diagnosis if I were you.

+1

If the only issue you're seeing is Firefox crashing on certain webpages, that would not suggest a processor issue. Sounds more like a software issue, but it's hard to say without getting the full picture (system specs, OS, Firefox version, etc).

I'm thinking he should run Memtest and test his memory for defects.
The current Celeron 333 works fine. Maybe Pentium uses the memory slightly differently. Maybe I'll try that Memtest when I have time. Thanks.
+1

If the only issue you're seeing is Firefox crashing on certain webpages, that would not suggest a processor issue. Sounds more like a software issue, but it's hard to say without getting the full picture (system specs, OS, Firefox version, etc).

Okay, I'll list the spec later.
I'm thinking he should run Memtest and test his memory for defects.

Memtest is a good idea, you should aim for 10 rounds without any errors.

I built a PIII 667 based computer & found that the processor + motherboard + RAM is finicky & originally led to many strange program crashes, etc. In that particular case I figured out that I had to clock down the bus from 133mhz to something like 119mhz in order to not get any memtest errors, which itself was sneaky b/c the memtest errors would only come up on the 6th or 7th round of memtest. System runs stable now.

This has got slightly off-topic, maybe I should start another thread, but re potential RAM probs: Not had a chance to use Memtest yet, but I did try each of the 3 RAM sticks individually and they all got Doc Watson coming up, on those web pages. 2 x Kingston 128mb and 1 x Crucial 64mb. My test makes it far less likely memory probs are the cause of Doc Watson, right? Unless all 3 sticks are faulty.

Aiming a fan at a heatsink isn't the same as a proper heatsink/fan combination. Celeron 333s didn't really need active fan combos but the 550 and 600MHz P3 were designed to use an active heatsink/fan combination.

Okay Fred, thanks.

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Posts

    • Sysinternals Suite 2026.17.06 by Razvan Serea The Sysinternals Suite is a comprehensive package of advanced Windows utilities created by Mark Russinovich, who launched the Sysinternals website in 1996 to share his system tools and technical resources. This suite combines a wide range of troubleshooting and diagnostic tools, including Process Explorer, Process Monitor, Sysmon, Autoruns, ProcDump, the PsTools collection, and many others. It provides everything IT professionals and developers need to manage, monitor, and troubleshoot Windows systems and applications. The Suite bundles all of the core troubleshooting utilities along with their help files. Non-troubleshooting extras—such as the BSOD Screen Saver or NotMyFault—are excluded. In addition to the well-known tools, it also includes AccessChk, Autologon, Ctrl2Cap, DiskView, Disk Usage (DU), LogonSessions, PageDefrag, PsLogList, PsPasswd, RegMon, RootkitRevealer, TCPView, VMMap, ZoomIt, and more. Sysinternals Suite 2026.17.06 changelog: Autoruns v14.3 - This update to Autoruns, a utility for monitoring startup items, adds bug fixes and improves the command-line application autorunsc. ZoomIt v12.1 - This update to ZoomIt, a screen magnification and annotation tool, adds image backgrounds, webcam background blur and microphone noise cancellation support. Coreinfo v4.01 - This update to Coreinfo, a tool that reports processor, socket, NUMA memory, and cache topology of a system, as well as processor features supported, adds support for new processor features. DebugView v5.02 - This update to DebugView, a tool for displaying both kernel-mode and Win32 debug output, adds Ctrl-Shift-A support for selecting all output, and agent skills support for the CLI utility. LiveKd v5.64 - This update to LiveKd, a utility that allows running the kernel debugger on a live system, fixes a debugging privileges issue. ProcDump 3.5.2 for Linux - This update to ProcDump for Linux, a tool for capturing process dumps, adds .NET counters and a custom core dumper. Process Monitor v4.04 - This update to Process Monitor, a utility for observing real-time file system, Registry, and process or thread activity, adds some bug fixes Sysmon v15.21 - This update to Sysmon, an advanced host security monitoring tool, adds some bug fixes. Download: Sysinternals Suite 2026.17.06 | 168.0 MB (Freeware) Download: Sysinternals Suite for ARM64 | 15.4 MB Link: Sysinternals Suite Home Page | Screenshot Get alerted to all of our Software updates on Twitter at @NeowinSoftware
    • Given only Volume license customers and specific resellers can obtain the LTSC versions legitimately it seems likely that this has been tinkered with quite a lot!
    • Apple CEO Tim Cook confirms looming price hikes due to memory shortages by Hamid Ganji Image via Apple Memory and chip shortages have led to significant price increases for electronics over the past year, and it seems that more hikes are on the way for upcoming smartphones and computers. Apple CEO Tim Cook has confirmed that the company is planning to increase the prices of some of its products due to the ongoing memory and storage shortages. In an interview with The Wall Street Journal, Cook confirmed the looming price hikes for Apple’s future products, adding that “Unfortunately, price increases are unavoidable.” He also said the company is doing its best to “mitigate the huge increases that are being passed to us, and we’ve been trying to shield our customers from the increases, but the situation has become unsustainable.” The Apple CEO also noted that the allocation of a large portion of memory chips to AI companies has contributed to shortages in the market, resulting in lower supply at a time when demand for devices remains high. “We definitely need memory pricing and supply to return to reasonable levels for consumer products. That’s the bottom line,” Cook said. Cook also added that Apple is ready to use its vast cash reserves to help boost supply in the market because additional production capacity is needed. While he declined to specify how Apple plans to do that, he said the company will not build its own memory and storage factories despite its financial resources and silicon expertise. Cook did not provide further details on the scale of the price increases or which Apple products would be affected, though iPads and Macs could see higher prices sooner than other products. Apple’s next product launch event is scheduled for September, when the company is expected to unveil the iPhone 18 Pro, iPhone 18 Pro Max, and its first foldable iPhone. It remains unclear whether the upcoming iPhones will be affected by the price increases, but given the current memory shortage, higher prices seem increasingly likely. There is currently no clear timeline for the end of the memory shortage. Samsung, one of the world’s three largest memory chip manufacturers, recently said the shortage could persist for several more years.
    • Downloads does not equal actual usage, even less when the app is pre-installed in some Galaxy phones.
    • +1000 to this, don't understand why they added that margin around the top bar, even the close button is a PITA to click without aiming. Ofc, this is just preview and hopefully they will revert such odd UX decision before hitting final version.
  • Recent Achievements

    • One Month Later
      eurospharma62 earned a badge
      One Month Later
    • Week One Done
      With What earned a badge
      Week One Done
    • Week One Done
      Harris Gilbert earned a badge
      Week One Done
    • One Month Later
      Vincian earned a badge
      One Month Later
    • First Post
      Jocimo earned a badge
      First Post
  • Popular Contributors

    1. 1
      +primortal
      542
    2. 2
      +Edouard
      167
    3. 3
      PsYcHoKiLLa
      85
    4. 4
      Steven P.
      64
    5. 5
      ATLien_0
      64
  • Tell a friend

    Love Neowin? Tell a friend!