ASHRAE - Series Series Counterflow For Central Chilled Water Plants
ASHRAE - Series Series Counterflow For Central Chilled Water Plants
ASHRAE - Series Series Counterflow For Central Chilled Water Plants
Life-Cycle Cost nif icant weekday cooling loads, the Fixed Parameters
To ensure that the chilled water plant centers meeting areas often are used on As in most central chilled water plants,
provided the owner with the best possible weekends. The loads are year-round. the initial costs for distributing the
value, lowest life-cycle cost was chosen However, airside economizers reduce the chilled water (pumps, piping) were con-
as the determining criterion for the plants need for mechanical cooling during sidered critical. Using a larger-than-con-
design. An independent financial con- many months of the year. ventional difference between the
sultant performed a cost analysis of each The consultant developed a detailed entering and leaving chilled water tem-
alternative to ensure an equitable com- spreadsheet to examine the life-cycle peratures permits a lower flow rate.
parison. Each of these analyses was based costs for the chillers, chilled water Smaller pipes and pumps can then be
on a load profile specific to the plant. pumps, condenser water pumps, and used to satisfy the same capacity.2,3 Based
The profile was developed from the ex- cooling tower fans in each proposed de- on that premise, the plant owner estab-
pected loads of the customer who had sign. The spreadsheet accounted for: lished these non-negotiable design pa-
signed the contract to purchase chilled Load profile, rameters for the chilled water system:
water. The primary cooling loads are from Ambient dry-bulb and wet-bulb Entering-chiller water temperature:
a convention center. In addition to sig- conditions (Load does not vary directly 55F (12.8C),
Leaving-chiller water temperature:
As can be the case when a third party per- disclosure of the original data is necessary to 37F (2.8C), and
forms an economic analysis, few details (in- verify the results and offer recommendations
cluding the actual plant load profile) were that might benefit the plant owner. About the Authors
made available to the authors. Readers may It may have been possible to achieve a Steve Groenke is the new equipment sales man-
find this frustratingso did the authors. Al- higher entering chiller temperature, but do- ager with Trane in Timonium, Md. Mick
though third-party analyses can ensure greater ing so was not within the plant owners direct Schwedler, P.E., is a senior principal applications
objectivity and fair play by the principals, full control. engineer with Trane in La Crosse, Wis.
Cond. Water
Temperature + Cond. Approach
Temperature
At saturation, these temperatures re-
late directly to the refrigerant pressures
in the evaporator and condenser.
It is clear from Table 1 that the series
series counterflow arrangement yields
Figure 2: Conceptualization of reduced lift.
the lowest full-load chiller power (about
14% lower than the parallelparallel
configuration). The dramatic reduction in chiller power oc- ity to produce chilled water at an elevated temperature.
curs because the upstream chiller in the seriesseries (When this article was written, the plant had not operated
counterflow arrangement operates at a higher chilled water through an entire cooling season.) How the plant should re-
temperature, which means that the refrigerant temperature spond to varying system conditions was discussed with the
and refrigerant pressure in the evaporator are also higher in design engineer, plant owner, and plant operators. For ex-
the upstream machine. Similarly, the downstream chiller ample, if the entering chiller water temperature did not reach
sees a lower condenser leaving water temperature design conditions, the operators could:
and therefore has a lower condenser refrigerant pressure Increase pump speed or the number of active pumps
than it would in a plant with the chiller condensers arranged to increase flow rates through, and fully load, the active
in parallel. chillers.
Figure 2 illustrates the concept of reduced lift using the Reset the setpoints of the upstream chillers to 55% of the
design parameters for this chilled water plant. Although lift is total temperature difference. Lowering the setpoint of the up-
the difference between the refrigerant pressures in the evapo- stream chillers as the result of a drop in entering-chiller water
rator and condenser, its magnitude can be approximated using temperature lessens the benefit of reduced lift. However, the
the difference between the water temperatures leaving the upstream chillers will always run at a higher evaporator
evaporator and condenser. Chiller power can be reduced by pressure than the downstream chillers, which saves energy
decreasing compressor lift. In this case, the difference in consumption and costs.
average lift at design is approximately 13%: Chiller sequencing was also discussed. It was determined
that the most cost-effective startup strategy would fully load
1 {[(54.3 + 53.8)/2]/61.9} = 0.126 one chiller module, and then activate the remaining chillers in
The reduction in lift provided by the seriesseries modules (pairs). Activating the upstream chiller and operating
counterflow arrangement also occurs at part-load conditions. it at the higher water temperature would take advantage of all
Why? The temperature of the water leaving the evaporator of of the available heat-transfer surface area without increasing
the upstream chiller is always warmer than the system water, the energy consumed by ancillary equipment.
and the temperature of the water leaving the condenser of the At the design conditions defined for the system, chiller per-
downstream chiller is always cooler than the system water. formance is well below the 6.10 COP (0.576 kW/ton) require-
Because each of the chillers in this design has two refrigera- ment set by ANSI/ASHRAE/IESNA Standard 90.1-2001, Energy
tion circuits (Figure 1), the reduced lift effect is multiplied. Standard for Buildings Except Low-Rise Residential Build-
Instead of two lifts, there are four (Figure 2). Therefore, the ings. At standard ARI rating conditions, each chiller module
difference in average lift at design for the system with four would operate with an efficiency of 0.445 kW/ton. The per-
independent refrigeration circuits in a seriesseries formance conditions for this application, however, were care-
counterflow arrangement exceeds 19%: fully selected to optimize the overall energy consumption of
the entire chilled water plant.1
1 {[(51.1 + 50.3 + 50.0 + 48.9)/4]/61.9} = 0.191
Standard ARI rating conditions are 44F, 2.4 gpm/ton (6.7C,
The upstream chiller need not be equally loaded at all 0.043 L/s per kW) for the evaporator and 85F, 3 gpm/ton (29.4C,
times; the anticipated savings come from that chillers abil- 0.054 L/s per kW) for the condenser.
Focus on the System the chiller evaporators in response to system load. The pump
The reduction in chiller power comes at the expense of penalty still exists at full load. However, it is significantly
increased pumping energy. With the utility meter measuring less at part load/partial flow because the differences in evapo-
the consumption of the entire plantnot the chillers alone rator pressure drop and chilled water pumping power decrease
the ultimate goal must be to reduce plant power. Would the rapidly as the flow rate slows. Combining variable primary
seriesseries counterflow arrangement reduce the compres- flow and a series arrangement of evaporators circulates twice
sor power enough to make up for the increase in pump power? as much water through the evaporator of each chiller.
Table 2 compares the overall power consumption for the This combination also creates a second advantage: each
chilled water plant based chiller can accommodate
on three different arrange- a much greater reduction
ments of six chillers. In Arrangement Pumps in water flow, which post-
Cooling Total
each case, the condenser- Chilled Condenser To w e r P l a nt pones the need for a sys-
Evaporator Condenser Chillers
water flow rate is 2 W a t e r W a te r tem bypass. (Use of
gpm/ton (0.0358 mL/J). Parallel Parallel 6,489 14 18 480 7,001 variable primary-flow
Series Parallel 5,827 60 18 480 6,385
Chilled Water Pumps. systems is increasingly
Series
The f ive chilled water Series Counter- 5,618 60 126 480 6,284 common and well docu-
pumps in this design are flow mented.) 46,9,1112,1415,17
piped in a manifold ar- Table 2: Comparison of power requirements (kW) at identical Condenser Water
rangement; one of the rates of condenser water flow. Pumps. The condenser
pumps is redundant. water pumps are piped in
Doubling the water flow through the evaporators created a a manifold arrangement. In the selected plant configuration,
water pressure drop that was significantly higher for the se- one active pump is provided per chiller module. Redundancy
ries arrangements than for the parallel configurations (Table is provided by two smaller pumps, which enables one of the
1). To minimize this penalty as much as possible, single- pumps to operate during low-load conditions and reduce
pass tube bundles were used in the series evaporators. Other- pumping energy consumption. This is described later.
wise, the pressure drop and resultant pump power would have Basing an economic comparison of design alternatives on
been even higher. Despite the high water pressure drop, re- life-cycle costs requires an overall summation of the total costs
ducing compressor lift by arranging the evaporators in series for power (demand) and energy (consumption). Therefore, al-
yielded chiller power savings that dwarfed the additional though the total power for the seriesseries counterflow plant
pump power needed at full load. was lowest (Table 2), it was also necessary to account for the
To further offset the economic impact of larger chilled wa- costs related to energy consumption.
ter pumps, the plant design also varies primary flow through Water flow rates and configurations for the chiller con-
The authors of this article speculate that the decisions made by the
plant owner represented a comfortable compromise between the
seriesseries counterflow arrangement, which was economically sen-
sible but unfamiliar, and other more conventional (and familiar) as-
pects of chilled water plant design.
References
1. ARI Standard 550/5901998, Standard for Water Chilling Pack-
ages Using the Vapor Compressor Cycle (Arlington, VA: Air-Condi-
tioning and Refrigeration Institute).
2. 2000 ASHRAE HandbookHVAC Systems and Equipment. Chapter
11, District Heating and Cooling. Advertisement in the print edition formerly in this space.
3. 2000 ASHRAE HandbookHVAC Systems and Equipment Hand-
book. Chapter 12, Hydronic Heating and Cooling System Design.
4. Avery, G. 2001. Improving the efficiency of chilled water plants.
ASHRAE Journal 43(5):1418.
5. Avery, G. 1998. Controlling chillers in variable flow systems.
ASHRAE Journal 40(2):4245.
6. Bahnfleth, W.P., E. Peyer. 2001. Comparative analysis of variable
and constant primary-flow chilled-water-plant performance. HPAC
Engineering, April, p. 50.
7. Coad, William J. 1998. A fundamental perspective on chilled
water systems. Heating/Piping/Air Conditioning, August, pp. 5966.
8. Demirchian, G.H., M.A. Maragareci. 1997. The benefits of higher
condenser water T at Logan International Airport central chilled water
plant. IDEA 88th Annual Conference Proceedings, pp. 291300.
9. Hartman, T. 2001. All-variable speed centrifugal chiller plants.
ASHRAE Journal, September.
10. Kelly, D.W., T. Chan. 1999. Optimizing chilled water plants.
Heating/Piping/Air Conditioning, January, pp. 145147.
11. Kirsner, W. 1996. The demise of the primary-secondary pump-
ing paradigm for chilled water plant design. Heating/Piping/Air Con-
ditioning, November.
12. Luther, K. 1998. Applying variable volume pumping. Heat-
ing/Piping/Air Conditioning, October, pp. 5358.
13. Rishel, J.B. 1996. HVAC Pump Handbook, pp. 109112.
14. Schwedler, M., B. Bradley. 2000. Variable-primary-flow sys-
tems: An idea for chilled-water plants the time of which has come.
Heating/Piping/Air Conditioning, April, pp. 4144.
15. Taylor, S. 2002. Primary-only vs. primary-secondary variable-
flow chilled water systems. ASHRAE Journal 44(2):2529.
16. Trane. 2001. Multiple-Chiller-System Design and Control, pp.
5859.
17. Waltz, J.P. 1997. Dont ignore variable flow. Contracting Busi-
ness July:133144.