• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

50c temps with water

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

marzocchiz2

New Member
Joined
Nov 28, 2003
can anyone help me figure this out? I have a dual athlon system, xp2400+ thorton core modded for mp, dangerden maze4 cpu blocks, eheim 1250 pump, dual heatercore with dangerden shroud and 120mm vantec stealth fans, home made res (msr 22oz fule bottle), all 1/2" id tubing. I can't get my temps below 49-50c except when the computer is first turned on (about 35c) but after a few minutes it will stay at 50c idle. about 55 load. I have tried distilled water with just water wetter (but it got a bit murky). de-ionized water mixed about 50/50 with toyota red anti-freeze and water wetter. both mixes give the same temps. I cleaned and reapplied AS5 on the cpus, no difference.

my system is:
tyan 2466-4m
dual 2400+ thorton
1 gig (2x512) mushkin ecc/reg
ati a-i-w 9800pro
promise ata100
4 7200rpm hds
sb audigy
pc power and cooling turbocool 510
lian-li pc62
 
Are the heatercores warm? If so, you might need more powerful fans. The maze4 isnt the best waterblock either, but it should be able to keep you cooler.
 
Hmmm.
I am not familiar with that mobo, or the newer chips- I would say the first thing to check is how accurate the temp reporting IS for that board.
Some tend to read considerably higher than others.

With the specs of the h2o stuff, I would have to expect that it should do just fine at cooling the system- a few questions though:
1) tube routing and the reservoir....
Kinks in the tubing can drastically reduce flow and cause high temps; so can any other flow restriction, so if the res you made is restrictive temps will be higher than they otherwise would be. Any problems there?
2) air in the lines/bubbles.
Any air in the system can also drastically reduce performance- any visible air bubbles or the sound of water in the system?

I now move to my last point:
Temperature monitoring has become even less dependable than it used to be thought.
With mobos reading temps at different points, and the desire to have either-
mobos that show low temps
mobos that never overheat
have caused manufacturers to do some strange thing with the code that translates temps so we can read them. ;)

My point? Comparing one system to another is very flawed.
What is far more important is how the system RUNS:
is it stable?
Does it overclock to a reasonably expected point?
Does it smell like something burning? (a dreaded thing!)

I have three systems with mATX boards that all read considerably higher temps than the cooling schemes I have installed should provide.
And I care not. ;)
They are fast, stable and overclock well...and they all run 24/7 with no problems.

The lowest temp of the three is usually around 53c load and the highest is normally around 58c.
Peak temps range from 56c to about 70c.

And I leave them running.

Check the temp accuracy that others get in similar systems (specifically the mobo) and take it from there.
 
attachment.php

the flow to and from the res is unrestricted, i don't think the res it self would cause any flow problems,
attachment.php

however things are tight around the ps and heatercore but not enough to cause much problem (i think).
attachment.php

i'm pretty sure all the air is out, the water wetter took care of most of it and i tilted the system a bit to get any left over air out of the heatercore. the system doesn't overclock (locked chips and no multi or fsb options in bios) however it does report the chips as 2600+mp on boot and in wcpuid, not sure why that happens. the heater core doesn't feel very warm but the res gets kinda warm (at least it seems that way). i can feel the air moving when i put my hand under the heatercore.
attachment.php
 

Attachments

  • sysmon.JPG
    sysmon.JPG
    46.8 KB · Views: 133
  • DSC00569small.jpg
    DSC00569small.jpg
    73.6 KB · Views: 92
  • DSC00571small.jpg
    DSC00571small.jpg
    60.5 KB · Views: 96
  • DSC00572small.jpg
    DSC00572small.jpg
    57.5 KB · Views: 91
I am still thinking about other possible causes....

In the meantime-
how is system performance and stability?

If it is running ok then I would say no sense of EXTREME urgency to solve the problem, yes?

I ask because I personally have no experience with AMD duallies, period, and haven't hooked up any duallie with H2O- yet. ;)
But we have members here who I am sure will be more familiar with a setup like yours, and may have info that will help- it just may take a few days before they are here and see this thread. ;)
 
no stability problems, i just expecting much lower temps after switching over to water. its about 5c lower at idle and 10c lower at load vs. air.
 
Well, the fact that temps ARE lower than they were with air is the way it should be, and the larger difference in load temp also follows what I typically see.

It does seem to me that the board you have is indicating temps higher than average- as I said, that really isn't a problem per se, often it is merely the coding of the bios and such.

One other thing that can lead to temps that seem off when water cooling- any physical temp sensor (not on-die sensors) will often actually show a Higher temp when water cooling is used: a hsf will often blow air ON the sensor and cool its surface somewhat, resulting in an inaccurate reading.

Unless you are having difficulties, I wouldn't worry about the indicated temps being higher than you expected, as long as system performance is where it should be- and stable. :)

Keep an eye on this here thread for a few days and see if some of the AMD duallie folks post up- you may want to <BUMP> the thread once a day for a few days to help make sure it gets seen. ;)
 
I'm no guru but one thing I can say is that I think 50/50 mix of antifreeze and distilled water is not the proper mix. I've done and read more like 10-25% mix.

Another thing would be perhaps you have not mounted the blocks properly? You could try and tighten them a little at a time and watch the temp readings. Tighten it a little and watch the temps to see if they go down.


Also it could be your mobo not reading the temps correctly?
 
bucken77 said:
I'm no guru but one thing I can say is that I think 50/50 mix of antifreeze and distilled water is not the proper mix. I've done and read more like 10-25% mix.

I missed that. ;)

Good point, and while it probably won't make a HUGE difference, it should drop temps a bit.
 
Doh!
I just noticed that this thread is in Cooling and should be in Water Cooling. :)

No biggie- I'll see if I can get it moved, but in the future, posting in the most closely related forum area will usually get the best results. ;)
 
I have this same board and I water-cool it as well.

The temperatures reported by this board are greatly affected by airflow within the case. In fact, if I place a regular 80mm fan in just the right spot, my CPU temps will drop by 10C. When you moved to water you removed the heatsink fans and affected the airflow in the case. Since the heatsink fans have such a great affect on temperature, you simply cannot perform a pre-water to post-water comparison of temperatures.

If you have quality equipment and a good setup then your real temperatures are probably as good as anyones.
 
Back