For many decades, engineers and some OEMs have been recommending implementing variable condenser water flow on water[1]cooled chillers to reduce system energy consumption.

Proponents point to many ASHRAE studies:

1) Waller (ASHRAE Journal, January 1988) recommended a condenser water flow rate of 1.5

2) Shelton and Weber (1991 ASHRAE Transactions) found that equipment selected for a condenser water flow rate of 2 gpm/ton saved 3.5% in system peak demand and 10.5% in annual energy consumption, compared to 3 gpm/ton.

3) Shelton and Joyce (June 1991 ASHRAE Journal) concluded that “the conventional practice of designing chilled water plants with condenser flow rates of 2.8 to 3.0 gpm/ton results in unnecessarily high condenser flow rates”.

However, there is another camp of engineers and OEMs that does not recommend this practice, and my experience in this industry leads me to agree with this camp – that lowering and/or varying the condenser flow in the chiller does not save energy, as the chillers’ increased power consumption outweighs any savings in condenser pumps or tower fans.

Dan Mizesko

Dacen Kinser, of the University of North Texas, in Denton, Texas, wrote in an ASHRAE paper: “My analysis does not support the generalization that a condenser water flow rate design of 2 or 1.5 gpm/ton will result in the lowest system full load power consumption for chiller, condenser water pump and cooling tower. I conclude that a 3 gpm/ton (0.054 mL/J) condenser water flow rate will most often lead to a smaller full load power draw for system components compared to 2 or 1.5 gpm/ton, and although a lower condenser water flow rate can often result in lower annual energy cost, this result is highly dependent on the specific site utility rates, pumping head, and the chiller.”

So, let’s take a look at what happens when we lower the condenser flow. Reducing the condenser flow will increase the condenser water temperature range; it will also increase the required condenser pressure. Increasing the condensing pressure on a chiller will result in increased chiller power cost and reduced performance.

Variable flow in the condenser is not recommended by some chiller OEMs, as it generally raises the energy consumption of the system by keeping the condenser pressure high in the chiller. Although reducing the condenser flow will improve the cooling tower LMTD and a smaller tower can be used, the savings from this strategy will not offset the increased cost of the chillers’ increased power consumption.

In addition, if implementing this on an existing plant that has been designed for 3 gpm operation, cooling towers typically have narrow ranges of operation with respect to flow rates and will be more effective with full design flow versus the lowered flow.

Another major consideration not generally mentioned by proponents of reduced condenser water flow is that the rate of fouling in the condenser will increase at lower water velocities associated with variable flow. This will not only increase maintenance costs but also increase power consumption tremendously.

In the May 2021 issue of Climate Control Middle East (Licence to Chill), I wrote as follows: “A chiller’s efficiency is affected most by its resistance to heat transfer, its (LTD) Leaving Temperature difference of its heat exchanger tube surfaces more than anything else. So, it is imperative that the chiller’s evaporator and condenser tube surfaces have the lowest resistance to heat transfer possible.”

In the same article, I also stated that “fouling of the chiller condenser tubes substantially impacts the power consumption of the centrifugal compressor. This is why the focal point of any water treatment programme should be the prevention of deposition.

A layer of scale that is one by sixty-fourth of an inch thick on the condenser tubes can increase electrical usage in a centrifugal chiller by as much as 33%. Biofilms can decrease condenser heat exchange efficiency to an even greater degree. Thus, it is of paramount importance to assure that chiller heat exchange surfaces are maintained in a clean, deposit-free condition”.

Fouled condensers are the major cause of increased chiller power consumption. Lowering the condenser flow will lead to increased condenser fouling, just the opposite of what you want to achieve.

If you are insistent on going the Condenser Variable Water Flow direction, I suggest you first baseline the current plant efficiency performance (plant annual kWh of chiller, cooling tower fans and condenser water pumps) and then divide this by the annual ton-hours of plant cooling. This is the annual performance in terms of kW/ton for the plant components (chillers, towers, condenser water pumps).

This will be your baseline pre-condenser variable- or lower-flow retrofit. You will then compare this versus post-retrofit varying condenser water flow rate to prove or disprove the savings. You will need to install Utility-grade BTU meters on the chillers and on the main plant header. You will also need to install Utility-grade power meters on the tower fans and the condenser pumps. I again suggest you get a full year’s worth of data before you perform the retrofit.

Before you even attempt retrofitting your plant to variable condenser flow, you need to be sure the condenser system flow rate stays above the minimum flow rate, as per the requirement of the tower and/or chiller OEM. If either of these minimum flows are close to the current system design flow, variable flow should not even be considered.

Last, if you are still considering variable condenser flow, I suggest you have an automatic tube cleaning system installed to overcome the increased fouling associated with low-flow conditions; even with this in place, though, you may not get the savings you think you will with variable condenser flow.

Dan Mizesko is Managing Partner/ President, U.S. Chiller Services International. He may be contacted at dmizesko@uscsny.com