Quote Originally Posted by Segei View Post
I'm concerned about SDT +55C. It will be condensing pressure 320 psig. As far as I know every ammonia refrigeration plant in Canada has PRVs set to 250 psig. 320 psig is new world for ammonia refrigeration plants.
Second concern is energy use. Most likely at 320 psig condensing pressure a plant will use twice more energy compare to 150 psig. Today everybody focused on energy savings. It is good if you will be able to save 10-20%. At 320 psig you will waste additional 100% of energy.
Hello, Segei!

Your concerns are quite valid. My two cents on the arguments would be to put things in a different perspective.

Designing refrigeration systems at 300 psig or more is not new. Industrial trans-critical CO2 systems already available over a 100 years ago were designed at 1500 psig (although they were not common in North America), Today, small commercial transcritical CO2 systems are designed at 2000 psig.

The need for designing ammonia systems for higher pressures has built gradually. Frick (US) introduced 300 psig besides 250 psig as standard in their ammonia offering about 20 years ago. And, companies such as RFF and Danfoss have industrial valves designed for 580 psig.

This is out of need. Today's ammonia heat pumps have design pressures between 580 psig to 750 psig to obtain water at +70C to +80C (we just commissioned a 2.5 MW ammonia heat pump in Switzerland supplying water at +70C ~ +158F).

Now, designing an air cooled system that can run at 320 psig does not mean that it will be running at 320 psig. It will just run at the pressure the ambient temperature allows it to condense to. Many systems are chosen based on the COP at "design conditions", which really happen just a few days during the year, running inefficiently the rest of the time. This is why A/C systems have moved to evaluation of SEER and similar measures. In EU, condensing units and chillers, A/C or refrigeration, have to comply with Eco-Design Directive, which sets a minimum acceptable level of efficiency at part load and different ambient conditions.

In practice, one of the first things which has to be looked at when designing a system, is the profile of the heat sink which will be used. For instance, ground bore pipes could represent a fairly stable temperature heat sink, while ambient air temperature could vary a lot. In most cases, air or water are used as heat sinks, and that means a yearly profile of the dry-bulb temperature, the wet-bulb temperature, or both, has to be generated. If possible a profile of the load should be created, too, and matched. Then, different plant concepts can be analysed to determine their energy usage. Normally, this requires some type of simulation. The total cost of ownership can then be estimated for each alternative, and a decision can be made with all facts available.

The importance of energy consumption is relative to the cost of energy, how valuable is the process, the risks associated with failure, the cost of plant space, the cost, direct and indirect of refrigerant charge and many others. Some examples:

1. Last time I was in Saudi, filling a normal saloon car with fuel was about 10 USD. So, customers in that part of the world were not particularly interested in energy consumption. However, water was (still is) a premium, and a big concern to them. In such place, they tend to like air cooled systems.

2. In many chemical and petro-chemical processes, refrigeration systems run non-stop for periods of up to two years or longer, regardless if there is load or not. The value of the product requires the systems to cool as soon as the process demands it. If a system is off, pushing the start button does not guarantee that the equipment will run. Having a back-up is of no use. Again, it cannot be guaranteed that the back-up will start when the button is pushed. However, a running system is running, and can be monitored in detail and be ready to cope with the duty when need calls. We recently supplied a project of this sort, where the customer would not allow us to touch the chiller to set up the economiser into operation as the cooling process was generating $750.000.00 of profits daily. They could live with the extra energy consumption.

3. Some of the most challenging systems to design are the chillers for critical operations in Navy ships (weapon systems, radar, communications, and others). The height of the decks are restricted, the footprint is pre-defined, there is limited power generation, and water pumping capacity for the heat rejection is also limited. In this case, an important issue is start-up currents, as the network is limited, too. However, if you lose the cooling, you may loose the weapons, radars and communications, lose a ship with the crew, lose a battle, and lose a war... So, reliability and envelope have to be balanced against energy efficiency, as the consequences are enormous.

4. Another real case project. A chiller to cool down the packaging machines of a chocolate factory. Each packaging machine packages six metric tons of processed chocolate per minute. So, if cooling is lost for a minute, it creates a expensive mess. Again, reliability has to be balanced against energy efficiency.

5. Jet-engine test facilities. Very large systems are required, but they may run only two weeks in a year. In this case, you have time to prepare and check the equipment, and be sure that the start button will do the trick. The capital cost to make a highly efficient system could have a very long payback time. We have similar customers in the food industry, where process equipment runs maybe two weeks in a year, and capital cost has to be balanced with energy efficiency.

At the end, as my International Law professor thought me, "the best answer for any question is 'it depends' ".

We did a distribution centre for a very large worldwide distribution company to be used by one of the top three food manufacturing companies in the world, and they chose independent ammonia air cooled chillers over a central machinery room (ambient temperatures between -10C/+35C) as the advantages were just overwhelming in that particular project: no machinery room, a fraction of the refrigerant charge (0.7 lb R717/TR), less challenges with local environmental authorities and emergency services, no need of water network or additional chemicals, factory assembled chillers delivered as "boxes" for outdoor location on simple plinths, in built redundancy (just added another chiller), great system part load efficiencies... The results just confirmed all expectations.

And, please don't get me wrong. We still have a contracting division which makes site installs with large machinery rooms, evaporative condensers and all the bells and whistles. But, every project has a different set of needs.

Best regards,

-Manuel.