• 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.

34C fluid temp, 49C processor temp?

E

EvilAngel

Gawd
Joined
Jan 9, 2002
Messages
597
Is there something wrong with my WC setup or is this normal?

Ambient temp: 28C
Water temp: 34C
Procressor IDLE temp: 49C

Ambient <--> Processor difference: 21C *
Water <--> Processor difference: 15C *
Ambient <--> Water difference: 6C

So my dual 120mm radiator seems to be working right. The pump is a Eheim 1048 (150gph)

Is this a waterblock problem? Poor contact maybe? Need lapping?
 
those temps sound quite high to me.

what waterblock?

what temp is your GPU monitoring software reporting?

if your GPU is reporting similarly high temps, i would think that your coolant is not circulating properly. check for blockage and that the pump is turning properly.
 
Make sure there isnt junk inside the block itself. Sometimes grime can build up, although this takes a while.

Remount your block. Lap if needed. Make sure pressure is evenly applied. Check to make sure you are using the right amount of thermal paste.

It might be possible that your pump is not up to snuff...do you have a gpu cooler also?

What tubing are you using? 1/2'' ID?

What block(s) are you using? Some are more restrictive than others.

Where is your radiator mounted?

Are you using elbows? Do you have kinks in your flow?

This is just a small list of things to check out, and report back to us so we can help you more. Hope it helps.
 
if your GPU temps are fine its got to be cpu contact

when applying AS5 or something similar, try to have as little on there as possible well still providing coverage

also you may want to look into temp sensor issues with your board
 
NF7 2.0 reports temperatures which are way too high when using water cooling.

I have the same problem.
 
mwarps said:
NF7 2.0 reports temperatures which are way too high when using water cooling.

I have the same problem.
Click to expand...
my NF7-S rev 2.0's socket readback was in very close agreement with my thermal probe..........

you know, before i baked it extra-crispy. :)
 
Hello, thanks everyone for the replys and sorry for the delay.

The block i think its a MCW462-U.

I dont have a GPU block.

The only elbows on my setup are the quick attachs on the radiator. im going to remove them and use clamps.

http://www.mikhailtech.com/articles/cooling/mcw/page4.shtml

I changed the thermal paste, made sure the contact was good but i didnt have artic silver at hand so i used... Artic alumina, temp went up 5C

When i get back home from work, ill re-apply artic silver and ill remove the elbows to see if i get better temps.

BTW:
Im using 3/8 tubing.
The radiator is outside the case.

Water temp is only 34C so i dont really think the rad is at fault.
 
i agree that with coolant temps like that, the rad is not likely to be the source of your current problems.

now that you have re-seated the heatsink, and gotten similar temps to before, then you have most likely eliminated poor contact as an issue.

if your coolant is not circulating properly, then that could give you good coolant temps and a hot processor. hence all of the blockage checks, and checking to make sure that the pump is turning.

if your coolant turns out to be circulating properly, then about all i can say is that you are using an older waterblock, that may not perform as well as you might hope.

maybe someone else will have some suggestions after you eliminate coolant flow as a potential problem.
 
Gotta be flow or contact. The idle temp should be the same or slightly higher than the coolant. My 3500+ newcastle idles at 32C, and is only 45C load, so even for an AXP your idle should be in the thirties. Good luck.
 
Well, i did a few experiments...lapped the water block... and used artic silver, contact is good. = same temps.

So i measured the true flow of the pump:

Instead of a reservoir used 1 gallon of water took 110 seconds for the pump to completely empty the gallon.

3600 (an hour) / 110 = 32.7GPH

With everything connected in my WC setup the pump pushes 32.7GPH or 123.8 Liters PH.

Could this be the problem not much flow?
 
Where is the temp sensore in your loop? If its right after the radiator, then youll definately see much cooler water temps than proc temps. That pump also seems way too weak, with a waterblock & rad restricting your pumps flow, youll be lucky to get 0.5 gph.
 
ikellensbro said:
Where is the temp sensore in your loop? If its right after the radiator, then youll definately see much cooler water temps than proc temps. That pump also seems way too weak, with a waterblock & rad restricting your pumps flow, youll be lucky to get 0.5 gph.
Click to expand...

I understand the water temp is going to be cooler specially since im reading the temp at the reservoir that is feeded from the radiator and then from the reservoir goes to the waterblock. But... the difference is too much and the processor temps are very high.

The "True" flow of the pump is 32.7GPH, thats with the radiator and the CPU block connected. The pump is rated at 160Gph (at 0 head 1/2 tubing i belive), im using 3/8".
 
I see you have quite an overclock on that cpu. What voltage are you pumping through it? That could be the reason :).
 
Oh, that makes things a little different now.
Id think a bad block off hand, or bad temp sensor(s), or bad cpu (running hot/core not flat).

Maybe try a plain heatsink and see what temps you get?
 
DarkenReaper57 said:
I see you have quite an overclock on that cpu. What voltage are you pumping through it? That could be the reason :).
Click to expand...

Right now...1.95V :p but..

still... if i go back to STOCK (1.525v) for the 2400+ Mobile i get high temps... i cant remember the exact temps, let me go back to stock.

OK, 42C idle @ stock 1.525v :(
 
so............even at stock speeds, you are well above your coolant temp............

now, your flow does not sound very good to me, but one way to try and eliminate that possibility would be to do a second coolant temp test close to the outlet of your waterblock. if the coolant there is almost as hot as your processor temp, then it is almost certainly a case of insufficient flow. if the coolant there is still cool-ish, then that pretty much leaves your block.

you say that your pump is rated for 160 GPH at zero head, but i am under the distinct impression that usually people using AC pumps get something with a zero head rating of well over 250 GPH. a new pump should come with a GPH vs head chart or graph. to compete with something like the danger den D4 pump, then you should actually look for a pump with around 250 GPH flow at 4' of head.

the only reason that i keep mentioning that the block as something that might be causing a problem, is that if i am remembering correctly, the wet side of the contact plate is pretty much just flat copper, with some groves cut into it. the design could just be showing its age and simplicity.
 
You must log in or register to reply here.
Back
Top