P4 Case Cooling Experiments - Block the PS Vent Holes?


S

Steve Sr.

I just put together a P4 based system in an Antec 1080 case. The case
comes with a True Power 430 watt power supply. The power supply has a
120mm fan on the bottom that draws air into the bottom of the supply
and an 80mm fan that exhausts air out the back of the case. Sthe power
supply also has an output to power case fans in parallel with the
one(s) in the PS and a fan speed monitoting output to feed the
motherboard.

In addition the case has room for 5 more fans, 2 in the lower front,
two in the middle of the rear and one on the side cover over the video
card. In this application I have the two in front and the highest one
in the rear populated.

Since heat rises I have tried to turn the case into a chimney to aid
the fans as much as possible. This draws cool air in the lower front
and exhausts hot air out the top rear. I have blocked off the other 2
unused fan openings so that convection doesn't get short circuited.

I ran the computer like this overnight running Memtest86 and the next
morning I noted CPU and MB temperatures of 46/33 degrees C
respectivly. I also noted that monitored PS fan was running about 1605
RPM.

During this test I noticed that the case and all of the components in
the case got warm to the touch especially the CD drives in the top of
the case. This got me to thinking about the source of a lot of this
heat. It was the power supply itself!

The existing fan arrangement in the PS with a big one drawing in and
smaller one drawing out was causing the the power supply heat to be
dumped back into the case via the vent holes in the power supply!

That heat should be getting dumped *outside* the case and not back
into it. In order to make this happen I tried an experiment and
covered all of the vent holes in the PS with clear packing tape. This
would force a single path for air through the PS.

I ran the same experiment and in the morning noted no change in the
CPU / MB temperatures. I surmise that the MB temperature is heavily
influenced by exhaust from the CPU fan. The real surprise is that the
top of the case and all of the other components inside it were now
cool to the touch!

Some may wonder that this arrangement would cause the PS to overheat.
I considered this possibility and was careful to monitor the PS fan
speed early during the test. When I didn't detect a marked increase in
speed I let the test run overnight. In the morning the PS fan speed
was only 1695 RPM or about 5.6% higher which indicated that the PS was
not appreciably hotter than when it was sucking recirculated hot air
inside the case.

Comments? Has anybody else ever tried this?


Regards,

Steve
 
Ad

Advertisements

P

Philip Callan

Steve Sr. said:
I just put together a P4 based system in an Antec 1080 case. The case
comes with a True Power 430 watt power supply.

My Antec Sonata case has a TruePower380W and unless the PSU registers 45C or
above, anywhere in the 1200-1800 rpm range for the fan is appropriate (my
asus MB keeps showing this figure in the BIOS in red, thinks its too low for
a PSU fan) this is designed as such.
 
H

harry wong

Sounds interesting. I have the same case and taped over the side vent only-
am using 1 fan in front (lower drive cage) and have kept the 2 rear. Been
pretty stable at 32C MB with the fan spinning at around 1560.

With your temps I assume you are using an Athlon?

harry
 
E

Ed

I just put together a P4 based system in an Antec 1080 case. The case
comes with a True Power 430 watt power supply. The power supply has a
120mm fan on the bottom that draws air into the bottom of the supply
and an 80mm fan that exhausts air out the back of the case. Sthe power
supply also has an output to power case fans in parallel with the
one(s) in the PS and a fan speed monitoting output to feed the
motherboard.

In addition the case has room for 5 more fans, 2 in the lower front,
two in the middle of the rear and one on the side cover over the video
card. In this application I have the two in front and the highest one
in the rear populated.

Since heat rises I have tried to turn the case into a chimney to aid
the fans as much as possible. This draws cool air in the lower front
and exhausts hot air out the top rear. I have blocked off the other 2
unused fan openings so that convection doesn't get short circuited.

I ran the computer like this overnight running Memtest86 and the next
morning I noted CPU and MB temperatures of 46/33 degrees C
respectivly. I also noted that monitored PS fan was running about 1605
RPM.

During this test I noticed that the case and all of the components in
the case got warm to the touch especially the CD drives in the top of
the case. This got me to thinking about the source of a lot of this
heat. It was the power supply itself!

The existing fan arrangement in the PS with a big one drawing in and
smaller one drawing out was causing the the power supply heat to be
dumped back into the case via the vent holes in the power supply!

That heat should be getting dumped *outside* the case and not back
into it. In order to make this happen I tried an experiment and
covered all of the vent holes in the PS with clear packing tape. This
would force a single path for air through the PS.

I ran the same experiment and in the morning noted no change in the
CPU / MB temperatures. I surmise that the MB temperature is heavily
influenced by exhaust from the CPU fan. The real surprise is that the
top of the case and all of the other components inside it were now
cool to the touch!

Some may wonder that this arrangement would cause the PS to overheat.
I considered this possibility and was careful to monitor the PS fan
speed early during the test. When I didn't detect a marked increase in
speed I let the test run overnight. In the morning the PS fan speed
was only 1695 RPM or about 5.6% higher which indicated that the PS was
not appreciably hotter than when it was sucking recirculated hot air
inside the case.

Comments? Has anybody else ever tried this?


Regards,

Steve

I have the Antec 1040 Tower/400w. I took out all the Antec fans.

For the PSU I taped off the side vents, (CD&DVD were getn hot) replaced
the bottom fan only and connected that to a voltage regulator I made so
I can adjust it's speed, 2500RPM was overkill.

For 80mm case fans I use one in front HDD bay and two in the rear, all
are thermally controlled type via sensors touching the CPU heatsink and
ram heatsink on the video card.

This has worked great for me for past 5 months, quite and cool.
Ed
 
S

Steve Sr.

Harry,

No, this PC is an Intel P4 2.6GHz. Remember that during this test the
CPU was not idle. It was runnimg Memtest86 and had stabilized
overnight.

Steve
 
S

Steve Sr.

For the PSU I taped off the side vents, (CD&DVD were getn hot)

Exactly my reasoning for trying this. From your experiment It doesn't
look like I'm creating a PSU reliability problem.

Steve
 
Ad

Advertisements

6

6bal

I just put together a P4 based system in an Antec 1080 case. The case
Comments? Has anybody else ever tried this?

Yes
You are way to concerned about your cases temp. ;)

I have that same PS and it just puts out a lot of heat which would
rise above the fan they have at the bottom of the PS so the top of the
case gets warmer. Yo could get a case with the fan in the top though
if it really bothers you.
 
Ad

Advertisements

6

6bal

My Antec Sonata case has a TruePower380W and unless the PSU registers 45C or
above, anywhere in the 1200-1800 rpm range for the fan is appropriate (my
asus MB keeps showing this figure in the BIOS in red, thinks its too low for
a PSU fan) this is designed as such.
You know you can change that value so it doesn't show red?
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top