A chiller plant is not a single machine — it is a system of interdependent components whose collective efficiency is determined almost entirely by design choices made months before the first weld is struck.
For most commercial and industrial facilities in India — offices, hospitals, data centres, pharmaceutical plants, hotels — the chiller plant is the single largest electrical load on the building. In a large air-conditioned building, chilled-water generation and distribution alone can account for 40 to 60 per cent of annual electricity consumption. That is not a statistic to note and move on from; it is the central financial argument for investing in careful, engineering-led plant design rather than the cheapest-capacity solution a procurement team can source.
This article examines the specific design decisions that move the needle on energy — and the pitfalls that cause plants, perfectly functional on commissioning day, to consume far more electricity than their specifications promised.
Understanding part-load reality: IPLV and NPLV
The most common error in chiller selection is optimising for peak kW per tonne of refrigeration (TR) at full-load design conditions. In practice, a chiller in India rarely operates at 100% load. Outdoor wet-bulb temperatures are only at their annual maximum for a fraction of operating hours; occupancy and internal gains vary continuously. The ARI 550/590 standard (now AHRI 550/590) introduced the Integrated Part-Load Value (IPLV) and Non-Standard Part-Load Value (NPLV) precisely to address this: they express a weighted average COP across four part-load conditions — 100%, 75%, 50% and 25% — reflecting the profile of typical operation.
A chiller with a strong full-load kW/TR but poor IPLV will cost more to run over a year than a unit with a moderate full-load efficiency but a high IPLV rating. MEP consultants specifying for Indian climates should model annual energy using the NPLV methodology with localised bin-hour data rather than specifying purely on peak condition performance.
A chiller that looks 8% cheaper to buy but carries a 12% inferior NPLV rating will, over a 15-year asset life, cost far more than the capital saving — even before accounting for maintenance and refrigerant costs.
Chiller sequencing and staging
Multi-chiller plants are almost universal above 300–400 TR. The question is not whether to install multiple units but how to operate them. Chiller staging — the logic that decides when to start or stop a machine — has a profound effect on efficiency.
Running one large chiller at 50% vs two at 50% each
The efficiency curve of a centrifugal or screw chiller is not linear. Most modern chillers reach peak efficiency at 70–85% part load, with efficiency falling sharply below 40%. Running a single large machine at 45% load may be less efficient than running two smaller machines, each at a higher load fraction. The correct staging sequence is plant-specific — it depends on the number of machines, their individual part-load curves, and the load profile of the facility — but it must be encoded into the BMS from commissioning, not left to the default logic supplied by a controls vendor.
Lead-lag rotation and equal run-hours
Beyond efficiency, staging logic must account for lead-lag rotation — distributing run-hours across all machines so that no single unit accumulates disproportionate wear. A BMS strategy that always starts the same lead chiller will present a failed unit at precisely the moment it is most needed.
Variable-primary-flow vs primary-secondary pumping
Traditional chiller plants in India use a primary-secondary (decoupled) pumping arrangement: constant-speed primary pumps maintain minimum flow through each chiller; a separate secondary loop, driven by variable-speed pumps, serves the building. This is a reliable, forgiving topology — but it has an energy penalty. The primary pumps run at constant speed regardless of system demand, and at part load the bypass bridge can short-circuit warm return water back to the chiller, degrading efficiency.
Variable-primary-flow (VPF) eliminates the secondary loop entirely. A single set of variable-speed pumps varies chilled-water flow through the chillers in response to system demand, subject only to a manufacturer-specified minimum evaporator flow rate. The energy saving in the pump motors alone can be 20–35% compared with a constant-primary arrangement. However, VPF demands more rigorous controls design: the BMS must manage flow rate and chiller staging simultaneously, and the chiller manufacturer’s minimum flow envelope must be respected to avoid evaporator freeze and surge events. When designed and commissioned correctly, VPF is the preferred topology for most new Indian commercial installations above 500 TR.
Condenser-water optimisation
For water-cooled chillers, the condenser-water circuit — cooling towers, condenser-water pumps, and the piping between them — is the second-largest electrical consumer in the plant. Three design decisions dominate this subsystem’s energy consumption.
Approach temperature and cooling-tower selection
The approach temperature — the difference between the cold-water temperature leaving the cooling tower and the ambient wet-bulb temperature — is a direct function of tower surface area, fill media and airflow. A smaller approach means colder condenser-water entering the chiller, which raises chiller COP substantially: a 1°C reduction in condenser entering water temperature typically improves chiller efficiency by 2–3%. Over-specifying tower size is often justifiable on a lifecycle basis, but it must be weighed against the capital cost and footprint constraints common on Indian rooftops.
Condenser-water setpoint reset
Many plants operate at a fixed condenser-water setpoint (typically 32°C) irrespective of ambient conditions. During cooler months, night-time operation, or periods of low outdoor wet-bulb temperature, the condenser-water temperature can be safely reduced well below setpoint. Condenser-water temperature reset — allowing the setpoint to float downward as wet-bulb drops — is one of the most cost-effective controls strategies available and requires only a BMS modification to implement on existing plant.
Cooling-tower fan VFDs
Cooling-tower fans sized for peak summer conditions spend the majority of their annual run-hours at part load. Installing variable-frequency drives (VFDs) on tower fans and controlling them to maintain the condenser-water setpoint (including any reset strategy) provides fan energy savings that follow the cube law: reducing fan speed to 80% cuts fan power to approximately 51% of full-load. In Indian climates with relatively long shoulder seasons, the payback period for tower-fan VFDs is typically two to four years.
The low delta-T syndrome
In a correctly designed chilled-water system, supply water leaves the chiller at 6–7°C and returns from the building at 12–13°C — a delta-T of 5–6°C. Low delta-T syndrome occurs when the actual return temperature is significantly lower than design, perhaps 9 or 10°C, reducing the delta-T to 2–3°C. The consequences cascade: to deliver the same cooling capacity, the system must circulate a much higher volume of water, which increases pumping energy sharply; the chiller operates at higher entering water temperatures, degrading COP; and in VPF systems, it may trigger premature staging of additional chillers.
The root causes of low delta-T are well understood:
- Oversized coils — AHU and FCU coils selected with excessive safety margins bypass chilled water at full flow even at low loads.
- Incorrect control valve sizing — oversized two-way or three-way valves operate in the cracked-open position for most of their duty, providing poor flow control.
- Poorly commissioned balancing — without proper hydronic balancing, high-resistance branches receive insufficient flow whilst low-resistance branches overflow, neither achieving design delta-T.
- Bypassing three-way valves — constant-bypass three-way valves should be replaced with two-way valves in variable-flow systems, or at minimum their bypass ports locked off where demand-based control is feasible.
Correcting low delta-T on an existing plant involves rebalancing, valve replacement and coil audit — it is not a commissioning shortcut that can be addressed after handover without cost.
Hydronic balancing
A balanced hydronic system ensures that each terminal unit (AHU, FCU, process coil) receives its design flow rate under all operating conditions. Without balancing, flow distributes according to the path of least resistance: nearby terminals overflow whilst remote units starve, leading to localised comfort failures and compounding the low delta-T problem described above.
Modern best practice uses pressure-independent control valves (PICVs) at each terminal, which combine a differential pressure regulator and a flow-limiting element in a single body. PICVs maintain a constant maximum flow rate regardless of system pressure fluctuations, greatly simplifying balancing and reducing the risk of flow-related problems as plant conditions change over the facility’s life. The upfront cost premium over conventional two-way valves is recovered through reduced commissioning time and improved system performance.
BMS, controls and monitoring
A chiller plant without sophisticated controls is a plant operating on manual instinct rather than system intelligence. At minimum, a well-configured Building Management System (BMS) should deliver:
- Automated chiller start/stop sequencing based on system load, with configurable lead-lag rotation.
- Condenser-water setpoint reset based on outdoor wet-bulb measurement.
- Cooling-tower fan speed modulation via VFD to maintain condenser-water setpoint.
- Chilled-water supply temperature reset at part load, where chiller manufacturer limits permit.
- Continuous monitoring of actual plant kW/TR (not just chiller nameplate), with alarm thresholds for degraded efficiency.
- Delta-T monitoring at each chiller and across the distribution system, with alerts when values fall below design tolerance.
Energy metering at the plant level — a sub-meter on the chiller electrical panel, cooling-tower fans, and primary/secondary pumps separately — provides the data needed to identify deteriorating efficiency before it becomes a significant cost problem. This is not an add-on luxury; it is the minimum instrumentation required for responsible asset management.
Right-sizing: the danger of oversizing
Oversizing a chiller plant is perhaps the most persistent problem in Indian commercial construction, driven by conservative load calculations, generous safety factors applied at every stage of the design chain, and the instinct that “bigger is safer.” An oversized chiller operating well below its efficient part-load range for most of the year will consume more energy per unit of cooling delivered than a correctly sized machine. It will also short-cycle — starting and stopping frequently — which increases compressor wear and maintenance costs.
Right-sizing requires accurate load estimation using hour-by-hour simulation where possible, realistic occupancy profiles rather than worst-case assumptions, and an honest appraisal of diversity factors across zones. It also requires the engineering confidence to resist last-minute capacity additions during value-engineering reviews that add tonnage without adding value.
Designing for Indian conditions
India’s climate zones vary substantially, from the hot-dry conditions of Rajasthan and Gujarat to the hot-humid profile of coastal Maharashtra, Tamil Nadu and West Bengal. Condenser-water approach temperatures, cooling-tower sizing, and the efficiency benefit of setpoint reset strategies all vary significantly between these climates. ASHRAE 90.1 and ISHRAE guidelines provide the analytical framework, but the specific design parameters — outdoor design wet-bulb, bin-hour distributions, utility tariff structures — must be localised. A chiller plant designed using generic data and then deployed in Ahmedabad or Chennai without climate-specific adjustments will underperform from day one.
What this means for facility and project owners
Every decision described in this article — IPLV specification, VPF topology, condenser-water reset, hydronic balancing, BMS controls — is made during design and commissioning. By the time a facility is occupied and the electricity bills arrive, the cost structure is fixed. The margin for retrofitting these strategies narrows and their cost rises sharply after handover.
This is why the choice of HVAC system designer and installer matters so much. An integrator who understands the interaction between chiller selection, pumping topology, controls strategy and hydronic performance — and who takes accountability for commissioning the whole system, not just supplying equipment to a specification — will deliver a plant whose energy consumption matches its design intent.
At ECS, our approach to chiller plant design and SITC treats energy performance as a contractual obligation, not an aspiration. From load modelling and equipment selection through to TAB, BMS commissioning and handover documentation, we engineer each plant as an integrated system. If you are scoping a new chiller plant or reviewing the performance of an existing installation, speak with our engineering team — we are glad to discuss your specific facility and climate context.