• 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.

Loop order: 2 blocks, 2 rads

K

Kninetik

n00b
Joined
Jan 15, 2005
Messages
59
I've seen other posts dealing with 1 giant radiator but my situation seems rare. Currently, Im simply using 1 CPU, 1 radiator. Next week I am recieving a GPU block and installing a second so I need to figure out the setup soon.

I see 2 ways to do this...

1) Pump => CPU => GPU => Rad => Rad2 => Pump
Seems to me that the CPU will have awesome temps because water just went through 2 radiators while the GPU will recieve the excess heat generated by the CPU and cause temperatures to go up.

2) Pump => CPU => Rad => GPU => Rad2 => Pump
This is what Im thinking is optimal. Cooled water goes into the CPU, comes out warm, gets cooled again, then goes into the GPU and the cycle repeats.
 
Dont worry about which way you order the loop. The water will be about 0.3 degrees hotter coming out of the CPU than it is going in, and even less coming out of the GPU. That difference is not enough to matter. What you want to do is run the routing so that you use the cleanest and least amount of tubing possible (while still avoiding kinks and elbow bends). This will provide you with the best performing setup.

You can either run the two radiators in series, or you could try to run the radiators in parallel (only the radiators). Since the radiators are identical they will get an equal flow, and running them in parallel will significantly reduce the restriction on the loop. But only do that if it is easily done.
 
I tried running them in parallel (assuming that this means you use a "Y" splitter to split the flow into two) to cool just the CPU before and that raised temperatures 2C as opposed to just running it off of one radiator, so I switched off back to one. I havent tried it in a series of 1 after another but I think it might work a ton better.

I think the reason temps went up is because the flow slowed down a good deal when split and by looking at Overclockers.com's numbers, flow matters a great deal when it comes to getting rid of excess heat. I might have split it incorrectly because one radiator was fixed on top of the case while the other on the bottom of the case. If you have an article to link me to, please do so :)

Also, you say water comes out less heated from the GPU than from the CPU. My 6800GT runs around 80C maximum with fans at full blast. There is a lot more heat generated by it so it should be heating the water a bit more.

Btw Eras: Thanks for the help you've been giving me the past week
 
Kninetik said:
Also, you say water comes out less heated from the GPU than from the CPU. My 6800GT runs around 80C maximum with fans at full blast. There is a lot more heat generated by it so it should be heating the water a bit more.
Click to expand...
Temperature != heat
Just because something is hotter does not mean that it outputs more heat. A GPU requires less energy to power, and therefore will output less heat than a modern CPU.

You didn't mention what pump you were running, the blocks you are using, or the radiators you are using. That info would greatly aid us
 
See Sig, new cooling system will contain

1/2" TDX, 2 BIP, MCP650 pump, DD Acetal GPU

My old parallel configuration that I had a problem with was just 3/8" Swifty 6000A, MCP650, 2 BIP

Current setup is just 1 BIP, which outperformed 2 BIP in parallel
 
You must log in or register to reply here.
Back
Top