Pump Block
 

for those that I have shared this idea with, here is a rought solid works model of what I was trying to say. What do U think Brad?
http://www.borderbuyer.com/fixittt/blocks/pumpblk.jpg

how would the impeller spin with the shaft in the middle of the flow? shouldn't the shaft be at the edge of the flow?

it will be driven by an external 12 v motor

what about emi? im sure emi would get to a processor when its that close... awsome idea though

the motor would be located away from the block, either by a straight shaft, or a flex shaft, would reduse alot of vibration that way as well

I am confused as hell on this??? Why would you want the pump to be on the block itself? To save room maybe? I don't see how you will get much flow or volocity with that samll of a pump. Enlighten me on the subject.:)

On another note I am making a block that looks very similar to that today, minus the pump deal. Guess I should get started. Testing the the water cooler on the mill's motor.:)

well jay, I was thinking........

how much turbulance do you think is inside the waterpump housing where the impellar is?????????

Why not get all that turbulant surging water right inside the block.
Where it is needed most. Now what I have designed is the same size imellor that is on a 120 GPH pump. So with a strong 12 v motor it should still pump more then adaquite volume, and it is contained inside the block. Now if it works any better, or at all, is yet to be seen, But hey................. If no one else is gonna do it... I might as well.
http://www.borderbuyer.com/fixittt/blocks/pumpblk2.jpg

If that render is close to what you intend to make it will not pump water, impeller pumps draw water into the center of the impeller and throw it out to the sides. Of course that will be tricky to do since the drive shaft has to be in the center too, may have to mount it on it's side in a higher block.

well I thought of that too, abd I think that if the inlet to the impellar was the lowest poiint in the system, then gravity should feed the impellar to the point to where water flow would then take over once it is running. Say have a small res inline above the block.

Interesting design fix. Maybe you should set that up as a flow meter instead.

Hey, I just fabed a aluminum top for my spiral to cure the cracked acrylic from the 220watt pelt mishap.

with the blades where they are, all its gonna do is churn the water.

the blades need to be offset from the passage like this.

bah!! how do you upload an image to the thread!?!

Gravity feed doesn't work in a loop. Otherwise you get a perpetual motion machine.

Even if you power that thing you will get NO flow, even with another pump in the system, unless you move it off centre. Water is incompressible and as you have it, it will transport as much water forward as it transports backward giving no net flow.

Edward

Yeah After it was said here, I did some research, am working in solid works right now to fix the problem....... That is why I post here, to bounce stuff off of peopl

thanks guys. I think (with a good sugestion from this thread) I have the problem licked.

Ill post pics when I am done

hehe, a thread that was basically aimed at me and I didn't even reply for so long ;)


why not have the pump rotor right before the cpu? I think that in the length the water has to travel in that current design the water may have a chance to go back into it's normal laminar flow mode.

why not have it just after the first 180 degree bend, just before the water goes over the cpu die?

Here is a new thought all together. Instead of the pump being in the water block itself maybe make it in the top peice instead?

The top peice would have the pump in the middle so it can still be mountable somewhat and it will have a channel miled into it aswell from the pump to the inlet of the water block, and the outlet of the water block with go through the top peice out to the rad.

Did that make any since?

I had thought about that, but would you like all that weight sitting on your cpu?

That was one of my thoughts also.

I still like the idea of it being a flow meter. If you could set it up with an rpm sensor you could set the motherboard to shut down if the flow stops.

If you use AL it won't be any heavier than a Maze 2. Copper is much heavier than AL.

I was thinking of the fact that a pump is hanging off the side of the cpu block

I don't think it would be any heavier than a Copper HSF like the SK6 or MC462. Use the 4 holes and weight shouldn't matter. You can even come up with a support on the back side of the mobo if needed.

Ok, I have dont some research and talked to an engineer. and we have come to the conclusion that this updated render of the original idea will work, if a few minor changes are made. I moved in inlet closer to the impellar, and with an angled hose barb directing the flow colser to the impellor and also, I dont know how to draw it yet in solid works, but the chanber will have to be more of an egg shape. The chamber that the impellar sits in. and after close imspection of a few pumps I have here, that is the case, it gives the fluid a path. Cyntrifical force will do the rest.

Again I am bouncing it off you U guys.

http://www.borderbuyer.com/fixittt/b...p%20block2.jpg

also here is another design I have been working on as sort of a failsafe. (I AM GONNA GET THIS RIGHT!)
What do U think?

http://www.borderbuyer.com/fixittt/blocks/pumpblkv2.jpg

Those look very interesting. it would be very nice to have the pump on the block. It will make it much easier to put everything into the case.

That second one there looks cool. Havn't figured out how it works but it looks good.

Jay,
The impellar would be driven by a 12 volt motor. Basicly what I want to do is SMASH the water into the block directly. with the second render, I will be basicly having the water inlet right over the impellar blades like a waterbump has.

It was more a play thing when I started, now its turning into a quest. The way I figure it, everyone is basicly doing the same thing with waterblocks. We have reached that point. I want to go in another direction. Maybe I am just spinning my wheels, but the way I figure, someone has to.......... to say it was done.