I spent some time over the weekend reviewing ASHRAE’s latest document on Real-time Energy Consumption Measurements in Data Centers. It is quite a comprehensive book in explaining and providing guidelines on energy, power systems, air and hydronic measurements.
One of the interesting topic explained in this ASHRAE book was related to how accuracy of sensors may influence the efficiency of chillers.
Chiller efficiencies are usually rated in terms of kW/ton, which is commonly used for larger commercial and industrial air-conditioning, chillers, heat pump and refrigeration systems. The term is defined as the ratio of energy consumption in kW to the rate of heat removal in tons at the rated condition. Hence, the lower the kW/ton the more efficient the system.
kW/ton = Ec / Er
where
Ec = energy consumption (kW)
Er = heat removed (ton)
Newer equipment nowadays have ratings in the range of 0.58 and 0.59 kW/ton. However, it is important to realize, that the specified kW/ton indicated in the manufacturers literature is for a specific chiller, under a specified load, and at pure design conditions (chilled temperature, delta temperature, specific flow rates), and furthermore with specific electrical characteristics.
Most of the time, if you do the calculation of actual rated tons vs actual voltage/amps/power factor or kW, it usually doesn’t add up. This is due to the actual electrical characteristics and power factor being different from the ones tested or used by the manufacturer. The manufacturer specs is, after all, a design specs that requires all parameters to be at the precise and exact optimum characteristics for the design specified kW/ton (which is very likely to differ on actual site).
If your objective is to calculate what are your percentage load and chilled/condenser conditions in which you can operate to obtain the maximum efficiency, then the calculation would be the same as how the manufacturers calculate it. You would just have to compare the efficiency at different load and conditions from one to another to understand the optimum level for maximum efficiency. Input parameters for such calculation are actual chiller gpm (flow rate) and delta temperature (delta-T) to obtain actual tonnage at different percentages of load and measuring amperage at each of those load conditions as well. This will give you actual site efficiency at each of the percentages of load where the readings were taken. However, to do this, you must have a fairly accurate set of sensors or measurement tools to measure the chiller pressure drop to obtain gpm, an accurate thermometer to measure delta temperature across chiller, then calculate actual tons. To measure the amperage, either use the chiller display if its micro processor based or a good amp meter. With these readings, you can then calculate where the best efficiency is being achieved.
As mentioned, chiller efficiency depends on the energy consumed, and is rated in kilowatts per ton cooling. This is usually for electric motor driven. Absorption chillers, on the other hand, are rated in fuel consumption per ton cooling. To determine a chiller’s maximum efficiency at which optimum load point or conditions, the actual kW/ton for the different levels of actual load have to be calculated. To calculate the energy consumed in kW:
Energy consumed, kW = [(1.732 x amps x volts) ÷ 1,000] x PF
where
1.732 = square root of 3
amps = actual running load amps
volts = actual voltage
PF = Power factor
To calculate tonnage, take the condenser delta temperature (delta-T), which is the difference between the condenser water temperature in and the condenser water temperature out, multiply by the condenser water flow gallons per minute (gpm), and divide by 24.
Heat removed, Tons = (delta-T x gpm) ÷ 24
Then, we also have the Energy Efficiency Ratio – EER is a term generally used to define cooling efficiencies of unitary air-conditioning and heat pump systems. The efficiency is determined at a single rated condition specified by an appropriate equipment standard and is defined as the ratio of net cooling capacity – or heat removed in Btu/h – to the total input rate of electric energy applied – in watt hour. The units of EER are Btu/Wh.
EER = Ec / Pa
where
EER = energy efficient ratio (Btu/Wh)
Ec = net cooling capacity (Btu/h)
Pa = applied energy (Watts)
This efficiency term typically includes the energy requirement of auxiliary systems such as the indoor and outdoor fans. Typically, the higher the EER, the more efficient is the system.
Lastly, we have the Coefficient of Performance – COP – which is another basic parameter used to report efficiency of refrigerant based systems. This parameter is the ratio between useful energy acquired and energy applied and can be expressed as
COP = Eu / Ea
where
COP = coefficient of performance
Eu = useful energy acquired (btu in imperial units)
Ea = energy applied (btu in imperial units)
COP can be used to define both cooling efficiencies or heating efficiencies as for a heat pumps. In terms of cooling, COP is defined as the ratio of of heat removal to energy input to the compressor, while for heating, COP is defined as the ratio of heat delivered to energy input to the compressor. COP can also be used to define the efficiency at single standard or non-standard rated conditions, or as a weighted average of seasonal conditions. The term may or may not include the energy consumption of auxiliary systems such as indoor or outdoor fans, chilled water pumps, or cooling tower systems. In general, the higher the COP, the more higher the efficiency. COP can be treated as an efficiency where COP of 2.00 = 200% efficiency.
So, how are kW/ton, EER and COP related? Their relationships are defined as follows:
kW/ton = 12 / EER
kW/ton = 12 / (COP x 3.412)
COP = EER / 3.412
COP = 12 / (KW/ton) / 3.412
EER = 12 / KW/ton
EER = COP x 3.412
If a chillers efficiency is rated at 1 KW/ton,
COP = 3.5
EER = 12
Therefore, a high efficiency electric chiller rated at 0.6 kW per ton rejects approximately 14,000 Btuh/ton of cooling.
Chiller efficiency increases with decrease in condenser water temperature. The condenser water temperature can be manipulated to increase chiller efficiency. The temperature of the water sent to the chiller condenser from the cooling tower is determined, largely by the ambient wet bulb temperature and the efficiency of the cooling tower (the amount of air drawn through the tower and the efficiency of air-water contact). Dry bulb temperature has only a minimal impact on cooling tower performance. The cooling tower is normally specified to meet the design wet bulb temperature in any geographic area — commonly 75°F to 78°F. The cooling tower manufacturer then designs the tower to produce 85°F water under this condition for the design heat rejection level and water flow. The temperature difference between the water sent to the condenser (i.e. coming off the cooling tower) and this wet bulb figure (say 78°F) is defined as the approach temperature, i.e. 7°F (= 85 – 78).
Under the right environmental conditions, the cooling tower can produce colder water than 85°F. And, most chillers will operate with reduced power input if this water temperature is reduced (down to a given limit as specified by the equipment manufacturer). This concept is called “floating the condenser” and holds the potential of conserving energy and reducing operating costs.
However, you will have to be a little careful that you don’t run the temperature too cold to cause chiller problems (i.e. beyond the limit specified by the equipment manufacturer), and definitely not to the point where you waste cooling tower fan horse power trying to get the condenser water down to an unachievable temperature. There are controls on the market or can be built that look at ambient wet and dry bulb conditions to determine the achievable condenser water temperature. From this determination, a condenser fan set point is developed and controlled for, without wasting fan power, and getting the best out of your tower and efficiency out of your chiller.
Selecting the condenser water parameters and cooling tower design is very complicated. Design of these systems requires experience, careful analysis, and consideration of initial investment and operating costs. The material presented here is simply an explanation of several design parameter opportunities. Please refer to the specific cooling design modules elsewhere in this information system for additional details.
Historically, plant engineers have kept chiller operating logs to measure chiller performance and determine causes of problems. The data collected includes readings taken from the chiller during scheduled inspections such as evaporator and condenser temperatures, pressures, flows, running load amps, volts, etc. This schedule varies from every two hours to once a shift, depending on the type of operation and more importantly, manpower constraints. In most facilities, logs are a vital tool in scheduling downtime, preventive maintenance, and inspections based on chiller run hours. Today, it is common for facilities to maintain logs, but they rarely get reviewed until there is a problem, which is too late.
As the chiller efficiency is calculated based on the operating conditions of the chiller, the accuracy of the sensors and measurement tools in collecting these data should be periodically calibrated to ensure accuracy. Otherwise, it may project the efficiency incorrectly. For example, a 2°F difference in the delta-T would infer a 20% inefficiency of the chiller.
Tags: Chiller, Data Center, Efficiency, PUE, Temperature