First Autocascade with R290/R744
03.25.07 - 07:54pm
With yesterday being Shutdown Day and all I decided it would be a prime opportunity to pursue a project that I have thought about for months now. Phase change cooling is easily the secondly most discussed talk topic on this website with overclocking snagging second easily. Usually I talk about my various single stage units that operate in a similar manner to standard refrigeration systems however this time a different system will get the spotlight, the autocascade. Autocascades are essentially tweaked single stages with a few extra bits to let them operate with normally useless blends of refrigerants.
How Autocascades Work
Single stage systems generally operate with a single gas that will condense at or above ambient temperatures with usable pressures. Some of these gasses include r134a, r22, r290, and r404a. These gasses are relatively easy to compress, condense around ambient, and generally aren’t considered to be high pressure gasses. Autocascades use one of these single stage gasses along with a high pressure gas to achieve temperatures that would normally only be possible with two compressor cascades. Below is the schematic used in my autocascade, it should be easier to explain this if you can see exactly what I’m talking about.
Starting at the compressor a mixture of the two gasses, in this case r290(propane) and r744(carbon dioxide), travel through the condenser where the vast majority of the heat due to compression is dropped off. Within the condenser the r290 portion of the mixture will begin to condense and drop out of the gas and since r290 is such an oil-loving gas it will easily pickup the small quantity of oil that is blown out of the compressor. This liquid/gas mixture then travels into the suction line heat exchanger(SLHX) which uses the cold suction gasses too further cool the refrigerant mixture. By further chilling the gasses the separation of the two gasses is assured which will make the next step easier. From the SLHX the mixture enters a phase separator which does just that, separates the phases. The liquid portion will drop down to the bottom of the separator while the gas will float to the top, effectively separating the two “phases”. Now that the gasses are separated we can begin to get some real work done, all these previous steps were done just to get ready for the main event.
The heat exchanger(HX) is essentially a closed condenser for refrigerants. In my particular system I used a tube in shell heat exchanger that contained 7 small pipes within a larger pipe. The r744 is still in a gaseous form and it is pumped through these smaller pipes while the liquid r290 is pumped into the surrounding larger shell. Within this large pipe the propane begins to evaporate off into a gas while sucking up all the latent heat within the HX, including any heat held by the gaseous r744. With the temperatures dropping the r744 begins to condense within it’s small pipes and it should exit the HX in a liquid form while the r290 will exit in a gaseous form. The r744 will then travel to a capillary metering device and then it will be dumped into the main evaporator and then change phases back to a gas and sucking up all the latent heat within the evap. This gaseous r744 is then combined with the gaseous r290 from the HX and this very cold mixture then passes through the SLHX and completes it’s journey back at the compressor. This entire assembly can be much smaller as you can cut out the SLHX and use the condenser as a combined condenser/phase separator but naturally the performance would suffer. Below is the current stats for the system and then a list of future improvements.
- Compressor: Rechi 1/2 HP Rotary
- Condenser: 10″ x 10″ 4 row
- SLHX: 10″ tube in shell gas/liquid HX
- Phase Separator: 12″ long 1″ diameter copper pipe filled with copper wool
- HX: 12″ long 1″ diameter copper pipe filled with 7 1/4″ diameter pipes
- Metering Device: HX- 8′ of 0.031″ capillary tube Evap- 7′ of 0.031″ capillary tube
- Evaporator: Cap in cap evap with 1&1/8″, 7/8″ and 1/2″ end caps brazed to a 1& 3/4″ copper plate
- Suction Line: 36″ flexible braided stainless steel 3/8″
This being my first build means that there is a lot of room for improvement. Next time I’ll be piping the colder evap suction gases through the HX to help subcool the r744 even more. The r290 capillary tubing will also be shorter, I think 6.5′ will do the trick and this will then be wrapped around the evap suction line to subcool it further than the HX could. I will either use a larger tube in shell HX or I’ll purchase a plate HX as I think my HX is also limiting my temperatures but I haven’t loaded the system either. This week I’ll be constructing a 225 watt dummyload so that I can perform tests on this system without risking my testbed.
Results and Conclusion
I wasn’t very impressed with my final results at all but I keep reminding myself that this is my first unit, most builders completely fail on their first build. While tuning the system I have seen evap temperatures as low as -58 celsius, currently it will plateau at -56.5 celsius. With better tuning I have a feeling -65 celsius should be possible but anything lower than that will be hard to achieve due to the -78 celsius boiling point limitation provided by r744. Once I have built the dummyload then I’ll have loaded temperatures, hopefully -50 celsius loaded will be possible with better tuning. Sorry about the crappy resolution on the pictures, need to purchase a digital camera.
isnt it easier to use a cpev instead of capilar? i think it would make tunning a lot easier… it’s expensive but is worth it
That is true, tuning would be considerably easier with a cpev however I can get 100 feet of capillary tubing for the cost of a single cpev. Soon I’ll using some TXVs that I scored for cheap, the hardest part is getting the valves with properly sized orifices.