[conspire] Fwd: (forw) Re: looking for good FAQ Websites for CPU heatsink/Fan hardware
rossbernheim at speakeasy.net
Fri Apr 4 20:38:01 PDT 2008
I did not look closely when I sent the reply first time.
Begin forwarded message:
> From: Rick Moen <rick at linuxmafia.com>
> Date: April 4, 02008 8:29:26 PM PDT
> To: Ross Bernheim <rossbernheim at speakeasy.net>
> Subject: (forw) Re: [conspire] looking for good FAQ Websites for CPU
> heatsink/Fan hardware
> I'll bet you meant to post that -- not send it in private mail.
> Forwarding it back for your convenience. (I've also rewrapped your
> text to fit it within standard 80-column lines.)
> ----- Forwarded message from Ross Bernheim
> <rossbernheim at speakeasy.net> -----
> From: Ross Bernheim <rossbernheim at speakeasy.net>
> To: Rick Moen <rick at linuxmafia.com>
> Date: Fri, 4 Apr 2008 20:06:57 -0700
> X-Mailer: Apple Mail (2.919.2)
> Subject: Re: [conspire] looking for good FAQ Websites for CPU
> heatsink/Fan hardware
> On Apr 4, 02008, at 3:16 AM, Rick Moen wrote:
>> Quoting Eric De MUND (ead-conspire at ixian.com):
>>> Speeze CPU fans are the only ones I recommended and used myself.
>>> Originally, it might even have been on the recommendation of
>>> CABALists Ross or Rick.
>> Probably Ross. I've not bought a separate CPU fan in a long while
>> for one of my own systems, anyway. I guess I'd look for one with
>> significant fins / heatsink (conductive effect) and _try_ to get a
>> with decent bearings, i.e., _not_ the usual cheap sleeve bearings.
>> aware that you need to make sure there's sufficient physical
>> (above and around the CPU) for whatever you buy.
> Not I. I only recommend that any fan have ball or roller bearings
> as it
> is the bearings that fail on over 9 out of 10 fans.
> Case fans, the larger the better as they can move more air with less
> noise and at lower rpms. CPU cooling is a number of things that have
> work together properly to remove the heat from the chip and keep it
> running reliably.
> With a CPU, the first thing is a thin coating of thermal transfer
> grease. The second is proper mounting of the heat sink to the
> CPU/thermal transfer grease. And here we run into mechanical factors
> that are contradictory. We want a large heatsink to soak up the heat
> and provide a large surface area to transfer the heat to the air. The
> weight of this heatsink places mechanical stresses on the CPU, CPU
> socket and the heatsink mount. Then we get to the fan. More air
> movement is better and ball or roller bearings only need apply on
> reliability basis.
> Generally most PC motherboards and cases are poorly designed for
> thermal mamagement. The original PC was put together by IBM in only a
> few weeks. Proper thermal design was not part of the program due to
> reduced schedule. As speeds and heat inside the case has increased,
> one has taken the time to do proper re-design of the motherboard
> and case design.
> There are now appearing some properly designed PC's that take thermal
> management into account. These are mostly the small systems such as
> Shuttle boxes where the whole system is designed by one company and
> small size means that they have to do it properly.
> I have built two of the Shuttle XPC boxes up with Ahtelon X64
> processors. The first one I did for work and it has been chugging
> nicely for the last couple of years now. I set it up as a dual boot
> system with the default to boot into Debian Linux and Windows XP SP2
> the other operating system. It is on a UPS and runs our PLM server
> is my Windows client for the PLM client software and for my client
> software for the accounting system. (I have to do Purchasing.) The
> boot lets me know when either Windows does something like updating and
> rebooting on its own or there is a failure and the system goes down
> when I am not there.
> The system was so nice I went out and bought the same hardware for
> home. The CPU heatpipe and big fan mean that it normally runs quietly
> and reliably.
> ----- End forwarded message -----
More information about the conspire