• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

I just don't get it.

E

[em]HEAT

Limp Gawd
Joined
Mar 13, 2004
Messages
312
I've been dealing with this W/C setup ever since I install the dam thing, for some reason I just can't get my system to idel below 45c. I've changed the pump to increase waterflow, I've lapped the waterblock and put a dual 120mm radiator and this dam box is still running hot. When playing games the temps jump up to 60c, then the auto shutdown activates. I don't understand what's going on. Are there other things that I can do to decrease the temps? The W/C setup is a Danger Den maze 4. What should I do?
 
Here's what everest is reporting and this is at idle.

everest.jpg
 
your problem man not be in your setup yet in your monitoring software. Some of the others can back me up on this because I am not sure what programs out there that fall under this category. But there are a lot of progams that do not talk to the mobo properly and register bad temps.

Also make sure you are using the correct amount of thermal compound. too much is bad but at your (presumed) level of expertese I am sure you already know that.
 
Seems like you have tried everything that I would have tried. What does your BIOS say your temps are? If they are not close to what you are getting from your software monitoring progam on idle then it's your software.
 
make shure your water block is completely level. It can be very hard to tell if its completely level as i have had some problems with my stock p4 H/F as alot of others have had too... even if they dident know it. people that complain about the prescott running hot well my pressy only runs at about 30-47c (on stock cooling) and i realy dont think thats out of the norm for stock air cooling

i would like to see what kind of mounting your water block has going on it.

now this is asuming that you are geting proper water flow threw the water block and that your w/c system is not clogged. also what is the ambient temps?

you could check to see if the water is warm or hot to the touch it should feel cold or at lest mildly warm at the very hottest.(my system is cold to the touch when at full load for 8 hours)

i have a 3.0e @ 3.9 water temps are never over 24.4c (76f) and cpu temps are from
28c-36c
 
but i would try using the CoreCell from MSI instead of that everest stuff not shure but i would give it a try :)
 
Yeah, what power supply are you pushing all that with? Its a pretty decent sized load...

MSI K8N-NEO 2 Platinum / 939
AMD 64 FX-55
4x512 OCZ PC3200 DDR400
BFG Tech. GeForce 6800 Ultra

If your 3.3v line is at 1.5v I dont think your computer would boot, and 10~v on the 12v line is way outta spec. If it is reading those voltages so far off, the temp could be as far off too.
 
mathot said:
Yeah, what power supply are you pushing all that with? Its a pretty decent sized load...

MSI K8N-NEO 2 Platinum / 939
AMD 64 FX-55
4x512 OCZ PC3200 DDR400
BFG Tech. GeForce 6800 Ultra

If your 3.3v line is at 1.5v I dont think your computer would boot, and 10~v on the 12v line is way outta spec. If it is reading those voltages so far off, the temp could be as far off too.
Click to expand...

It's an Ultra X-Connect 500w
 
[em]HEAT said:
It's an Ultra X-Connect 500w
Click to expand...

I might invest in a multimeter(they dont cost that much) to test what the actual voltages are under load, cause if you are getting 10.5V on the 12V rail under load, return that POS PS and get yourself a real power supply.

Also, you might want to test the temperature yourself with a temp probe (make sure it is as close to the CPU as you can get w/o actually coming between the cpu and the waterblock) Because even BIOS readings can often times be way off.
 
Hiya Heat :D

The first thing I would do , is to take the actual temperature of the coolant , both before , and after the rad.This will help determine the issues.........
 
BrainEater said:
Hiya Heat :D

The first thing I would do , is to take the actual temperature of the coolant , both before , and after the rad.The will help determine the issues.........
Click to expand...

That wont tell you diddly, other than the fact that yes your radiator is dissipating heat (which you already knew because your water wasn't boiling)
 
erm....actually it tells you a lot...with these two measurements , and the 'actual measured cpu temp' you can determine precisely what part of the heat exchange circuit is not working properly.
 
This came up in another thread. The temp difference would be so small there is no way you'd be able to even measure it. Think about how fast the water is moving over your proc, how much heat can really be absorbed in 0.3 of a second? 0.5 of a second? Less? The water temperature is constant throughout.

Unless of course you've tried it and have actual evidence to the contrary...
 
mathot said:
This came up in another thread. The temp difference would be so small there is no way you'd be able to even measure it. Think about how fast the water is moving over your proc, how much heat can really be absorbed in 0.3 of a second? 0.5 of a second? Less? The water temperature is constant throughout.

Unless of course you've tried it and have actual evidence to the contrary...
Click to expand...


While I could certainly build a test rig to prove this in a few hours , I won't bother.I'll just try to explain it.

Think about what you are saying : "The water temperature is constant throughout"

Umm...ok....If the water temp is constant everywhere , the water cooling is not working at all.The rad is dissapating nothing.Obviously , your statement , umm, 'holds no water' ;)

Secondly , when you are searching for empirical data on the performance of a cooling system ,how do you think you get that data ? you get it by measuring it....not guessing , not feeling it with your hand , but by measuring it.period.

The main values required to test a single loop watercooling system are :

The cpu temp (actually measured with a real thermometer)
The temp before the rad (liquid hotside)
The temp after the rad (liquid coldside)
The ambient air temp


OK......
The cpu temp value will always be the highest.The difference between the cpu temp and the liquid hotside gives you an indication of the efficiency of the cpu to water interface (waterblock,thermal compound etc).The difference between the liquid hotside and the liquid coldside gives you the efficiency of the rad.

Now I will submit : you need a good thermometer.you can't be usin granny's oven thermometer.The one I use is an industrial digital thermometer capable of tenths of a degree.(and thats barely good enough)

.....

Now there's a lot more to this - flowrates , airspeeds etc....but knowing these 4 temps as you experiment with the cooling setup is where you start..A good description of watercooling/rad testing can be found Here .
 
a better place to measure would be the waterblock hotside and coldside, then you can find the C/W curve of your waterblock and see if your measurements match up, that would let you know if you are getting thermal transfer problems between the block and the CPU.
 
BrainEater said:
Think about what you are saying : "The water temperature is constant throughout"
...your statement , umm, 'holds no water' ;)

Now I will submit : you need a good thermometer.you can't be usin granny's oven thermometer.The one I use is an industrial digital thermometer capable of tenths of a degree.(and thats barely good enough)
Click to expand...

Right, my point was mearly that the water moves so fast in a closed loop, mixing with a res and going into the rad, etc, the water temp will differ in the tenth of a degree at most. Which puts you where Erasmus said, that cooling is working, but we already knew that because the water is not boiling.
A 120mm fan through a rad will not chill water moving at 50-100 gph to a very measurable extent. At least give me that...
 
Erasmus354 said:
a better place to measure would be the waterblock hotside and coldside, then you can find the C/W curve of your waterblock and see if your measurements match up, that would let you know if you are getting thermal transfer problems between the block and the CPU.
Click to expand...

Absolutely.
The more data you have about the temperature differentials in the system the better.I would suggest , for most setups , this will be the same as the rad hotside/coldside....the pump/resevoir/plumbing doesnt really dissapate very much.


mathot said:
Right, my point was mearly that the water moves so fast in a closed loop, mixing with a res and going into the rad, etc, the water temp will differ in the tenth of a degree at most. Which puts you where Erasmus said, that cooling is working, but we already knew that because the water is not boiling.
A 120mm fan through a rad will not chill water moving at 50-100 gph to a very measurable extent. At least give me that...
Click to expand...

Sorry bud , I can't give ya that. :( (not entirely anyways)
A properly working radiator will cool the liquid by many degrees , even tens of degrees.The reason you need the tenth-of-a-degree accuracy , is to document the subtle changes that you make while 'tweaking' the system......diff waterblock ,thermal compound/pump/rad/etc.

-----------------

The bottom line is this :

[em]Heat needs to get a thermometer.Buy/rent/borrow.....First thing to check is the 'actual' temp of the cpu...see if it's just bad software or whatever (although the thermal shutdowns suggest not).....failing that , it's time to start experimenting , and the first step in that is to measure everything to determine the 'baseline data'...

alternately , ya could just throw money at it.Buy top of the line everything , that'll help ..... :rolleyes: ....hehe
 
Many people have had mis-reads with temps on the Neo2. I know I certainly do get misreads all the time. I don't trust what this board tells me. When I startup folding or something, my temperatures jump up 12c as soon as I start the program up. That is not possible at all. So don't really trust what it is telling you.
 
You must log in or register to reply here.
Back
Top