temps question - (pic 56k warning)

oswald · Nov 26, 2004

hey there. I've just rebuilt my comp with the WC setup and I'm concerned about the temps. My hardware is in the sig and i would really appreciate somebody with similar config (especially the same mobo) to tell me his temps as mine seem a little too high (CPU 35-48 load, CASE 27-34).

Anyone?

here's a pic of my setup

Aphex_Tom_9 · Nov 26, 2004

some things-
MB temp monitors are never really dead-on accurate.
Are you overclocked?
Is the waterblock on tight and the thermal paste isnt too much or too little?

oswald · Nov 26, 2004

well i think the thermal paste is all right. I've heard that temps are never dead on, but what concerns me that my last aircooling fan - zalman CPS7000CU with a 92mm fan was just around 8-10C hotter than this setup...

Aphex_Tom_9 · Nov 26, 2004

oswald said:
well i think the thermal paste is all right. I've heard that temps are never dead on, but what concerns me that my last aircooling fan - zalman CPS7000CU with a 92mm fan was just around 8-10C hotter than this setup...

well a 10c temp drop is pretty nice, but not that amazing. i think the discrepancy is caused by the lack of airflow around the socket that WC causes. some people have seen even less temp drops with WC because they switched from air which helped the temp readings more, but not the actual cpu temp

gingo · Nov 26, 2004

I went from air to water and saw only a 1-2C load temp drop. Like Tom said though, I beleive it is just from lack of airflow around the socket area.

Senater_Cache · Nov 27, 2004

Aphex_Tom_9 said:
well a 10c temp drop is pretty nice, but not that amazing. i think the discrepancy is caused by the lack of airflow around the socket that WC causes. some people have seen even less temp drops with WC because they switched from air which helped the temp readings more, but not the actual cpu temp

He is right.
my old setup was at 49C load with 6 80mm through out the case.
I now have your setup minus the pump, and have moved the old setup to a new case as a server.
It has 1 80mm intake and 1 80mm exhaust in the PSU.
Everything else i sexcactly the same.
load is now 64C !! ( its ok, its a badly ventilated Thunderbird)

shows you the difference ambient air-temp and socket air-flow can have on temps.

Masakari · Nov 27, 2004

As others have mentioned above the higher temps are caused by the lack of airflow around the CPU socket.

I performed an experiment a while ago where I held a 80mm fan close to the CPU socket. The temps reported for the CPU dropped 4C in only a few minutes.

My motherboard uses a socket thermsistor so this would not apply if your motherboard reads the temps from the thermsistor located on the CPU package itself.

oswald · Nov 28, 2004

thanks for the input everybody. after all i'm happy with the noise (or rather with the lack of it) in my system.

Nebulous · Nov 28, 2004

Hmm, you should have better temps.

1st - your rad looks blocked from the pic. Gotta cut all that metal off so the rad gets air

2nd- The way you have your loop is another concern. It should be set up like so:

Pump>rad>cpu block>(res)pump.

This way the cpu block will get the cooler water directly from the rad.

This is my observation.

Senater_Cache · Nov 28, 2004

Nebulous said:
Hmm, you should have better temps.

1st - your rad looks blocked from the pic. Gotta cut all that metal off so the rad gets air

2nd- The way you have your loop is another concern. It should be set up like so:

Pump>rad>cpu block>(res)pump.

This way the cpu block will get the cooler water directly from the rad.

This is my observation.

LOL

no,

order of components will not make a difference of even 1C . Far less actually. So little that it is difficult to measure with average equipment.

He is right about the rad however, it should have easy access to out-of-case air.

Another thing you may improve would be to turn your pump around.
Why did you place it that way anyway? If you have a ligitimate reason, then its ok.
Note that turning it around will eliminate that tubing bend under the vid-card, and thus shorten your loop by about 5-6" or more, which is always nice as tubing creates resistance.

Where is your PSU ??

Nebulous · Nov 28, 2004

Senater_Cache said:
LOL

no,

order of components will not make a difference of even 1C . Far less actually. So little that it is difficult to measure with average equipment.

He is right about the rad however, it should have easy access to out-of-case air.

Another thing you may improve would be to turn your pump around.
Why did you place it that way anyway? If you have a ligitimate reason, then its ok.
Note that turning it around will eliminate that tubing bend under the vid-card, and thus shorten your loop by about 5-6" or more, which is always nice as tubing creates resistance.

Where is your PSU ??

Sry Senater, but I've got better results with the rad b4 the wb, than rad after the block, and this is with several setups. maybe you got poor results, but not i

Also i just noticed he's missing the psu. i see wires, but no psu.

Legion2005 · Nov 29, 2004

Whats up with those colorful zipties? I have never seen anything like that.

I would say definately cut out that part of the case by your radiator. There isnt enough air for sure.

temps question - (pic 56k warning)

oswald

Registered

Attachments

Aphex_Tom_9

Member

oswald

Registered

Aphex_Tom_9

Member

gingo

Member

Senater_Cache

Member

Masakari

Registered

oswald

Registered

Nebulous

Dreadnought Class Senior

Senater_Cache

Member

Nebulous

Dreadnought Class Senior

Legion2005

Member

Similar threads