A new FM engineer joins the team and asks a simple question: “Where does the cold air in the building actually come from?” You point at an FCU in the ceiling. They ask what feeds it. You say the chilled water system. They ask what that is. And suddenly you realise you’ve been maintaining a system for years without ever stepping back to explain the whole picture — from the chiller plant to the air coming out of the grille.
Understanding the chilled water system as a complete picture — not just the individual components you maintain — changes how you troubleshoot, how you communicate with clients, and how you spot problems before they become failures. A chiller that’s running fine but producing poor ΔT means nothing unless you understand the full loop it’s connected to.
This guide explains the chilled water system completely — how it works, what each component does, the design principles behind it, the key operating parameters your team should know, and the most common performance problems that come from poor understanding of how the system works as a whole. Written for FM engineers, maintenance technicians, and anyone responsible for a building with central cooling.
What Is a Chilled Water System and Where Is It Used?
A chilled water system is a centralised cooling system that uses water as the medium to transfer heat from a building to the atmosphere. Instead of refrigerant pipes running to every air conditioning unit (as in a VRF or split system), a chilled water system uses a central chiller to cool water, then pumps that cold water through pipework to air handling units and fan coil units throughout the building.
The core principle is straightforward: water absorbs heat well and can be transported efficiently over long distances. A single chiller in the basement plant room can serve hundreds of zones across a 30-storey tower, a sprawling hospital campus, or a large shopping centre — places where individual refrigerant-based systems would be impractical, inefficient, or impossible to maintain.
Chilled water systems are the standard cooling solution for:
• Commercial office buildings — typically above 5–10 floors or 5,000 m² floor area
• Hotels — where central plant with FCUs in each room is the most practical approach
• Hospitals and healthcare facilities — where air quality control and system reliability are paramount
• Shopping centres and retail — large floor areas with high heat loads from lighting, people, and equipment
• Universities and campuses — where district cooling serves multiple buildings from a central plant
• Data centres — where precision cooling of high-density server rooms requires the reliability of central plant
How a Chilled Water System Works: The Complete Cooling Loop
To understand the chilled water system, you need to understand two separate but connected water loops: the chilled water loop (which cools the building) and the condenser water loop (which rejects the heat to atmosphere). Both loops pass through the chiller, which is the bridge between them.
The chilled water loop — how the building gets cooled
Cold water leaves the chiller at 6–7°C and is pumped through insulated pipework to every air handling unit and fan coil unit in the building. At each terminal unit, warm room air blows across the chilled water cooling coil. The air gives up its heat to the water — the air cools down and is delivered into the room; the water warms up.
The water returns from the terminal units at 12–13°C — it has picked up 5–6°C of heat from the building. This return water goes back to the chiller, which extracts the heat and cools it back down to 6–7°C, ready to go around again. That 5–6°C temperature difference — the delta T (ΔT) — is one of the most important performance parameters in any chilled water system.
If the ΔT is lower than 5–6°C — say, only 2–3°C — the chiller is running but delivering a fraction of its design cooling capacity. This is Low ΔT Syndrome, one of the most common and most expensive performance failures in chilled water systems.
The condenser water loop — how the heat gets rejected
The chiller extracts heat from the chilled water using a refrigeration cycle. That heat has to go somewhere — it goes into the condenser water. Condenser water enters the chiller condenser at typically 29–32°C and leaves at 35–37°C, carrying the heat away.
The condenser water is then pumped to the cooling tower on the roof, where it cascades over fill media while fans draw air through. The water gives up heat to the air, cools back down to 29–32°C, and returns to the chiller. This is why the cooling tower condition is so critical to chiller performance — if the tower can’t cool the condenser water back down to the design temperature, the chiller’s head pressure rises, efficiency drops, and eventually the chiller trips.
| 💡 Key insight: The cooling tower is as important to chilled water system performance as the chiller itself. Most chiller high-head-pressure faults trace directly back to a poorly maintained cooling tower — not the chiller. |
Chilled Water System Components: What Each One Does
Here is every major component in a typical chilled water system and what it actually does — written for the engineer who needs to understand why each component matters, not just that it exists:
| Component | What It Does and Why It Matters |
| ❄️ Chiller | The heart of the system. Removes heat from chilled water using a refrigeration cycle. Output: chilled water at 6–7°C. Size ranges from 50 kW for small buildings to 10,000+ kW for large campuses. |
| 💧 Chilled Water Pump (Primary) | Circulates chilled water through the chiller evaporator. In a primary-secondary system, this pump serves the chiller only — not the building distribution. |
| 💧 Chilled Water Pump (Secondary) | Distributes chilled water to AHUs, FCUs, and other terminal units across the building. Variable speed drive (VSD) control modulates flow to match building demand. |
| 🌡️ Cooling Tower | Rejects heat from the condenser water circuit to atmosphere. Essential for water-cooled chillers. Cooling tower condition directly affects chiller head pressure and efficiency. |
| 💧 Condenser Water Pump | Circulates condenser water between the chiller condenser and the cooling tower. Fixed speed in most installations — flow is constant to maintain chiller performance. |
| 🔧 AHU / FCU (Terminal Units) | The building-side heat exchangers. Cooling coils in AHUs and FCUs transfer heat from supply air to chilled water, cooling the air before delivery to occupied spaces. |
| ⚙️ Control Valves | 2-way or 3-way valves on each terminal unit control chilled water flow to match zone demand. Stuck-open valves are the primary cause of low ΔT syndrome. |
| 📊 Building Management System (BMS) | Controls the entire system: chiller staging, pump speed, valve positions, setpoint reset, and alarm management. The brain of the chilled water system. |
Chilled Water System Configurations: Primary-Only vs Primary-Secondary
Not all chilled water systems are designed the same way. The hydraulic configuration — how the pumps are arranged — determines how the system responds to varying load and how energy-efficient it is. The two main configurations are primary-only and primary-secondary.
Primary-only variable flow system
A single set of variable speed pumps circulates chilled water through both the chiller and the building distribution. When cooling demand decreases, the pumps slow down and reduce flow. Simple, low capital cost, fewer pumps to maintain.
The limitation: modern chillers have a minimum flow requirement for the evaporator — typically 30–40% of design flow. A primary-only system that modulates below this minimum risks flow instability and chiller trips. This configuration works well when the cooling load never drops below the minimum chiller flow — common in buildings with relatively constant loads like data centres.
Primary-secondary variable flow system
Two separate pump sets: primary pumps that circulate water through the chiller at a constant flow rate (meeting the minimum chiller flow requirement), and secondary pumps with variable speed drives that distribute water to the building at varying flow rates based on actual demand.
A bypass leg (common header) connects the primary and secondary circuits, allowing the flow rates to be different. When secondary demand is low, some primary water bypasses the building and mixes with return water. When demand is high, secondary pumps draw from the primary circuit at higher rates.
This configuration is the most common in large commercial buildings because it allows variable speed pumping (energy saving) while protecting the chiller from low-flow conditions. It’s more complex and has more components to maintain, but the energy savings over a system lifetime are substantial.
| 📌 For chiller troubleshooting that connects directly to these operating principles, see the Chiller Troubleshooting Guide. |
Key Operating Parameters Every FM Engineer Must Know
These are the numbers that define whether your chilled water system is performing correctly. If you don’t know these parameters for your building, pull the commissioning data and find them. They’re your baseline for every maintenance decision:
| Parameter | Design Value | What It Tells You |
| Chilled water supply temp | 6–7°C | Chiller output temperature. If above 8°C at design load: chiller not performing or setpoint issue. |
| Chilled water return temp | 12–13°C | Heat picked up from building. Target ΔT = 5–6°C. Lower return = low ΔT syndrome. |
| System ΔT | 5–6°C | The key performance indicator. Below 4°C = investigate. Below 3°C = system has a serious problem. |
| Condenser water supply temp | 29–32°C | Water entering chiller condenser from cooling tower. Above 35°C = cooling tower issue. |
| Condenser water return temp | 35–37°C | Water leaving chiller to cooling tower. ΔT across condenser should be 5–6°C. |
| Chiller COP at full load | 4.5–6.5 | Energy efficiency of chiller. Declining COP over time = maintenance issue or system degradation. |
| Pump differential pressure | Per design (kPa) | Pressure across the pump set. Rising ΔP at constant flow = system resistance increasing (strainer, valve). |
| System flow rate | Per design (L/s) | Total chilled water flow. Significantly below design = pump problem, strainer blockage, or valve issue. |
The Most Common Chilled Water System Performance Problems
Most performance problems in a chilled water system are not mysterious — once you understand how the system works, the symptoms point directly to the cause. Here are the most common ones:
Low Delta T Syndrome
The single most common and most damaging performance problem in chilled water systems. Design ΔT is 5–6°C. The system is producing 2–3°C. The chiller runs continuously but the building never reaches setpoint. Energy bills spike. Nobody can work out why because the chiller appears to be running normally.
The cause is almost always control valves that are stuck open or operating at minimum position — they allow chilled water to bypass the terminal unit coil, mixing warm return water back into the chilled water circuit and reducing the apparent return temperature. Walk the building and feel the chilled water pipework near each terminal unit. Return pipes that are cold (close to supply temperature) indicate a stuck-open valve mixing cold supply water into the return before it picks up heat.
For a detailed diagnosis guide, see the Chiller Low Delta T Syndrome .
Chiller high head pressure
Chiller trips on high condenser pressure. The instinct is to investigate the chiller — check the refrigerant, check the compressor. In 80% of cases, the cause is the cooling tower, not the chiller. A tower with blocked fill, a failed fan, or inadequate water treatment cannot cool condenser water back to the design temperature. The condenser water returns too warm, the refrigerant condensing pressure rises, and the chiller trips to protect itself.
Always check the condenser water temperature before investigating the refrigerant circuit. If condenser water return is above 35°C, the problem is the cooling tower. For the full diagnostic approach, see the Chiller Troubleshooting Guide.
Air in the chilled water system
Air pockets in the chilled water system cause noise (gurgling, banging), poor flow at high points, and reduced heat transfer at terminal unit coils. The symptom is usually zones that never cool properly despite the chilled water supply temperature being correct. The fix is systematic bleeding — starting at the high points and working down. Air problems are most common after system shutdown and refilling, after pipework modifications, or in systems without automatic air vents at high points.
Strainer and filter blockage
Every chilled water system has strainers on pump inlets and at terminal units. As the system ages, scale, corrosion products, and debris accumulate. A blocked pump inlet strainer causes the pump to cavitate — you hear a crackling sound from the pump, flow drops, and differential pressure across the pump increases. Strainers should be checked quarterly and cleaned when the differential pressure across them is more than 20% above the clean value.
Chilled Water System Maintenance: What Your Team Must Do Monthly
A chilled water system doesn’t have a single dedicated maintenance checklist — it’s maintained through the individual PPM programmes for each component. But these are the monthly checks that apply to the system as a whole:
• Log all operating parameters — CHW supply and return temperatures, condenser water supply and return temperatures, chiller amps, pump differential pressures. Trend these over months. A steady drift in any parameter is an early warning of developing problems
• Calculate and record system ΔT — this single number tells you more about system health than any individual component reading. If it’s dropping month by month, investigate immediately
• Check expansion vessel pre-charge pressure — cold system pressure should match the pre-charge pressure (typically 1.0–1.5 bar). A system that regularly loses pressure has a leak. A system with incorrectly charged expansion vessel risks pressure fluctuations that trip chiller safety devices
• Inspect all pipework insulation — damaged insulation on chilled water pipework allows condensation and heat gain. Replace any damaged sections immediately — heat gain on the distribution pipework reduces effective cooling capacity and increases energy consumption
• Check pump seal condition — drip from a pump mechanical seal is normal (1–2 drips per minute). Excessive leaking or spray = seal replacement required. Log seal condition monthly and schedule replacement before failure
| 📌For the complete HVAC preventive maintenance programme that covers all system components, see the HVAC Preventive Maintenance Guide. |
Frequently Asked Questions
What is the difference between a chilled water system and a VRF system?
A chilled water system uses water as the heat transfer medium — a central chiller cools the water, pumps distribute it to terminal units (AHUs and FCUs) throughout the building, and the water picks up heat and returns to be cooled again. A VRF (Variable Refrigerant Flow) system runs refrigerant directly from an outdoor unit to multiple indoor units — no water is involved in the distribution. Chilled water systems are better suited to large buildings where long pipework runs make refrigerant distribution impractical, and where centralised plant maintenance is preferred. VRF systems are more common in medium-sized buildings and applications where zone-by-zone flexibility is the priority.
What is the design chilled water temperature?
The standard design chilled water supply temperature is 6–7°C, with a return temperature of 12–13°C, giving a design ΔT of 5–6°C. These temperatures are a balance between chiller efficiency (warmer chilled water = higher chiller COP = less energy) and terminal unit performance (colder water = smaller coils = lower capital cost). Some modern systems use higher chilled water temperatures (7–9°C supply) to improve chiller efficiency, particularly in mild climates, but require larger terminal unit coils to compensate for the reduced temperature differential.
Why does my chilled water system have low delta T?
Low ΔT syndrome — where the system produces 2–3°C instead of the design 5–6°C — is almost always caused by control valves that are stuck open or operating at minimum position at terminal units. These valves allow chilled water to pass through the coil without picking up significant heat, then mix cold supply water into the return, reducing the apparent return temperature. The fix is identifying which valves are stuck open (by checking which return pipes are cold when they should be warm) and repairing or replacing them. For a detailed diagnostic guide see the Chiller Low Delta T Syndrome article.
What is the purpose of the bypass leg in a primary-secondary chilled water system?
The bypass leg (also called the common header or decoupler) connects the primary and secondary circuits at a low-resistance point between the chillers and the building distribution. Its purpose is to allow the primary and secondary circuits to operate at different flow rates — the primary pumps maintain constant flow through the chillers, while the secondary pumps vary flow based on building demand. When secondary demand is less than primary flow, excess primary water bypasses through the decoupler and mixes with return water. When secondary demand exceeds primary flow, water from the decoupler supplements the primary supply. A decoupler that is too long (high resistance) defeats the purpose of the primary-secondary arrangement.
How do I know if my chilled water system is sized correctly for my building?
The clearest indicator of correct sizing is the system ΔT under peak load conditions: a correctly sized system achieves 5–6°C ΔT at design load with chilled water supply at 6–7°C. If the system struggles to cool the building during peak conditions (high ambient temperature, full occupancy) despite the chiller running at 100%, either the chiller is undersized for the actual load, or the distribution system has problems (low ΔT, high strainer resistance, failed pumps) that reduce effective cooling capacity. Always compare against the original commissioning data to determine which is the case.
Conclusion
A chilled water system is not a collection of separate assets — it’s one interconnected loop where every component affects every other. The chiller performance depends on the cooling tower. The ΔT depends on the control valves. The pump performance depends on the strainer condition. Understanding these connections is what separates an engineer who can troubleshoot a chilled water system from one who can only call the specialist.
For the most common chilled water system fault, see the Chiller Low Delta T Syndrome For the complete chiller fault diagnosis guide, see the Chiller Troubleshooting Guide. Use the PPM Planner to build your chilled water system maintenance schedule.
