EXTREME Overclocking Forums
Home | Reviews | Forums | Downloads | RealTime Pricing Engine | Folding Stats Contact Us


Go Back   EXTREME Overclocking Forums > General CPU Cooling > Phase Change, Dry Ice, LN2, Other Crazy Stuff

Welcome Guest Visitor! Please Register, It's Free and Fun To Participate!
The EXTREME Overclocking Forums are a place for people to learn how to overclock and tweak their PC's components like the CPU, memory (RAM), or video card in order to gain the maximum performance out of their system. There are lots of discussions about new processors, graphics cards, cooling products, power supplies, cases, and so much more!

You are currently viewing our boards as a "guest" which gives you limited access to view most discussions. You need to register before you can post: click the register link to proceed. Before you register, please read the forum rules. By joining our free community you will have access to post topics, communicate privately with other members (PM), respond to polls, upload your own pictures, and access many other special features. Registration is fast, simple, and absolutely free! To start viewing messages, select the forum that you want to visit from the selection below.

After you have registered and read the forum rules, you can check out the FAQ for more information on using the forum. We hope you enjoy your stay here!

Note To Spammers: We do not allow unsolicited advertising! Spam is usually reported & deleted within minutes of it being posted, so don't waste your time (or ours)!


Please Register to Post a Reply
 
Thread Tools
Old 01-10-2013, 01:02 PM   #1
technogiant
Overclocker
technogiant's Avatar
Senior Member
 
Posts: 335
Last Seen: 09-15-2014
iTrader: 0 / 0%
subambient phase change submersion pc project.

I've started this new thread here because this fully deserves the title of "other crazy stuff"

It's really an evolution of my window ac chillbox build but it is going to be a whole new build and so I thought should have a new post.

As some of you may know I've been kicking ideas around for months to further evolve my chillbox. Some recent ideas I've had, I think, have moved this project forward to the point where I can start building, I'm just after some confirmation that my ideas aren't completely off the wall and at least have the possibility of working.

The build will essentially again be an ac unit chillbox except the pc will be submerged in liquid.

The liquid will not be the usual mineral oil or 3M Novec 7000 but a refrigerant gas that has been cooled into a liquid in the chillbox. The most promising candidate for this is Hfc227ea with a boiling point of -16c.

The major problem I've had with thinking this through is the problems of managing a large quantity....about 10 liters of liquified refrigerant which would become a volume of >2000 liters if not contained and allowed to return to ambient temps and pressures.

After much thought I'm fairly sure I've come up with a means of moving the refrigerant back and fore between the chillbox chamber and a pressure bottle so that it can be contained safely when not being used in the pressure bottle and easily transferred to the chillbox when required for use.
The best bit of this method is that it is very simple and involves no pumps being purely pressure, temperature and gravity driven.

Okay, it's a cyclic process and this is how I propose (and hope) it would go.

At the starting point I have a pressure bottle of refrigerant at room temp, feed the gas into the chillbox via a flexible line bottom fed into the chillbox.
The ac unit cools and condenses it to a liquid and this process is continued until there is sufficient to submerge the pc.

That's the easy bit......the hard bit is to get the refrigerant back into the pressure bottle when you've finished.

What I'm planning is to place the supply pressure bottle, flexible line still attached to the chamber but now with the valve closed into a freezer.
When the temperature of the pressure bottle is lower than the boiling point of any refrigerant still contained in it then the pressure within it will drop below atmospheric.....If I were then to open the valve connecting the pressure bottle to the chillbox as the latter is at atmospheric pressure the the fluid will be drawn out from the chillbox via the bottom fed line back into the pressure bottle, this would be driven both by suction and gravity the chillbox being positioned above the pressure bottle.

This is all experimental stuff to me and I'd be grateful for peoples views on whether this would possibly work......it all seems a little too simple and too good to be true.....it would make for a very simple build.

Last edited by technogiant; 01-10-2013 at 01:30 PM.
United Kingdom  Offline
    Register to Reply to This Post
Old 01-10-2013, 02:08 PM   #2
Witchdoctor
"Think"
Witchdoctor's Avatar
Senior Member
 
Posts: 6,750
Last Seen: 04-03-2017
iTrader: 17 / 100%
to dangerous IMO,

Not saying it would woks but when you have that kind of expansion if something fails you coulds get hurt or hurt someone else.

what temp[s do you hope to acheive and how much better could they be than say a simple single stage. And you always could look into a two stage unit if single stage does not meet your temp requirments. I use single stage and love it, good for testing stuff out prior to freezing it. for 24/7 stuff though a conservative single stage is a nice option

I am intrested in your theroy and will be watching this thread

Good luck
United States  Offline
    Register to Reply to This Post
Old 01-10-2013, 02:35 PM   #3
technogiant
Overclocker
technogiant's Avatar
Senior Member
 
Posts: 335
Last Seen: 09-15-2014
iTrader: 0 / 0%
As regards temps....well I'm basically after the best I can achieve using my ac unit.....currently in my chillbox it holds chamber temps at -30c at idle and -22c at full system load.

I want to use the most efficient form of cooling so that I can get my loaded component temps as close to those temps as possible. That's why I went with the submersion/phase change idea as it offers, I feel, the most efficient way of linking my component cooling to the ac units cooling power.
I've also got some pretty novel ideas for direct de-lided cooling of th cpu/gpu which would only work with the submersion/phase change method.

So it's not so much I have a temperature target but just want to extract the most I can out of the ac unit and still keep the pc suitable for regular use.

I understand your thoughts about possible danger.....but tbh as all the liquid will be contained in an insulated chamber even if the ac unit failed it would only boil off slowly as heat ingresses, and it will have an externally vented pressure release.

I think the worst case scenarios would be if the chamber base was to crack/fracture while full of fluid, or if the line connecting the chamber to the pressure bottle was to break while transferring liquid from the chamber to the pressure bottle.
I need to build some fail safes into the design to guard against that sort of thing, as you say could be dangerous.
But thanks for expressing your opinion on that, I'm after feedback to try and second guess these potential problems.

Last edited by technogiant; 01-11-2013 at 01:00 AM.
United Kingdom  Offline
    Register to Reply to This Post
Old 01-10-2013, 03:14 PM   #4
Witchdoctor
"Think"
Witchdoctor's Avatar
Senior Member
 
Posts: 6,750
Last Seen: 04-03-2017
iTrader: 17 / 100%
Have you thought about a simple chiiler ? dump the coil in a cooler full of a water glycol solution and pump it through blocks ?
United States  Offline
    Register to Reply to This Post
Old 01-11-2013, 01:58 AM   #5
technogiant
Overclocker
technogiant's Avatar
Senior Member
 
Posts: 335
Last Seen: 09-15-2014
iTrader: 0 / 0%
Well I've considered putting a water loop inside my current chillbox, but tbh it's all been done before (as have most things), I like to do things differently and think what I'm considering here would be a real first (perhaps for good reason :-/ ) and the most efficient cooling method.

Additional Comment:

Going back to the safety issue and negating the major risks, I'm going to make a double walled chamber so the chances of both failing simultaneously are negligible.

I'm also wondering what type of hose would be best to use to link the chamber to the gas bottle. I need something that will be flexible as I need to move the bottle in and out of the freezer and not crack when flexed at low temperature.

The candidates I'm currently thinking of are normal gas bottle hose, hvac tube, air tool pressure line hose and hydraulic hose.

The gas bottle tube would be the obvious choice as it would be the correct diameter for any fittings I may use...but I'm not sure how resistant to cracking it is if flexed at -20c?

The most durable would probably be hydraulic hose as its designed for high pressure use in situations where it's required to flex and is reinforced with steel mesh.....just not sure if it would be flexible enough for my uses?

Last edited by technogiant; 01-11-2013 at 01:58 AM. Reason: Automerged Doublepost
United Kingdom  Offline
    Register to Reply to This Post
Old 01-11-2013, 02:50 AM   #6
Witchdoctor
"Think"
Witchdoctor's Avatar
Senior Member
 
Posts: 6,750
Last Seen: 04-03-2017
iTrader: 17 / 100%
http://www.bahcallrubber.com/pdf/cold_blue.pdf
United States  Offline
    Register to Reply to This Post
Old 01-11-2013, 03:09 AM   #7
technogiant
Overclocker
technogiant's Avatar
Senior Member
 
Posts: 335
Last Seen: 09-15-2014
iTrader: 0 / 0%
That cold blue looks great....just what the "Witchdoctor" ordered...lol....not sure how it would be as regards chemical compatibility though being made of pvc

Also found this while browsing, has nitrile rubber inner may be more compatible for hydrocarbon/hfc type chemicals as it is also used for oil kerosene and fuel oil....

http://www.goodyearep.com/productsde...7&taxid=315731

Has a lower temp rating but doesn't mention flex at low temp.

Last edited by technogiant; 01-11-2013 at 03:40 AM.
United Kingdom  Offline
    Register to Reply to This Post
Old 01-11-2013, 03:58 AM   #8
Witchdoctor
"Think"
Witchdoctor's Avatar
Senior Member
 
Posts: 6,750
Last Seen: 04-03-2017
iTrader: 17 / 100%
That looks like a WIN, anything will loose flexibility to some point as it gets colder, especially my joints .............

You could call them up and ask how well it maintains flex and what temps does it start getting hard AKA brittle
United States  Offline
    Register to Reply to This Post
Old 01-11-2013, 06:03 AM   #9
lundrog
Turbo VMware Nerd.
lundrog's Avatar
Senior Member
 
Posts: 402
Last Seen: 09-19-2016
iTrader: 1 / 100%
when you start working on this , you should setup a forum webcam... so we can watch.
United States  Offline
    Register to Reply to This Post
Old 01-11-2013, 06:07 AM   #10
technogiant
Overclocker
technogiant's Avatar
Senior Member
 
Posts: 335
Last Seen: 09-15-2014
iTrader: 0 / 0%
So now I know the sort of product I'm after actually exists I'll do some preliminary tests. It all pivots on being able to recover the liquid gas from the chamber back into the pressure bottle.
I'm just going to get an empty butane gas cylinder, attach a hose from it to a container filled with 10 liters of water/antifreeze, close the bottle valve, sling it in a chest freezer for a few hour and then open the valve with the container above the bottle to see if it is all sucked in, even if it just drains back more slowly by gravity that will be fine provided it doesn't take too long.

Nice idea lundrog...I was toying with keeping a video log rather than just still pics.....its going to be a slow build though as I have to sort out the room it's going in first, but I can get on planning and building the chamber to a point in the mean time.

Last edited by technogiant; 01-11-2013 at 06:17 AM.
United Kingdom  Offline
    Register to Reply to This Post
Old 01-12-2013, 04:50 AM   #11
Orthello77
Overclocker
Senior Member
 
Posts: 410
Last Seen: 12-19-2015
iTrader: 0 / 0%
TG , got to love a new build , will seriously be one unique build if this comes off. Like anything extreme not done before theres no manual for this sort of stuff. At least you don't have to worry about pumps !!! ...

Subscribed as usual ;-)
New Zealand  Offline
    Register to Reply to This Post
Old 01-14-2013, 01:45 AM   #12
technogiant
Overclocker
technogiant's Avatar
Senior Member
 
Posts: 335
Last Seen: 09-15-2014
iTrader: 0 / 0%
Thanks Orthelllo, I'm still adapting and evolving my ideas, come up with a way that won't require me to move the pressure bottle back and fore, will save putting strain on connections and possible pipe cracking which could be catastrophic.

I'd require a recovery pressure bottle with a twin tap outlet like these

http://i21.geccdn.net/site/images/n-.../RIT_95007.jpg

With my original plan the pressure bottle when being used to fill the chamber with gas would have to have been lifted higher than the developing fluid level, otherwise fluid would have just run back into the bottle and effectively just served to chill the bottle down. So it meant quite considerable movement of the bottle.

I've done a quick sketch of the new setup.....my drawing is rubbish but it's the method that's important.



Uploaded with ImageShack.us

So this is again a cyclic process so we start at ambient temps, freezer off and both valves off.

Turn on the air con unit and open valve/line 2......refrigerant gas enters the chamber is condensed and pools in the chamber bottom as valve/line 1 is still closed.

Whilst this is occurring the pressure bottle will of course decrease in temperature which will have the effect of pre-chilling the freezer. It may be that it would take the freezer temp down too low and prevent vaporisation from the bottle....that's something I'd have to experiment with, may even have to add warmth to the bottle to get sufficient fluid in the chamber in a reasonable time.

Anyhow, once there is enough fluid turn off valve/line 2 and turn the freezer on so it will be cold enough for when you want to shut down and recover the liquid.

All the while the chamber is at atmospheric pressure due to a background gas and expansion chamber....either air of nitrogen....I may go for bottled nitrogen as its cheap enough and dry.

When your ready to shut down just open valve/line1, if the freezer temp has dropped lower than the liquid boiling point the pressure in the bottle will lower than atmospheric and the liquid will be sucked back in. If not then more gas will enter the chamber and it may be that the ac unit will have to be left to run to condense that, all the while liquid will run back into the bottle by gravity helping to lower its temp below the boiling point of the liquid so ultimately its all back in there, it may be that opening valve/line 2 will help the liquid run back as it will allow the displaced gas to go back to the chamber.

It's going to take a bit of experimenting to work this out but think it will do what I want.

I've two concerns about it though:-

firstly what sort of aperture is in these gas bottle valve's? Would it be sufficiently wide to allow liquid to flow back under gravity at a decent rate?

secondly what happens if some of the background gas gets into the recovery bottle, its obviously not a volatile gas and while it will compress under pressure it will not liquify....would it cause too high a pressure in there?
Not that I think much would get in there any way really as the bottle is being refilled from the chamber sump so it would get the liquid first followed by the denser refrigerant gas, the nitrogen would tend to be at the top of the chamber being less dense.










Additional Comment:

Well I've been thinking more about the process of transferring the liquid back into the pressure bottle as this is the step that I'm not sure will work.

I wanted to experiment with an old gas bottle to see if water would flow back into it but can't get my hands on one.

So I've just been experimenting with some plastic containers and plastic tubing.

I've found that water will flow back into the sealed bottle from a chamber raised above it, but it does so intermitently.

It flows initially until the pressure builds up in the bottle then it stops with a column of water in the pipe blocking it.

That column gradually drips into the bottle until it suddenly breaks down and a flurry of air bubbles rush from the bottle into the chamber and the flow resumes until pressure builds again, and so the process continues.

I'm thinking that in the real set up the bore of the pipe will be larger than what I was experimenting with so the flow will be better.

Also the liquid I'm using will flow better than water. Going back to my chemistry lessons water is a polar molecule and experience relatively strong interactions between themselves. So water molecules have quite strong cohesive forces pulling them together which allows it to do things like hold together in an unsupported column of water.

The liquid refrigerant molecules are non polar and very weakly held together by something called van der wall forces, so the liquid will have very low cohesion and the liquid column should break down very easily and allow pressure equalization with the bottle and chamber.

So on that basis alone flow back will work, and when you consider that at least initially there will be a lower pressure in the gas bottle than in the chamber then initially liquid will be sucked back in until pressure equalizes, although I'm not sure how much will be sucked back that way.

I think that using as large a recovery bottle as possible will help the situation, as a lower pressure in a large volume bottle will suck back more fluid before pressure equalizes than a smaller bottle would.

Also once the pressure has equalized the pressure would build more slowly in a larger bottle than in a smaller one.

Additionally its possible to disrupt the column by agitation or adding pressure at one end.

So all in all I'm as sure as I can be that the recovery process is going to work adequately.......now to put my money where my mouth is and get on with the build.

Last edited by technogiant; 01-14-2013 at 02:27 AM. Reason: Automerged Doublepost
United Kingdom  Offline
    Register to Reply to This Post
Old 01-14-2013, 03:09 AM   #13
Orthello77
Overclocker
Senior Member
 
Posts: 410
Last Seen: 12-19-2015
iTrader: 0 / 0%
Thats interesting stuff about water been a polar molecule .. prolly why this koolance flowmeter FM18 i have won't work at all with the methanol mix i have (mixture is too soft possibly) .. either that or its just not working. I can blow through it and it will report an RPM but my liquid mix through it just reports nothing.

Re the pipe flow you could have a pipe in a pipe design , eg outer pipe larger than inner pipe , that would reduce flow a bit but would allow air to escape. Not sure if it would work with your valves though - maybe not. If the liquid flows better than water its probalbly a null issue.

These sort of builds require $$ i had someone look at mine the other day .. too scared to tell them what it cost - even i don't want to know lol .. oh well .. and if you did it without mistakes / learn't experience .. probably cost half. So good luck with the $$ side of it :-)

If fully immersed what is that refrigerant going to be like to clean up say on an upgrade .. much of a clean up job ?
New Zealand  Offline
    Register to Reply to This Post
Old 01-14-2013, 08:56 AM   #14
technogiant
Overclocker
technogiant's Avatar
Senior Member
 
Posts: 335
Last Seen: 09-15-2014
iTrader: 0 / 0%
I don't think that would be the problem with your koolance flow meter, more likely as it is reporting an rpm the water flow is rotating something in there, my first guess would be that the "rotating something" is running on fluid bearings which are freezing up and stopping rotation.

That's a nice idea about the pipe in a pipe....but I've been trying to think how I could adapt that to my usage scenario....but don't think it's going to adapt to it.

But I may well use a pipe in a pipe just as a safety measure in case the transfer pipe should fail. I'll have a larger diameter pipe around it vented off to the expansion chamber and ultimately external vent.
So basically the build will have a double walled chamber and a double skinned transfer pipe.....that should all but negate what I think would be the worst possible case scenario of having either of those fail and 10liters of liquified refridgerant dumped on the floor, could cause fatal asphyxiation.

Don't think there will be any clean up necessary after submersion....the fluid will just evaporate away, the one I hope to use is Hfc227ea better known in fire extinguisher use as FM-200. It's a replacement for halon gas as it has less environmental impact and is used to protect area where high value delicate stuff is housed, like server rooms etc....it causes no damage and leaves no mark. But tbh I think that any refridgerant gas will just evaporate off not just hfc227ea.

As regards costs....well I cringe at the cost of building my original unit....and I'm going to build another!!!!
Having said that I'll be able to reuse a lot, the ac unit for instance and that was about half the cost of the original total build, the fans, fan control unit and cable extenders will all transfer over so it it will mainly just be material costs...plus >100 for the recovery bottle and >100 for a freezer...and the real biggy >350 for a 13.5kg bottle of hfc227ea!!! (may even need 2x of those)

Last edited by technogiant; 01-14-2013 at 09:37 AM.
United Kingdom  Offline
    Register to Reply to This Post
Old 01-18-2013, 07:57 PM   #15
ORL
Bring on the Liquid
ORL's Avatar
Senior Member
 
Posts: 2,141
Last Seen: 07-11-2017
From: Missouri
iTrader: 0 / 0%
You are going to have extreme chill the exit tube from the bottle reservoir. If you plan to use a subzero bp liquid. The problem you will encounter is called vapor locking. As the fluid exits the bottle the temperature will chance drastically and evaporation will occur creating a pocket of gas in the exit tube which will lock the fluid flow until enough rear pressure has been built to push it back out. This will not happen if you do not have a one way valve/pump assembly. The evaporation will cause equilibrium inside the assembly and halt all flow.

There are a few safety features we need to discuss first though. One, you need adjustable relief valves, and not just one. You need these all over this system. A failure when you have only a single relief valve will cause a catastrophic pressure release and then the tank will become a rocket if the pressure flow locks out the one way valve and trys to force liquid out of the release. The fluid will transition faster than the fluid will flow causing ruptures.

Next all of this work should be performed behind a blast shield. If you had a pressure failure the liquid will be ejected at a velocity that can actually cut a finger off. I seen a man lose the tip of his index finger to a pressure washer he had modified to help remove grout faster in out doors stone work.

I am going to spend some time reviewing these posts and your ideas and will provide any help I can.
United States  Offline
    Register to Reply to This Post
Old 01-19-2013, 01:47 AM   #16
technogiant
Overclocker
technogiant's Avatar
Senior Member
 
Posts: 335
Last Seen: 09-15-2014
iTrader: 0 / 0%
Thanks for your views Orl.....I'm looking for such input as I want to do this safely and if I feel it to be inherently unsafe will not go ahead....after all its only a computer cooling system and not worth taking risk to life and limb for.

The point you raise about a vapor lock I don't think will be an issue for several reasons.

Firstly the exit/return pipe is going to be mounted vertically, there will be no U bends or anything like that to trap vapor. As I've already found with some limited experimentation the problem is not vapor locking in the tube but back pressure in the bottle caused by fluid return prevents the fluid column from flowing back....after a short time period the fluid column breaks down and vapor rushes from the bottle to the chamber.....but my view on this is that as the pipe bore will be wider than that which I experimented with together with the fact that the liquid will have lower cohesive properties than water then the return rate will be greater than that which I found through experiment. Additionally agitating the fluid column made it break down easily so I don't think that will be a problem....but of course this is experimental and all we can do is make our best guess.

Secondly there should not be an issue with fluid vaporizing in the return tube for two reasons, firstly the tube will already be full of fluid from the chamber above as the valve will be at the bottom of the tube on the recovery bottle, so it will be at sub boiling point, and additionally the bottle and return pipe will be housed in the freezer which will be at sub boiling point temperatures during the return process.

As regards the valves, I'm not quite sure I understand the concern....perhaps you could elaborate please.

The system as I envisage it would only use the two valves on the recovery bottle one attached to the fill line (line 2) and one attached to the exit/return line (line 1)....I do not envisage using any pressure regulator on the fill line as I will have a large volume expansion sac either 65 liter of 240 liter.....so gas entry would be manually regulated by turning the recovery bottle valve on and off to alternately fill the expansion sac and then waiting for the gas to be condensed down to liquid then repeating the process.

Is the recovery bottle manual turn valve really likely to fail or lock open during the fill process where it will just be venting gas as it was intended?

.....if it did I do intend to have an externally vented breather pipe just sealed with a stopper that would pop off under pressure....but whether that would cope with full bottle pressure and a fully open locked valve would be doubtful.....perhaps a secondary inline valve on the filling line would negate this concern.

As regards a blast shield.....well if I thought that was required I'd not be doing this project at all.....

The beauty of this system is that it is all done at atmospheric pressure with next to no risk of explosive pressure release.
The only pressure is in the gas bottle which was designed for it.

The low temperature of the chill box is what is constraining the liquid refriderant not pressure.

The only event that could lead to pressure build up would be if the air con unit failed while the chamber was filled with fluid and simultaneously I was for some reason also unable to recover the fluid to the recovery bottle.

Even in those circumstances pressure build up would not be explosive....as the liquid is contained in an insulated chamber it would only boil off slowly at a rate determined by the ingress of heat from the environment through the insulated chamber and into the fluid.

Any boil off in those circumstances would I'm sure be handled by the externally vented breather tube.

The worst case scenario I can envisage would be either the chamber or exit/return pipe cracking or splitting while full of fluid.....but as described earlier I intend to have a double skinned chamber and return pipe, this outer layer will also be thermally insulated to prevent potentially explosive thermal boil off and will also be vented to the external breather tube, so this should not be an issue.

Last edited by technogiant; 01-19-2013 at 05:46 AM.
United Kingdom  Offline
    Register to Reply to This Post
Old 01-19-2013, 02:22 PM   #17
ORL
Bring on the Liquid
ORL's Avatar
Senior Member
 
Posts: 2,141
Last Seen: 07-11-2017
From: Missouri
iTrader: 0 / 0%
Here is my main concern, if your cooling fails, the system is unable to handle the pressure at a full ambient and it fails. This will result in an astronomical change in pressure and as it decreases the evaporation quickens in the chambers. This will eject all of the coolant in either liquid or gas form from the systems rupture until its able to reach a temperature/pressure balance. If this hits flesh in liquid form its instant frost bite. Although this is no different than the risks with other forms of extreme cooling.

Techno, I am not sure you will be able to get this to return to liquid state in a timely enough manor without a compressor setup on the line either, it depends a lot on how cold your freezer can go.

Edit, On this subject consider a condensing coil inside the freezer for the Gas return line.

Also another thing to consider, you will have to run the bottle in the Freezer upside down. Or the returning gas will keep entering the tank and not let fluid return through the other valve. If you flip the bottle upside down the Gas will flow to the top of the tank and in return force the fluid down the other tube. But you will still need to install a tube inside the tank that guides the gas to at least mid way up the tank above the fluid level, or the fluid level will once again cause a lock.

I do like the idea of running a multiple stage system using essentially an enclosed pot capable of recycling the liquid. This is something I have considered before but never tried. I want to see this work, I am just offering thoughts.

Additional Comment:

Another thought as well, You are going to need to put the exit valve from the PC chamber at the max level you want the coolant to reach. As the gas expands it will attempt to force all fluid from the Freezer tank into the PC chamber, by placing the exit right there it will start recycling liquid instead of just forcing all liquid from cooling tank into chamber.

Last edited by ORL; 01-19-2013 at 02:22 PM. Reason: Automerged Doublepost
United States  Offline
    Register to Reply to This Post
Old 01-19-2013, 03:21 PM   #18
technogiant
Overclocker
technogiant's Avatar
Senior Member
 
Posts: 335
Last Seen: 09-15-2014
iTrader: 0 / 0%
Again thanks for your views and comments Orl, but I don't think I've explained the system properly, some of the things you've mentioned don't fit with the usage scenario unless I'm misunderstanding you.

Quote:
Originally Posted by ORL View Post
Here is my main concern, if your cooling fails, the system is unable to handle the pressure at a full ambient and it fails. This will result in an astronomical change in pressure and as it decreases the evaporation quickens in the chambers. This will eject all of the coolant in either liquid or gas form from the systems rupture until its able to reach a temperature/pressure balance. If this hits flesh in liquid form its instant frost bite. Although this is no different than the risks with other forms of extreme cooling.
The chamber itself is not meant to nor will it be able to contain any pressure.

In the event that the air con unit fails after the the chamber has been filled with fluid the first course of action would be to open the valve on line 1, that would suck/drain the fluid back into the bottle.

If that failed for some reason then the temperature of the liquid would gradually raise. This is a slow process as the chamber is thermally insulated. My current chamber takes many hours to return to ambient once powered down.

Once the liquid reached boiling point it would not all boil off immediately and explosively. The action of boiling off would in itself maintain the liquid temperature at the boiling point as it removes heat from the liquid and the gas it produced would be vented off. The rate of gas production would be limited by the rate that heat can enter the chamber, which is insulated and as indicated by my currents chambers slow return to ambient is only a slow process.
So provided the breather tube was able to vent the gas at a sufficient rate then pressure build up in the chamber would be minimal.

Think of it like an insulated flask containing liquid nitrogen, that has no means of cooling and is not a pressure vessel and yet it does not boil off explosively, it just boils off at a rate determined by the heat ingress through the insulated flask.

Quote:
Originally Posted by ORL View Post
Techno, I am not sure you will be able to get this to return to liquid state in a timely enough manor without a compressor setup on the line either, it depends a lot on how cold your freezer can go.

Edit, On this subject consider a condensing coil inside the freezer for the Gas return line.
This is where I don't think I've explained myself properly, the freezer and gas bottle are not part of the operational cooling system. They are solely to contain the gas between uses. Once the chamber has been filled with gas from the cylinder and that condensed by the ac unit to give sufficient liquid to submerse the components then both valves to the gas bottle will be turned off.
There is no circulation or exchange between the chamber and the bottle/freezer while the system is operating, gas generated by evaporation on the pc components is re condensed by the ac units evap and drips back into the submersion pool. The ac unit has a cooling power of 3.6kw and so should be ample to re condense the evaporant in timely manner.
The freezer is not involved in the cooling task.

The reason for including the freezer is just to allow the liquid to be returned to the gas bottle between uses. Once the gas has been condensed into the chamber and the gas valves closed then the freezer will lower the temperature of the now sealed gas bottle below the boiling point of the refrigerant, once below that point the pressure in the bottle will fall below atmospheric pressure. This is important as when you want to power down and return the fluid to the bottle then opening line/valve 1 will suck the fluid from the bottom of the chamber which is at atmospheric pressure. If the bottle was not chilled its contents would be at high pressure and simply blow gas into the chamber and not suck fluid back.

Quote:
Originally Posted by ORL View Post
Also another thing to consider, you will have to run the bottle in the Freezer upside down. Or the returning gas will keep entering the tank and not let fluid return through the other valve. If you flip the bottle upside down the Gas will flow to the top of the tank and in return force the fluid down the other tube. But you will still need to install a tube inside the tank that guides the gas to at least mid way up the tank above the fluid level, or the fluid level will once again cause a lock.
Again this will not be part of the cooling cycle and will only be of relevance when shutting down and returning liquid to the gas bottle. Initially as the gas bottle will be chilled below the boiling point of the liquid then the gas bottle pressure will be lower than atmospheric and so it will initially suck fluid back through line/valve 1. This will of course tend to equalize the pressure until such time that the liquid flow back is impeded by the back pressure. This is what I've been doing some limited experiments on and for the reasons I mentioned previously believe that the fluid will still drain back adequately just through gravity once the suction has been used up, possibly opening line/valve 2 may help by allowing the pressure to equalize more readily...that's something I'd have to experiment with.

I'm thinking about your suggestion to put the gas cylinder upside down. It would have the disadvantage that if the low pressure in the bottle was unable to suck back all of the fluid then the remainder would not drain back as there would be no route for gas to escape from the upturned bottle.
Also you would always be left with an amount of liquid in the pipes at the same level as the fluid level in the bottle as they would form "U" bends as they come down then back up to the valves. This could have a large gaseous volume when returned to ambient.

Lastly I don't think gas bottles are safe to store upside down.

Last edited by technogiant; 01-19-2013 at 04:19 PM.
United Kingdom  Offline
    Register to Reply to This Post
Old 01-19-2013, 06:24 PM   #19
ORL
Bring on the Liquid
ORL's Avatar
Senior Member
 
Posts: 2,141
Last Seen: 07-11-2017
From: Missouri
iTrader: 0 / 0%
Ah see I had your features and functions mixed up. I was under the impression you were designing an active cooling solution using temperature changes as the means of transport via evaporation/pressure propulsion.

I assume then you are simply trying to make a fully submersed pot with capability of not wasting the used fluid/gas. In my head I was thinking of a mix between phase change/contained pot with active circulation for dropping the temperatures.
United States  Offline
    Register to Reply to This Post
Old 01-19-2013, 11:52 PM   #20
technogiant
Overclocker
technogiant's Avatar
Senior Member
 
Posts: 335
Last Seen: 09-15-2014
iTrader: 0 / 0%
I'm doing both......the pc will be cooled by submersion in cooled liquid, but the main cooling feature will be phase change of the liquid to gas.....the gas generated will be re condensed by the ac unit and drip back into the submersion pool.

The gas bottle and freezer are just to generate and recover the liquid before and after use. The propulsion being from pressure temperature and gravity.

I think I'll re do my diagram in more detail.....If you missed my meaning Orl then I'm sure many others have also.
United Kingdom  Offline
    Register to Reply to This Post
Sponsored Links:
Please Register to Post a Reply

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 -6. The time now is 10:47 PM.

Copyright ©2000 - 2016, Jelsoft Enterprises Ltd.
Powered by vBulletin
Copyright ©2000 - 2016, EXTREME Overclocking