Logo - CCME
Digital Issue - CCME

Oh no, not my chillers!

Equipment operators and plant owners are in denial or just plain unaware that most chillers are operating above their design efficiency, says Dan Mizesko

| | May 6, 2021 | 8:50 pm
Share this story

The evidence is clear, the majority of chillers and chilled water plants operate above their intended design efficiency (kW per ton). Most chillers operate above efficiency design, and I’m not talking about one per cent or two per cent here. I’m talking about 20%, 25%, 30% and even higher. I have seen chillers operating 60% above their kW per ton design.

Dan Mizesko

What’s most frustrating is that the equipment operators and owners do not even realise their equipment is operating inefficiently and consuming excessive power or are in denial. Now, I know some of you reading this are saying, “Oh no, not my chillers!” or “Not my plant”. You are saying, “We have the most efficient chillers, and we have a CMS system and a SCADA system, we do all the maintenance as per the OEM manual, we have VFDs on all our pumps, towers and chillers.” Unfortunately, the bad news is that just because you have purchased the most efficient chiller in the world does not mean it’s operating efficiently. Just because you are performing the maintenance as per the OEM manual does not mean your chiller is running efficiently.

It takes substantially more than the recommended maintenance procedures to ensure your chillers operate as efficiently as possible. Isn’t that the goal to operate chillers and chilled water plants as efficiently and reliably as possible? Now, let me explain how I can state why most chillers do not operate as efficiently as designed.


U.S. Chiller Services from 2003 to 2012 analysed extensive data from operating chillers and plants in the Gulf region, from the newly constructed and commissioned plants to five-year-old plants. All were electric-driven centrifugal plants. What we determined was that the operating range of these chillers was from 0.98 to 1.8 kW per ton. Water-cooled electric centrifugal chillers in the Gulf region are generally selected and designed to operate in the range starting from 0.687 to 0.760 kW per ton. So, it was clear these chillers were operating 30% above design efficiency at minimum. Again, some of these chillers were just commissioned and put into beneficial service by the OEM field service department.


A survey report by the U.S. Department of Energy found more than 120,000 chillers in the United States are expending more than 30% in additional energy through operational inefficiencies. This is not a misprint – 120,000 chillers operating more than 30% above design! A survey report by San Diego State University Energy Engineering Institute on 31 chiller plant sites across the United States found chillers that should be operating in a range starting from 0.65 to 1.00 kW per ton, according to the OEM specifications, were often actually operating in a range starting from 1.5 to 3.0 kW per ton and that most chiller plant operators had no idea how well their chiller plants were performing.

A report on 14 chiller plants by the Northwest Energy Efficiency Alliance found a substantial variation in plant energy efficiency ranging from 0.55 to 1.80 kW per ton. A report by the California Energy Commission addressed the problem that buildings and chilled water central plants do not perform as well in practice as is anticipated during design.

The reasons cited were:

  • Improper equipment selection and installation
  • Lack of rigorous commissioning and proper maintenance
  • Poor feedback on operational performance and energy performance


Many companies who purchase centrifugal chillers for their facilities demand a Factory Witness Test of the chillers they are buying before accepting delivery and payment. The OEM Sales Engineer will select the chiller with a software selection program to predict the chiller’s performance. The buyer agrees to purchase the chiller based on the software selection program’s predictions of the chiller performance, including capacity part load and full load and also kW per ton, to name just a few of the parameter performance predictions. The Factory Witness Test confirms that the actual chiller performance matches the predicted performance, and the results serve as a benchmark.


The witness testing standard requires specific types of instrumentation and specifies chiller testing procedures and tolerances. It also requires that accurate measurement is an integral part of Witness Performance Testing. All of the following instrumentation must be calibrated and traceable to NIST, the National Institute of Standards and Technology: Power meters, temperature sensors, flow meters and pressure transmitters. This ensures the instrumentation meets the accuracy requirements of AHRI Standard 550/590.

A copy of the latest calibration report is made available to the attendees of the Witness Test. As per industry standards, the factory test loop must have redundant measurement devices designed into test loops to verify test data. Redundant sensors are used to measure evaporator water flow, evaporator water temperatures, condenser water flow, condenser water temperatures and chiller power consumption. Agreement between the calibrated sensors confirms that the data taken is accurate.

Some companies even require a Zero[1]Tolerance Witness Test to be performed. In other words, the chiller MUST perform at its predicted design performance and, if not, the buyer can cancel the sale. I have been involved in many Factory Witness Tests over the years, and I can tell you it takes the factory days to prepare the chiller for the test, with the objective of getting everything right, so that the chiller performs as designed. Even with the days of preparation, the chiller will often not perform to design kW per ton. More days of adjustments must be done to finally achieve design performance, sometimes including changes to the chiller’s design. As and when the chiller is certified and meets design, a printout is given to the buyer to benchmark that the chiller performs as predicted.


Does anyone reading this think that when a chiller is started and commissioned in the field, it receives the same amount of attention, care and diligence as during a Factory Witness Test? Are the proper and precise tools and equipment utilised in the field, as during the Factory Witness Test? Is the same data looked at during the field startup as during the Factory Witness Test?

Is the chiller kW per ton even checked during a field startup as it is during a Factory Witness Test? Finally, is the Field Startup Technician as trained and aware of how the chiller should perform as the Factory Witness Test personnel? Of course, the answer to all the above questions is NO. After answering all the above questions with a “NO”, how could chillers ever operate as efficiently as designed? As the data shows, chillers do not operate as efficiently as designed. The primary reason is ineffective start-up and commissioning from day one of beneficial usage.

The above is a small section taken from an OEM commissioning manual. I want the reader to understand the design condition that the startup technician is to record. Does it mention kW per ton? No. Also, what’s interesting is that the flows are to be determined by pressure drop readings. The Factory Witness Test is performed with a calibrated flow meter; however, the startup technician sets design flow with an instrument that is not nearly as accurate as a flow meter.

My point is that energy efficiency is not a priority or even checked during the initial startup and commissioning of the chillers. With that being the case, if the chiller is not started and commissioned in the field with the same level of expertise, tools, equipment and data with an emphasis on efficiency, how can you expect the chiller to operate as efficiently as designed? The fact is that the chiller operated at design efficiency at the factory, and that was the last time it did.

Next month, I will explore chiller diagnostics tools that identify chiller inefficacies, restorative opportunities and optimisation to restore proper chiller operation and overcome the lack of rigorous start-up and commissioning.

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

Share this story

Feedback for this story

Your email address will not be published. Required fields are marked *