A computer components & hardware forum. HardwareBanter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » HardwareBanter forum » Processors » Overclocking
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

Tentative experiments



 
 
Thread Tools Display Modes
  #1  
Old July 29th 03, 10:44 AM
David Maynard
external usenet poster
 
Posts: n/a
Default Tentative experiments

Stephen Howard wrote:
Any learned bods care to give an opinion?

My old Chaintech 6ATA2 board with its Celeron 466 chip are pretty much
nearing the end of their days ( well, they work fine, but you know what I
mean! ), and during a regular dust of the internals I decided to have a bash
at overclocking the CPU and quietening the machine down.

To this end I set the FSB to 75Mhz, resulting in the chip showing 525Mhz at
boot.

I then spun up a few fans I had lying about and found that a 24v power supply
fan run at 12V is very quiet indeed. So I bunged it in the power supply, and
orientated it to exhaust ( as normal ).

For good measure I bunged another 24v fan in the rear of the case, and a
larger 12v at the front - and made them switchable ( in case they were needed
).

MbProbe reports the CPU running at 32 degrees centigrade when running
Photoshop and Dreamweaver ( which is about as stressed as this machine gets
), and in two weeks running 9 hours a day there have been no problems with
only the PSU and CPU fans operating ( and it's sooo much quieter ). MbProbe
reports the Moboard as at least two degrees cooler than the CPU, and the case
ten degrees.

Can this be too good to last??


Not necessarily but there are a few risks in there. For one, 12 volts is below
the guaranteed start voltage for a 24 volt fan so there is a chance that, some
day, one of them just won't spin up when you turn it on. That's not so bad for
the case fan as you can monitor case (or CPU) temp, and you have a second one,
but it could be a serious problem if the one in the PSU fails to start.

Btw, reverse the front and rear fans. It's better to have the stronger one in
the rear plus front noise is easier to hear than rear noise so it'll sound a bit
quieter if the noisy one is in the rear.

The second risk is you've lowered the PSU airflow so it's not going to cool as
well. You're probably getting away with it because it's not operating at full
load but, since you aren't monitoring PSU temp, you don't really know.

It's safer to use the 12 volt fans but, for quiet, on 7 volts (fan white molex
plug ground [black wire] rewired to +5 [red]) as the typical guaranteed start
voltage is 7. Although, for the PSU, I'd stay with the stock fan unless you can
check temps to ensure you're not cooking it.


  #2  
Old July 29th 03, 01:33 PM
Stephen Howard
external usenet poster
 
Posts: n/a
Default

On Tue, 29 Jul 2003 04:44:43 -0500, David Maynard
wrote:


Not necessarily but there are a few risks in there. For one, 12 volts is below
the guaranteed start voltage for a 24 volt fan so there is a chance that, some
day, one of them just won't spin up when you turn it on. That's not so bad for
the case fan as you can monitor case (or CPU) temp, and you have a second one,
but it could be a serious problem if the one in the PSU fails to start.


I noted that problem with one of the 24v fans - it refused to go
without a bit of a nudge, and was cast aside in disgust.

The current PSU fan ( 24v ), although quieter, makes a sufficiently
audible whirr, which is enough to detect if it's operating - but I
appreciate the warning. Will have a think about what I can do about
that.

Btw, reverse the front and rear fans. It's better to have the stronger one in
the rear plus front noise is easier to hear than rear noise so it'll sound a bit
quieter if the noisy one is in the rear.


I haven't needed to run the case fans, not according to the temp.
gauge, but I'll bear that advice in mind when I next open the case up.

The second risk is you've lowered the PSU airflow so it's not going to cool as
well. You're probably getting away with it because it's not operating at full
load but, since you aren't monitoring PSU temp, you don't really know.


An excellent point.
What d'you reckon of the feasibility of bolting the large ( 6 inch? )
12v fan to the underside of the PSU? I'd have to block the side vents
I imagine.
This would obviously blow case air through the PSU, but as the fan was
held inside the case it should prove quieter ( plus it runs quieter
than the standard 12v PSU fan anyway ).

Would I be right in assuming that specially built quiet PSUs are
fitted with slower fans, and built to work at higher operating
temperatures?

It's safer to use the 12 volt fans but, for quiet, on 7 volts (fan white molex
plug ground [black wire] rewired to +5 [red]) as the typical guaranteed start
voltage is 7. Although, for the PSU, I'd stay with the stock fan unless you can
check temps to ensure you're not cooking it.


That sounds ideal - but let me get this clear.... I leave the positive
feed as is, remove the negative feed and rewire it to a positive 5v
supply?
The difference across the terminals would be what powers the fan at 7
volts?

Might all sound a bit fussy, but other than the computer my working
environment is completely silent - so any acoustic improvements I can
make would be a bonus.


Many thanks,



--
Stephen Howard - Woodwind repairs & period restorations
www.shwoodwind.co.uk
Emails to: showard{whoisat}shwoodwind{dot}co{dot}uk
  #3  
Old July 30th 03, 12:09 PM
Stephen Howard
external usenet poster
 
Posts: n/a
Default

On Tue, 29 Jul 2003 19:07:01 -0500, David Maynard
wrote:

Stephen Howard wrote:



The current PSU fan ( 24v ), although quieter, makes a sufficiently
audible whirr, which is enough to detect if it's operating - but I
appreciate the warning. Will have a think about what I can do about
that.


Also note that it's starting might get more problematic as the fan ages and
contamination, I.E. dust, could affect it more.


Another good point - my workshop can get very dusty at times, hence my
need to regularly clean out the 'puter.


PSU fans are pulling case air anyway but the problem is in understanding the
airflow inside the PSU. I.E. They don't just slap vents at random on them;
they're designed to create the proper airflow through the various components
with the stock fan and just how well something else tacked on would work isn't
immediately obvious.

I considered that - the PSU has internal vents on the bottom and on
the side opposite the fan. I figured that a fan of sufficient 'oomph'
would drive air right round the box if there was only one exit point.
Of course, that depends whether incoming air from below has access to
the top of the circuit board - I could even be better off bunging the
fan on the side.

Having said that, I have a number of PSUs knocking about, and quite a
few of them only have vents on the side opposite the fan, which
implies that cooling over the top of the circuitry is more important
than below ( makes sense I guess ).

Would I be right in assuming that specially built quiet PSUs are
fitted with slower fans, and built to work at higher operating
temperatures?


There are a number of possibilities. One is, as you suggest, designing for
higher temps but then one would potentially be using different components to
accommodate the higher temps which, as but one example, might be a more
efficient heatsink so that even though the overall PSU temp is higher the actual
component temp isn't. Another trick is the 'sometimes noisy' temperature
controlled fan approach. It runs a quiet low speed, unless you use the power
where it speeds up to keep it cool.


The controlled fan was what made me wonder just what these quieter
PSUs actually contain. Seems you pay a premium for them, and if it's
just a bit of jiggery-pokery with a fan..well...

That sounds ideal - but let me get this clear.... I leave the positive
feed as is, remove the negative feed and rewire it to a positive 5v
supply?
The difference across the terminals would be what powers the fan at 7
volts?


Correct.

Btw, that isn't a 'proper' engineering solution either as it doesn't do anything
to ensure the 7 volt is really 7 volt. I.E. the 12 volt rail could be running a
bit low, within spec, and the 5 volt could be running a bit high, within spec,
and you wouldn't have 7 volts on the fan so *it* would be 'out of spec' but, in
practice, the odds are better than starting with one 2 volts under spec to begin
with (the 24 volt on 12 [which, as noted, could be running low]).

snip

Thanks for the further clarification - I can see the sense behind it.

On my athlon I ducted the CPU heatsink to a single case fan in the rear which
does double duty as case and CPU cooling. It works quite well, especially with
the fan on 7 volt quiet mode. My next step was to duct the CPU to the bottom fan
of the dual fan PSU (where the real noise now is) to cut another fan out but I
haven't gotten around to that yet as the present solution is 'good enough'.

If you want to mess around with PSU fans (like you hinted at above) you could
consider using a large, slow RPM, fan in-between the CPU heatsink and PSU.

snip
Depends on just how much work you want to expend on it. Rewiring a few
connectors doesn't take much.


Sounds promising, thanks.

There is one other possibility for me. I've toyed with the idea of
mounting a fan outside the workshop and ducting in fresh air.
This would mean I could use a mains powered drum fan - which would be
very quiet ( as well as being outside the building ), but it would
mean the 'puter would have to be fitted with ducts. Not too big a
problem, it never gets moved and the existing fans could be made
switchable.

I would assume that the best arrangement would be to pump fresh air in
at the bottom rear of the case and have it vent through the PSU as
normal - and the mains powered fan could take a filter too ( wouldn't
want any stray birds/cats/children caught up in the 'puter ).

Might sound like overkill for the current setup, but I've got my eye
on a new board and chip, which will no doubt require much more
efficient cooling.

Cheers,



--
Stephen Howard - Woodwind repairs & period restorations
www.shwoodwind.co.uk
Emails to: showard{whoisat}shwoodwind{dot}co{dot}uk
  #4  
Old August 1st 03, 11:18 AM
Stephen Howard
external usenet poster
 
Posts: n/a
Default

On Thu, 31 Jul 2003 21:47:37 -0500, David Maynard
wrote:


Sure, you could do that. Anything that moves the air appropriately through the
things in need of cooling will work but a bottom front entry is better as that
moves air from front to back and should cool everything in between whereas
bottom rear will tend to hug the rear on it's way to the PSU.

Ah, that would prove to be a bit messy - unless I knocked up some
internal ducting....

I beginning to think I might be better off with a laptop - but where's
the fun in that eh??

Regards,



--
Stephen Howard - Woodwind repairs & period restorations
www.shwoodwind.co.uk
Emails to: showard{whoisat}shwoodwind{dot}co{dot}uk
  #5  
Old August 3rd 03, 01:45 AM
David Maynard
external usenet poster
 
Posts: n/a
Default

Stephen Howard wrote:
On Thu, 31 Jul 2003 21:47:37 -0500, David Maynard
wrote:



Sure, you could do that. Anything that moves the air appropriately through the
things in need of cooling will work but a bottom front entry is better as that
moves air from front to back and should cool everything in between whereas
bottom rear will tend to hug the rear on it's way to the PSU.


Ah, that would prove to be a bit messy - unless I knocked up some
internal ducting....


Might not be as hard as it seems, depending on the case. I mean, if you HAVE a
'bottom rear' entry (wondered about that when you said it) you just need to
baffle it on in a bit. Ever see that clear plastic case cooling addon (I forget
the name) that goes on the bottom front fan inlet and just ducts air to the
bottom middle of the case? Same thing in reverse.


I beginning to think I might be better off with a laptop - but where's
the fun in that eh??


Hehe. Welcome to the wonderful world of engineering.

Frankly, the 'external fan' seemed needlessly complex and a bit like overkill to
me. I'd go with the internal fan and ducting, like I suggested, as I favor self
contained systems. That or the 'simple' solution of putting the existing fans on
7 volts.


Regards,





  #6  
Old August 4th 03, 12:33 PM
Stephen Howard
external usenet poster
 
Posts: n/a
Default

On Sat, 02 Aug 2003 19:45:04 -0500, David Maynard
wrote:

Stephen Howard wrote:


Ah, that would prove to be a bit messy - unless I knocked up some
internal ducting....


Might not be as hard as it seems, depending on the case. I mean, if you HAVE a
'bottom rear' entry (wondered about that when you said it) you just need to
baffle it on in a bit. Ever see that clear plastic case cooling addon (I forget
the name) that goes on the bottom front fan inlet and just ducts air to the
bottom middle of the case? Same thing in reverse.


I do have a bottom rear entry ( pause for giggles ), and plenty of
space in the case to build a duct....


I beginning to think I might be better off with a laptop - but where's
the fun in that eh??


Hehe. Welcome to the wonderful world of engineering.

Frankly, the 'external fan' seemed needlessly complex and a bit like overkill to
me. I'd go with the internal fan and ducting, like I suggested, as I favor self
contained systems. That or the 'simple' solution of putting the existing fans on
7 volts.


I shall experiment later with 7V across a 12V fan - though once I have
the dosh for an upgrade my feeling is that a passive CPU cooler and a
quality PSU would be a good place to start.

Regards,



--
Stephen Howard - Woodwind repairs & period restorations
www.shwoodwind.co.uk
Emails to: showard{whoisat}shwoodwind{dot}co{dot}uk
 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT +1. The time now is 11:07 PM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 HardwareBanter.
The comments are property of their posters.