Why a mini fridge won't work, Revisited. FREEZERS!

Ok, I read the article about the Minifridge and the ExtremeOverclocking forum, but I was wondering about some different ideas.

The Idea I was thinking of Is getting a Freezer, and putting my whole case in it. I would drill a whole and put in a few USB extensions plus audio cables. I would take care of sealing this whole.

Now I assume the Freezer alone wouldnt cool this system to sub ambient temps, and I also know that I run Watercooling. I would create a Bypase valve that would hook up to the freezer, and that would re-direct my water through a few phasechange based watercoolers mounted on the outside of the freezer. On the outside of the case. I hear thats what a few coolers from water fountains may be usefull.

If I have subambient air temps in the freezer, and the freezer has auto defrost, I would assume that I could then not worry about condensation or frost, and wouldnt need thermal grease on my mobo cpu socket etc.

The freezer air would help keep the entire system cold, even the parts that are not normally cooled. And because of the waterchillers they would be able to add additional cooling to the water loop. Thus maybe eliminating the extreme stress of such a system.

If you could continue to drop the temps of the system, and add more and more phasechange coolers, you could get the liquid to drop below or at the temp of current phasechangers. Yet you would have the freedom to remove the case and run it at lower settings for lan parties.

Of course you would have to switch from water to a subambient cooling solution (like liquid Freon), except that it would have to remain liquid at around 110f.

The liquid issue is one area of concern, because it would have to be non-corrosive.

Another area of concern is which pumps could handle such low temps.

I am wondering what others feel about a system. Honestly I dont think I will build it, unless I can find some parts in Craigslist and/or junkyards, and if I could do so cheaply. It also seems like allot of work, and may not be physically possible.

I think such a system would be 100% more efficent then a peltier system, and you wouldnt have to worry about any condensation due to the fact that at 0c their will be no water in the air.

I hear they use AirConditioners, which would work if placed mounted in a custom box. I could then make this box to fit the exact size of my case. I again dont know what the overall power of these systems could reach.

Tell me what you think...
Mike

Again for the size of this, and the fact that INSULATION IS NOT HARD PEOPLE!! why wouldn't you just save money on a normal phase system.

I think you'd burn out the compressor on the freezer, which isn't meant to run continuously (which it would have to do to keep up with that heat load). I guess you could replace it with a better one.

Why put the whole case in? just put the mobo and components in. or gasp, just remove the compressor and hook it up that way. But, ya, i dont think a normal freezer can handle the heatload of a modern computer thats on for more then 30 seconds.

Basically what you are trying to do is put the case into a small, very cold air conditioned room. You shouldn't have any issues with condensation, except when you shut the power down on the AC. Then you will have a pc that is colder than the surrounding air(if you were to take it out of that small space into the room you are in). The main problem with a setup like this is that the chilled components MUST remain separated from the warmer, moisture laden air outside the enclosure. You would want to remotely mount your Optical drive. Otherwise if you open the door to change a disk....wham, nice warm, moisture filled air just hit the system. Second, is the fact that these small fridges aren't designed to run in a continuous duty cycle. You would be better off with a unit designed for AC(say a small 5000 BTU window unit) which adds to the size and complexity of the project. Also, those things really suck some power, prepare for outrageous electric bills. I don't think that adding in a WC loop is the answer you need, as a high end air cooler will do just fine. Since it will have much lower air temps to work with, it will give you much lower working temps.

From a $$$$ viewpoint I think that you are going to spend a ton for a small return. The complexity of such a system is IMHO not worth it. Especially since it sounds like you want the best of both worlds(subambient and portability).

A freezer is tuned to the same heat load capasity as a refigerator if not less (considering the expectation that some of the product is already frozen most of the time) and is only tuned to reach a different on load temperature.

In other words, your dealing with the same functional problem as you are with a refigerator.

greenmaji said:

A freezer is tuned to the same heat load capasity as a refigerator if not less (considering the expectation that some of the product is already frozen most of the time) and is only tuned to reach a different on load temperature.

In other words, your dealing with the same functional problem as you are with a refigerator.

Click to expand...

Its just such a good idea...
Lower Ambient temp bellow 0c.
Then Lower Liquid Cooling Temp WAY down.

Its so easy to cool liquid with phase change.

I think of this because if someone could make a case that holds your CD-Drives, etc outside. Then provides all of your computer parts inside (plus HardDrives), then just provides extensions to all of the wirers from the MoBo with external wire jacks.

Use an External WaterPump + Water Cooler and Airconditioning. All pre-fabbed up nice.

The only issue would be to control the noise the system makes outside. Inside HD's, Fans, ETC would all be silented by the insolation.

The heat load is less is the problem, food gets cold and stays cold unlike computer components that produce heat in operation.

In any of these types of schemes (I'll use 2 freezers for this and a pelt for this and bla bla bla) all you have to do to see if it will work is look at the heat load.

How much power does your computer consume at full load? Most of us have a 400+ watt power supply, and they arn't all that efficient.

Show me a freezer/watercooler/fridge that can handle that heat load continuously, and I will show you a retail DirectDie system that costs less. It will also work much better, be quieter, more efficient, and won't take up your entire garage.

Freezers get colder, that doesn't mean they have more capacity. In general they actually have less capacity because to reach colder temps they run with less refrigerant under more vacume. They are also designed to be opened and have warm things put in them less often than a fridge.

GreenJelly said:

Ok, I read the article about the Minifridge and the ExtremeOverclocking forum, but I was wondering about some different ideas.

The Idea I was thinking of Is getting a Freezer, and putting my whole case in it. I would drill a whole and put in a few USB extensions plus audio cables. I would take care of sealing this whole.

Now I assume the Freezer alone wouldnt cool this system to sub ambient temps, and I also know that I run Watercooling. I would create a Bypase valve that would hook up to the freezer, and that would re-direct my water through a few phasechange based watercoolers mounted on the outside of the freezer. On the outside of the case. I hear thats what a few coolers from water fountains may be usefull.

If I have subambient air temps in the freezer, and the freezer has auto defrost, I would assume that I could then not worry about condensation or frost, and wouldnt need thermal grease on my mobo cpu socket etc.

The freezer air would help keep the entire system cold, even the parts that are not normally cooled. And because of the waterchillers they would be able to add additional cooling to the water loop. Thus maybe eliminating the extreme stress of such a system.

If you could continue to drop the temps of the system, and add more and more phasechange coolers, you could get the liquid to drop below or at the temp of current phasechangers. Yet you would have the freedom to remove the case and run it at lower settings for lan parties.

Of course you would have to switch from water to a subambient cooling solution (like liquid Freon), except that it would have to remain liquid at around 110f.

The liquid issue is one area of concern, because it would have to be non-corrosive.

Another area of concern is which pumps could handle such low temps.

I am wondering what others feel about a system. Honestly I dont think I will build it, unless I can find some parts in Craigslist and/or junkyards, and if I could do so cheaply. It also seems like allot of work, and may not be physically possible.

I think such a system would be 100% more efficent then a peltier system, and you wouldnt have to worry about any condensation due to the fact that at 0c their will be no water in the air.

I hear they use AirConditioners, which would work if placed mounted in a custom box. I could then make this box to fit the exact size of my case. I again dont know what the overall power of these systems could reach.

Tell me what you think...
Mike

Click to expand...

Bad Idea from start to finish unlese you have a huge industrial feezer or dont want much in terms of temperature, in which case why bother anyway.

Also in case you do have a huge industrial freezer it is still not tuned to remove a large load very fast, so you will have to cut it up, recover refrigerant,respec and replace parts, evac and recharge. Hope you have money tools and a licence (if you want to get any refrigerants, other wise you are limited to r134a and propane).

To cool the whole computer is not only pointless but is also increadibly stupid.

First of all your computer as a whole outputs much more heat then just your cpu/gpu so the load is alot larger further driving the point that unlese you have a huge industrial freezer this isnt really going to work.

Second, its stupid because cooling all the other parts besides your pcu/gpu will gain you nothing but will however cost you something. That something is what you could have had had you not wasted your cooling on things that dont need cooling and cooled things that actually benifit from more cooling.

Do you even know what auto defrost does? Yes you will have condensatrion problems, what are you thinking? If you go below subambient you will have condensation probalems. How badly is a factor of humidity and temperature but it doesnt take much to kill your hardware. At 0C you wont have water in the air because it will be in the form of ice ON your hardware. However before it freezes it will be condensing as a liquid ON your hardware and will short it out. And when it melts it will again.

You want to switch water for "liquid freon"? I sugest you do alot of reading before you kill yourself or someone else because you are absolutly clueless. Freon refers to refrigerants, which are gasses at standart temp/pressure, like propane for eaxample. Do you know what it takes to make propane liquid? Alot of pressure and a sertain temperature range. You want to run a liqid cooling loop at 100-200psi?? ..... not to mention all those substances are infinetly worce at moving hear compared to water. What you would use if you knew what you were talking about would be a water and methanol or ethanol mix.

Iwaki md20 RLZ is good, panworld 40px, iwaki RD20/30 are great.

Phase change baised refrigeration is more efficient then pelts but dont be an idiot and thow around numbers you pulled out of your a$$ like "100% more efficent", especialy when they are that far off.

Btw, the way you are talking about throwing freezers everywhere are you paying the electricity bill? Do you have any remote idea how much all this will cost you to run?

It looks to me like you understad nothing and will fail very fast. I sugest you do alot more reading, maybe spend a few months here or at xtremesystems.org and actually understand how things work and what you are getting in to.


PS:

You dont need to add freezers to keep lowerig the temperature of your coolant, given if they all pull down to lets say x degrees you dont go lower then x no matter how many you add. You can however have a cascade that can pull the heat exchanger down to the bp of your last stage gas, lowest done on XS was -170C. But thats not a beginers project by any means.

http://www.xtremesystems.org/forums/forumdisplay.php?f=80

Though I dont know if frost will form on the motherboard, but I do know that the temp of the components will be warmer then the temp of the air. If this is above 0c, and the air is bellow 0c, I dont really know what would happen.

Usually if your components are above the temp of ambient, the water in the air evaperates, instead of collecting. If the temp of ambient is above the temp of components then it condensates.

Once you get past 0c on both your components and in the air, and we do not add any more air to the system, I assume that their is NO moisture in the air.

I do know that Moisture will form on the cooling components as they push the air too 0c. I just dont know what happens to the properties of water and air after they hit the 0c mark. I would assume that their is no water in the air.

People do cool their entire PC's like I am talking about. Its called air conditioning. unfortunately they do not go as cold as I want to go.

So I ask this question... If its cheaper to cool the individual components then the entire system, then why are mainframes always cooled with air, which is then cooled with an airconditioner?

The reason I think this idea is so attractive is that it could be a very safe way to reach subambient temps.

I hate phasechange PC units, because they require a seperate box, are expensive, are fragile, and require modifications to the motherboard and other areas. I just dont want to add silicone and thermal grease, with all these other components to my system.

Peltiers also require modifications, are also expensive, require an additional water cooling loop, are inefficent and very power hungry, and if the cooling of the hotside stops... they can catch on fire...

Mike

BTW: They cool the air, cause it is cheaper then cooling the individual parts.

Wait a minute...

People... I assume we are people... and not some alien species...

Why do we tear this idea apart, instead of thinking of a way to make it work... We should be brainstorming ideas, and coming up with solutions. I think we should be positive, and not negative. If we really look at the problem, with the idea that it could be done, then we may end up finding out that it is possible.

Just my 2 cents.

I don't think cooling the whole computer is pointless, the gains would just be very small relative to the expense and work. I would suggest a special phase change system to remove the heat from the cpu to outside the freezer, and replace the stock compressor/whatever else needs to be replaced with components up to the task.

If you want to use forced air AC it would be much easier to build one for that purpose rather then entertaining the idea of a freezer intended for food

You seem to have already come to the concludion that air is a very poor conductor of heat, and have desided you can live with that. That would be the biggest disadvantage of using air.

greenmaji said:

If you want to use forced air AC it would be much easier to build one for that purpose rather then entertaining the idea of a freezer intended for food

You seem to have already come to the concludion that air is a very poor conductor of heat, and have desided you can live with that. That would be the biggest disadvantage of using air.

Click to expand...

unfortunately we are currently stuck one way or another using air to cool our systems. Even water cooled and peltier systems require air to cool... Phasechange also requires air.

The difference in all of the proposed solutions "extreme" solutions (AC, Phasechange, TEC) is that they make the difference of air temps to heat exchanger larger then that of a pure water cooling system.

I cant beleive there are no freezers that are made for such a heavy load. Even an air conditioner with a sub 0c temp range would be great. With a large capacity to handle a large number of BTU's. I wonder if such a system would be giant, or if their are smaller options.

Using air as the thermal conductor..
No, phase, water nor even "air" cooling uses it.

GreenJelly said:

unfortunately we are currently stuck one way or another using air to cool our systems. Even water cooled and peltier systems require air to cool... Phasechange also requires air.

The difference in all of the proposed solutions "extreme" solutions (AC, Phasechange, TEC) is that they make the difference of air temps to heat exchanger larger then that of a pure water cooling system.

I cant beleive there are no freezers that are made for such a heavy load. Even an air conditioner with a sub 0c temp range would be great. With a large capacity to handle a large number of BTU's. I wonder if such a system would be giant, or if their are smaller options.

Click to expand...

There are, but they will run you $5000, and then $1000 per month for electricity bill. Instead of contemplaing the possibilities of the univerce of a forum go and do some work yourself. Go look and see if you can find a freezer for your load. Then go see how much it costs and see how much it will cost you to run it. You people come here understading nothing made no effort what so ever to do research or read books or actually go around and see what things cost and how they work and expect everything to be served up on a silver platter.

Saying that cooling air is more efficient then idividual parts is again an empty statment and means nothing because that is dependant on the situation. Understand that maybe cooling every cpu on a server farm with thausands of computers may not be "practical" so the rooms are air conditioned. However on a scale of one pc its the other way around.

Personaly I am not a fan of direct die myself. Chillers are the best option.

GreenJelly said:

unfortunately we are currently stuck one way or another using air to cool our systems. Even water cooled and peltier systems require air to cool... Phasechange also requires air.

Click to expand...

actually thats wrong. Green makes a good point as the contact pointis never air. But I think you are talking about it in a difference sence anyway. Point being you can move the heat in to a water loop then evaporate the water, that can be done even on the moon where there is no air, so no you dont have to have air. We just use it on the end because its convinient, however you clearly lack chemistry and physics knowledge. Using air to cool a heatsink and to cool a condensor is the same thing in essence yes but the result is very different, one is 40C load temp, other is -40C load temp. You have to look at it an an energy problem.

GreenJelly said:

Wait a minute...

People... I assume we are people... and not some alien species...

Why do we tear this idea apart, instead of thinking of a way to make it work... We should be brainstorming ideas, and coming up with solutions. I think we should be positive, and not negative. If we really look at the problem, with the idea that it could be done, then we may end up finding out that it is possible.

Just my 2 cents.

Click to expand...

The reason the idea is torn apart as you say because its flawed from the start, there is no point to making it work as YOU describe it. There are variations but far from you sugest. The reason its torn down is the same reason ford would tear you down if you come down to them with a proposition to make a car that runs on chocolate. You can but given the alternatives we have are better in every single way its just moronic.

You seem to turn down alternatives simply because you have only misconseptions and misinformation about them. If you want to cool air even though its a moot point given the alternatives you can, but it will be more expencive, harder, and with much less results.

You say you dont like phase change units because they need an extra box? If your freezer is not an extra box I dont know what is. You may have 1 box instead of 2 however it will be 4 times as big so whats the point. Also phase chance direct die units are make in the same case with the pc all the time.

You dont like phase change direct die units because of the price? You have not read what I wrote. Let me repeat, a freezer will need to be retuned, which will involve cutting it up, which will cost you much more money then a used vapo for example. And will require skill and very good understanding of the concept which you clearly lack.

You will need to insulate your case as waster will condense ont he outside when it gets cold and will make a mess in yoru room.

You will need to make a closed loop circulation system to move the air.

You will need to replace the air with 100% dry air, or better yet something like dry nitrogen or agron, those are the best options. Otherwise you will have condensation on your hardware I gurantee it.

GreenJelly said:

Though I dont know if frost will form on the motherboard, but I do know that the temp of the components will be warmer then the temp of the air. If this is above 0c, and the air is bellow 0c, I dont really know what would happen.

Click to expand...

I do know. Things dont happen instantly. First the water in the air will condense on your hardware and short it out, eventually it will freeze in to ice. And once the cooling is turned off it will melt. Yes you will have condensation if you go roughtly below 15C (dependant on your humidity, below 15C is a givem pretty much though). And if you go below 0C you will have ice.

GreenJelly said:

People do cool their entire PC's like I am talking about. Its called air conditioning. unfortunately they do not go as cold as I want to go.

Click to expand...

Well you just made our point for us. Theyd ont do it because its not smart or reasonable. Stop trying to outsmart the world, its not not done because we havent thought of it here or it hasnt been tried yet. Its not done because it has been tries and it has been established to be a waste of time and money.

Gnufsh said:

I don't think cooling the whole computer is pointless, the gains would just be very small relative to the expense and work. I would suggest a special phase change system to remove the heat from the cpu to outside the freezer, and replace the stock compressor/whatever else needs to be replaced with components up to the task.

Click to expand...

Actually it is pointless but let me define it. By pointless I mean to say cooling your capacitors for example will not net you anything. Further more if you do lower the resistance overall maybe you could potentialy gain 0.00000000001mhz from it but lets say any gains that small are = to pointless.

Further more to prove its absolutly pointless I offer this again. If you cool everything you are wasting cooling instead of using it on parts that can benifit from it like say your cpu, your gains actually Decrease if you cool everything since the same resourses could have been used to get you may more bang for your buck.

Wow, Farnsworth is one of thoose guys that critisizes everyone and everybody. I happen to be having fun in this conversation. If you dont like it, then dont partake. Allot of what you say is correct, but allot of it is so wrong. You assume and belittle people. "You People...", "...its just moronic.", etc...

I beleive that not only the quest can be accomplished, but that it can be done in a reasonable way. Everytime someone tries to do something, their are people who say "It cant be done"... Yet countless times the critics where WRONG! Infact every conceviable idea has critics and skeptics. Often the critics prevent people frome even triing. To assume that it is impossible is to simplify the world to an extent that just isnt reasonable. I love people who justify why things cant be done because it hasnt been done in the past.

If you dont want to try and make something that is "Impossible" into "Possible", then maybe you shouldnt respond.

GreenJelly said:

I beleive that not only the quest can be accomplished, but that it can be done in a reasonable way. Everytime someone tries to do something, their are people who say "It cant be done"... Yet countless times the critics where WRONG! Infact every conceviable idea has critics and skeptics. Often the critics prevent people frome even triing. To assume that it is impossible is to simplify the world to an extent that just isnt reasonable. I love people who justify why things cant be done because it hasnt been done in the past.

If you dont want to try and make something that is "Impossible" into "Possible", then maybe you shouldnt respond.

Click to expand...

You know the funny thing is yes our technology is often advanced by people who are criticized. But living in a scientific community let me explain something to you, this is how it works. If everything new wasn’t scrutinized there would be a lot of misinformation and incorrect conclusions floating around. That being said the "criticism" is fought by an "educated" person in defense of their theory on some biases. You on the other hand have no education even close to this area, if any at all. Have no experience, no information and no understanding of the concept. You are in no position to educated those who have been here doing this all this time.

You come here for technical help, so either take it or leave it. We arent here to offer you counsing or worry about your feelings.

Now get it right already, no one is saying it cant be done, everyone is saying it shouldnt be done because there are better ways to go about it. Stop being so defencive about it, entering this community you better be ready to listen to people when they tell you no this isnts the smart way to go about it, they know better then you they actually tried this before.

GreenJelly said:

Wow, Farnsworth is one of thoose guys that critisizes everyone and everybody. I happen to be having fun in this conversation. If you dont like it, then dont partake. Allot of what you say is correct, but allot of it is so wrong. You assume and belittle people. "You People...", "...its just moronic.", etc...

Click to expand...

Farnsworth is one of those guys who has been here actually doing this for the last few years, who is a mechanical engineer major, who actually knows what he is talking about as opposed to a what apperars to be a sensitive child who is convinced he is going to outsmart the world.

If I said something wrong go ahead and proove me wrong I am waiting.

GreenJelly said:

I beleive that not only the quest can be accomplished, but that it can be done in a reasonable way.

Click to expand...

Then go try your theory out. We won't go where you live and try to stop you. Just don't come to us crying if it fails.

GreenJelly said:

Everytime someone tries to do something, their are people who say "It cant be done"... Yet countless times the critics where WRONG!

Click to expand...

That may be true, but on many more occasions the critics are right.

Many people learn that this idea is impractical by attempting it or quickly seeing the amount of reasources it takes to successfully accomplish this task.

GreenJelly said:

Infact every conceviable idea has critics and skeptics. Often the critics prevent people frome even triing. To assume that it is impossible is to simplify the world to an extent that just isnt reasonable. I love people who justify why things cant be done because it hasnt been done in the past.

Click to expand...

I don't see anyone who stated that this is impossible. Your argument has no basis nor relevance in this case.

The best arguement has yet to be made. Air is a poor conductor of heat. Answering that question raises another, "If your going to go to all that effort to get around the fact that air is a poor conductor, then why not try idea X that is much more practical?"

The truth is that these ideas are not new in any way nor uncommon. The reason that the thread "Why a mini fridge won't work" got stuck, wasn't because someone got bored and messed around. They did it because so many people have had pretty much the same idea that turns out not to work well if at all.

As Matttheniceguy said, If you find a way to do this, we will show you a cheaper way to do it better.