overheating help

  • Thread starter Thread starter Steven Campbell
  • Start date Start date
It is possible to have too many fans. One optional case fan I added in
started making a lot of noise, so I removed it, and the effect on PC
temp was virtually nil.
Click to expand...

Your airflow (cfm) thro the case was sufficient for the internal wattage,
so adding more cfm had no benefit for that environmental temperature.

In summer, under full load, you might see a difference since the delta-T
between the environmental air & heatsink is less. It's quite small though
since the heatsink is /still/ much hotter than the environmental air.

Biggest problem with PC cases is their often crude punched grill,
and the even cruder air intake at the front in the plastic fascia. You
have 1900mm^2 or less hole, and 9000mm^2 min from 2x 80mm.
Fitting a side-front grill can make a big difference to cfm, so you
can use quieter fans for the same or even lower temperature.

PCs will become a pain when we hit 200W/CPU, re heavy duty
dual-CPU graphics box, multi-HD, PSU - ie, 600W room heater.
 
Steven Campbell said:
I built my daughter a PC last year and have never been able to solve the
overheating problems.
It is an Athlon 1400 on a ASUS a7v8x board, 256Mb, CDrw & DVD the case
dimensions are (DxWxH)mm 476x185x443 (cheap case from)
<http://www.ebuyer.com/customer/products/index.html?action=c2hvd19wcm9kdWN0X
292ZXJ2aWV3&product_uid=44509>
It has a Thermaltake Silent boost with Arctic Silver 3
An exhaust fan. I cut out the mesh on the back of the case and installed a
wire grill as I felt previously the hot air was bouncing off the mesh and
back into the PC.
I have an 80mm fan at the front / bottom but when I have it running it
increases the CPU temp and decreases the Motherboard temp.
Disconnected it decreases CPU and increases MB temps.
MBM shows the temps with the bottom fan connected sometimes as high as CPU
70, MB 32 not under full load.
With the fan disconnected it shows CPU 62, MB 40 again not under load.
This is all with the side of the case partly off.
The PC is in the corner of the room which is really the only place she can
have it. I don't think that is helping as the hot air has no where to go.
The set up was improved after I added the Silent Boost to it just before
Christmas but gradually it is getting worse. It has now got to the point
where it freezes occasionally.
The fans are all blowing the way they should and the Silent Boost is
positioned correctly.

Can anyone give me some ideas? I thought maybe the cheap case wasn't helping
but am reluctant to shelling out a lot of cash to find it doesn't help. If
it was guaranteed then it would be no problem.

cheers

Steven.
Click to expand...

Thanks for all the advice to my problem above. To be honest I didn't know
where to start as a lot of the advice was conflicting.
To clarify a couple of points some posts brought up:
The temps I had quoted were measured using Mother Board Monitor 5 and
checked with the temps stated in the BIOS. The temps were also just running
temps for the PC i.e. The PC was put on and left to run with no programs
running apart from the processes in the background. It hasn't been
overclocked. It is a standard 1400 AMD

I opened the PC up and the inside was particularly dusty. The heatsink on
top of the CPU had a layer of fluff over it. The CPU fan itself was working
and seated properly.

The space for the front fan was just little holes punched out the case so I
opened this up into one big hole slightly smaller than the 80mm fan.
I took everything out the case, MB, CPU, DVD, CD, fans etc and cleaned them
and the case. I also took the power supply to bits and cleaned it. It only
has one fan (extractor)(it is AMD Approved). I put the CPU back using
Arctic Silver 3 with the minimum amount possible just covering the chip.
Re-seated the Thermaltake Silent Boost correctly and resembled everything
else. I also tidied the wires and re-routed the sound wires from the DVD, CD
as they were previously running between the Power Supply and the CPU (I
don't think that would have made any difference as they are so thin anyway).
I then reversed the front fan so it was an exhaust fan, as suggested in a
previous post. So I had 1 exhaust at the lower front and 1 exhaust at the
middle rear (Just opposite the CPU). I also made sure that the case
overhung the board that it sat on on the floor so air could escape from the
front.

Switched it on and I was getting average running temps of MB 36 CPU 61.
I then changed the front fan to be blowing air in and average temps are now
MB 30 CPU 57.
I haven't tested with the front fan disconnected yet. I don't think that
would help any!

At least the temps are a lot better than before. But if anyone can suggest
something else to bring it down further, it would be much appreciated.

Steven.
 
|| It is possible to have too many fans. One optional case fan I added
|| in started making a lot of noise, so I removed it, and the effect on
|| PC temp was virtually nil.
|
| Your airflow (cfm) thro the case was sufficient for the internal
| wattage, so adding more cfm had no benefit for that environmental
| temperature.
|
| In summer, under full load, you might see a difference since the
| delta-T between the environmental air & heatsink is less. It's quite
| small though since the heatsink is /still/ much hotter than the
| environmental air.
|
| Biggest problem with PC cases is their often crude punched grill,
| and the even cruder air intake at the front in the plastic fascia. You
| have 1900mm^2 or less hole, and 9000mm^2 min from 2x 80mm.
| Fitting a side-front grill can make a big difference to cfm, so you
| can use quieter fans for the same or even lower temperature.
|
| PCs will become a pain when we hit 200W/CPU, re heavy duty
| dual-CPU graphics box, multi-HD, PSU - ie, 600W room heater.

Let us hope that improvements in CPU design and manufacturing technology
will keep this more easily manageable.
Not so long ago, AMD brought about an improvement in this area by
designing their CPUs to run at lower core voltages than before. This lead to
lower power consumption and thus less heat dissipated. For example, the
Thunderbird core used in 1.0 and 1.4 GHz Athlons consumes more power, and
therefore produces more heat, than the Thoroughbred core used in the XP2400.
Kevin.
 
| || I built my daughter a PC last year and have never been able to solve
|| the overheating problems.
|| It is an Athlon 1400 on a ASUS a7v8x board, 256Mb, CDrw & DVD the
|| case dimensions are (DxWxH)mm 476x185x443 (cheap case from)
||
|
<http://www.ebuyer.com/customer/products/index.html?action=c2hvd19wcm9kdWN0X
|| 292ZXJ2aWV3&product_uid=44509>
|| It has a Thermaltake Silent boost with Arctic Silver 3
|| An exhaust fan. I cut out the mesh on the back of the case and
|| installed a wire grill as I felt previously the hot air was bouncing
|| off the mesh and back into the PC.
|| I have an 80mm fan at the front / bottom but when I have it running
|| it increases the CPU temp and decreases the Motherboard temp.
|| Disconnected it decreases CPU and increases MB temps.
|| MBM shows the temps with the bottom fan connected sometimes as high
|| as CPU 70, MB 32 not under full load.
|| With the fan disconnected it shows CPU 62, MB 40 again not under
|| load. This is all with the side of the case partly off.
|| The PC is in the corner of the room which is really the only place
|| she can have it. I don't think that is helping as the hot air has no
|| where to go. The set up was improved after I added the Silent Boost
|| to it just before Christmas but gradually it is getting worse. It
|| has now got to the point where it freezes occasionally.
|| The fans are all blowing the way they should and the Silent Boost is
|| positioned correctly.
||
|| Can anyone give me some ideas? I thought maybe the cheap case wasn't
|| helping but am reluctant to shelling out a lot of cash to find it
|| doesn't help. If it was guaranteed then it would be no problem.
||
|| cheers
||
|| Steven.
|
| Thanks for all the advice to my problem above. To be honest I didn't
| know where to start as a lot of the advice was conflicting.
| To clarify a couple of points some posts brought up:
| The temps I had quoted were measured using Mother Board Monitor 5 and
| checked with the temps stated in the BIOS. The temps were also just
| running temps for the PC i.e. The PC was put on and left to run with
| no programs running apart from the processes in the background. It
| hasn't been overclocked. It is a standard 1400 AMD
|
| I opened the PC up and the inside was particularly dusty. The
| heatsink on top of the CPU had a layer of fluff over it. The CPU fan
| itself was working and seated properly.
|
| The space for the front fan was just little holes punched out the
| case so I opened this up into one big hole slightly smaller than the
| 80mm fan.
| I took everything out the case, MB, CPU, DVD, CD, fans etc and
| cleaned them and the case. I also took the power supply to bits and
| cleaned it. It only has one fan (extractor)(it is AMD Approved). I
| put the CPU back using Arctic Silver 3 with the minimum amount
| possible just covering the chip. Re-seated the Thermaltake Silent
| Boost correctly and resembled everything else. I also tidied the
| wires and re-routed the sound wires from the DVD, CD as they were
| previously running between the Power Supply and the CPU (I don't
| think that would have made any difference as they are so thin
| anyway). I then reversed the front fan so it was an exhaust fan, as
| suggested in a previous post. So I had 1 exhaust at the lower front
| and 1 exhaust at the middle rear (Just opposite the CPU). I also
| made sure that the case overhung the board that it sat on on the
| floor so air could escape from the front.
|
| Switched it on and I was getting average running temps of MB 36 CPU
| 61.
| I then changed the front fan to be blowing air in and average temps
| are now MB 30 CPU 57.
| I haven't tested with the front fan disconnected yet. I don't think
| that would help any!
|
| At least the temps are a lot better than before. But if anyone can
| suggest something else to bring it down further, it would be much
| appreciated.

Steven,
Glad to hear that you have alleviated the problem and brought the running
temperatures down to something a little more reasonable. I would suggest
that with your current fan arrangement - front intake, rear exhaust and
power supply you should be getting more than adequate airflow through the PC
case. I am currently using a 1.5 GHz Athlon Thunderbird to write this and my
reported temps are CPU = 60, m/board = 45.
Given that you have now made sure that the HSF is correctly mounted on the
CPU, cleaned away the dust build-up (which can't possibly have been helping)
and made sure you have a more than adequate airflow, I would suggest that
you have gone about as far as you reasonably can without major changes which
might bring about little further benefit.
The fact that you are running a similar system to this one, at slightly
lower temperatures, and this one is perfectly stable and reliable - even
when running continuously for several days - leads me to suggest that you
have no further cause for concern.
Two points I would suggest you attend to:
1) Given the dust build-up you discovered, it would be worthwhile
periodically inspecting the inside of the PC case and cleaning away dust as
'preventative maintenance. Except in a very dusty environment, you should
find that once every six to twelve months is sufficiently frequent for this.
2) Set the CPU and m/board temperature alarms on the system (in the BIOS,
if provided) to a reasonable level above your current running temperatures
to alert you of any future problem. I would suggest that ten degrees above
the current temps would be a sensible point to have the alarms set.
Kevin.

|
| Steven.
 
Steven Campbell said:
<http://www.ebuyer.com/customer/products/index.html?action=c2hvd19wcm9kdWN0X

Thanks for all the advice to my problem above. To be honest I didn't know
where to start as a lot of the advice was conflicting.
To clarify a couple of points some posts brought up:
The temps I had quoted were measured using Mother Board Monitor 5 and
checked with the temps stated in the BIOS. The temps were also just running
temps for the PC i.e. The PC was put on and left to run with no programs
running apart from the processes in the background. It hasn't been
overclocked. It is a standard 1400 AMD

I opened the PC up and the inside was particularly dusty. The heatsink on
top of the CPU had a layer of fluff over it. The CPU fan itself was working
and seated properly.

The space for the front fan was just little holes punched out the case so I
opened this up into one big hole slightly smaller than the 80mm fan.
I took everything out the case, MB, CPU, DVD, CD, fans etc and cleaned them
and the case. I also took the power supply to bits and cleaned it. It only
has one fan (extractor)(it is AMD Approved). I put the CPU back using
Arctic Silver 3 with the minimum amount possible just covering the chip.
Re-seated the Thermaltake Silent Boost correctly and resembled everything
else. I also tidied the wires and re-routed the sound wires from the DVD, CD
as they were previously running between the Power Supply and the CPU (I
don't think that would have made any difference as they are so thin anyway).
I then reversed the front fan so it was an exhaust fan, as suggested in a
previous post. So I had 1 exhaust at the lower front and 1 exhaust at the
middle rear (Just opposite the CPU). I also made sure that the case
overhung the board that it sat on on the floor so air could escape from the
front.

Switched it on and I was getting average running temps of MB 36 CPU 61.
I then changed the front fan to be blowing air in and average temps are now
MB 30 CPU 57.
I haven't tested with the front fan disconnected yet. I don't think that
would help any!

At least the temps are a lot better than before. But if anyone can suggest
something else to bring it down further, it would be much appreciated.

Steven.
Click to expand...

Having re-read you OP it occurred to me that your Asus MB could be a
contributing factor. I once had a 1.4 Athlon on an Asus board and found it
very difficult to cool because Asus deliberately overclock their boards at
manufacture by setting the vcore slightly high.

At the time I was having my problem there was no BIOS release for it
available that allowed an appopriate decrease in the vcore setting and so I
returned the board to Ebuyer who tested it and agreed that there was little
that could be done.

I know it's already been suggested somewhere in the quagmire of "answers"
that your post created but you don't mention having tried it, it could be
the root of your evil problem. Don't know whether I'm right in saying that
the 1.4 Tbird was the worst heat producer of them all?

Your system temps are exactly what I would have expected now BTW and Asus
MBs have never had a good rep for having the most accurate sensors either.
 
Kevin Lawton said:
2) Try blowing with your mouth into free air, and then
at or through an empty PC case fan grille. Notice how much
noise the air makes as it impinges on grille. Notice also
how not all of the air you are blowing passes through
the grille, but bounces off the metal instead.


In this case I was illustrating a point that the PC case
fan grilles increase noise by causing turbulance as they
disturb the passage of air from the fan. Precise empirical
measurement was not absolutely necessary in order to
demonstrate the principle,
Click to expand...

In the second part I was addressing heat, not noise, and so were you,
unlike your quote. Temperature can be measured accurately with a
thermometer placed in front of the fan.
 
| |
|| 2) Try blowing with your mouth into free air, and then
|| at or through an empty PC case fan grille. Notice how much
|| noise the air makes as it impinges on grille. Notice also
|| how not all of the air you are blowing passes through
|| the grille, but bounces off the metal instead.
||
||| It's good that you're not one of those people who obsesses
||| about cooling for the sake of cooling and who performs
||| proper systematic testing and doesn't use ad hoc methods
||| that are subject to great bias.
||
|| In this case I was illustrating a point that the PC case
|| fan grilles increase noise by causing turbulance as they
|| disturb the passage of air from the fan. Precise empirical
|| measurement was not absolutely necessary in order to
|| demonstrate the principle,
|
| In the second part I was addressing heat, not noise, and so were you,
| unlike your quote. Temperature can be measured accurately with a
| thermometer placed in front of the fan.

Well, fine.
I was also addressing the problem of the greater noise which will be
generated if a fan is forcing air through a grille of small holes instead of
a fan-sized open gap. Do you really need precise measurements when you can
clearly hear the difference ?
Kevin.
 
Steven,
Glad to hear that you have alleviated the problem and brought the running
temperatures down to something a little more reasonable. I would suggest
that with your current fan arrangement - front intake, rear exhaust and
power supply you should be getting more than adequate airflow through the PC
case. I am currently using a 1.5 GHz Athlon Thunderbird to write this and my
reported temps are CPU = 60, m/board = 45.
Given that you have now made sure that the HSF is correctly mounted on the
CPU, cleaned away the dust build-up (which can't possibly have been helping)
and made sure you have a more than adequate airflow, I would suggest that
you have gone about as far as you reasonably can without major changes which
might bring about little further benefit.
The fact that you are running a similar system to this one, at slightly
lower temperatures, and this one is perfectly stable and reliable - even
when running continuously for several days - leads me to suggest that you
have no further cause for concern.
Two points I would suggest you attend to:
1) Given the dust build-up you discovered, it would be worthwhile
periodically inspecting the inside of the PC case and cleaning away dust as
'preventative maintenance. Except in a very dusty environment, you should
find that once every six to twelve months is sufficiently frequent for this.
2) Set the CPU and m/board temperature alarms on the system (in the BIOS,
if provided) to a reasonable level above your current running temperatures
to alert you of any future problem. I would suggest that ten degrees above
the current temps would be a sensible point to have the alarms set.
Click to expand...

Kevin, thanks for the advice.
I think the reason it was so dusty in the first place was due to me removing
the side of the PC and letting some cool air in. It may have helped in the
short term but long term just caked the inside with dust!!!

cheers

Steven.
 
Having re-read you OP it occurred to me that your Asus MB could be a
contributing factor. I once had a 1.4 Athlon on an Asus board and found it
very difficult to cool because Asus deliberately overclock their boards at
manufacture by setting the vcore slightly high.

At the time I was having my problem there was no BIOS release for it
available that allowed an appopriate decrease in the vcore setting and so I
returned the board to Ebuyer who tested it and agreed that there was little
that could be done.

I know it's already been suggested somewhere in the quagmire of "answers"
that your post created but you don't mention having tried it, it could be
the root of your evil problem. Don't know whether I'm right in saying that
the 1.4 Tbird was the worst heat producer of them all?

Your system temps are exactly what I would have expected now BTW and Asus
MBs have never had a good rep for having the most accurate sensors either.
Click to expand...


I missed that suggestion!!!
I had a look in the BIOS and the CPU Vcore was set to AUTO 1.75 so I
changed it to manual but the lowest it could go was 1.75 so I re-set it to
the AUTO setting.

Further on in the BIOS under Hardware Monitor the Vcore Bios was being
displayed as between 1.82 and 1.84 yet in a previous screen as stated above
it was set to "Auto 1.75"

Not knowing anything about Overclocking, would reducing the Vcore reduce the
speed of the CPU?

cheers

Steven.
 
Steven Campbell said:
so either.


I missed that suggestion!!!
I had a look in the BIOS and the CPU Vcore was set to AUTO 1.75 so I
changed it to manual but the lowest it could go was 1.75 so I re-set it to
the AUTO setting.

Further on in the BIOS under Hardware Monitor the Vcore Bios was being
displayed as between 1.82 and 1.84 yet in a previous screen as stated above
it was set to "Auto 1.75"

Not knowing anything about Overclocking, would reducing the Vcore reduce the
speed of the CPU?

cheers

Steven.
Click to expand...

This is exactly the problem I had, have you checked to see if there is an
updated BIOS available that may allow you to lower this? There may even be
an option to set vcore by jumpers on the board, have a look in your manual.

Lowering the vcore will not effect the speed of the cpu.
 
Kevin Lawton said:
2) Try blowing with your mouth into free air, and then
at or through an empty PC case fan grille. Notice how much
noise the air makes as it impinges on grille. Notice also
how not all of the air you are blowing passes through
the grille, but bounces off the metal instead.


In this case I was illustrating a point that the PC case
fan grilles increase noise by causing turbulance as they
disturb the passage of air from the fan. Precise empirical
measurement was not absolutely necessary in order to
demonstrate the principle,
Click to expand...
Well, fine.
I was also addressing the problem of the greater noise
Click to expand...

No, you were again addressing the problem of heat, and your claimed
45C case temperature is rather bad, even without a case fan, unless
the room is 35C or hotter.
 
| |
|| 2) Try blowing with your mouth into free air, and then
|| at or through an empty PC case fan grille. Notice how much
|| noise the air makes as it impinges on grille. Notice also
|| how not all of the air you are blowing passes through
|| the grille, but bounces off the metal instead.
||
||| It's good that you're not one of those people who obsesses
||| about cooling for the sake of cooling and who performs
||| proper systematic testing and doesn't use ad hoc methods
||| that are subject to great bias.
||
|| In this case I was illustrating a point that the PC case
|| fan grilles increase noise by causing turbulance as they
|| disturb the passage of air from the fan. Precise empirical
|| measurement was not absolutely necessary in order to
|| demonstrate the principle,
||
||| In the second part I was addressing heat, not noise, and
||| so were you, unlike your quote.
|
|| Well, fine.
|| I was also addressing the problem of the greater noise
|
| No, you were again addressing the problem of heat, and your claimed
| 45C case temperature is rather bad, even without a case fan, unless
| the room is 35C or hotter.

In some of my previous posts in this thread I discussed the issue of the
noise made fans and the roaring or whooshing sound made by turbulant air.
This is relevent within the broad subject area of PC cooling as the most
commonplace method of cooling a PC is by moving air through it by using one
or more fans and fans make noise. Many people dislike the noise made by fans
and would prefer to take steps to avoid it if possible. Creating less
turbulence is one way of bringing down this noise to a degree, which is why
I intriduced it. Creating turbulence will not increase airflow - it will act
to diminish it to a small extent.
I also introduced into this discussion the concept of drawing the air into
the PC case via a filter in order to cut down on the dust and dirt deposited
on heatsinks and fans - as this can reduce their efficiency.
Given that the OP has reported back that, since cleaning inside his PC and
ensuring an adequate airflow, the overheating problem has been alleviated to
a degree this shows that it was relevent to include dust and airflow in the
discussion.
Unlike yourself, I was attempting to assist the OP to resolve their
overheating problem by making use of my experience of building and
maintaining PCs over the past two decades.
At no time did I claim to have a case temperature of 45C, though I can
agree with you that a PC is likely to run hotter in a hotter environment.
Kevin.
 
Kevin Lawton said:
I was also addressing the problem of the greater noise


In some of my previous posts in this thread I discussed
the issue of the noise made fans and the roaring or whooshing
sound made by turbulant air.
Click to expand...

The originator of this thread was addressing heat, as did you in your
first reply to me.
Unlike yourself, I was attempting to assist the OP to resolve their
overheating problem by making use of my experience of building and
maintaining PCs over the past two decades.
Click to expand...

You must have missed my reply to him, where I told him to try
reversing his case fan and quit running his computer with the side
cover removed. I also asked him why his CPU was cooler when the case
was hotter. But you're right that when I replied to him, unlike you I
didn't make use of YOUR 20 years of experience of maintaining and
"building" PCs.
At no time did I claim to have a case temperature of 45C,
though I can agree with you that a PC is likely to run hotter
in a hotter environment.
Click to expand...

You wrote, to Steven Campbell:
I am currently using a 1.5 GHz Athlon Thunderbird to write
this and my reported temps are CPU = 60, m/board = 45.
Click to expand...

Since few motherboard temperature sensors can be relocated far from
the motherboard, without soldering, their readings are usually
considered the case temperatures.
 
| |
|| I was also addressing the problem of the greater noise
||
||| No, you were again addressing the problem of heat, and
||| your claimed 45C case temperature is rather bad, even
||| without a case fan, unless the room is 35C or hotter.
||
|| In some of my previous posts in this thread I discussed
|| the issue of the noise made fans and the roaring or whooshing
|| sound made by turbulant air.
|
| The originator of this thread was addressing heat, as did you in your
| first reply to me.
|
|| Unlike yourself, I was attempting to assist the OP to resolve their
|| overheating problem by making use of my experience of building and
|| maintaining PCs over the past two decades.
|
| You must have missed my reply to him, where I told him to try
| reversing his case fan and quit running his computer with the side
| cover removed. I also asked him why his CPU was cooler when the case
| was hotter. But you're right that when I replied to him, unlike you I
| didn't make use of YOUR 20 years of experience of maintaining and
| "building" PCs.

Yes, of course not !
I mentioned it only to mark the point that my suggestions were based on
practical experience of solving and avoiding such problems - checking the
simplest and most obvious things first can be a very effective fault finding
method, particularly when dealing remotely and with limted data. Like I
said, I was trying to help the OP, not criticising anyone.

|| At no time did I claim to have a case temperature of 45C,
|| though I can agree with you that a PC is likely to run hotter
|| in a hotter environment.
|
| You wrote, to Steven Campbell:

Well, it sort seemed appropriate, as he was the OP with the overheating
problem.

|| I am currently using a 1.5 GHz Athlon Thunderbird to write
|| this and my reported temps are CPU = 60, m/board = 45.
|
| Since few motherboard temperature sensors can be relocated far from
| the motherboard, without soldering, their readings are usually
| considered the case temperatures.

Given the prevailing points being made at that time by another poster to
this list concerning the temperature of the air inside the case and how a
large number of fans should normally be unneccessary to ensure sufficient
airflow for adequate cooling, I took your 'case temperature of 45C' comment
to refer to the inside air temperature. The 45 degrees temperature I had
noted on my own system (for comparison) was that reported as the temperature
of the northbridge chip on the motherboard. Okay, I know that this would be
the temperature of the outside of the IC package, and not of the actual
silicon core, but it served to illustrate the point well enough. A PC case
internal air temperature of 45C - a tad hotter than a cup of tea - would be
approx 25C above a normal room temperature and indicate a very serious
airflow problem indeed.
Kevin.
 
Back
Top