Hi all,
I ran some calculations in order to find the theoretical performance characteristics of various refrigerants compared to SO2. The results were fascinating.
Here's the raw data:
docs.google.com/spreadsheets/d/1hz0QPtUV1ne_HpCLKmGZTigN2RGRT9qWciNln4wwqcg/edit?usp=sharingI picked an operating point of a 15F evaporator, and 100F condenser, and used a volumetric flow approximately the same as a DR3. By doing this, I could calculate various operating characteristics such as discharge temperature, COP, power consumption, refrigeration capacity, etc for each refrigerant.
I used
CoolProp to solve the equations of state for each refrigerant, e.g. to get the enthalpy, entropy, pressure, etc at the various operating points.
The calculations assume an ideal compressor with no losses, so the results are only useful for comparing the relative performance characteristics of one refrigerant to another.
The refrigerants
------------------
- SO2: The baseline. Class B1 (toxic, but not flammable)
- R134a: Commonly available, has been tried successfully in monitor tops, but found to operate and sound best with a restricted suction line
- R152a: Commonly available as air duster. Class A2 (flammable). Has been used with success
- R227: Most often used as a fire suppressant. Interesting because its pressure characteristics are somewhat similar to SO2
- R1234ze: A new HFO-type refrigerant, starting to see use as a blowing agent (especially in Europe) due to its very low global warming potential. Its pressure characteristics are somewhat similar to SO2. Class A2L (slightly flammable)
- R12: Banned due to ozone depletion, but can still be found. Has been tried in monitor tops, with reports of objectionable noise level.
- R1234yf: Another new HFO-type refrigerant, considered to be a low global warming replacement for R134a. Class A2L (slightly flammable)
Relative mass flow
-----------
This takes into account the density of the refrigerant and the volume that passes through the compressor. Reading through some literature, it appears that this directly affects sound pressure levels coming from the compressor.
- SO2: 1
- R134a: 3.3
- R152a: 1.9
- R227: 3.6
- R1234ze: 2.6
- R12: 4.2
- R1234yf: 4.1
If compressor noise is truly proportional to mass flow, then these numbers would suggest that R12 is the noisiest replacement refrigerant, and R152a the quietest. Still, R152a has nearly twice the mass flow as SO2 - so I wouldn't really call is close. R134a is pretty bad as well.
Relative power draw
----------------------
All refrigerants except for R227 would draw more compressor power than SO2. R227 is nearly identical, R1234ze draws only 12% more. The rest are roughly 50% more than SO2.
- SO2: 1
- R134a: 1.5
- R152a: 1.4
- R227: 0.9
- R1234ze: 1.1
- R12: 1.5
- R1234yf: 1.5
Relative capacity
-------------------
All refrigerants would result in greater refrigeration capacity than SO2. R134a and its modern replacement R1234yf result in the greatest increase in capacity (roughly 75% greater refrigeration BTUs).
- SO2: 1
- R134a: 1.7
- R152a: 1.6
- R227: 1.2
- R1234ze: 1.3
- R12: 1.7
- R1234yf: 1.8
Normalizing for capacity
------------------------
Now things get interesting. If we wanted a modern refrigerant to behave as close to SO2 as possible in terms of run times and power draw, one way to do that is to adjust the mass flow of refrigerant until the refrigeration capacity matches SO2. The most straightforward way to do that (from the computation standpoint) is to "shave the piston" so to speak; reducing the volumetric efficiency of the compressor until it pumps the mass flow that we want. So if we reduced mass flow such that each refrigerant had the same refrigeration capacity as SO2, we'd see the following:
- SO2: 1
- R134a: 1.9
- R152a: 1.2
- R227: 3.0
- R1234ze: 2.0
- R12: 2.5
- R1234yf: 2.3
.. if we believe that mass flow rate is proportional to compressor noise, then R152a comes out as the clear winner, with only 20% more mass flow than SO2 for the same refrigeration capacity. R227 has the highest mass flow, followed by R12; they would be the noisiest by far. Interestingly, the normalized R134a value is almost same as the non-normalized R152a value. So R134a with restricted mass flow ought to sound about the same as unrestricted R152a.
As far as power draw, restricting the mass flow will lower the watts consumed. All replacement refrigerants end up drawing 10-20% less power than SO2 when mass flow is normalized for refrigeration capacity:
- SO2: 1
- R134a: 0.87
- R152a: 0.91
- R227: 0.82
- R1234ze: 0.86
- R12: 0.89
- R1234yf: 0.85
These figures are theoretically what would happen by *shaving the piston*. I think a flow restrictor would result in slightly different final values for mass flow and power draw, but I'm really not sure at this point. The thermodynamics of that situation are more complicated to work through.
COP
----
Interestingly, all modern refrigerants are slightly more efficient than SO2.
- SO2: 5.6
- R134a: 6.4
- R152a: 6.2
- R227: 6.8
- R1235ze: 6.5
- R12: 6.3
- R1234yf: 6.6
Real-world COPs are less than half that.. probably closer to 25% in our monitor tops.
Discharge temperature
------
There's other interesting bits in the raw data. For example, SO2 has a really high discharge temperature compared to all other refrigerants, at about 220F. I did a double take, but looking at some other refrigeration texts seems to confirm that SO2 generally has very high discharge temps. R152a is the next highest at almost 130F, while the HFOs and R227 were barely above saturation temperature. John Higdon once remarked at the apparent lack of attention to cylinder head cooling in the monitor tops. It looks like just about any modern refrigerant would be a dramatic improvement compared to SO2.
