CO2 Cascade Ammonia Efficiencies

There are many students of thermodynamics that are versed in Mollier diagrams and the refrigeration cycle as it relates to industrial refrigeration. You’ve heard it probably more than once that, theoretically, CO2 efficiency doesn’t kick in unless you’re below -20°F.

This is simply not true. There are several reasons for this.

First is the mechanical work expended on compressing CO2 versus ammonia. Most of you know by now that comparable CO2 compression displacement of compressors is 1/8 to 1/10 the cfm for CO2 versus the comparable work you are required to provide with ammonia compressors. You need to let this “soak in” a little, because while theoretically the cycle may show ideal efficiencies, the reality is you have to expend eight times more mechanical work (compression efficiency losses) to perform the same work with ammonia than you do with CO2. Then you ask, “What percent of the horsepower does that represent?” If you conservatively say it’s 5% for reciprocating compressors – and I’ve seen it as high as 36% more for screw compressors – then you start to realize, “Oh! I have this one compressor pumping CO2 that would require eight more if it was ammonia!” Then you want to say, “Well, Hank, I could just use one bigger ammonia compressor.” Yes, you could, but guess what? That mechanical efficiency loss is eight times greater for a bigger machine than for a smaller machine. To make it clear: for mechanical work, efficiency loss will be eight times more for ammonia than for CO2. Taking that conservatively at 5%, that will represent a 40% savings on the low side compression of the CO2 cycle. You are going to say, “Hank, that just can’t be.” Imagine looking at one 6-cylinder CO2 reciprocating compressor handling a 200,000 sq ft freezer warehouse. Now, if it was ammonia, you would need eight more of the compressors running, and guess what? Each of those compressors has expended machinery work (friction losses) to provide the additional cfm needed for ammonia.

Secondly, there are many who expound the merits of economized screw compressors. While this has merit, in some applications such as a constant load spiral freezer, it becomes very costly where the high side loads vary and basically hardly ever match the economizer port position for the compression or suction required. Years ago I was schooled on the merits of side ports, and the very fact that rotary screw lobes, as they pass the side port, cause compression and decompression as gas blows by the rotary lobe as it passes the port opening. These losses are conservatively 5% of the compression work. In many applications, as in refrigerated warehouses where the high side loads may basically diminish to almost zero, particularly in Northern climates, the economizer port could be way off the mark for the total load. You will then say, “Well, it does do liquid subcooling,” which does effect a thermodynamic savings, however liquid subcooling is a small percentage compared to the loss of efficiency when you don’t have a high side load. The whole concept of thinking economized is a misnomer in the refrigeration cycle. What we are trying to do is ensure we have liquid subcooling via a reduction in sensible heat in the liquid as it passes to the low side by removing some of the sensible heat with the high stage of compression, or supposedly at the economizer port position, we would expend less horsepower. While the concept is correct, to jointly use the economizer port with other high side loads that vary significantly is a major flaw in economizing. The high side load and liquid subcooling can be provided far more efficiently via a two-stage system that will first of all match the load without wasting compression, and hopefully be able to cycle on and off as the high side load profile requires.

The third area of savings is in the low side “heat of compression.” This is the heat produced as the gas is compressed, ideally isentropically, but in reality the mechanical friction losses heat also heat the gas (as, in theory, the Joule effect kicks in). If you want to dramatize this, put your hand on the head of a reciprocating ammonia booster compressor, and you will find it’s normally in the 150° to 170° range for ammonia compressors. Guess what? When you put your hand on the head of a CO2 compressor it actually feels cool. Call it what you want, the “heat of compression” is simply not there compared to ammonia booster compression.

Taking all these factors into account, we estimate the low side compression using CO2 is somewhere around 50% less than comparable ammonia compression, and for single stage economizing screw compressors, probably near, like, 60% less just on the low side of the compression cycle.

The high side compression efficiency remains unchanged, noting that you have less “heat of compression” passing to the ammonia high side from CO2 cascade systems, the bottom line being somewhere between 20-25% better overall compression efficiency for CO2 cascade over the two-stage ammonia. You can look at theoretical charts all day long and you will not see a difference in the mechanical efficiency that in the real world exists. To dramatize that a little, when you then consider maintenance, guess what? You have 1/8 as many compressors to maintain on the low side for CO2 than you do for ammonia, and that’s quite undisputable, since it takes 1/10 the displacement of compressors to do the work. So for all the old guard ammonia guys that are preaching -20°, I’m here to tell you that any suction up to +10° to +15° operating off CO2 cascade is going to deliver with less energy and with less maintenance of the low side of compression.

One other unique benefit in the CO2 cascade system is that with CO2, high temperatures such as 32° docks can operate using recirculated CO2 that returns to the intermediate condenser heat exchanger and never sees a compressor suction directly. I call it the compressorless cycle. There are several occasions where this also can work to your advantage since the air units can be placed more strategically for air distribution, whereas if it was ammonia, there would be concern about air unit placement. With multiple air units, convection cooling can be used on docks and staging areas in lieu of long air throws by ammonia units, and will maintain a better humidity on docks or staging areas, preventing moisture from reaching the freezer doors, etc. This is provided by cycling the fans rather than the liquid feed valve of the dock units, which of course saves energy as well.

So, now you know the rest of the story.