I hope you're not referring to the windshield thing posted a little while back!

To take advantage of the 70 degree thing in the cupped design would require elliptical shaped cup holes, not to mention carving and drilling the jet tubes at that angle, which can be done, but requires quite a bit more effort and cost. For regular (and even not so regular) machining mortals, the 90-degree jet impingement works pretty damn well, and it's easier to boot.

Jaydee, the holes/cups on my design aren't quite as small as 1/16", although they were that small on the free-jet (no tubes) implementation I first tried.

I thought I might have been close. I was staring at the pic of the base and trying to calculate it in my head. I am suprised they are a little over though. Still be difficult to to anything special inside that hole. Going to be another challenge to work out the holes in the straws. I bet they are around 1mm though. I am moving forward with that REV. 1 design with the pits and multi hole jets that didn't work right. I got some freash insight from you and a few other things in mind. Might actually work. :D

I threw out some ideas at ocau a while back for Cathar, I'll post them here for comments.

diagrams on this page

Along with this comment:

The tubes would have to be at the cup depth and halfway through the cup so that there would be no lip at the transition point where the water leaves the cup and would naturally flow around the tube wall up and out on opposing sides of the cup. With this longer loop the main force of the exit flow would never hit the jet but slam into the nozzle wall above. In theory it should allow for better development of the stagnation area.

Adding the tubes would require a different approach to the baseplate, think smaller hunk of copper. We are talking very small dia. holes with extreme acuracy to make a prototype. As for the nozzle plate, just get centrifugal microtubes, I found some for a thousand for $16US. Snip them off and place them in conical shaped holes drilled into the nozzle plate and you now have protruding jets. I'm suprised nobody thought of this already, milling them jets sounds like no fun, and there's always more than one way to skin a cat.

How does one hold them in place?

Flare the ends? How to do that to something smaller than 1mm with 0.1mm walls without splitting it, bending it, snapping, etc?

Or do they already come flared for you?

Do they even come that small?

I did a quick google search and see down to 0.2ml capacity.

One of the jet tubes on the Cascade has a capacity of 0.025ml.

I do not believe a flare is necessary. They are tapered already. Insert them into a tapered hole. If the tapered hole was in 1/8" thick area of poly, then a matching 1/16" plate (with matching hole pattern) could be epoxied over to hold them in place (one dab at the four corders should do. Or maybe even epoxy each one in place. The tapered nature of the joint will be extremely strong so they should stay very straight.

The capacity is for the whole tube, you would only need the very tip.

Check my edit above g_f if you haven't seen it already...

hey G_f I was thinking of doing an idea like you are going to. Do you have a link to the tubes that you found?

I went to my local hobby shop and found some copper tubes that were Either 1/16" ID or OD (They where not clearly labeled but I believe them to be 1/16" OD)

Hi everybody !:)

I'm creating my first WaterBlock following Cathar's Cascade model.:) Actually, I'm designing the nozzle :
http://perso.wanadoo.fr/webzeb/Buse.png
http://perso.wanadoo.fr/webzeb/Tranche.png
It's composed of 61Holes (I've not yet chosen their diameter : 0.5mm up to 1.25mm:confused:). I hope 1mm is a good choice.;)Six of them are partially - 2/3 - Hidden by the pipe...

What do you Think about this design ? How it could be improved ?

thanks !:)

Welcome, or should I say: Bienvenue!

The design you presented is just the top, right? Assuming that the baseplate has the same pattern of holes, how do you propose to have the flow clear the area? Will there be tubes?

LeeJSmith made a similar block. We had to tune the tubes because the water was exiting faster than it was entering, reducing the jet effect. you ought to take that into correction ;)

Hole diameter : 1mm
"Cone" max. diameter : 1.5mm
Distance between two holes (centers) : 1.5mm
Distance max. between two holes : 12mm
Number of holes : 61 (6 partially hidden)
"Injection" pipe diameter : 12mm
Nozzle size : 20mm (cf. drawing)

bigben2k

Thanks !:)

You make an allusion about water evacuation ?

hi ben, webzeb

I still cant get the block to work any better with the smaller tubes.
to re-cap

I have a 3mm base plate with 2.5mm cups spaced 3mm apart. Each cup is 2.5mm deep with a flat bottom.

the tubes are 1/16x0.014 and go 1.75mm into the cups.

i have made a couple of changes since the pictures below and even made a new top to make sure the tubes are spaced the same as the cups and the tubes sit correctly in side them.


http://homepage.ntlworld.com/lee_smith/hexblock1.jpg http://homepage.ntlworld.com/lee_smith/hexblock2.jpg
http://homepage.ntlworld.com/lee_smith/hexblock3.jpg http://homepage.ntlworld.com/lee_smith/hexblock4.jpg

Yes.

You state that your "cup" diameter would be: max 1.5 mm. What's needed is the ID (inner diameter) and OD (outer diameter) of the tube (if any) that will project the water inside the cup.

To put it shortly, the hydraulic equivalent of the outlet should be greater than that of the inlet, so you have to make sure that the gap between the outside of the jet tube and the cup, is greater than the inner diameter of the jet tube.

Otherwise the coolant gets sucked right out of the cup, without any jet inpingement effect.

I think I should use copper capillaries : what do you think about this idea ?

That's what I recommended in the first place ;) Brass would be good too.

The only problem is that they can be tricky to cut: they will be crushed under a pair of regular scissors. I don't have a solution for that yet.

The solution i pretty easy.:)

Fill the tube with a steel wire, and them cut it with a tungstene cutter.;)

i used a diamond glass cutting wheel in a dremel. It cut's through the brass tubes i have very easy but you will need to file and de-bur after that is a real fiddly job i lost most of the skin of my finger and thumb in the process.

you can ge the tubes from here in the us. i get them from a model shop

http://www.specialshapes.com/brasstubing.asp

My design tuned with the modifications who seem to be needed :
http://perso.wanadoo.fr/webzeb/Buse2.png
http://perso.wanadoo.fr/webzeb/Tranche2.png
You can also see the opper entry tube...

I think you guys are putting way to many holes in these things. I just made one with 9 pipes/holes and it is working out very well. I just made the holes bigger to make up for the lack of quantity. With a 170GPH pump it is kinda restrictive but the temps are pretty good. I can only imagine what my Hydrothruster 500GPH pump will be able to do with it. Anyway will post more on that when I can get my damn capture card to work again....

How about building a "bathtub" for the block. A sort of watercooling the watercooling block.

Do you think this one could be better ?
http://perso.wanadoo.fr/webzeb/Buse3.png
19Holes (diameter : 1.5mm)

As long as it covers the area of the core of the CPU, it'll be OK. You can make it "future proof" by making it larger ;)

Good Idea.;)

In fact, I've designed it in order to cover the exact area of my tube...

And what about waste of charge ? Sprays' speed ? Efficiency ?