Go Back   Pro/Forums > ProCooling Technical Discussions > General Liquid/Water Cooling Discussion > Water Block Design / Construction
Password
Register FAQ Members List Calendar JavaChat Mark Forums Read

Water Block Design / Construction Building your own block? Need info on designing one? Heres where to do it

Reply
Thread Tools
Unread 02-03-2003, 11:11 AM   #1
bigben2k
Responsible for 2%
of all the posts here.
 
bigben2k's Avatar
 
Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
Default Jet nozzle (for LiquidRulez)

(this thread is meant for LiquidRulez and I to discuss nozzle designs)

You mentionned needing a nozzle "shaped like an 180 deg. arc to fit the inside circumference of the 1/2" NPT fitting".

Now I'm thinking you mean this: am I right?
Attached Images
File Type: jpg lr nozzle.jpg (3.2 KB, 519 views)
bigben2k is offline   Reply With Quote
Unread 02-03-2003, 11:27 AM   #2
Roscal
Cooling Savant
 
Roscal's Avatar
 
Join Date: Apr 2002
Location: North of France
Posts: 198
Default

A hyperbolic shape sould be better to kill pressure drop in a nozzle like attachement. Parabolic and hyperbolic are in any way better than a brutal change of section...
Attached Images
File Type: gif nozzle.gif (1.9 KB, 493 views)
Roscal is offline   Reply With Quote
Unread 02-03-2003, 11:30 AM   #3
bigben2k
Responsible for 2%
of all the posts here.
 
bigben2k's Avatar
 
Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
Default

thanks. I was going to advance this standard:
Attached Images
File Type: jpg nozzle spec 4b.jpg (54.4 KB, 495 views)
bigben2k is offline   Reply With Quote
Unread 02-03-2003, 04:53 PM   #4
SysCrusher
Cooling Savant
 
SysCrusher's Avatar
 
Join Date: Dec 2002
Location: Florida
Posts: 256
Default

Quote:
Originally posted by Roscal
A hyperbolic shape sould be better to kill pressure drop in a nozzle like attachement. Parabolic and hyperbolic are in any way better than a brutal change of section...
This is the best way to do it. Mine right not is just a change of size that doesn't change gradually which doesn't help pressure drop one bit but is all I got right now until I find something better.

Ben, try to keep away from a to steep of an angle on the inside of the orfice outlet. To much of angle and you end up with a spray pattern. I tried it with a nozzle from mcmasters like that and like you said it didn't work. You need a straight stream. Height of the nozzle from the base is very important depending on the pump pressure. Also, to far away it will allow the back pressure of swirling heated water to get recycled back into the impingement. In other words, if it to far away not all the water will hit the base and flow across it, some gets bounced back or swirls back. A submerged impingement acts quite different than the ordinary run of the mill impingment.

Last edited by SysCrusher; 02-03-2003 at 05:06 PM.
SysCrusher is offline   Reply With Quote
Unread 02-03-2003, 05:46 PM   #5
bigben2k
Responsible for 2%
of all the posts here.
 
bigben2k's Avatar
 
Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
Default

"submerged" is the key word.

The above is an ASME spec for nozzles. There's a 10 degree taper on the inside lip, and I believe that it may be for air, but I could be wrong. Either way, it should work fine.

I also believe that the ASME spec should reduce the length of the nozzle (entrance to outlet) to a minimum, instead of randomly applying a curve: one must remember that the longer it is, the more restrictive it will be.

In Radius, I was looking at a huge pressure drop, when considering a 3/16 nozzle size, but I made the calculation as an estimate, by calculating the pressure drop over 1 inch of 3/16 tubing: one inch would be too long, my pump couldn't push that kind of flow through it.

The pic in the first post would be the absolute worst kind of nozzle, as it would indeed spray the coolant everywhere. The desired effect is straight down, for jet inpingement.

In going over the ASME specs, I found that the outlet pressure can be expected to be 80% of the inlet pressure. In an orifice plate (flat piece, drilled straight), the pressure differential can range from 60 to 80.
bigben2k is offline   Reply With Quote
Unread 02-03-2003, 06:38 PM   #6
SysCrusher
Cooling Savant
 
SysCrusher's Avatar
 
Join Date: Dec 2002
Location: Florida
Posts: 256
Default

Quote:
Originally posted by bigben2k
"submerged" is the key word.

The above is an ASME spec for nozzles. There's a 10 degree taper on the inside lip, and I believe that it may be for air, but I could be wrong. Either way, it should work fine.
I doubt the 10 degree taper will hurt anything either.
Quote:
Originally posted by bigben2k

In Radius, I was looking at a huge pressure drop, when considering a 3/16 nozzle size, but I made the calculation as an estimate, by calculating the pressure drop over 1 inch of 3/16 tubing: one inch would be too long, my pump couldn't push that kind of flow through it.
What pump you got? The nozzle in mine right now is 1/8. As for the flow, I couldn't tell you what it is now but I suspect it's been cut in half atleast. It's working though much to my amazement. But you have the fins in the center to so thats different, I forgot again.:shrug:
Quote:
Originally posted by bigben2k

The pic in the first post would be the absolute worst kind of nozzle, as it would indeed spray the coolant everywhere. The desired effect is straight down, for jet inpingement.
Kind of like mine now with the first one but more flat on the inside. Adding more flow through it and pressure it will start to get a pattern spray to it much like a garden hose. Not bad though with the hydor pump. Second is the way to go.
Quote:
Originally posted by bigben2k

In going over the ASME specs, I found that the outlet pressure can be expected to be 80% of the inlet pressure. In an orifice plate (flat piece, drilled straight), the pressure differential can range from 60 to 80.
Thanks for sharing and pointing me to the ASME specs! I'm really interested in the nozzle you found. You'll be amazed when you try it out. I have a feeling your radius will knock the socks of them with impingement. Block I'm using now has NO fins/channels. I was shocked to see the results with just my crude version of jet impingement though mine is off center of the die! I didn't realize the cpu die was offset.

Last edited by SysCrusher; 02-03-2003 at 06:59 PM.
SysCrusher is offline   Reply With Quote
Unread 02-03-2003, 07:05 PM   #7
bigben2k
Responsible for 2%
of all the posts here.
 
bigben2k's Avatar
 
Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
Default

Thanks!

I've got a Little Giant 2-MDQ-SC. It'll do > 500 gph at 1 foot of head.

http://www.lgpc.com (look under magnetic drive pumps)

The spec I posted, is actually for when the outlet diameter is less than half of the inlet diameter. I've got the complete page from the book. PM me an addy for it (I've got a scan on file, somewhere...)
bigben2k is offline   Reply With Quote
Unread 02-03-2003, 10:32 PM   #8
LiquidRulez
Cooling Savant
 
LiquidRulez's Avatar
 
Join Date: Sep 2002
Location: In Hell
Posts: 322
Default

No. I meant 180 degrees as in the impingement nozzle itself. Kind of like a cresent moon without points. but tangent arcs instead.
But unlike picture below(didnt have time to change), I could cut it with a tapered end mill so that it is hyperbolic as mentioned above.
But for now, Id just like to be able to know how to calculate the area of the impingement nozzle as shown in the attachment, so that it creates the proper amount of impingement for a 1/2"NPT barb. Then Ill go from there.
Attached Images
File Type: jpg part2.jpg (42.0 KB, 434 views)

Last edited by LiquidRulez; 02-03-2003 at 10:38 PM.
LiquidRulez is offline   Reply With Quote
Unread 02-04-2003, 11:24 AM   #9
bigben2k
Responsible for 2%
of all the posts here.
 
bigben2k's Avatar
 
Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
Default

Ok, I'm still not entirely clear on your goal, but let's see:

First, your opening (in your pic) would more than likely be blocked by the material of the barb, so you'll have to adjust it, or use a bigger barb (5/8?).

The main problem with jet inpingement is that it requires a lot of pressure, so a good pump. Because of the type of pumps we use, we're seriously limited in the size of the nozzle we can use.

Your arc may have the equivalent opening as a 1/4 round hole, and your pump may not be powerfull enough to make a jet that will be fast enough.

Now, for what I didn't get into... Why on earth would you use an odd shape like that? If you want to keep your design secret, that's fine, but you might consider using multiple smaller jets. It's an option.

The hyperbolic curve would be impossible to implement on that curved opening: it would cut into the barb's thread. You might consider using a much larger barb, so that there's room for a nice curved entry point. Maybe 1 inch? (add a reducer). I've used 1 1/2

I'll run the numbers, with the information that you gave me.
bigben2k is offline   Reply With Quote
Unread 02-04-2003, 11:47 AM   #10
LiquidRulez
Cooling Savant
 
LiquidRulez's Avatar
 
Join Date: Sep 2002
Location: In Hell
Posts: 322
Default

The inlet and nozzle are not to scale and are just for representational purposes only. The nozzle opening will be cut to accomodate for the barb itself, so it will be flush with the inner bore.

I considered using multiple smaller jets also. Just thought it would focus more of the impingement into the outside where its most important being designed that way. Also to avoid any dead spots in my design.

A hyperbolic cut in that nozzle design can be cut with a tapered end mill.
Ive concidered making the top plate thick enough to accomodate the threaded part, and taper it from the bottom of the threads to the end of the nozzle. Im trying to go about it as easy as possible to mill.

You got PM BEN
LiquidRulez is offline   Reply With Quote
Unread 02-04-2003, 02:10 PM   #11
bigben2k
Responsible for 2%
of all the posts here.
 
bigben2k's Avatar
 
Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
Default

Ok, the area of the opening can be calculated as follows (but that's not exactly relevant (see below):

Area of outer circle (i.e. barb inlet area) minus area of inner circle. This'll give the area of a ring. Now I assume you're fixing to make this opening with an endmill, and so, if it runs in a 180 degree circle, you'll be left with an opening that's half of the calculated area (above), plus the area of the tip of the endmill.

Of course we're trying to get to where we know what size endmill to use.

I'll have to calculate the hydraulic equivalent of a slit: the flow restriction isn't the same as the corresponding sized round opening. Once I know, using the flow specs calculated (from above), I can determine the expected flow rate, and then calculate the slit size.

I'll try to optimize it all for your pump to be at its highest efficiency.

So now I need to know the material thickness, through which this slit will be cut. Also, I am assuming that there are no other restrictions, from the slit's outlet, and the block's baseplate...
bigben2k is offline   Reply With Quote
Unread 02-04-2003, 05:28 PM   #12
LiquidRulez
Cooling Savant
 
LiquidRulez's Avatar
 
Join Date: Sep 2002
Location: In Hell
Posts: 322
Default

Well, I,d like to make the mat. thickness for the nozzle around 0.10", but may go with 0.20". So if you could give me calcs on both, that would be a great start.
There is nothing under the nozzle but the baseplate.
LiquidRulez is offline   Reply With Quote
Unread 02-05-2003, 09:22 AM   #13
bigben2k
Responsible for 2%
of all the posts here.
 
bigben2k's Avatar
 
Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
Default

Great, no problem. I've got all the info I need. I'll get back to you before the weekend's over.
bigben2k is offline   Reply With Quote
Unread 02-07-2003, 11:49 PM   #14
LiquidRulez
Cooling Savant
 
LiquidRulez's Avatar
 
Join Date: Sep 2002
Location: In Hell
Posts: 322
Default

I was also kind of hoping that Cathar would have some input on this nozzle..............hint, hint
LiquidRulez is offline   Reply With Quote
Unread 02-08-2003, 03:13 PM   #15
koslov
Cooling Savant
 
Join Date: Jul 2002
Location:
Posts: 264
Default

I made a plexi nozzle yesterday in 1/4" acrylic, just to see if I could do it. It is relatively hyperbolic, but obviously it is hard to tell from the pic.

Bottomside view:


Top view:



Blackeagle is sending me some 1/2" Lexan, which I plan to tap into the top .25", then drill the nozzle below it in the remaining .25". Should make for a relatively simple nozzle design without using multiple layers.
koslov is offline   Reply With Quote
Unread 02-09-2003, 09:36 AM   #16
bigben2k
Responsible for 2%
of all the posts here.
 
bigben2k's Avatar
 
Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
Default

Some progress...

I had to chart the PQ curve for what I assume is your pump: Danner MagDrive 3.

I then used the pressure drop graph from Bill's radiator roundup, for the Big Momma, to calculate the remaining pressure available (for the block).

Observations:
#1: that pump has an optimal power range in a low pressure setup, and that range happens to be way outside what you can expect in a loop/rig for a PC. How attached are you to that Danner pump?

#2: Here's the data I've put together:
Pump pressure at 4.75 gpm: 1.5 psi (red)
Pump pressure at 3 gpm: 2.6 psi (yellow)
Pump pressure at 2 gpm: 3.0 psi (green)
Pump pressure at 1.5 gpm: 3.3 psi (dark blue)
Pump pressure at 1 gpm: 3.6 psi (light blue)

Radiator pressure drop at 4.75 gpm: very high
Radiator pressure drop at 3 gpm: very high
Radiator pressure drop at 2 gpm: 1.6 psi
Radiator pressure drop at 1.5 gpm: 1.0
Radiator pressure drop at 1 gpm: 0.3

Remaining pressure at different flow rates:
2 gpm: 1.4 psi
1.5 gpm: 2.3 psi
1 gpm: 3.3 psi

Given that the power is calculated as dP x flow rate:
2 gpm: 2.8 (unknown units)
1.5 gpm: 3.45
1 gpm: 3.3

So my conclusion is that your target flow range should be 1.5 gpm, and preferably closer to 1 gpm, if not 1.5. This will give you the maximum amount of "hydraulic power" for your nozzle.
Attached Images
File Type: jpg danner3.jpg (14.7 KB, 310 views)
bigben2k is offline   Reply With Quote
Unread 02-09-2003, 12:27 PM   #17
SysCrusher
Cooling Savant
 
SysCrusher's Avatar
 
Join Date: Dec 2002
Location: Florida
Posts: 256
Default

I think your wrong on that or looking at it the wrong way. What you calculated as target flow rate is what the flow rate that nozzle will produce with the chosen pump plus any restrictions before it. The idea is to take advantage of the pumps pressure just before it can no longer over come the back pressure which is not much based on what I found out. Even at my 1/8th nozzle I still see signs of pump cavatation. Use the pump outlet to go straight to the block so there is no restrictions before the block. Keep the inlet of the pump as less restrictive as you can.

Check out the Little Giant 5MDSC compared to the one you got now. The 5MDSC is made for pressure applications.
SysCrusher is offline   Reply With Quote
Unread 02-09-2003, 02:52 PM   #18
bigben2k
Responsible for 2%
of all the posts here.
 
bigben2k's Avatar
 
Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
Default

Call me dense, but I don't understand what you mean.

I took the pump's curve, and reduced the curve by what the heatercore would do, leaving the rest of the pressure available for this nozzle (well, it includes all the tubing and fittings, but I'm assuming that that will be relatively negligeable).

If myv65 was here, I'm sure he'd show us how to do it the right way (but he's probably still busy laughing ).

The next step at this point is to calculate what the nozzle should be like.
bigben2k is offline   Reply With Quote
Unread 02-09-2003, 04:51 PM   #19
murray13
Cooling Savant
 
murray13's Avatar
 
Join Date: Sep 2002
Location: Omaha, NE USA
Posts: 216
Default

BB2k don't start doubting yourself, you did it right!
murray13 is offline   Reply With Quote
Unread 02-09-2003, 05:07 PM   #20
SysCrusher
Cooling Savant
 
SysCrusher's Avatar
 
Join Date: Dec 2002
Location: Florida
Posts: 256
Default

Quote:
Originally posted by bigben2k
Call me dense, but I don't understand what you mean.

I took the pump's curve, and reduced the curve by what the heatercore would do, leaving the rest of the pressure available for this nozzle (well, it includes all the tubing and fittings, but I'm assuming that that will be relatively negligeable).
I understand now what your saying and that makes sense to me.I just mistook what you said. Your right.

Quote:
Originally posted by bigben2k

If myv65 was here, I'm sure he'd show us how to do it the right way (but he's probably still busy laughing ).
LOL I'm sure he is. I love to hear his input though.

Next is will the nozzle be able to over come the surface geometry of the block.
SysCrusher is offline   Reply With Quote
Unread 02-09-2003, 09:54 PM   #21
SysCrusher
Cooling Savant
 
SysCrusher's Avatar
 
Join Date: Dec 2002
Location: Florida
Posts: 256
Default

Quote:
Originally posted by bigben2k
Call me dense, but I don't understand what you mean.
Hey ben. I totally see what you meant and understand now. It just slapped me in the face. A pump produces it's max pressure at 0 flow. Of course you don't want zero flow.
SysCrusher is offline   Reply With Quote
Unread 02-10-2003, 09:37 AM   #22
bigben2k
Responsible for 2%
of all the posts here.
 
bigben2k's Avatar
 
Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
Default

Kewl

I'll be getting into the nozzle calculation shortly, then I'll be converting it to an odd-shaped equivalent.
bigben2k is offline   Reply With Quote
Unread 02-10-2003, 02:51 PM   #23
LiquidRulez
Cooling Savant
 
LiquidRulez's Avatar
 
Join Date: Sep 2002
Location: In Hell
Posts: 322
Default

Thanks Ben
Cant wait.

I was concidering using the Danner MAgdrive pump becuase it puts out less wattage than my Little Giant does.
I used to use it in my system, but found that it kept my temps about 5-7c higher. So i yanked it and replaced it with the Danner.
LiquidRulez is offline   Reply With Quote
Unread 02-15-2003, 10:39 AM   #24
bigben2k
Responsible for 2%
of all the posts here.
 
bigben2k's Avatar
 
Join Date: May 2002
Location: Texas, U.S.A.
Posts: 8,302
Default

Here's a link to a calculation of the hydraulic equivalent:
http://www.lmnoeng.com/PipeDuct.htm

I'll be posting more info as I go through the nozzle calc today.

Last edited by bigben2k; 02-15-2003 at 04:40 PM.
bigben2k is offline   Reply With Quote
Unread 02-15-2003, 04:35 PM   #25
LiquidRulez
Cooling Savant
 
LiquidRulez's Avatar
 
Join Date: Sep 2002
Location: In Hell
Posts: 322
Default

Quote:
Originally posted by bigben2k
Here's a link to a calculation of the hydraulic equivalent:
http://lmnoeng.com/pipeduct.htm

I'll be posting more info as I go through the nozzle calc today.
Whats it under on the homepage Ben? Because all I get is the " you dont have permission to acces this page" errror
LiquidRulez is offline   Reply With Quote
Reply


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -5. The time now is 04:40 PM.


Powered by vBulletin® Version 3.7.4
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
(C) 2005 ProCooling.com
If we in some way offend you, insult you or your people, screw your mom, beat up your dad, or poop on your porch... we're sorry... we were probably really drunk...
Oh and dont steal our content bitches! Don't give us a reason to pee in your open car window this summer...