New block and impingement nozzles

[SysCrusher]

What does every one think of this one? Any questions or input to add? I really value other's comments and input. Only way to learn. Still needs some sanding though. Try a palm sander turned upside down rubbing the block back and forth. Works great.

With impingement I'v seen great results so it's basically trying different ideas with impingement.

[bigben2k]

What are the innards like, again?

[SysCrusher]

Originally posted by bigben2k
What are the innards like, again?

LOL very ugly. No micro-channels. A flat 4mm base that tapers gradually to 1mm in the center over the die area. I should of took pics. Same shape as Jaydee's. That oval shape. This one had channels in the beginning but I couldn't get the flow right so I dremmiled them out. Ever try micro-channels with a dremmil? I killed nearly every cut off disk and I don't have a mill handy to try ideas. :shrug:

A three piece design would work better with flow. Just like your doing with yours. Trying to go side to side with a radius design just doesn't work well for even flow.

[bigben2k]

Well, it's good!

You've got a few advantages going for you: using aluminium, you'll have a thin baseplate, and combined with a jet inpingement nozzle, you should be able to get nice results.

How did you come up with this design? What's your idea behind it?

[SysCrusher]

Originally posted by bigben2k
Well, it's good!

You've got a few advantages going for you: using aluminium, you'll have a thin baseplate, and combined with a jet inpingement nozzle, you should be able to get nice results.

How did you come up with this design? What's your idea behind it?

It started with my first block idea. Trying to make one that was easy to do. In my first block I used a 3/4 inch drill and drilled 3 holes then connected the three with a drimmil. I thought about the slight taper of the bit where I could get closer to the base which is at 1mm now. Then I started reading about studies of impingement and that got me thinking. So I put together a rough jet impingement that hit hit dead center and the impingement would flow up the sides in a radius. That first try didn't work I got no improvements. Then I thought more and figured it was submerged and recycling the heated water back into the impingement. So I decided to make the nozzle fit right into that taper so it was sitting 2 mm above. Thats where I seen better results. The reason why the mutliple nozzle is tapered but it just doesn't fit with this new block. Currently with my old block I get 8C above my water temp at load. Have no idea what this one will do.

The idea behind the impingement I have is to use the jet impingement like you normally would but to sit it as close to the base as I can around 1mm above the base plate. Then it would be impinging against the sides around the main jet impingement hitting the base which makes the stagnation point larger and keeps the water from flowing back into the jet. The taper idea came from looking at bridges and how they used archs and triangle shapes to hold the bridge up. I used that same idea using a slight taper to get a thinner base directly over the die while retaining strength. I'm just using that same shape upside down.

I have a good drawn out block idea to mill. Just looking into the cost of it.

[SysCrusher]

Here's a pic ben to help explain my idea.



Adding channels would likely be an improvement. Making the stagnation point larger in any way can do nothing but help.

[SysCrusher]

Some links for reading if anyone is interested.

http://www.electronics-cooling.com/h...01_may_a2.html


nice heat transfer calc.
http://www.coolingzone.com/Content/D...as/fcalc10.htm

http://web.cvut.cz/cp1250/fme/k212/p...ta/h06%5Ea.htm

http://home.icpf.cas.cz/vejrazka/web...ew_booklet.pdf

Anyone else have reading material? I need to visit the library.

[SysCrusher]

The new block ended up being to restrictive. The nozzle diameter at 1/16 was to small. So I'll try drilling it out to 1/8 and give it another try. After that, I doubt it will get better without a stronger pump. It's even with cathar's but I don't think it can get better than that.

[Gulp35]

More Pics Please!! from what I have read it looks like this.
[edit] I can't get the server to work right now will post later[edit]I got it to work[/edit][/edit]

BTW how do you get that middle skinnny area to bond to the top (sorry for the not too specific description)

[SysCrusher]

That pic is exactly what I want milled for a block.

I don't know what you mean by middle skinny area. If it's what I think it means, you can mill that top piece to fit. Or, what I did was to drill the three holes in the bottom plate. Then drill your holes for the top plate. Then Take a 1/8-28 male tap and thread the inside of a 1/4 threaded 1/2 barb. Then take a copper pipe 3/8 outside diameter and use a 1/8-28 male tap on both ends. Then use a brass cap with the same threads drill a 1/8 hole in the center of that cap. Then screw that cap on to one end of the copper pipe. Then insert one the other end of that pipe through the bottom of the plate up into the hose barb and screw it in till it's flush with the bottom of the plate. Then you can attach your bottom plate to the top plate and the brass cap should sit just 2mm away from the base so that it sits inside that tapered hole.



I made this new block thinner to do away with the brass cap. I don't know if it will perform better but it might since there is less room for the water to pool around the jet. See the nozzle in the pic above? I did the same with it and threaded the barb the same way but using a 1/8 threaded nozzle and cut the nozzle tip off and drilled the hole to 1/8. 1/16 was to small. So when I give that a try again, I'll take pics of my old block for you to see. The pic you just did would be the best way to do it considering if we had a mill.

By the way, that multiple jet you see didn't work. It simply doesn't create the same velocity and pressure as a single 1/8 jet. Would work well with a high pressure pump.

[SysCrusher]

Well the new block performed worse than my previous attempt. Why? I have no idea yet.

[Cathar]

Originally posted by SysCrusher
Well the new block performed worse than my previous attempt. Why? I have no idea yet.

Jet needs some distance from the base-plate to work properly, otherwise the flow doesn't gets all mashed because you're trying to force too much flow without enough of an "outlet". How close was your barb to the base?

I would not recommend getting closer than 4mm for most any setup/nozzle configuration, and depending on how large your nozzle is, you may see better results being a little further away (say up to 8mm).

I'm no expert on this, but from everything I've read the distance of the nozzle from the base-plate can have a dramatic effect, and it's not simply a case of closer is better.

[SysCrusher]

Originally posted by Cathar
Jet needs some distance from the base-plate to work properly, otherwise the flow doesn't gets all mashed because you're trying to force too much flow without enough of an "outlet". How close was your barb to the base?

I would not recommend getting closer than 4mm for most any setup/nozzle configuration, and depending on how large your nozzle is, you may see better results being a little further away (say up to 8mm).

I'm no expert on this, but from everything I've read the distance of the nozzle from the base-plate can have a dramatic effect, and it's not simply a case of closer is better.

Yes to close and it effects the flow as it hits the base. To far can be worse also but not as worse as being to close.

This time it was 2mm away. 1mm away was to close. Obviously 2 mm is to close also. From the baseplate to the bottom surface of the top plate is only 4mm. So I can't go no higher than that. I might if I thread the inside of the barb higher up. I think your right though. It looks like 4mm is where it needs to be. The distance does play a big role. I also found if your channel your water pools in is to deep then it will counter effect the impingement.

My previous block performs well at 2mm. Further away is worse, atleast at 7mm anyways. The channels are almost 7mm deep so I think it's because of the extra water around the impingement circulating back into it. The reason why I used the large cap to fill the space. With out it, it wouldn't perform good no matter if it was at 2mm or 7mm distance. I didn't try anything inbetween at the time because it was a pain to do but I got a easy way of testing distance now - A barb threaded on the inside. I seen no change between the multi-jet and the single jet.

I also wonder if I got the center inlet right. Looks to be directly over the die after looking at the impression on the bottom of the block. Then mounting and remounting the block can be an issue. I'm limited to .5C accuracy too. I'll give 4mm a try when I get the time and see what happens. Just a matter of screwing the nozzle further into the barb. The new inside threaded barb makes it easy.

I can say both blocks perform worse without impingement.


My post are to long. <- my new sig.

[Les]

Originally posted by Cathar
I would not recommend getting closer than 4mm for most any setup/nozzle configuration, and depending on how large your nozzle is, you may see better results being a little further away (say up to 8mm).

Be interested to see the work on which this recommendation is based.
For 1.5mm nozzles results* indicate that h increases as the nozzle/plate*** seperation(H) is decreased from 3mm to 1.5mm.
For other nozzle IDs I keep an open mind until I see some data (or undertand the theory).
However all the calculations I have done with Flomerics** show an increase in h with decreasing H ("down to" and "below" the calculator's lower limit of H=2D)

* http://widget.ecn.purdue.edu/~eclweb/jet_benchmark/
** http://www.coolingzone.com/Content/D...as/fcalc10.htm

*** Ceiling is lowered "along with nozzle" to 1.5mm (In light of Cathar's 12.13 am post)

[Cathar]

There needs to be 4mm of clearance from the base to the "ceiling" of the block. If I read what you're saying, you intend to push the nozzle further up the inlet, but the distance from the base to the ceiling of the block will still be 2mm? You will want true 4mm of clearance over about a 25mm diameter circle around the nozzle.

Again, just my thoughts, maybe not fact.

[SysCrusher]

Originally posted by Cathar
There needs to be 4mm of clearance from the base to the "ceiling" of the block. If I read what you're saying, you intend to push the nozzle further up the inlet, but the distance from the base to the ceiling of the block will still be 2mm? You will want true 4mm of clearance over about a 25mm diameter circle around the nozzle.

Again, just my thoughts, maybe not fact.

The distance from the base to ceiling is always 4mm. I just screw my nozzle up into the barb more to adjust the height of the nozzle. In my old block, base to ceiling was 7mm but I had to sit the nozzle 2mm away and use a cap to fill the space(a long explanation if you want to hear it. Has to do with eddy formation). Only way it worked in that block. But, I have a better way to set the height now.

[SysCrusher]

Originally posted by Les
Be interested to see the work on which this recommendation is based.
For 1.5mm nozzles results* indicate that h increases as the nozzle/plate*** seperation(H) is decreased from 3mm to 1.5mm.
For other nozzle IDs I keep an open mind until I see some data (or undertand the theory).
However all the calculations I have done with Flomerics** show an increase in h with decreasing H ("down to" and "below" the calculator's lower limit of H=2D)

* http://widget.ecn.purdue.edu/~eclweb/jet_benchmark/
** http://www.coolingzone.com/Content/D...as/fcalc10.htm

*** Ceiling is lowered "along with nozzle" to 1.5mm (In light of Cathar's 12.13 am post)

I have been looking through that and end up with the same results. So I think I was right going down to 2mm. But think of this. With our pumps that produce low 3 PSI pressure, velocity won't be that much so the distance won't matter until you get to far away. Say 7mm? Use a high velocity in that calc. Then you start to see a BIG difference with distances. I'm thinking as long as it's between 2mm and 6mm you will be ok with our typical pumps and anything inbetween 2mm - 6mm won't make a visual difference. Don't get me wrong, the distance will make a difference but not a difference we will ever notice. Atleast with our pumps. Nozzle diameter does make decent difference.

Now I tried the two blocks again. They both perform different with one being worse. I took both jets out. They both got worse but the difference between the two stayed the same. So this weekend I'll keep trying the nozzle distance from 1mm and up just to see. I have a hunch nothing will make a change because the effect at this velocity is so low to even notice.

After this I see about changing shapes of the nozzle and it's angle.