Spray nozles

CheeseBall · 03-05-2003, 11:24 PM

If there is one thing more I can do to my soon to be water rig, it's a nozle on the intake. Now I am totally uninformed on the subject so I turn to you guys. How should I go about doing this? Should I machine the nozle as part of the the top half of my block (so when the barb is screwed in it meets w/ the nozle part)? or make my own custom hose barbs/nozle?

Now I don't want anything that is going to take weeks. But I would like at least SOMETHING to help the intake velocity.

If you know a good article or something please post it.

Te conozco mosco,
Bryan

crane · 03-05-2003, 11:53 PM

Heres what I did...you know spray tips like go on a pressure washer or a misting type sprayer? I found one that makes a spiral type, high velocity jet spray. The tip had 1/4 npt male threads so I ended up soldering the nozzle part to the upper block plate.
I made sure of flow,spray and jet contact with the base plate before I soldered it in place. If I remember correctly the tip needed some modification to acheive the proper pattern of spray.
Just an idea to get you started.

bigben2k · 03-06-2003, 07:46 AM

The purpose of a nozzle is to increase flow velocity within a smaller diameter. The reason for using one, is to propel this high speed flow against the baseplate, creating a very local area of turbulent flow, which improves the heat transfer to the coolant.

Here's a hint:

Tuff · 03-06-2003, 09:02 AM

Bigben..that is only 2/3 correct...you need a flare at the opposing end aswell to atomize the water...It is not the turbulence causing the extra efficiency in the block...its the atomization of the Water.

Here is a hint:

The idea behind WaterBlock spraying is rooted firmly in thermodynamics. As water changes phases from liquid to gas an enormous amount of heat energy is absorbed. We want to use the heat absorbing properties of the water to gas phase transformation to cool our waterblocks.

Steam=gas
Atomized water=gas

BUT...something is missing in the equation

I know what it is..Do you?

I will be doing a prototype soon, when my machinist gets done with my base plate. I will have more soon.

Tuff

myv65 · 03-06-2003, 09:13 AM

Tuff,

Let me get this straight. You think you're vaporizing a portion of the water with your nozzle? Are you serious? How is this possible? You are somehow maintaining an air pocket in the block?

You can't convert water from liquid to gas when you have a closed pocket completely filled with liquid water.

Even if you could, the only way you'd see a benefit is to allow the water to leave the block in a vaporized state. Otherwise the heat of vaporization is completely contained within the block and of no net benefit. (Liquid in, liquid out = no good, liquid in, gas out = heat carried away in form of heat of vaporization)

Only a few things really matter here. One is the pressure available as a nozzle converts pressure into velocity. Two is the nozzle shape as rapid diameter changes convert pressure to velocity inefficiently. Three is the pattern formed on the baseplate by the nozzle.

bigben2k · 03-06-2003, 11:12 AM

I don't see how vaporizing water is part of anything here but...

There is a small flare specified at the nozzle outlet. It's purpose is to keep the flow as close to straight as possible.

"Spraying" is not desired here: the purpose of a nozzle is for jet inpingement, which directs a high-speed flow towards the baseplate, creating a highly localized turbulent area, which in turn improves the cooling (aka it reduces the thermal resistance from the block to the fluid).

The reason to use a nozzle is to take maximum cooling advantage of the power provided by a pump, rather than using a much larger pump, in (what I'll call) a raw high flow design.

We've come a long way, since the original Maze 1 design. "Spraying" or spreading the flow doesn't really serve any efficient purpose, in many cases.

Now if you were talking about a phase change system, that a whole other ball game.

Tuff · 03-06-2003, 11:53 AM

Reread my post...it says something is missing in the equation.

Figure it out

Tuff

bigben2k · 03-06-2003, 12:15 PM

Sorry, but you lost me. What do you mean by "atomization of the water"?

This is starting to sound esoteric...

myv65 · 03-06-2003, 12:57 PM

Quote:

Originally posted by Tuff
Reread my post...it says something is missing in the equation.

Figure it out

Tuff

OK, how about reality? Atomized water is a rather broad definition. My previous statement still stands: It aint gonna happen in a closed space (block interior) where no air space exists.

Rather than attempting to be coy, how 'bout you simply state what you believe you've got happening?

You state it's "based firmly in thermodynamics". I'll be completely blunt and entirely unkind. Until you start putting down something that shows you have actually opened a thermo text and understood its contents I'm viewing all you say with a high degree of skepticism.

Where do I come off sounding so cruel? Easy, I actually have a BSME from a real college and have spent (egad, the time flies) nearly twenty years on engineering stuff. My reputation is solid here and on other forums and I've written a few articles to back this up.

If you're really on to something, great, more power to you. Otherwise I've seen far too many people posting claims of greatness that are built on sand.

BillA · 03-06-2003, 01:41 PM

careful Tuff
the thermal contraction will shear the pins on the CPU package
you need a heater

myv65
you violated your own sagacious words
carefully evaluate the first post and if they are a 'gamer', move on
this one has shit for brains

8-Ball · 03-06-2003, 02:24 PM

Quote:

you violated your own sagacious words
carefully evaluate the first post and if they are a 'gamer', move on
this one has shit for brains

ROFL

Laugh so much it hurts

Seriously though, if I've got a bit more time, I'll explain why this isn't possible.

8-ball

bigben2k · 03-06-2003, 02:56 PM

...and while you do that, I'll try to cover the topic!

Cheeseball: if you're where I think you are, you're ready to try to implement "jet impingement".

The pic I posted will give you the info you need to do it.

For the rest of us here's the original block

myv65 · 03-06-2003, 03:09 PM

Quote:

Originally posted by unregistered
you violated your own sagacious words
carefully evaluate the first post and if they are a 'gamer', move on
this one has shit for brains

LOL. Ah come on Bill. Combine this thing with that abrasive stuff from the other thread that you replied to and toss in that dartboard for good measure. Yeah, I'm beginning to see how places like CPUFX stay in business. . . . .

BillA · 03-06-2003, 03:12 PM

Ben
you do not know what you are talking about
the image you have posted above is indeed for a nozzle
where 'nozzle' is defined as the fitting used for a penetration by a small cylinder into a larger one
- as in a pipe into the sidewall of a tank
-> and it was suggested BEFORE that you look up the definition of an 'ASME nozzle'

see those angles on the small end ?
any idea what they are about ?
try welding

yes, the radius of curvature is intended to minimize the flow impediment, but so what ?
where is the flow into a wb going from a flat surface normal to the direction of inlet flow ?

arrghh, what are you 'covering the topic' with ?

CheeseBall
make a smooth taper, 3 or 4:1 is fine

EDIT: added avatar just for Dave
(lots of masochism apparent in these froums)

SysCrusher · 03-06-2003, 03:35 PM

This is what Tuff is talking about "I think". Here's a couple of links.
Link 1

link 2

link 3

It's totally impractical for our use. Where would you get 20m/s velocity, 4ml/min flow rate from.:shrug: Then you have to consider that it would be submerged which changes the whole game.

Cheesball:
A nozzle with a gradual taper (hyperbolic shape) is what you would want just as Bill stated. How you can machine that I have no idea as I'm no pro metal machinist. You could also make it what ever shape you want too as long as it has that gradual taper. Of course, you could just plug the end of a barb then drill, cut what ever hole in it and be done with it. It'll work but not as good. The height, diameter of the nozzle, and the surface geometry on the base of the block play important roles and need to be considered. Also, don't forget the pump and it's performance as that will dictate what you can actually achieve with a nozzle. I have no sure answers for you but just to experiment for your own learning adventure. I'm sure myv65 and other more educated people in this field here could help you better. BillA also has some good info from his testing that can help you make a slight educated guess. All I know is what works best with a flat, slight convex base.

03-05-2003, 11:24 PM	#1
CheeseBall Cooling Savant Join Date: Jan 2003 Location: Colorado Posts: 176	Spray nozles If there is one thing more I can do to my soon to be water rig, it's a nozle on the intake. Now I am totally uninformed on the subject so I turn to you guys. How should I go about doing this? Should I machine the nozle as part of the the top half of my block (so when the barb is screwed in it meets w/ the nozle part)? or make my own custom hose barbs/nozle? Now I don't want anything that is going to take weeks. But I would like at least SOMETHING to help the intake velocity. If you know a good article or something please post it. Te conozco mosco, Bryan

03-05-2003, 11:53 PM	#2
crane Cooling Neophyte Join Date: May 2002 Location: so cal Posts: 82	Heres what I did...you know spray tips like go on a pressure washer or a misting type sprayer? I found one that makes a spiral type, high velocity jet spray. The tip had 1/4 npt male threads so I ended up soldering the nozzle part to the upper block plate. I made sure of flow,spray and jet contact with the base plate before I soldered it in place. If I remember correctly the tip needed some modification to acheive the proper pattern of spray. Just an idea to get you started. __________________ dirt rules

03-06-2003, 11:12 AM	#6
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	I don't see how vaporizing water is part of anything here but... There is a small flare specified at the nozzle outlet. It's purpose is to keep the flow as close to straight as possible. "Spraying" is not desired here: the purpose of a nozzle is for jet inpingement, which directs a high-speed flow towards the baseplate, creating a highly localized turbulent area, which in turn improves the cooling (aka it reduces the thermal resistance from the block to the fluid). The reason to use a nozzle is to take maximum cooling advantage of the power provided by a pump, rather than using a much larger pump, in (what I'll call) a raw high flow design. We've come a long way, since the original Maze 1 design. "Spraying" or spreading the flow doesn't really serve any efficient purpose, in many cases. Now if you were talking about a phase change system, that a whole other ball game. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

03-06-2003, 12:15 PM	#8
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Sorry, but you lost me. What do you mean by "atomization of the water"? This is starting to sound esoteric... __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

03-06-2003, 02:56 PM	#12
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	...and while you do that, I'll try to cover the topic! Cheeseball: if you're where I think you are, you're ready to try to implement "jet impingement". The pic I posted will give you the info you need to do it. For the rest of us here's the original block __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

03-06-2003, 09:02 AM	#4
Tuff Cooling Savant Join Date: Feb 2003 Location: Canada Posts: 141	Bigben..that is only 2/3 correct...you need a flare at the opposing end aswell to atomize the water...It is not the turbulence causing the extra efficiency in the block...its the atomization of the Water. Here is a hint: The idea behind WaterBlock spraying is rooted firmly in thermodynamics. As water changes phases from liquid to gas an enormous amount of heat energy is absorbed. We want to use the heat absorbing properties of the water to gas phase transformation to cool our waterblocks. Steam=gas Atomized water=gas BUT...something is missing in the equation I know what it is..Do you? I will be doing a prototype soon, when my machinist gets done with my base plate. I will have more soon. Tuff

03-06-2003, 09:13 AM	#5
myv65 Cooling Savant Join Date: May 2002 Location: home Posts: 365	Tuff, Let me get this straight. You think you're vaporizing a portion of the water with your nozzle? Are you serious? How is this possible? You are somehow maintaining an air pocket in the block? You can't convert water from liquid to gas when you have a closed pocket completely filled with liquid water. Even if you could, the only way you'd see a benefit is to allow the water to leave the block in a vaporized state. Otherwise the heat of vaporization is completely contained within the block and of no net benefit. (Liquid in, liquid out = no good, liquid in, gas out = heat carried away in form of heat of vaporization) Only a few things really matter here. One is the pressure available as a nozzle converts pressure into velocity. Two is the nozzle shape as rapid diameter changes convert pressure to velocity inefficiently. Three is the pattern formed on the baseplate by the nozzle.

03-06-2003, 11:53 AM	#7
Tuff Cooling Savant Join Date: Feb 2003 Location: Canada Posts: 141	Reread my post...it says something is missing in the equation. Figure it out Tuff

03-06-2003, 01:41 PM	#10
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	careful Tuff the thermal contraction will shear the pins on the CPU package you need a heater myv65 you violated your own sagacious words carefully evaluate the first post and if they are a 'gamer', move on this one has shit for brains

03-06-2003, 03:12 PM	#14
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	Ben you do not know what you are talking about the image you have posted above is indeed for a nozzle where 'nozzle' is defined as the fitting used for a penetration by a small cylinder into a larger one - as in a pipe into the sidewall of a tank -> and it was suggested BEFORE that you look up the definition of an 'ASME nozzle' see those angles on the small end ? any idea what they are about ? try welding yes, the radius of curvature is intended to minimize the flow impediment, but so what ? where is the flow into a wb going from a flat surface normal to the direction of inlet flow ? arrghh, what are you 'covering the topic' with ? CheeseBall make a smooth taper, 3 or 4:1 is fine EDIT: added avatar just for Dave (lots of masochism apparent in these froums) Last edited by BillA; 03-06-2003 at 03:22 PM.

03-06-2003, 03:35 PM	#15
SysCrusher Cooling Savant Join Date: Dec 2002 Location: Florida Posts: 256	This is what Tuff is talking about "I think". Here's a couple of links. Link 1 link 2 link 3 It's totally impractical for our use. Where would you get 20m/s velocity, 4ml/min flow rate from.:shrug: Then you have to consider that it would be submerged which changes the whole game. Cheesball: A nozzle with a gradual taper (hyperbolic shape) is what you would want just as Bill stated. How you can machine that I have no idea as I'm no pro metal machinist. You could also make it what ever shape you want too as long as it has that gradual taper. Of course, you could just plug the end of a barb then drill, cut what ever hole in it and be done with it. It'll work but not as good. The height, diameter of the nozzle, and the surface geometry on the base of the block play important roles and need to be considered. Also, don't forget the pump and it's performance as that will dictate what you can actually achieve with a nozzle. I have no sure answers for you but just to experiment for your own learning adventure. I'm sure myv65 and other more educated people in this field here could help you better. BillA also has some good info from his testing that can help you make a slight educated guess. All I know is what works best with a flat, slight convex base.

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