[Script Kiddie]

Get yerself a 5gal bucket and two via1300's
set both pumps in the bottem with the exhaust going out fo the bucket. Both working together in parrallel doubles the flow.
How about in series?

[SkiFletch]

I dont know how to prove it, but I can say that two working in series will NOT doulbe the flow. increase, maybe, but not double. when you say in parallel, do you suck from one bucket, split it with a y, go to the two pumps, and re-merge with a y? or do you leave out the second y? I would think that if you re-merged it, they wouldnt be quite double, but close...

[athlonnerd]

parrallel does double flow, series will increse the pressure. but make shure you have identiacl pumps, if ones trying to pull more outh of the other than it can deliver, then your wasting energy, if ones pumping more in to the other than it can handle, then the faster pump will effectivle speed up the slower one, which also wastes energy. however, you would haave to step the tubde diameter down after the last pump, other wise that pressure wouold be lost, i do believe.

[JudgeDredd]

Parallel willl get you faster waterflow, but not exactly double.

[nikhsub1]

It also gives you double the heat.

[Diggrr]

I setup my system as a series of a Gemini spiral, and BeCooling's chipset and gpu blocks, on an ViaAqua 1300 and 3/8 tubing.
Flow dropped so amazingly much, that I re-introduced my Danner 250gph pump back in, in series on the opposite side of the loop.
Flow increased 30% (same test, same method, same water, with the system complete).

I wouldn't expect double flow with them in parallel, they would give each other backpressure at the "Y". Maybe it would be more than series and maybe not, but I haven't tested that, I don't have "Y"'s.

[Silversinksam]

Personally my opinion is that dual pumps are a good idea as its nice to have a backup. I used to be in avaiation and everything has a backup....I was going to run dual pumps but ultimatly just decided on using a reliable Eheim.

[Script Kiddie]

Oh yea, where they joined would be back pressure... hmm.

[#13]

Couldn't you just run the thing in parrallel throughout? Maybe share a reservior or radiator, but otherwise be indepentant. Sure, you'd have twice evervthing, and an engineering nigtmare of a waterblock, but hey, who said anything about "practicality"? Just of the top of my head, would two independant, smaller pumps introduce less heat into the system, while delivering comparable flow? I know they'd both produce heat, but they would also be separate. Would the heat gain still be cumulative in the same way that it would be w/ inline pumps? Just a thought...

[athlonnerd]

i think you can reduce back pressure, by making the final tube in the y, double the crossectional area as the two joining tubes.

[gone_fishin]

Something like this? The Y would only need to adapt from two half inch id tubes to one 3/4" id tube, continue through the block and to the resevoir. It would accomodate double the volumetric flow without additional back pressure. And since there are two pumps why not two rads to take care of the additional pump heat.

Block diameters of barbs and area of each in inches,

.785 1
.44156 3/4
.30664 5/8
.19625 1/2
.15025 7/16
.1104 3/8
.07666 5/16
.049 1/4
.0276 3/16
.0123 1/8

So two 1/2" and one 1/4" is exactly equal to one 3/4" barb.
Anyone try something like this?

[#13]

I don't know if anyone has tried it -- maybe if a few more people looked at this thread...

[overdoze]

Pump in series is better than in parallel. I have tested in parallel and the flow almost did not increase. So I put the pumps before and after the rad instead. Got about 30% increase flow

[JudgeDredd]

Originally posted by gone_fishin
Something like this? The Y would only need to adapt from two half inch id tubes to one 3/4" id tube, continue through the block and to the resevoir. It would accomodate double the volumetric flow without additional back pressure. And since there are two pumps why not two rads to take care of the additional pump heat.

Block diameters of barbs and area of each in inches,

.785 1
.44156 3/4
.30664 5/8
.19625 1/2
.15025 7/16
.1104 3/8
.07666 5/16
.049 1/4
.0276 3/16
.0123 1/8

So two 1/2" and one 1/4" is exactly equal to one 3/4" barb.
Anyone try something like this?

You probably couldn't find a waterblock with 3/4" fittings, and probably not wide enough channels to take advantage of it. It'd probably be better to use 1/2" where your 3/4" hose were, and 3/8" where you had the 1/2". Very good idea though, it'd get the CPU close to ambient temps. The problem is, with a setup like that, you'd probably want to watercool the chipset and GPU as well, so would you put those in series or parallel? I'd probably put the CPU on one pump and the GPU and chipset on the other... Hmmm...

[masitti]

You could also use one pump to cool GPU/NB/HD, and the other to cool your CPU.

[gone_fishin]

Originally posted by JudgeDredd


You probably couldn't find a waterblock with 3/4" fittings, and probably not wide enough channels to take advantage of it. It'd probably be better to use 1/2" where your 3/4" hose were, and 3/8" where you had the 1/2". Very good idea though, it'd get the CPU close to ambient temps. The problem is, with a setup like that, you'd probably want to watercool the chipset and GPU as well, so would you put those in series or parallel? I'd probably put the CPU on one pump and the GPU and chipset on the other... Hmmm...

That would be no problem with my custom waterblock and seeing as it has four outlets, it is perfect for branching off to the chipset, northbridge, and two other locations. It could be a good solution to go from 3/8" to half like you say with a couple of cheap pumps.