Cooling Manifold vs inline system?

Which is better and why?

If i was to wc again id probably go for a 2 port manifold tbh rather than an inline system when it goes from 1 block to the next.
I love the look of JTFs stuff too - quality stuff.

Dunno, thoughts?

 

Subscribed! I am interested too, because I am planning a combination type flow, where it goes to CPU first, then to the manifold to be split off to the GPU and Northbridge. At least this way the CPU gets nice cold water.

 

I guess it depends on what your aiming for. If simple reliability is your aim then inline will suit perfectly.

I dont know how manifolds work but i'm guessing they split a single source flow into any number of different outputs. You may have trouble with keeping equal flow through all blocks and may get standing water.

P.S. Kong - have u tried making any water blocks with your CNC machine yet?

 

Yea, the inlet barb is slightly bigger letting more water in - but the water pressure from the pump (cause its best to use it straight from the pump i think) is high enough to force it more or less equally thru all of them.

Iirc the ehiem output barb is bigger than the inlet anyhow.

Dunno, i was looking at the manifolds at coolercases (?) that's all, no doubt that dude will pop in later.

 

The outlet barb on my Ehiem 1250 is smaller than the inlet - I think it's that way to increase the pressure...

I really like the idea of manifolds, but it has been pointed out in the past here, as well as being discissed extensively at the procooling forums, that the number of right angle turns in a system employing manifolds, such as those sold at coolercases reduces water flow rates fairly drasticly IIRC...

 

Not quite. The correct way to look at it is that the inlet is larger than the outlet. This is because there will be an underpressure (relative to the rest of the circuit) of water at the inlet. If this differential gets too large (i.e. less water manages to enter the pump than leaves it) the liquid will start to assume a gaseous state, i.e. bubbles will occur (this is called cavitation, if you want to bang a name on it). And bubbles, as we all know too well, are bad: the pump rattles, water wetter foams, bubbles travel through the circuit... you get the picture.

So to compensate for this, and make sure plenty of water manages to get in, the inlet is larger than the outlet. This is also why in my PC (Metaversa, see project logs) the tubing from res to pump inlet is 3/4" ID, while the rest of the system is 3/8" ID.

 

Indeed... that's pretty much how I was thinking about it I guess... I also have a larger bore tube to the inlet, although I am using 10mm barbs on my aquatube... which is probably causing problems in itself. Nexxo, how did you seal the new threads on your aquatube in the end?

 

Ta mate

I used manifolds as I prefer everything being in parallel, so a block with high flow restriction (i.e. my chipset block) doesn't hamper the flow in the rest of my system.

 

Hey I got to say JTF after having a look at your Watercooling Setup.................that is a nice job man

Its the first pics I've seen of the manifolds been put into use....keeps everything looking well neat and tidy.

You keeping the forth port for HD cooling or you got something else up your sleeve???

Like it

 

Teflon tape is all it takes, as the barbs are tapering. The thread on the flow switch is parallel, but some Plumber's goop in the thread combined with teflon tape again did the trick.

 

PSU or HDD ... haven't decided yet

The manifolds are great though, do you guys actually make them?

 

We don't actually make these ones but we have a company presently making some prototypes out of Perspex/Plexi for us along with some other interesting bits, which all being well should be available soon...........fingers X'd.

I'll get some pics up just as soon as I can!

There was one other thing I thought of JTF...........you tried those "O" rings for hose clamps rather than jubilee clips. They are permanent fixture but once you have your set-up sorted there is no reason not to use them and they really finish things of nice............just a thought mind you!

 

tbh, id prefer a copper version over a plexi (iirc the ones u sell are blue alu anodised?). Apart from keeping it all 1 metal, i dont really trust plastic stuff still, especially if ur putting half a dozen barbs in it and high pressure water from the pump.

 

yeah Bindibadgi.....the ones we got are blue anodised aluminium and very sweet looking they are too

It was a guy who goes by the name of Bladerunner who it has to be said is a total water cooling genius..............(his site for those who haven't heard of him ZeroFanZone) that gave me the idea.

I'm talking a big chunk of polished Plexi so there would be little fear of 10/12psi bursting it...................never heard of a Lucite top Danger Den block exploding under water pressure!

 

Yea ive spoken to bladerunner before about his stuff

Its not so much exploding - its cracking around the barbs, of which there are more of than a waterblock. DD had problems with cracking and leakage and ocpc went out of business cause they had to replace loads of blocks (ok it probably wasnt the main reason).

Im just cautios

 

hey Bindibadgi.......nothing wrong with being cautious man!

/.....thats basically why we are having a few prototypes made so we can test em to make sure nothing like that happens.

Just thought they would look nice with LED's embedded into them.

Strangely enough i am also looking at other materials as well including copper

see how it goes!

 

Cool.

Instead of LEDs (cause EVERYTHING is sold with an LED in it nower days - id just use UV reactive stuff in the water so you can see it go thru the block Although a purple LED might look good if the block was slightly frosted too

Copper block would be nice - but youd have todo something to the surface cause fingerprints come up read bad on copper and it goes grey/brown easily

 

hows about uv dye and a uv led

 

well i was thinking purple is the UV end of the spectrum so u get both.

How much are UV leds compared to normal? I would have thought more cause the respective electron drop gap (from population inversion to ground state) needs to be significantly smaller that anything else since the classic LEDs are red which is the other end of the optical spectrum? And blue/purple leds which are the UV end cost a shed load as it is.

 

dunno i got a bunch of blue and uv ones for a couple of quid from mcmad, who got them from hong kong lol
now i just need a bunch of 68ohm resistors to run them off the 5v line
but i cant bring myself to ask my dads credit card to spent 20p on resistors from maplins