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

General Liquid/Water Cooling Discussion For discussion about Full Cooling System kits, or general cooling topics. Keep specific cooling items like pumps, radiators, etc... in their specific forums.

Reply
Thread Tools
Unread 08-01-2002, 03:16 PM   #101
airspirit
Been /.'d... have you?
 
airspirit's Avatar
 
Join Date: Jul 2002
Location: Moscow, ID
Posts: 1,986
Default

Trial 1: Clockwise setup.
Full bucket - no swirl
1/2 bucket - center dimpling, barely any swirlage
1/4 bucket - noticable swirlage
1/8 bucket - massive swirlage

Trial 2: Counter-clockwise setup.
No noticable differences throughout experiment

It looks like it works. From observations, I would definitely want to widen the circular chamber in the block. It seems that the higher the water column is in relation to the width, the slower the circular flow. With a wider (and therefore relatively flatter) chamber, I should get massive swirling in the block.

It'll work! Woo hoo!

And the guys in detail weren't even upset I screwed up their bucket.

Edit: Since I didn't have sparkles, I just used cigarette ashes from the ashtray out back. Sometimes we must deal with nastiness in the advancement of science.

Last edited by airspirit; 08-01-2002 at 03:18 PM.
airspirit is offline   Reply With Quote
Unread 08-01-2002, 03:32 PM   #102
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, you've completed the test, but what's the result?

I mean, you had a mass of water standing still, on top of the exits. In the block, the water will be coming in as fast as it's going out, right?
bigben2k is offline   Reply With Quote
Unread 08-01-2002, 03:42 PM   #103
airspirit
Been /.'d... have you?
 
airspirit's Avatar
 
Join Date: Jul 2002
Location: Moscow, ID
Posts: 1,986
Default

In the block the water will be coming in as fast as it goes out. This will not cause any change in results. It will be like sticking two hoses into the bucket while the two on the bottom are emptying. What I demonstrated was that the force of the exiting water was enough to cause the water to spin. That was all I needed. Inside the water block there will be water forced in, but there will also be larger suction on the outlets. It'll balance out, and you WILL get a whirlpool effect. I just learned that the circular chamber will need to be wider to make this work right.

Quick and dirty: this will work.
airspirit is offline   Reply With Quote
Unread 08-01-2002, 03:52 PM   #104
gone_fishin
Cooling Savant
 
gone_fishin's Avatar
 
Join Date: May 2002
Location: Da UP
Posts: 517
Default

If the inlet is over the cone, you will not get a whirlpool. I believe you will observe something like this with a tilted axis where the flow splits directions.
Attached Images
File Type: jpg blockb.jpg (10.6 KB, 262 views)
gone_fishin is offline   Reply With Quote
Unread 08-01-2002, 03:59 PM   #105
airspirit
Been /.'d... have you?
 
airspirit's Avatar
 
Join Date: Jul 2002
Location: Moscow, ID
Posts: 1,986
Default

The cone will facilitate the swirling effect, methinks. You gotta remember where the suction force is going, and the vector on that force. It will force the swirling and the slight cone will facilitate that.
airspirit is offline   Reply With Quote
Unread 08-01-2002, 09:54 PM   #106
Khledar
Cooling Savant
 
Khledar's Avatar
 
Join Date: Mar 2002
Location: North Vancouver BC
Posts: 234
Default

The cone does not facilitate a swirl. The cone forces water to flow more evenly away from the tip of the cone, it does not do anythign to help swirl the water. To prove: take a hose, take a cone, turn hose on above the cone, watch water go away from center of the cone. Now the outlets, they don't cause spinnage either, they cause water to be pulled towards them, straight. To prove: Take a bucket with water (sprinkles if you want too hehe) punch some holes, watch the water flow out the holes.

Put these two separate things together, and presto, you get even water distribution, and bilocational water collection; I believe gone_fishin's diagram would be fairly acurate.

The main thing is that, not all the water flows in the same direction, this is why no swirl will occur. By using a clock, outlets at 2 and 8, if you want clockwise rotation, look at 3 and 9 the water in those zones will be going the wrong way! (counter-clockwise) the same problem occurs with the opposite direction.

My .02.
__________________
"mooooo" said the cow.
ERTW - UBC

P4 2.4B @3.01Ghz 167FSB :: Abit IC7 :: 2x256MB HyperX PC3700 :: ASUS 9600XT :: WD Raptor 2x36GB RAID0
Khledar is offline   Reply With Quote
Unread 08-01-2002, 11:01 PM   #107
Can O' Beans
Cooling Savant
 
Join Date: Aug 2002
Location: Pa - USA
Posts: 264
Default

I think your cone/swirl block would benefit from fine "swirled" channels/ridges, whatever going down the surface of the cone. Think of the hand-held orange juicers for an idea, but a bit more swirl.
Can O' Beans is offline   Reply With Quote
Unread 08-02-2002, 12:38 AM   #108
sunblade
Registered User
 
sunblade's Avatar
 
Join Date: Jul 2001
Location: USA
Posts: 156
Default

Quote:
Originally posted by gone_fishin
If the inlet is over the cone, you will not get a whirlpool. I believe you will observe something like this with a tilted axis where the flow splits directions.
Dunno if this has been discussed, but what if the inlet(s)/outlets were swapped. Water could leave directly above the core, and the water entering from both sides could swirl over the core and up the outlet. It would be like an upside down drain?
sunblade is offline   Reply With Quote
Unread 08-02-2002, 01:19 AM   #109
gone_fishin
Cooling Savant
 
gone_fishin's Avatar
 
Join Date: May 2002
Location: Da UP
Posts: 517
Default

Quote:
Originally posted by sunblade


Dunno if this has been discussed, but what if the inlet(s)/outlets were swapped. Water could leave directly above the core, and the water entering from both sides could swirl over the core and up the outlet. It would be like an upside down drain?
If the two areas I shaded red were the main target for heat transfer, then yes the coolest entry waters highest velocity hitting these areas would be good but they are too far away from the core and the cool entry waters initial velocity is wasted on these surfaces in a reverse scenario, that is my theory for now.

What is the infatuation with the whirlpool anyway? What are its percieved benefits? If the water must travel a longer distance by spiraling around before exit, then it is not entering and leaving as fast as possible, would this not be considered a flow restriction?
Just curious to peoples thoughts on the whirlpool (remember there is no air in the block like an open whirlpool).
Attached Images
File Type: jpg block65.jpg (9.8 KB, 230 views)
gone_fishin is offline   Reply With Quote
Unread 08-02-2002, 03:00 AM   #110
Dix Dogfight
Cooling Savant
 
Dix Dogfight's Avatar
 
Join Date: Aug 2001
Location: Stockholm Sweden
Posts: 128
Default

Iv'e only studied the basics of thermodynamics and nothing at all on the subject of fluid dynamics. So I have to ask this question.
(maybe it totally stupid but what the heck)

If the water is run in both directions through this wb wouldn't the "flow vectors" at every point be the same but with a opposite sign?
If so, then per gone_fishin's latest pic the whirpool effect would take place.

Enlighten me please.
__________________
If it ain't broke, fix it.

Setup:
Dual Duron 1100 | Voodoo 3 2000 | Addtronics W8500(WTX) | Eheim 1250
| Car radiator | 2 Innovatech WB |
Dix Dogfight is offline   Reply With Quote
Unread 08-02-2002, 11:46 AM   #111
airspirit
Been /.'d... have you?
 
airspirit's Avatar
 
Join Date: Jul 2002
Location: Moscow, ID
Posts: 1,986
Default

Kheldar, I did that experiment. It takes a bit more work to simulate the system, though. With what you are describing, the force vectors from the outlets are pointing toward the center of the bucket, hence no swirling. If the outlets are cut in an oval shape and hoses are strung into the bucket and taped down to the sides (similar in layout to the block picture), the force vectors are now forcing th flow in a circular pattern.

The whole idea of this block is to solve the problems with EVERY block on the market.

1) The coldest flow needs to hit closest to the core. By having the entry directly over the core being forced downward, we are achieving that.

2) Stagnant water flow, or "slow spots" need to be eliminated. Even in the elegant design of MeltMan (see his post, he has an easy and outstanding design idea) keeps the fastest flow on the top of the block and the flow nearest the core is slower. By causing a swirling in the block, ALL areas of the block are in motion at all times eliminating slow spots.

The caveat is that the water stays in the block a fraction of a second longer (depending on flow rate), but since the coldest flow is hitting the block, this would possibly facilitate heat xfer.

3) You want minimal resistance in your block. There is no switchbacks in this block. It shouldn't slow flow much more than an elbow. You'll probably have as much slowdown in the Y adapter after the block as you will in the block.

4) Wh1rlp00ls are n33t. Talk about bragging rights! Heh. While this is unsuitable for peltier use (since the cooling is only going to be spread over about a 30-35mm diameter circle, it won't evenly cool the pelt), it should ensure maximum cooling of the core.

Upon further thinking, you could build arc shaped channels parallel to the flow of current to ensure circular flow, but it would increase manufacturing complexity. True, it could provide for more revolutions in the water before it exited the block, but even if the water only completes 1/2-3/4 revolutions, it serves its purpose. The entire purpose of the swirling is to agitate the entire coolant mass in the block to keep it at maximum speed. Any more swirling than that might hamper flow and efficiency. I'm trying to figure out an easy way to make something like this (I've got a good idea on that).
airspirit is offline   Reply With Quote
Unread 08-02-2002, 12:04 PM   #112
airspirit
Been /.'d... have you?
 
airspirit's Avatar
 
Join Date: Jul 2002
Location: Moscow, ID
Posts: 1,986
Default

Here she goes:
Attached Images
File Type: jpg block final spec.jpg (5.9 KB, 200 views)
airspirit is offline   Reply With Quote
Unread 08-02-2002, 12:12 PM   #113
airspirit
Been /.'d... have you?
 
airspirit's Avatar
 
Join Date: Jul 2002
Location: Moscow, ID
Posts: 1,986
Default

The top block is just a block that is drilled through. The top can be threaded for use with a nipple.

The bottom block is a pressed central cylinder ... the trick is retaining the center cone (if you wish, as other people said, it may not be necessary, but I feel that it may guarantee maximum cold water contact due to increased surface area. You be the judge). The two side exits are drilled and threaded for nipple use (straight would be best, but I think elbows should cause the flow pattern to remain the same inside the block, though you'll get lower rotational force, methinks). The bottom block is also channeled for use of a O-ring. This can be bored to be connected together by hex nuts, the bottom block can also be extended for direct mounting by use of the motherboard holes.

I'm going to try to shut up now ....

edit: I can't spell today.
Attached Images
File Type: jpg block final specb.jpg (4.7 KB, 195 views)
airspirit is offline   Reply With Quote
Unread 08-02-2002, 03:00 PM   #114
gmat
Thermophile
 
gmat's Avatar
 
Join Date: Mar 2001
Location: France
Posts: 1,221
Default

Just my 2 euro-cents in this very interesting thread. I saw Coriolis effect mentioned. You can forget it. Coriolis effect is millions times weaker than forces, pressures and velocities at work in our systems.
The common idea of water spinning down the drain because of Coriolis effect is a total myth. Spinning (swirls, etc) is provoked by fluid motion if it's already moving, or assymetry / irregularities of containing surfaces.
More details ? Read this:
http://www.ems.psu.edu/~fraser/Bad/BadCoriolis.html

Now back to our regular scheduled program
gmat is offline   Reply With Quote
Unread 08-02-2002, 03:23 PM   #115
airspirit
Been /.'d... have you?
 
airspirit's Avatar
 
Join Date: Jul 2002
Location: Moscow, ID
Posts: 1,986
Default

I kind of figured that was the case, but I just wanted all bases covered (a forum of skeptics we are). Note that there was no change in the results of my experiment when I swapped inlet direction. The only change was the direction of the swirl.
airspirit is offline   Reply With Quote
Unread 08-02-2002, 04:16 PM   #116
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

Quote:
Originally posted by sunblade


Dunno if this has been discussed, but what if the inlet(s)/outlets were swapped. Water could leave directly above the core, and the water entering from both sides could swirl over the core and up the outlet. It would be like an upside down drain?
Actually, we did cover it. Maskedgeek considered that approach, but the centripedal forces would induce more restriction, and the swirl wouldn't come into effect anyways.

Airspirit, your concept is good, but your application is just not there. In a bucket, the water that stands still, waiting to exit (for quite some time, I might add) will have a circular motion induced. In a waterblock however, with 100 gallons-per-hour flow rates, the water isn't there long enough to be carried into a swirl: at best, it won't head for the exits in a perfectly straight line, but that's about it.

In order to achieve a multiple turn swirl in a block, you're going to have to force the water to go that way, and that means a channel, and that means a spiral block...
bigben2k is offline   Reply With Quote
Unread 08-02-2002, 04:40 PM   #117
airspirit
Been /.'d... have you?
 
airspirit's Avatar
 
Join Date: Jul 2002
Location: Moscow, ID
Posts: 1,986
Default

Multiple turn = bad, because it's in the block too long
1/2-3/4 turn = good, because it's out quick and the entire block is cleared of water.

I don't want it to spin too much. That's the point. After 1/2 turn, the effectiveness starts to go down. All I'm trying to do is scrape all the water out of there eliminating slow spots. If it was to turn, say 4 times, the whole thing is a slow spot because it's in the block too long. See what I mean?

Oh, and the cone in the middle, after further reflection, is probably necessary. Like gone-fishing demonstrated, it'll mash the flow in all directions. That'll allow fresh flow to hit the outside walls, allowing maximum water clearance from them. That doesn't make too much sense when I read it, but what I am shooting for is for the water to be pulled from the sides causing spin, not from the middle.

Kind of like this. It demonstrates the areas covered by each outlet. Sloppy, but that's what I get for using jpegs in paint. The orange arrows is the water being mashed against the cone. The red arrows are the direction of force in the water being sucked. The two colored regions are the areas that each outlet pulls water from. This is essentially what I'm trying for.
Attached Images
File Type: jpg blockd.jpg (6.2 KB, 183 views)
airspirit is offline   Reply With Quote
Unread 08-02-2002, 05:10 PM   #118
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

I see where you're headed, and it's a lot like a previous design I discussed (deeply hidden in one of morphling's threads).

What I dreamed up though, was a center inlet, with fins that are threaded to accept a barb, then another large barb (about 2") which would cause the coolant to return in the 2" tube. Lots of fins all around, in a radial pattern.

The 1/2 in. tube would be inside the 2" tube, and they'd be connected to some kind of res to seperate the flow.

The problem, of course, is mounting a 2" barb...
bigben2k is offline   Reply With Quote
Unread 08-02-2002, 07:42 PM   #119
Khledar
Cooling Savant
 
Khledar's Avatar
 
Join Date: Mar 2002
Location: North Vancouver BC
Posts: 234
Default

With regards to the last flow image posted, that flow would work well but I think it would be difficult to achieve. The area of concer is circled in red in image 1 below. What I think the problem would be is illustrated with green arrows in image 2. I think the flow going out would pull the water in unwanted directions. As to fixing this, I think any added obstructions would cause the dead spots you are trying to eliminate, but there surely is a way.


__________________
"mooooo" said the cow.
ERTW - UBC

P4 2.4B @3.01Ghz 167FSB :: Abit IC7 :: 2x256MB HyperX PC3700 :: ASUS 9600XT :: WD Raptor 2x36GB RAID0
Khledar is offline   Reply With Quote
Unread 08-03-2002, 11:30 AM   #120
airspirit
Been /.'d... have you?
 
airspirit's Avatar
 
Join Date: Jul 2002
Location: Moscow, ID
Posts: 1,986
Default

That will be taken care of by the momentum of the water (after the system is "charged" after a second or two ... momentum must be built) and the force applied by the inlet. There will be no issue there. I wish I could hook this up to a simulation program and post the results in mpeg format for you guys to watch. I just don't know where to get one that would work.
airspirit 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 08:35 AM.


Powered by vBulletin® Version 3.7.4
Copyright ©2000 - 2024, 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...