DIY water block Finally a good one
 

Here's something every one can do easily. It's currently getting me 6C idle over my water temp. I hope once I use copper it'll do better. Now if I can get my radial micro finned version to do the same! The nozzle puts out a good stream. The draw back I found is bleeding. Once an decent size air bubble goes through the nozzle, it gets ablitterated into tiny micro bubbles. LOL With a res it bleeds pretty easy though. Hopefully some one else can try it out and tell me their findings. Open to all critisism(sp?). Tell me what you think.

Take two .25 - .5 inch copper plates. I'm using aluminum right now because it's free for me to test with it. 3/4 - 1 inch drill bit, 1/8 drill bit, 3/16 drill bit, 7/16 drill bit and chamfer(think that's what it's called). 1/4 npt tap and 1/4 tap. 3/8 copper pipe. 3/8 brass cap with a rounded top.

Drill your 3/4 to 1 inch hole dead center over the die area down to 1mm within the bottom. Then take your chamfer of the same size so that you end up with 1mm in the center and 4mm from the edges of the hole. Then drill your 2 3/4 to 1 inch holes on either side. Then cut the remaining metal between the holes. Take your top and drill your 7/16 holes dead center over your bottom plate holes. Tap them with a 1/4 npt tap. Screw in your barbs. Then take a piece of 3/8 copper tube about 1 inch in length. Take the 3/8 brass cap drill a 1/8 hole in the center. Flux the end of copper tube and fit your brass cap and fit it to one end of the tube. Then solder it in place. Insert the other end through through the bottom of your top plate into the barb so it's nice and snug. The top of the brass cap should be no higher than 2 mm from the bottom of your plate center hole. Might need to tap it in. I found if you set your top plate ontop of your bottom plate while your drill your 3/16 mounting holes and screw holes you will get better lined up holes. Then your done.

Cheap MS paint pics. Wish I had what you guys use.

http://www.comebackandplay.com:8080/~jay/blockdraw1.jpg

http://www.comebackandplay.com:8080/~jay/blockdraw2.jpg

I like your Idea, but I only get red "x"'s for pics.

I don't know why. Maybe right click on the pics and copy the address from the properties.

i see the pics, although i dont really get the idea.

The idea was to use jet impingement. The dome you see at the bottom of the top plate is a nozzle with a 1/8 hole. It's just a brass cap I drilled a hole in the center. It has to be a rounded cap though. I tried a flat cap and it didn't work well probably because it needed a flat base.

I see the pics now, but have a question. Why do you need to shrink down to 3/8" from 1/2" and then having the cap.

At this time it is the only way I can affix the cap. It doesn't hurt flow since your going to 1/8 inch hole anyways. If you thread the center barb hole completely through you should be able screw in a nipple that has been cut to size with the cap soldered on to it or the cap screwed on to the nipple. You might need to file the threaded side of the cap down to get the correct height if needed.

If you have 1 inch holes a bigger cap and threaded copper pipe that fits the barb holes you could do it easily just using the threaded barb hole. That's my next step in this block. Then the height of the nozzle can be completely adjustable. Then I like to add micro channels with the center cut out so that the cap will fit in perfectly directing the flow. I will round the corners of the cap first. Problem is all I got is a dremmil.:confused:

The cap keeps the flow of heated water from swirling back and getting recycled into the impingement. So it helps keep a constant cooler water where it needs to be.

The slight taper also keeps the base from flexing or warping. The same reason builders use arches and triangles when building bridges - it's just upside down in this case. It allows a thinner base directly in the center.

idle temps tell us nothign, even load temps tell us nothing if you dont specify your mode of measurement and room temp or even better yet, water temperature.

Yeah......what he said x2

Sorry but:
That DIY means?
I have used the tool of search in the forum but I don't find meaning.
My English is sucks and Altavista Translations it´s more sucks even.
:(

Ketchak.
www.devilmaster.cjb.net

LOL Hey take it for what it's worth. If you think it's purely "BS", then so be it. No concern of mine. :mad: I'v been laughed at and passed off since I started this one. I really don't care - It's my hobby not a job. I do this for fun and post my results for the simple guy to use. It's easy to get good results without getting to fancy using simple basic jet impingement the rest is common sense. My 6C over water temp is the DeltaT. That's the difference. Temps taken from a thermal probe accurate within .5C stuck just before the inlet of the block. The difference jumps between 5 - 6C, so that tells me it's somewhere within that .5C that I can't read. Temps taken from cpu die using a soldered on temp reader from the cpu pins.

Room air = 25C
Water temp= 28C
Cpu die temp = 34C
Cpu full load = 37C

Doing this with a XP1700 @ 1875 150FSBX12.5 1.75V .
For a 2v volt add 1C more to the readings.

DIY = Do it yourself. :) Slang sucks sometimes don't it.

Thanks SysCrusher
Your core is Palomino or Tbred?

Ketchak.
www.devilmaster.cjb.net

Palomino AGOIA. I like it so i'll stick with it for sometime. My memory isnt to good. I need to get some DDR333 -400. Memory get's unstable past 150 -155FSB. All I have is DDR 266.

Nozell like this?

http://www.customcooledpc.com/nozzel.jpg

Yes, just like that.

I pm'd you a pic of what I was trying to say about milling a weird nozzle design.

Mill should be able to this. Don't see why not. Hell you could probably do it will a drill press and a swiveling vice with an end mill.

is that nozzle that you show sort of like a parabola or is it more of a slope down toward the exit?

It is just a slop. Like a standard funnel.

parabola or hyperbolic shape would be best if you can find a way to do it.

If you wanted to drill a funnel shape, I think there are drill bits with a sharp pointed end (I'm not sure but I think that I have seen one). But if you need a Parabolic Nozzle you could use multiple bit drilling down on the same spot.

I'll try to describe it:
First use the width of the nozzle exit and drill althe way through the nozzle peice.
Second move to a larger drill bit and drill down not as far as the first but just so.
Third, repeat trying to make a parabolic shape (It would actually be staggered).
Fourth, Take your dremmel or some grinding tool and smooth out the stagering.

I'll post a pic later, (I have a history paper to write:(:()

or a tapered mill bit. I'm in the process of trying to make the block thinner. As it is now, my block is a full one inch thick. I have two pieces of .25 inch Aluminum to try out. Think I'll use a piece of lexen for the jet or find a way to plug up a hose barb so that it tapers a bit and drill a 1/8 hole in the center. The end of the nozzel has to be 2 mm from the base with an extrusion from the top plate to fit in the recess of the bottom plate. The trick is to keep heated water from recycling back into the impingement.

Pic to help explain.

http://www.comebackandplay.com:8080/~jay/blockidea2.jpg

I'm thinking if I go to the .25 inch bottom plate I won't have to use my hacked job jet impingement like my first post above with a full .5 inch plate.

I know what you mean, but i think there are ways to do it all in one step. They also make "counter sinking" bits that makes the ovel shape on the material for the screw to fit flush with the top. I have one but I think it's angle is not steep enough.

Here is the pic
http://tech.thewalkerschool.org/vanderclip/nozzle.jpg

I was wondering SysCrusher if you have thought of an asymetric block to keep water flowing in one direction instead of splitting to flow and causing a chance of unbalanced flow.

My current block that I am working on is a micro channel idea, mainly because it is easier to model on the computer. Right now I have the block with 9 , 3mm deep, 1mm wide channels over the die. with this I could handle a 520 W Processor with a 1C water temperature rise. and this is only a side to side flow block!!!

Next block I'm working on including nozzles.

http://www.comebackandplay.com:8080/~jay/blockrev3.jpg

It's mainly a experiment with jet impingement plus to make a thinner block. Can fit the multiple jet nozzle in this block yet. My current block it will fit but havn't tried it yet.

I can't see the pic. If you want, email it to me and I can host it for you.

I considered an asymetric design but from my experience and results the nozzle should be the shape of the design your using. So i'm stuck with round nozzle shapes. I have a slight problem with unbalanced flow with my new block as I could tell because more water seemed to flow from one side more. Actually shooting out one side more. I could hear the impingement hissing. It could be because I had trouble tapping the outlet holes - they got a little crooked. .25 inch isn't easy to get a tap straight in it.

I have no way of milling micro-channels nicely. If I could, I would be doing that and trying to incorporate my impingement ideas into that.

Your block idea sounds like it's coming along nicely. Sounds like a performer.

I may make a jet impingement block just to see how it performs, btw my pics may be going down more often because my school's server ( where i host my pics) is having troubles.

Hey I see your pic now. That's exactly the idea I had just without my tapered base which i'm wondeing if the taper base helps any. The only thing I would change is move that nozzle closer to the base. What happens is that the surounding water that the jet is passing through effects the volocity of the jet stream which slows it down and lowers the pressure against the base making the stagnation area smaller. Plus the water that just hit the base is now heated up and will tend to swirl up and back into the jet stream and recyclng that warm water. Closer to the base and that will be avoided. Also make the nozzle so that it's flat at the tip. That will keep old water from getting into the stagnation area and help keep the volocity up after it hits the base. Plus it helps guide it across the base. That was my first mistake on my first block.

Doing a submerged jet inpingement in a water block is different. My first time I looked at as a water jet hitting a plate and forgetting it was submerged. If you could look at the jet impingement magnified you would see small vortexes(sp?) being created right at the surface of fast moving water the impingment creates. The water surrounding it effects it greatly. After the jet hits the base it thins or eliminates the boundary layer and spreads out with high velocity. Called the stagnation area. This fast or higher velocity water will actually create a vortex of heated water from the water sorrounding the jet impingement that will help heat the new cooler water hitting the base. We don't need that. So it's best to keep the nozzle close to the base, make the tip of the base flat or the same shape as the base so that it covers the area just above the stagnation area. This will make sure that the impingement is not effected by any old heated water and it's velocity won't be effected by surrounding water.

The idea is to keep all stagnate water away from the impingement.

I'd put a jet impingement block done right on par with performance to a micro-channel block any day. Mixing the two just right might be even better.

My pic was only a demonstration of a was you could build a Parabolic Nozzle. after reading all of this I think that It seems easier to build one of these blocks than the block that I am going to. I'll just buy extra Copper to make expirimental Blocks.

The reason for this is I believe that performace on par with cathar's block could be accieved without the reletively huge price.

I'll post a pic tonite of a impingement style block that I had in mind.