Flooded Evaporators

Back in college during the days of CFC's, we covered industrial refrigeration in a class with the creative name of Refrigeration 4. We also covered absorption refrigeration which everyone confused with Hydronics and Steam but that is another story. Nick Reggi, if you are reading this, thank you for your patience, you are awesome.

Some of the things I remember thinking during this was that industrial refrigeration took liquid injection into the evaporator a step beyond and that superheat was not something to be tolerated. Pumping way more refrigerant into an evaporator than what you expected to actually evaporate was the name of the game here. This went against everything we had been taught up to that point. Thinking back this is likely the defining part of any learning curve once you have advanced far enough along it.

To start, let's look at metering devices as this is an obvious difference between direct expansion (DX) and flooded coils. When we learned about metering devices in class the basics were that metering devices maintained the high/low pressure areas of the system and that they were sized to meter enough refrigerant to meet the load without flooding the compressor. In other words, a certain amount of superheat was required to ensure there was dry vapor leaving the evaporator. This meant there was less refrigerant entering the evaporator than what would be optimal from a performance standpoint. Why? Because superheating a vapor takes away volume in the evaporator that could be used to change liquid to a vapor and absorb many times the amount of heat. As a result, superheating the vapor in the evaporator reduces the heat absorbed and lowers efficiency as a trade off to protecting the compressor.

A flooded evaporator on the other hand, is just what it's name suggests. It is an evaporator that is completely filled with saturated liquid and a limited amount of vapor. As the vapor is produced, it is pulled off before it can become superheated and before it forms a sizeable volume in the evaporator. In fact, it is not uncommon for the flow of refrigerant into a flooded evaporator to be 200% or more of the actual load requirements. Think about that for a minute. Most people would think that adding just 10% or 15% more refrigerant would be enough and this seems logical from a purely superheat standpoint. However, although it may be enough to get rid of the superheat, the evaporator would still contain a large amount of vapor and vapor is a poor heat transfer medium. In fact the limitation on how much refrigerant is fed into a flooded evaporator is based on how much heat is absorbed per pound versus the energy required to pump the refrigerant with the best ratio in favor of heat absorbed being optimal. This is about as far from a dry DX system philosophy as you can get.

The result of using a flooded coil is that the average evaporator coil temp is reduced which means the saturation temperature of the evaporator can be increased, reducing the compression ratio. (A similar effect is seen when using an EEV set to maintain the minimum stable superheat for the evaporator.) It also reduces the discharge temperature of the compressor which is the Achilles heel of using ammonia*. So do not expect to see a TXV in a flooded evaporator application. What you will see is a valve that can be controlled by the liquid level in a tank that holds the left over liquid leaving the evaporator. A float opens the metering device when the liquid level starts to drop and closes it when the liquid levels starts to rise. This makes sense as there will be less liquid leaving when the load on the evaporator increases and more liquid when the load decreases.

Another result of using a flooded evaporator is the need to deal with the oil that likes to (Hopefully) travel with the refrigerant. The challenge is ensuring the oil does not become trapped in the liquid reservoir or some other spot and fails to return to the compressor. One solution is to greatly reduce the amount of oil entering the system and an oil separator is the preferred method here. By removing 80+% of the oil before the refrigerant enters the condenser, the oil separator makes sure it is far easier to deal with whatever minimal oil makes it through. It also increases heat transfer as oil in large quantities can interfere with the transfer of heat between the refrigerant and the heat exchanger walls. However, even a minimal amount of oil being trapped will still add up and the ability to siphon off whatever oil does become trapped and returning it to the compressor is essential.

By pulling off the saturated vapor that is being produced, the compressor has a much better chance of being properly cooled if it is a semi hermetic design as the returning suction gas will be limited in superheat. This results in optimal winding cooling for the given density of returning vapor and also reduces the discharge temperature of the gas. When I consider the number of compressors I have seen torn down and were burned up with black windings and carbon deposits from oil breakdown, there is value in this approach. When you consider even an HFC system working at extra low temp for an ice cream freezer, superheat even 15 or 20 degrees above what it could be at the compressor suction will have a considerable negative effect on winding and oil life. It is important to remember that there is no precise temperature when motor winding life suddenly drops. It is a sliding scale with a cooler winding lasting longer than a hotter one.

One take away from this even if you never see a flooded evaporator in your work is to make sure superheat is never higher than it needs to be. Properly sized components with heat transfer surfaces kept clean, insulation on the suction line thick enough to limit non functional superheat and ensuring proper airflow for both the evaporator and condenser will ensure the compressor runs cool and for a long time**. Not to mention the substantial resources saved by the customer and the boost to your reputation.

* Ammonia is an amazing refrigerant with more heat absorption than even water for each pound evaporated. However, it has one drawback (Other than being dangerous to inhale but you will never not know it is there!) and that is it has a high discharge temperature compared to say R404a for a given compression ratio. Because of this, for low temperature applications, limiting condensing pressures as much as possible and ensuring there is very minimal superheat at the compressor suction are essential design requirements for using ammonia. For fun, run a compressor simulation software using ammonia and play around with the parameters taking note of the predicted discharge temps.

** Don't forget about proper voltage and current values as well!