[Hexus] JetCool CPU Waterblocks 10x Better at Heat Transfer - Page 4 - Overclock.net - An Overclocking Community

Forum Jump: 

[Hexus] JetCool CPU Waterblocks 10x Better at Heat Transfer

Reply
 
Thread Tools
post #31 of 43 (permalink) Old 07-13-2019, 01:01 PM
Robotic Chemist
 
Asmodian's Avatar
 
Join Date: Aug 2009
Location: San Jose, California
Posts: 2,505
Rep: 182 (Unique: 121)
Quote: Originally Posted by Drewminus View Post
The more I think about this, the less sold on the concept I am. Sure direct die's better, whatever, but I can't figure out how this "jet cooling" will be any better. To achieve higer flow velocity in their water block they must be restricting the flow, so I just cannot imagine how higher flow rates vs slower but more liquid produces significanty better results. I'm also pretty suss on water achieving a transfer coefficient of 500,000 w/m2 K. It's been a while since I've done physics so I'm not going to math it out, but nothings really adding up for me here.
The boundary effect is pretty limiting on heat transfer. This is why turbulence is so beneficial in water blocks. A jet into a surface is better than most other techniques, this is why normal water blocks use jet plates and similar too.

A jet directly into the back of the die is pretty ideal, I am not sure about 500,000 W/m² but it is going to be better than microchannels without the jets.
Asmodian is offline  
Sponsored Links
Advertisement
 
post #32 of 43 (permalink) Old 07-14-2019, 05:17 AM
New to Overclock.net
 
Join Date: Aug 2007
Posts: 2,203
Rep: 347 (Unique: 263)
I think Im going to patent an air cooler on the same design. Why have a large bulky air cooler when you can use an air compressor that blasts microjets of air directly onto the die. Much safer than water as well. Im sure air at 0.024 W/mk thermal conductance doesnt need much surface area just like water at 0.6 w/mk doesnt need much surface area when cooling hot spots on die. Traditional thinking of a large copper air cooler spreading heat instantly at 400 w/mk to 1000's times surface area via fins before attempting to transfer heat at a very slow pace of 0.024 W/mk for air... that is all nonsense, just like waterpins/channels doing same for waterblocks is all nonsense. The real issue is this added copper material preventing air from directly hitting the die.

My air compressor will also replace that water cooling jet as well, because water isnt the final heat dissipation, air is via the radiator. That water and radiator using that surface area myth again...

opt33 is offline  
post #33 of 43 (permalink) Old 07-14-2019, 11:10 AM
New to Overclock.net
 
prjindigo's Avatar
 
Join Date: Mar 2011
Posts: 1,701
Rep: 152 (Unique: 89)
This isn't a "10x" improvement, at best it's about 2.5x what we have now because what we have now is a narrow little line of fins with centripetal flow going through a well against the center of the die; if you look inside these water coolers we have they're either the "flat block fin" or the crescent flow of closed loop kits.

The truth is that even jets against the cooling plate don't work well because of turbulence and effect, once the jet leaves the aperture there's no guarantee of laminar flow and it just becomes a mess of cooled water mixing with hot water and all of it flowing sideways out of the system.

Fluted conical pins extending up from the block's heat spreader that have flow coming down them actually works quite a bit better: the flow stays laminar down the flutes and is in contact with the metal the whole way producing a flow that then jets out to the sides, a mild cupping at the base of the conical pins which turns the flow back up (cupped around to the cups around all the other fluted pins) *would* actually produce a much better contact solution. But now we're delving into the range of "fuxit, lets just use a $150 air conditioner's components and jet actual refrigerant onto the plate" territory.

Water cooling is good for what it's good for but if you want real super high performance cooling you need to go ahead and use a compressor system with refrigerant in a closed loop. My $136 window unit behind me is 450w consumption and can move 5000btu out the window, a 250w processor isn't gonna produce more than about 600btu of heat. The unit could easily keep my computer frosty if I piped refrigerant up to heatsinks on the CPU/GPU and had a little ambient cooling system inside the case.. I could even hermetically seal the puppy and run closed loop.

Seriously, water just isn't worth the hassle.
prjindigo is offline  
Sponsored Links
Advertisement
 
post #34 of 43 (permalink) Old 07-14-2019, 02:47 PM
Robotic Chemist
 
Asmodian's Avatar
 
Join Date: Aug 2009
Location: San Jose, California
Posts: 2,505
Rep: 182 (Unique: 121)
Quote: Originally Posted by opt33 View Post
Im sure air at 0.024 W/mk thermal conductance doesnt need much surface area just like water at 0.6 w/mk doesnt need much surface area when cooling hot spots on die. Traditional thinking of a large copper air cooler spreading heat instantly at 400 w/mk to 1000's times surface area via fins before attempting to transfer heat at a very slow pace of 0.024 W/mk for air... that is all nonsense, just like waterpins/channels doing same for waterblocks is all nonsense.
You seem to only be thinking of water and air as static thermal transfer materials. This is wrong. You are not using the water to transfer heat to something else, you are heating up the water directly. If you tried to simply cool the CPU with a static water bath it would not work very well (convection would allow it to work much better than the 0.6 W/mK would suggest, but still not very good).

Air has very little thermal mass to heat up and has a low heat capacity per mass, so it is incapable of absorbing much heat energy. Water is much better at absorbing heat and jetting it into the back of the core allows it to absorb the heat directly, not needing to transfer it. This is why the jets are important in all water blocks.

Per volume air has about 0.0153% of the ability to absorb heat compared to water. Or stated another way, compared to air the same volume of water will absorb ~6500x as much energy when heating up one degree.

Quote: Originally Posted by prjindigo View Post
The truth is that even jets against the cooling plate don't work well because of turbulence and effect, once the jet leaves the aperture there's no guarantee of laminar flow and it just becomes a mess of cooled water mixing with hot water and all of it flowing sideways out of the system.
This is ideal, turbulence is desired to transfer as much thermal energy to the water as possible. Laminar flow requires a much higher surface area for the same energy transfer due to the boundary effect and the low thermal conductance of water.

Test is some time. If you hold your hand in a stream of cold water and compare how cold it feels when flowing the water along your hand parallel to the stream or with it impacting your hand directly, perpendicular to the stream. Especially significant (and relevant) when cooling a fresh burn.

Last edited by Asmodian; 07-14-2019 at 03:08 PM.
Asmodian is offline  
post #35 of 43 (permalink) Old 07-14-2019, 04:13 PM
New to Overclock.net
 
Join Date: Aug 2007
Posts: 2,203
Rep: 347 (Unique: 263)
Quote: Originally Posted by Asmodian View Post
You seem to only be thinking of water and air as static thermal transfer materials. This is wrong.
No. Its called sarcasm. Ive tested different waterblocks for 20 years in some reviews, with and without jet plates, im well aware of the effects of jet plates.

What Im thinking is that even if you maximize flow and turbulence such that any water molecule that has absorbed any heat is instantly replaced with one that has not....water with its low thermal conductance still requires adequate surface area for a given power density.


Last edited by opt33; 07-14-2019 at 04:27 PM.
opt33 is offline  
post #36 of 43 (permalink) Old 07-14-2019, 05:56 PM
New to Overclock.net
 
Andrew LB's Avatar
 
Join Date: Apr 2013
Location: Peoples Republic of Kalifornia
Posts: 345
Rep: 23 (Unique: 18)
Quote: Originally Posted by Asmodian View Post
Air has very little thermal mass to heat up and has a low heat capacity per mass, so it is incapable of absorbing much heat energy. Water is much better at absorbing heat and jetting it into the back of the core allows it to absorb the heat directly, not needing to transfer it. This is why the jets are important in all water blocks.

depends on how much air we're talking about. Ive got a 160psi air compressor in my garage and using the blow gun can cool two freshly welded (with a Mig) pieces of 1/8" steel from a few thousand degrees celcius to cool to the touch in seconds. I can't imagine how noisy air jets would be. Would make Delta or San Ace fans seem quiet.




Andrew LB is offline  
post #37 of 43 (permalink) Old 07-15-2019, 10:01 PM
Data Recovery Engineer
 
Lord Xeb's Avatar
 
Join Date: Jan 2009
Location: Behind you
Posts: 20,215
Rep: 1342 (Unique: 999)
I just see marketing ****. Lets see if it is true when someone actually tests it.


Only fools do not BACKUP their DATA!

"Everyone is an expert in something. Knowledge is acquired not earned. Always be humble and wise. Never look down on others ignorant in your ream of expertise."
Hsvkinyodh
(15 items)
CPU
3770k @ 4.5 1.384v
Motherboard
Asus Z77 Sabertooth
GPU
GTX 1070 Gaming 2126/2249 w/ Hybrid cooler
RAM
TridentX 16GB 2400MHz
Hard Drive
Samsung 850 Pro 1TB
Hard Drive
Seagate 2TB
Power Supply
Seasonic X 760W
Cooling
H105
Case
CM 690 II Adv
Operating System
Windows 10
Monitor
Asus MG279Q - 24" 144Hz 1080p
Keyboard
Filco Majestic II w/ Cherry Reds
Audio
AKG A701
Audio
JDS Labs Objective 2
Audio
JDS Labs ODAC
▲ hide details ▲


Lord Xeb is offline  
post #38 of 43 (permalink) Old 07-16-2019, 06:22 AM
New to Overclock.net
 
feznz's Avatar
 
Join Date: Sep 2011
Posts: 1,958
Rep: 74 (Unique: 61)
https://www.overclock.net/forum/134-...r-cooling.html

direct die cooling will never work unless you start pumping liquid metal.
The is a reason for an IHS (integrated heat spreader)is there, the thermal conductivity simply is not enough for any known liquid feasible to transfer the heat directly off the die.
The solution the put a slab of copper to spread the head over a larger area to transfer the heat to and because excluding silver it ranks higher thermal conductive charts of compounds and elements.

But if this was a great idea not too sure why when it was first talked about 15 years ago it isn't mainstream now I think the biggest obstacle was the jets becoming blocked with foreign matter.
http://www.electronics-cooling.com/2...ir-and-beyond/

LG 38UC99-Asus Apex X z370-8600k@ 5.4Ghz-2AVX-Asus 1080Ti Strix+Asus GTX770-16Gb RGB G-Skill 3200Mhz-256Mb 960evo OS-1Tb 860evo Programs-Corsair RM1000i-1.5+2+4Tb HDD- Custom Water loop
feznz is offline  
post #39 of 43 (permalink) Old 07-16-2019, 02:59 PM
Robotic Chemist
 
Asmodian's Avatar
 
Join Date: Aug 2009
Location: San Jose, California
Posts: 2,505
Rep: 182 (Unique: 121)
Quote: Originally Posted by feznz View Post
direct die cooling will never work unless you start pumping liquid metal.
The is a reason for an IHS (integrated heat spreader)is there, the thermal conductivity simply is not enough for any known liquid feasible to transfer the heat directly off the die.
This is simply not true. In the paper you link they suggest the highest thermal transfer using direct Water Jet Impingement.

Quote: Originally Posted by feznz View Post
But if this was a great idea not too sure why when it was first talked about 15 years ago it isn't mainstream now I think the biggest obstacle was the jets becoming blocked with foreign matter.
http://www.electronics-cooling.com/2...ir-and-beyond/
Because it was unnecessary and too expensive? Using some decently clean DI water should be fine, the jets do not need to be that small. We probably do not need the super small versions that paper talks about.

Also, is your point that it doesn't work due to thermal transfer or that the jets clog? If it doesn't work in the first place, due to poor thermal transfer, I don't see how the jets clogging is an obstacle.
Asmodian is offline  
post #40 of 43 (permalink) Old 07-16-2019, 04:56 PM
New to Overclock.net
 
Join Date: Aug 2007
Posts: 2,203
Rep: 347 (Unique: 263)
That paper is using uniform heat sim dies as does all experimental testing, hence term uniform heat flux. Modern cpus, like 9900k, have hot spots where heat is concentrated on small sections of the die which decrease the effective surface area dramatically. Which is why you can have a 15-25C gradient just through the die. Copper water blocks or copper IHS spread heat at rapid 400 w/mk to a more uniform temp (intel claimed in past white papers to within 1-2C spread vs up to 10X higher temp spread vs bare die hot spots).

Even if there is a multi-jet design (like the one IBM abandoned 15+ yrs ago even though they milled microchannels in die to deal with hot spots) that can cool a uniform heated sim chip 10-15% better than a 15 year old design waterblock (and maybe as well as a modern waterblock), that would not suffice. It would have to perform multiple times better than a cold plate, because without spreading the heat via copper it will have multiple times lower surface area which will will proportionally lower its effectiveness.

Ill believe it when I see it tested independently on an actual intel cpu dissipating 200+W at load. So far its been 15 years, havent seen it yet.

opt33 is offline  
Reply

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may 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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off