|
|
General Liquid/Water Cooling Discussion For discussion about Full Cooling System kits, or general cooling topics. Keep specific cooling items like pumps, radiators, etc... in their specific forums. |
Thread Tools |
11-12-2004, 03:52 PM | #1 |
Cooling Neophyte
Join Date: Sep 2004
Location: Oklahoma
Posts: 43
|
Question regarding pumps
I currently am using a Hydor L35 pump in a single RBX block/'86 Chevette HC setup.
I am looking into doing a couple changes/upgrades to my system and havent been able to find the following information, any input would be helpful. I am looking at a couple different pumps to upgrade my system. The pump I am considering now is a Panword 40PX (13.5ft head, 6 gallon per minute rating). I know in my current loop getting over 2 gallons per minute through my RBX is at best highly improbable. It seems like most of the RBX/Cascade/White Water blocks all sustain a sub 2 gallon a second flow rate according to most of the Pump PQ/Block curve charts I have seen (assuming I am reading them correctly which isnt exactly guaranteed). I am wondering how much of the pumps capability is actually wasted currently due to the restriction of a waterblock? My second upgrade will be a GPU block. I have a 6800 card and it seems the Maze 4s will actually work fine for most 6800s. The flow rate through the Maze 4s is fairly close to the throughput for my RBX which leads me think of putting in two loop branches- one with a block and rad for the CPU, one with a block and rad for the GPU. The idea is to harness more of the wasted pumps capability while not being forced to send the coolant heated by one block over the other block. If my flow rate right now is ballpark sub 2 gallons a minute and the Panworld has the capability to pump up to 6 gallons combined with twice the head of my current L35....would this negatively or positively effect my overall CPU cooling if I used the dual branch loop? I am also trying to save internal space by avoiding using 2 pumps and 2 reservoirs (I prefer having the additional fluid via a res in my loop in case of a slow leak over a T line). The rad/heater cores (paired '77 b-ville cores) would be mounted externally however. Thoughts? |
Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
|
|