Exploring the complexities and economics of absorption refrigeration systems.

A recent industry news headline caught my eye. It stated that more than two-thirds of the world market for chillers is based on absorption refrigeration. It is worth noting at this point that the refrigeration business divides neatly into two camps that, like oil and water, don’t mix well or for long. All of my career has been spent with system that use a compressor to persuade the refrigerant in the system to condense at atmospheric temperatures.

I have always been aware of absorption as an alternative way to achieve the same end. I even have the temerity to think that I understand how it does it, like a kid in the crowd reckoning he sees how the magician made the elephant disappear. But, like the kid in the crowd, I have never done it myself, and I have a sneaky feeling that it is not as straightforward as they like to make it seem.

The theory is simple enough. Take two substances that have a strong attraction to each other, such as ammonia and water. Arrange a circuit so that the two are forced apart by heating the mixture to boil off one of the two, for example the ammonia. Allow the gas coming off the boiling mixture to recondense by letting the ambient cool it down. Use this liquid to cool the chilled fluid by evaporation and allow the evaporated gas (ammonia) to be absorbed by the other substance (water) that was left behind in the boiling process. So far, so good—but the ammonia in the condenser needs to be at high pressure, and in the evaporator it needs to be at low pressure, just like in a vapor compression system. This can be achieved by pumping the liquid from the absorber (at low pressure) to the regenerator (at high pressure) where the ammonia will be sufficiently pressurized to condense above the atmospheric temperature. This is a neat trick, although maybe not in the elephant league, because the power required to raise a certain mass flow of liquid through a pressure difference is much less than the power required to deliver the same mass flow if it is a gas. This is because the liquid is “incompressible,” so you get a lot of pressure rise for a relatively small effort. Gas is eminently squashable so needs a lot more work to achieve the same thing.

This is the point at which diehard compression guys like me start to lose our confidence. How do you size an absorber to make sure that it does enough absorbing, and does it fast enough to keep up with the rest of the system? How do you make sure that a desorber (the bit that separates the two component fluids of the system) puts the right stuff in the right place? Truly, there is more than a touch of the Moonshiners’ craft in this alchemy.

The economics of absorption are also, well, absorbing. The system uses heat, not electricity, and it’s often said to be “waste” or “free” heat. Not many people burn stuff just for the fun of it—they’re called arsonists and pyromaniacs—so I am skeptical about the “free heat” tag. The truth is that it cost somebody something to make the heat, and to conserve it, and to deliver it to the desorber. What’s more, because rather a lot of heat is required, the condenser of the absorption system will be way biggerthan would be needed on a compression system. So the cooling towers are bigger, their fans are bigger and their condenser water pumps are bigger. Which raises another point: the absorption system doesn’t have a compressor to use electricity but all the pumps and fans and pumps and more pumps don’t run on heat; they are wiredinto the mains.

Also, the absorption system provides a flow of cold secondary refrigerant, but an industrial installation might otherwise have profited from direct use of the refrigerant. So the economics are complicated and are very subject to the balance between the cost of fuel and the cost of electricity. If the heat is an outflow from a process that would not be more efficient without such a flow at such a temperature then absorption is a truly fantastic proposition. Totally absorbing.