What's new
  • Please do not post any links until you have 3 posts as they will automatically be rejected to prevent SPAM. Many words are also blocked due to being used in SPAM Messages. Thanks!

Post your WaterCooled PC pics here thread

vinister

Well-known member
Joined
Feb 1, 2008
Messages
297
Location
Vancouver
Geez, you guys have to stop and realize that I have been using this build for months and I know what I am doing.

The Fusion plumbing was backwards, this was just an early 'dry-fit' photo.

Temps under full load are below 30C on CPU and below 40 on ALL THREE gpus.

Whoever thinks only the top or bottom cards will get flow need a quick lesson in fluid dynamics:

Parallel water cooling works perfectly well if all the flow resistant paths are the exact same, which they are. If you still really don't believe me, go to the library and get a book on fluid dynamics, or simply try it yourself and you shall learn =)

The energy you guys put into your MS paint skills needs to be redirected to your thought process. The left side of the 3 cards forms a "high pressure" area because the only way out is through the blocks. Because you have a high pressure area, with 3 equal-resistance paths to the other side (the low pressure area), equal amounts of fluid will flow through the 3 equal paths. The water is not pushing down, it is pushing in every direction equally, this is how pressure works. It will just as readily flow into the top block as the middle, as the bottom, the locations of the 'openings' do not matter to the water, it is simply pushing out in every direction. Just think, you could drill a hole in there anywhere you want, water will come shooting out whether the hole is on the top, bottom, middle, side, it just doesn't matter, pressure is pressure and it pushes, period.

Parallel resistance in electronic networks is the exact same, as long as the branches have equal resistance, equal current will flow.

All that being said, it is VERY important that all 3 flow paths have the same resistance to flow. If one of the barbs sticks too far in to the block, partially blocking the flow through the block, the whole system falls apart, and that card would quickly overheat. If one of the blocks was a different model or production run, it would also quickly alter the dynamics. The reason the results are so drastic is because for every 2 drops of water that do NOT flow through one block, an extra drop flows through the other 2, so the temperature deltas start to really get noticeable.

You CANNOT parallel a cpu, NB, and gpu blocks. The resistances are not equal, all the flow will take the easiest path. YOU CAN parallel identical blocks. I don't know what the blockage is in people's brains that stops them from understanding this. Just remember that pressure pushes in EVERY direction, even UP, gravity is negligible (think of a drinking fountain in the park, the water is shooting UP). Equal current will flow into equal resistance paths. The water wants out, it can't go back because it is being pushed on. If you put a pin-hole in your tubing, a tiny bit of water will flow out of it, the rest will still be pushed through your blocks. If you put 3 pin-holes in the tubing, equal water will come out of the 3 holes, whether they are on the top, side, or bottom of the tubing. Keep going up in scale, and you arrive and a high pressure zone with 3 ways out - all equal.

Please don't argue about this with me here - this is a photo thread, I want to see photos in it. PM me if you want, or better yet, just go to the library.
 
Last edited:

Mibs

Well-known member
Joined
Jun 5, 2007
Messages
311
Location
St. John's, NL
Well vinister I don't think anybody meant to offend you, we're all friends here. The loop looks great and some of us obviously just don't have the knowledge that you do in fluid dynamics, me being one of them. After reading your post it makes much more sense now, I understand how it works. That's all I was asking. I never questioned your knowledge on it. So thanks for the info, and again beautiful loop!
 

vinister

Well-known member
Joined
Feb 1, 2008
Messages
297
Location
Vancouver
Yeah sorry thats the whole reason I don't post too many pics of my setup, every time I get 2 pages of people telling me how it wont work, when it actually works extremely well.

Also there are 2 rads, one front, one back, the flow goes as follows:

Pump ----> Fusion V2 ----> Rear Rad ----> Parallel GPUS ----> Front Rad ---> Pump

Squee - your mspaint drawing that says "bad" on it is missing only the arrows flowing between the upper and middle cards. Maybe you have not seen one of these GPU blocks in person, there is a flow path across the block between the 2 in/out barbs. That would make the "bad" into "great!"
 
Last edited:

Squeetard

"Quote This..."
Joined
Nov 15, 2007
Messages
3,984
Location
Hell
The resistances are not equal. The paths through the blocks are far more resistant than the straight shot right through the tubing. Also, the last block does not match the first 2 at all in terms of setup. You would need a u-shaped piece of tubing out the back of it.

Your fluid dynamics theory is valid. But you don't have it. The water inside the blocks is getting flow, but much like an eddy or pool in a fast flowing creek.
 

vinister

Well-known member
Joined
Feb 1, 2008
Messages
297
Location
Vancouver
Im' sorry squee, you're just wrong here man.

Straight through the tubing? And where does that go? To more equal resistance blocks. There is no 'path out' through the tubing. There is no exit out the bottom. The only way in is at the top left. The entire left side is 'high pressure'. The only way out is at the top right, making the entire right side 'low pressure'. The only way the water can go from the high pressure side to the low pressure side is through 3 equal blocks, there is no tubing connecting them. There is no 'easier path'.

Be the water... you go into the left half. You are thinking, I am smart, why waste my time going through blocks when there is tubing straight ahead. So you keep going until you get to the bottom, where there is a plug. So now what? There are 3 doors out, the 3 blocks. All 3 are equal resistance. Now you think, I'm already at the bottom, it MUST be easier to go through the bottom block. But not all the water is 'at' the bottom yet. Since you stopped, all the water drops lined up behind you stopped as well, as if they had hit a 'plug'. Pressure builds because you are all being pushed on, and the only ways out have resistance. An equal number of you will go through each door, because you are all being pushed on by the same force.

Think of yourself in a lineup of people. You are all ushered into a hallway. The hallway has 3 doors out, which can fit only one person at a time. You all go in there and people at the input are still pushing, cramming you in harder and harder. Eventually you will have to start going through doors. Do you think you'll all get pushed out the last door? No way! people near the first door will say 'I'm outta here', and take off, same with the 2nd and third doors. As long as the pressure in the hallway is pushing harder than the flow will allow, an equal number of people will come through each door.

I don't know how many ways I can explain this - I think you may be one of the few people that won't be able to understand this no matter how anyone puts it.

I have built it and proved it. Build one and prove me wrong!
 

Dashock

Well-known member
Joined
Nov 22, 2007
Messages
1,921
Location
Toronto
If the temps are low and hes happy thats all that matters nice looking rig Vinister and him using it for a couple months now and nothings happened its allll goood in the hood.
 
Top