# I Freezer without refrigerant

Tags:
1. Sep 20, 2017

### Frenemy90210

How about using air inside the freezer as a refrigerant instead of some refrigerant gas ( such as Freon/Ammonia.). We can Compress the air inside the freezer in the compressor, loose the heat and leave the thus cooled air back in the freezer. ; and the cycle continues.

What are the problems with such a system ?

2. Sep 20, 2017

### Staff: Mentor

What are the factors (related tot he refrigerant) that affect the efficiency of an actual refrigerator?

3. Sep 20, 2017

### sophiecentaur

Basically, efficiency would be low, using convenient compressor technology and operating temperatures. Refrigerant materials tend to be chosen for their change of state within the conditions in the cycle.

4. Sep 20, 2017

### Staff: Mentor

To add: the phase change allows a smaller amount of refrigerant to carry a vastly larger amount of energy and transfer it more efficiently due to the constant delta-T

5. Sep 20, 2017

### Khashishi

It sounds like this freezer could not run in a steady state to keep things cold. It would have to compress the air, making it hot inside. The air would eventually cool off to room temperature. Then it would decompress, making it cold inside. The air would eventually warm to room temperature. And you'd have to repeat.

6. Sep 20, 2017

### sophiecentaur

How is this different in principle from any refrigeration cycle?

7. Sep 20, 2017

### Khashishi

Well, in a normal refrigerator, you compress the refrigerant, not the air in the freezer. So the contents of the freezer don't get hot.

8. Sep 20, 2017

### sophiecentaur

I re-read the OP. I guess I interpreted it in a way that would actually make sense but, you could be right; it could be suggesting that the sausages and milk would also get hot and cold every cycle. But I think he means to imply that you could cool air, which would then be passed into / through the food compartment.
PS You wouldn't need to recycle air; it could come from outside (along with the spores and bacteria from the room!)

9. Sep 20, 2017

### Frenemy90210

Well, I meant take the air that is inside the freezer, compress it outside, all the heat will be lost outside, thus not warming the insides of the freezer, and then release back the compressed air ( which is at room temperature) inside the freezer; As the air expands, it will cool.

Also I was thinking about re-circulating the air instead of taking in external air for efficiency reasons.

I understood the reasons of efficiency for using refreigerant, since those gases are easily compressible compared to air.

10. Sep 20, 2017

### sophiecentaur

Having thought about this again, I think there is a difference. Bear in mind that any refrigerant will leave the 'cold side' at a lower temperature than room temp. I think recycling it could be less wasteful than introducing fresh, warmer air from the room into the cycle. A bigger area of cooling pipes could be needed, though, to shift as much heat out from a lower temperature into the room.

11. Sep 20, 2017

### Asymptotic

When I was studying the basics of refrigeration an illustration showed the simplest possible (open loop) arrangement.
• An insulated enclosure.
• A coiled pipe called an evaporator coil inside the enclosure, with the lower end leading outside where it is connected to a pressurized refrigerant cylinder. The upper end of the coil exits the box.
• A needle valve is placed in line to adjust refrigerant flow rate.
Most of the heat transferred out of the enclosure is due to changing the phase of the refrigerant from a liquid to a gas (boiling it, as it were). The needle valve is adjusted so only vaporized, and not liquid refrigerant, flows out of the evaporator coil.

This would be extremely wasteful (and perhaps toxic or explosive as well, depending upon the refrigerant used), and the needle valve would have to be continually adjusted to keep it so only vapor passed as the amount of heat in the box changed. Not very practical, but it's enough to cool down whatever is in the enclosure.

An actual refrigerator closes the loop by compressing the gaseous refrigerant, then moving the heat added both by the load, and generated through compression to some place other than the inside of the enclosure. This is done with a heat exchanger called a condenser (so called, because it condenses the hot refrigerant gas back to a liquid form) with this heat typically expelled to ambient air. The needle valve is replaced by an expansion valve, which basically is a specialized, automatically controlled needle valve.

My quick gloss isn't 100% accurate and leaves out a lot of important minutiae, but that's the general gist of it.

Can air be liquefied? Sure.
At what temperature does liquid air boil?
How much energy does it take to transform atmospheric air into a liquid?

Answer these two questions, and you'll be close to knowing why air isn't a commonly used refrigerant.

12. Sep 20, 2017

### sophiecentaur

Showing that it is just not handy do do it in your home with air.

13. Sep 20, 2017

### Frenemy90210

Asymptotic : I think compressing the air till it becomes liquid is not necessary. Even halfway through the process should be suffice. I think many textbooks are wrong on this issue. Many of them say that refrigerant MUST be liquefied, which IMO is wrong.

14. Sep 20, 2017

### Staff: Mentor

It isn't required to phase change, it just reduces the required mass flow rate.

...we could easily design the cycle and fund out what it takes...

15. Sep 20, 2017

### jartsa

An air compressor sucks air from the fridge, air goes to some pneumatic tool, cool air comes out of the tool and goes into the fridge.

The compressor could be a screw compressor, the tool could be a screw motor, the motor-screw could be turning the compressor-screw.

Like this:
//////////---------------////////////

Screw compressor on the left. Screw motor on the right. Axle at the middle. Left out are the electric motor and the casing.

The reason for the size difference of the screws is that otherwise it does not work

Last edited: Sep 20, 2017
16. Sep 21, 2017

### Asymptotic

By using a compressed air vortex cooler, perhaps? I've used them for enclosure cooling when absolutely necessary, but regardless of their pros, they aren't my first (or second, or fifth) choice due to their one big con, inefficiency.

To put numbers on it, in an eBay auction a Super King model SKF27B freezer using 7.5 oz. of 404A Freon refrigerant is rated 115V at 5.0 amps for 2045 BTU/hr of cooling. 115V at 5A is 575 VA. For this exercise, let's assume unity power factor (PF=1.0) and $0.10 USD/kWh. 575 watts at$0.10/kWh is $1.38 per day at 100% operation. Assuming 1:1 scaling, if this freezer were available in a 2500 BTU/hr rating it would require 1.22 times the power and energy cost, or about$1.69/day.

A Vortec model 7135 vortex cooler is rated 2500 BTU/hr at 35 SCFM compressed air flow.

How much a CFM of compressed air costs depends in part on compressor efficiency. Screw compressors are widely used due to their good efficiency, and a high output, single stage airend produces approximately 4.95 CFM/HP at 100 PSI. Efficiencies of smaller compressors, and in particular, small reciprocating piston jobs are worse, and go down to about 2 CFM/HP.

Our 2500 BTU/hr at 35 CFM vortex cooler requires about (35/4.95), or 7.07 HP, or 5.27 kW of air compressor power which translates to an energy cost of about $12.65/day at$0.10/kWh versus \$1.69/day for an equivalent amount of cooling from a refrigeration system.

17. Sep 21, 2017

### Frenemy90210

[QUOTE="Asymptotic, : Thanks for the analysis; But I was actually thinking of standard refrigeration cycle, not the vortex cooler. My guess is that standard refrigeration cycle should work as well for this half way compression. Its just that each cycle will have half as throughput.

18. Sep 21, 2017

### sophiecentaur

On top of all this theory stuff there is the massive factor of the practicalities of a real fridge system. The compressor unit and basic refrigerant idea has developed to work uninterrupted for several decades and has proved itself. The mixture of oil and refrigerant in the 'pod' is truly inspired imo. Why change if it's not going to work better?

19. Sep 21, 2017

### Asymptotic

Except that it comes back to the original question.
My experience regarding refrigeration was solely with vapor compression equipment similar to that used in a household fridge, except dramatically scaled up for industrial use (several stages spanning 70 to 250 cooling tons), and used to chill process water rather than air. If a plant throws out a lot of easily recoverable waste heat (ours didn't) another option is equipment based on vapor absorption.

However, I can't think of any examples of gas cycle cooling (which is what you have described) used in an industrial setting. BTU for BTU, a gas cycle moves more mass than a vapor compression cycle, and in doing so requires larger equipment and more energy. Imagine how long an engineer would remain employed after going to the boss, and proposing cooling machinery that costs more to run, and takes up more floor space ...

20. Sep 21, 2017

### Staff: Mentor

Ok, here's my design/analysis of the cycle:

The effect you are trying to harness is the Joule-Thomson effect, via a throttling process, which seems at first glance like it makes sense but is actually a pretty strange bird.

https://en.wikipedia.org/wiki/Joule–Thomson_effect

One might think from harnessing our intuition that PV=nRT (the ideal gas equation) implies that a reduction in pressure will have a corresponding reduction in temperature, but it actually isn't that simple because volume is also a variable in this equation. What you actually get in an ideal gas reducing pressure reduces volume by the same proportion and no change in temperature results.

The J-T effect is thus a real gas effect, based on the non-ideal behavior of gases. This turns out to be important for our example because of the resulting strangeness: sometimes when you expand a gas the temperature drops, but sometimes it rises. As it turns out, for air (well, nitrogen at least), we run into this region and that prevents a single-stage freezer from working using air. So rather than complicate this by trying to design a two- stage system, I'm going to punt and just design a refrigerator, not a freezer. Turns out, the wiki article does most of the work for us in this sample problem:
https://en.wikipedia.org/wiki/Joule-Thomson_effect#Throttling_in_the_T-s_diagram

The sample shows that if you start with 300K air at 200 bar(!) and expand it to 1 bar, you get 270K air. If we assume the refrigerator needs the air to be supplied 5K below the resulting fridge temperature, that's perfect. If we assume a refrigerator uses half the energy of a freezer and provides half the cooling (because its warmer), that means @Asymptotic's example would be a 287.5 watt power input for 300 W (1022 BTU) of cooling. Airflow required based on the 5K delta-T and the specific heat of air is 178 standard cubic meters per hour (105 CFM). Note: I believe @Asymptotic's vortex cooler capacity is rated versus ambient temperature, not versus freezer/fridge temperature, which is why I our results are out of line (why think the 35 CFM quoted is way low).

Now, 178 standard cubic meters per hour compressed to 200 bar is a monster compressor. Here's the thermodynamics:
https://en.wikipedia.org/wiki/Compressor#Effect_of_Cooling_During_the_Compression_Process

Let's assume roughly isothermal compression 300k (in reality it will start at the fridge temp, but I don't think that's a significant difference). The input power required for compressing 178 m3/hr from 1 to 200bar is 29 kW.

Last edited: Sep 21, 2017