[Home]	Search:
Theme: Select one Steely Old Modern Midnight Ultramarine Swiss Chocolate Traditional

Issue : July-September 2001

Evaluation of Overall Chiller Performance Characteristics

By M. Nadeem
Ener Save Consultants Pvt. Ltd.
New Delhi

Nadeem is a M. Tech from IIT Delhi and works as a consulting engineer with Ener Save whose head office is in Canada. He has been involved in the design of some innovative projects in India such as Geothermal Heat Pumps and Solar Run air conditioning system, for the past 10 years.

A water chilling machine is the most important component of a central airconditioning system. There are several types of chillers available viz reciprocating, rotary screw, centrifugal, centrifugal with VSD etc. All these chillers behave differently at varying operating conditions (cooling load and coincident cooling fluid temperatures), which are application specific. Thus, for determining the most efficient type of chiller for a particular application, the year round performance, and not just the performance at the design conditions, needs to be evaluated. The criteria as laid down by ARI - Standard 550/590 (1998) for determining the overall performance of chillers is generalized, and as a result can only predict nonapplication specific performance behavior. See Table 1 and Figure 1.

Chiller Performance Evaluation Criteria

Most often, the flow through the condenser and evaporator of a chiller are kept constant. Thus, condenser cooling fluid temperature and part loads become the primary parameters for determining the performance of the chiller. See Figure 2. These two parameters are application specific and are variable round the year. Moreover, different types of chillers behave differently under these two variable parameters. Under the same operating conditions one particular type may be more efficient than another type but the reverse may be true for the same chillers under some other operating conditions.

[top]

At any time, a chiller can operate at the following two different operating conditions:

• Part load with lower cooling fluid temperatures.
• Part load with higher cooling fluid temperatures. As per ARI Standard 550/590 (1998), a chillers' part load values are evaluated as follows:
• Determine the part load energy efficiency at 100%, 75%, 50% and 25% load points at the conditions specified in Table 1.
• Use the following equation to calculate the IPLV or NPLV.

Table1 : Part Load Conditions for Rating as per ARI Standard 550/590(1998)
	IPLV	NPLV
Evaporator (All Types) 100% Load LWT 0% Load LWT Flow Rate (gpm) Field fouling allowance (F.F.A.)	44°F [6.7°C] 44°F [6.7°C] 2.4 gpm/ton [0.043 L/s per kW] 0.0001	Selected LWT Same as 100% load Selected gpm/ton [L/s per kW] As specified
Condenser (Water Cooled) 100% load EWT 75% load EWT 50% load EWT 25% load EWT 0% load EWT Flow rate (gpm) [L/s] F.F.A.	85°F [29.4°C] 75°F [23.9°C] 65°F [18.3°C] 65°F [18.3°C] 65°F [18.3°C] 3.0 gpm/ton [0.054L/s per kW] 0.00025	Selected EWT Linear Variation upto 65°F 65°F (18.3°C)} 65°F (18.3°C) 65°F [18.3°C] Selected gpm/ton [L/s per kW] As specified
Assumptions: The flow rates are to be held constant at full load values for all part load conditions. For part load entering condenser water temperature, the temperature should vary linearly from the selected EWT to 65°F for 100% to 50% loads, and fixed at 65°F for 50% to 0% loads. Chiller Operates at: 100% load for 1% time per annum 75% load for 42% time per annum 50% load for 45% time per annum 25% load for 12% time per annum LWT – leaving Water (liquid) temperature EWT – entering Water (liquid) temperature IPLV – Integrated Part Load Value, a single number part-load efficiency figure calculated as per ARI standard rating conditions. NPLV – Non-Standard Part Load Value, a single number part-load efficiency figure calculated at conditions other than ARI standard rating conditions. COP – Coefficient of Performance, ratio of the cooling capacity in Watts(W), to the total power input in Watts(W). EER – Energy Efficiency Ratio, ratio of the cooling capacity in Btu/h to the total power input in Watts(W). F.F.A – Field Fouling Allowance, provision for anticipated thermal resistance due to fouling accumulated on the heat transfer surface during use specified in h.ft2 °F/Btu[m2 °C/W]. Note: The intent of part load ratings in terms of IPLV and NPLV is to permit the development of part load performance over a range of operating conditions. NPLV instead of IPLV is used to develop application part load performance at conditions other than ARI standard rating conditions.

For COP and EER
IPLV or NPLV = 0.01A + 0.42B + 0.45C + 0.12D
where:
A = COP or EER at 100%
B = COP or EER at 75%
C = COP or EER at 50%
D = COP or EER at 25%

For kW/Ton
IPLV or NPLV = 1 / {(0.01/A) + (0.42/B) + (0.45/C) + (0.12/D)}
Where:
A = kW/Ton at 100%
B = kW/Ton at 75%
C = kW/Ton at 50%
D = kW/Ton at 25%

The prevalent criteria for the performance evaluation of chillers, as laid down by ARI has the following limitations:

1. ARI standard assumes that when the cooling load is low, the cooling fluid temperature is correspondingly lower. This assumption is seldom true, since cooling load does not depend solely on the ambient conditions but it has additional dependence on solar radiation, occupancy, ventilation, lighting, equipment etc. See Figure 3.
2. ARI assumption of part load operation times and values is based on the weighted average of the most common building types and operations using average weather in 29 U.S. cities.
3. ARI standard does not take into account variations in internal heat loads.
4. Combined operation of more than one chiller is not considered.

Click to view the clear picture

[top]

Due to the above assumptions, criteria laid down by ARI for the performance evaluation of chillers can only give non-application specific results.

Thus, for determining the type of chiller which will be most efficient for a particular application, year round evaluation of the following two parameters needs to be done:
• Year round hourly cooling load variation.
• Determination of condenser cooling fluid (water or air) temperature coincident with each cooling load value.

Case Study

An existing prestigious five star hotel located in New Delhi had their centrifugal chillers installed in 1978. These chillers were selected with R- 11 as refrigerant and had already served their useful life, as a result had high operation and maintenance cost. Our firm was asked to suggest a suitable replacement for the chillers. Since, as energy audit consultants for the same hotel, we already had detailed cooling load patterns for various areas, we undertook the exercise of year round performance evaluation for various types of chiller options considered and made our recommendation accordingly.

The hotel comprises of approximately 30,000m2 of covered area on eight floors. There are a total of 296 guest rooms in the hotel apart from restaurants, banquet halls, shops, lobby etc. In addition, there is a separate arrangement to provide conditioned fresh air to guest rooms for ventilation.

The hotel has the following features, which are typically true for large hotels in India:

• The block cooling load is low during summer and monsoon (on account of lowest occupancy and hence least area usage). Thus, for such an application, the cooling load is low when the cooling fluid temperatures are high. See Figure 4.
• During any typical day, the cooling load is maximum during evening hours when cooling fluid temperatures are lower.
• Most of the time, there is a combined operation of more than one chiller.

The peak block cooling load for the existing hotel is 800 tons and it has two working chillers of 400 tons each (centrifugal water cooled type). The year round cooling load variations along with coincident cooling fluid temperatures were analyzed and evaluated (for 8760 hours) using the following procedure:

1. Calculation of hourly building structure cooling loads for individual areas and the net effect on the block cooling load.
2. Calculation of hourly building solar cooling loads for individual areas and the net effect on the block cooling load.
3. Calculation of hourly building ventilation cooling loads for individual areas and the net effect on the block cooling load.
4. Calculation of hourly occupancy loads (as per the occupancy patterns - see Figure 4 and usage hours for various areas of the hotel) for individual areas and the net effect on the block cooling load.
5. Calculation of hourly building lighting loads for individual areas and the net effect on the block cooling load
6. Hourly net effect of fixed cooling load on the block cooling load.

Click to view the clear picture

[top]

Using the above procedure, the hourly net block cooling loads were evaluated for 8760 hours (365 x 24). The performance of each chiller type was worked out based on the following procedure:

• Calculation of hourly coincident cooling fluid temperatures for 8760 hours. See Figure 5.
• Evaluation of year round power consumption for each chiller type using hourly block cooling loads and coincident cooling fluid temperatures.

None of the chillers operated when there was no cooling load requirement i.e. during winters, taking into account the effect of ambient conditions, internal and external heat transfer etc.

For each chiller type, considering flow rate through condenser and evaporator as constant, the efficiency dependence in terms of Entering Cooling Water Temperature (ECWT) and part load is known from various manufacturers. Thus, at each value of block cooling load (hence, part load on the chiller) and coincident cooling fluid temperatures, hourly power consumption for each chiller type was obtained from the manufacturers. Typical part load power consumption at various entering cooling water temperatures for 400 ton Centrifugal, Rotary Screw and Centrifugal Chiller with VSD are as shown in Tables 2, 3 & 4. At each hour, block cooling load and hence, part load on each chiller is evaluated while simultaneously obtaining coincident cooling fluid temperatures from the hourly weather data. This combined information is utilized to determine hourly specific power (kW/Ton) and hence the total power consumption for each chiller type from the available part load performance characteristics (refer Power Consumption Tables 2, 3 & 4).

[top]

Results of Case Study*

The specific and annual power consumption for different types of chillers viz rotary screw, centrifugal and centrifugal with VSD were worked out using the procedure as described above and ARI standard 550/590 (1998)

1. Results based on hourly cooling load and coincident cooling fluid temperatures.
   Annual cooling load = 1,812,825 ton hours (as obtained from the hourly simulation).

Type of Chiller	**IkW / ton	Total Energy Consumption per Annum (kWh)
Rotary Screw Chiller Centrifugal Centrifugal with VSD	0.603 0.589 0.516	1,093,634 1,067,853 935,852

2. Results based on ARI standard 550/590 (1998):
   Annual cooling load = 1,812,825 ton hours (as obtained from the hourly simulation)

Type of Chiller	**IkW / ton	Total Energy Consumption per Annum (kWh)	Variation w.r.t. hourly simulation procedure
Rotary Screw Chiller Centrifugal Centrifugal with VSD	0.499 0.544 0.378	904,962 986,177 685,610	– 17.25% – 7.65% – 26.74%

[top]

Conclusion

Based on the above case study for the hotel, it is concluded that overall performance characteristics of different types of chillers differ significantly than as predicted by ARI standard 550/590 (1998). See Figure 6 and 7. This difference is primarily owing to higher cooling fluid temperatures at lower part loads and vice versa. Moreover, it is also inferred that, generally, screw chillers are less efficient than normal centrifugal chillers at part loads when the cooling fluid temperatures are higher. Also, at any given cooling fluid temperature, all positive displacement type of chillers are less efficient at part load than at full load, see Figure 2. However, non positive displacement type of chillers have their best efficiencies at approximately 80% part load.

[top]