Temperature difference. What do you think?

[Mayfair]

Hello all,

Firstly if you finish reading this, You deserve a cookie!

Finished my first water-cooling build last week (29/08/11) which is the following:

EK Supreme HF Full Nickel CPU Block
EK-FC580 GTX+ - Nickel Plexi Waterblock
Laing D5 Vario MCP655 12V DC Pump
EK-Multioption RES X2 - 250 Advanced
XSPC RX480
XSPC RX240

All was running excellent was very pleased with the results. Overclocked my i7 2600k to 4.7GHz with a random fixed vcore of 1.35 just for testing purposes which I will be soon lowering to find the lowest stable.

The temps was pretty good imo (54,57,60,55) on 100% load. My GTX 580 was maxing out on 100% load at 40c which was also excellent as it was reaching 87c on air.

As I said I was very pleased with the results of my first water-cooling build, Until the next day I noticed a very very small minor leak coming from under the port of the RX240 rad, It was a very small leak I checked it with a torch and noticed around the port there was a very fine opening where XSPC didn’t weld it correctly. So I partly drained my loop and took out the dual rad and got it RMA'd (which I must thank welsh Tom and Paul(XSPC) for the excellent speedy RMA, Dispatched a new rad the same day)

I got my new rad Friday and put the loop back together, I did re-use the fluid.

Now here is my problem, The temps I was getting before I partly drained the system and put the new rad in was CPU(54,57,60,55c) and GPU(40c) now I have the system re-built and with the new rad and re-used the fluid temps are now CPU(60,64,66,62c) and GPU(47c) that was last night's testing.

As you can see the temps have increased by around 6-7c, Now I was thinking this could be just room temp difference but I don’t think it is as 6-7c is quite a difference in room temp.

Do you think somehow when I re-used the fluid it could have got a few little bits in it and has clogged up somewhere in the loop? Which is causing the temp difference since I’ve re-built the loop.

Here is a screenshot from when I was testing before I changed out the radiator.



Here is a screenshot from last night with the loop re-built with the new radiator.



And today I’ve ran prime95 and the temps are around the same as last night all but 1c which I would account for room temp difference.

What do you guys think could be the cause of this temp difference since I’ve re-fitted the new radiator, Do you think it could be something in the loop that has clogged a little?

I don’t fancy taking the whole loop apart and cleaning right now, If you guys think it could be something in the loop that has clogged can it do any damage if I left it until I change the fluid in around 3months time?

Also do you think these temps are about right for my setup?

Many thanks in advanced for your help!

 

[JeffyB]

Did you make sure the loop was fully bled again?

 

[Mayfair]

Did you make sure the loop was fully bled again?

Yes, There are no bubbles I can see in the tubing and the is no bubbles at all in the res. I've also tilted the case etc and can’t see no bubbles or anything.

I also left the system running 24hours leak testing etc.

Although every now and then I can hear like a little clicking or somthing from the pump, Which was the reason I thought maybe there was somthing in the loop that got in when i re-used the fluid.

 

[RollerBall]

Yes, There are no bubbles I can see in the tubing and the is no bubbles at all in the res. I've also tilted the case etc and can’t see no bubbles or anything.

I also left the system running 24hours leak testing etc.

Although every now and then I can hear like a little clicking or somthing from the pump, Which was the reason I thought maybe there was somthing in the loop that got in when i re-used the fluid.

Where did you store the liquid? The only real way to check is the take the loop apart and do it again. I personally wouldn't have reused the liquid as there's more chance of it being ionised and picking up nasties when it was outside the loop.

Out of interest, is that all internal and what case are you using? Got a build log?

 

[Avathar77]

Seems like a problem with the new rad?
I have read some articles about some RADs not properly circulating the coolant properly and just passing through via the chambers near the G1/4 holes.

 

[Mayfair]

Did you make sure the loop was fully bled again?

I was tilting the case etc and no bubbles etc like I said, But you got me thinking and I started squeezing(pinching) the top highest tube(from CPU -> Res) and that was causing quite a lot of bubbles. I managed to get so when I squeezed the tube no bubbles was visible.

After doing this and leaving the pump running on its own for another night while I went to bed, I tested the temps the next morning and they seemed around the same as before fitting the new rad all but 1 or 2 degrees which I would account for room temperature.

So it seemed you was right after all

Although when putting both side panels on the case(TJ07) the temps were reaching around 65-68c under prime95 IIRC at 4.7GHZ 1.35V. Which I’m not happy with so I will probably tear down the loop and system and clean everything and make sure the cpu block is seated correctly and hopefully this will reduce the cpu temps. As when I tested with a NH-D14 on an open air bench it was reaching 70c max under prime, So comparing the two the temps I’m getting at the moment doesn’t seem right..

I'm happy with the gtx 580 temps, 26-27c on idle and 41-44c max load. This makes me think either the cpu block isn’t correctly seated or its got a few bits in it.

What do you think?

Tearing down the system etc will also give me a chance to add a koolance drain valve at the bottom of the loop, as draining the system to fit the replacement rad was a right pain in the *** so this time I'm going to make sure to fit a drain port!

Where did you store the liquid? The only real way to check is the take the loop apart and do it again. I personally wouldn't have reused the liquid as there's more chance of it being ionised and picking up nasties when it was outside the loop.

Out of interest, is that all internal and what case are you using? Got a build log?

It was drained in a clean(rinsed with de-ionised) ice cream tub then put back in the original mayhems bottles and was on the side for a while then I put it in a dark cupboard.

Yes it's all internal, I'm using a TJ07. I took some pictures along the way, But haven’t had time going threw them and making a log as of yet. Now I need to tear down the system etc going to be pretty busy

Seems like a problem with the new rad?
I have read some articles about some RADs not properly circulating the coolant properly and just passing through via the chambers near the G1/4 holes.

Never even thought of that! Didn't know some rads did that.

Thanks for all the replies so far, once I order a few extra bits including new mayhems coolant I will tear it down and re-build it and let you know what I find.

Also just out of interest, for 7/16 tubing what tube sealing plug do I need for the drain valve, 1/2" ID?