Home / DC Solutions / Technical Manuals / Chiller Plant

Technical Manual · Cooling Plant

Chiller Plant — P&ID Methodology & Engineering Basis

Full engineering methodology for the chiller-plant SCADA HMI: ISA P&ID symbol conventions, primary/secondary chilled-water loop design, lead-lag chiller sequencing, CHW flow derivation, and the display/status engine anchored to js/conv-engine.js. This is an honest account — the page is a partial-calc display engine, not a full thermo-hydraulic simulation.

Engine conv-engine.js v2.0.0 Basis CONV_CALC v2.0.0 CHW Basis 19.4 / 27.0 °C Plant Config 7/10 · N+1
▶ Open the live Chiller Plant HMI

01 Purpose & engineering basis

The chiller-plant page is an operator-grade SCADA HMI mimic — an ISA-style P&ID rendered in the browser. Its job is to show plant topology (primary/secondary loops, chiller train positions, pump duty/standby), deterministic simulated status (run/fault, lead-lag state, flow animation), and setpoint annotations (CHWS supply temperature, differential-pressure setpoints, N+1 headroom).

Truth boundary: current telemetry is the adopted 30 MW site-IT scenario from CONV_CALC v2.0.0. The 40 MW site-IT figure is the governed four-hall design boundary. Current plant status is 7 of 10 water-cooled centrifugal chillers running, 31.25 MW evaporator duty, 36.4034 MW condenser/tower heat rejection, 35 MW running capacity, and 45 MW post-one-unit-loss capacity.

What the page is not: it is not a dynamic thermo-hydraulic simulator. Pump speed, valve position, pressure and per-machine annotations are bounded deterministic HMI values. The 943.0 and 982.3 L/s figures are calculated reference flows, not measured header flow. Evaporator duty, tower rejection, chiller input, COP, kW/RT and N+1 margin descend from the governed engine snapshot.

P&ID tags follow ANSI/ISA-5.1. ASHRAE 90.4 is cited as the data-centre energy-performance framework; it is not used here to invent a universal CHW setpoint or COP. The project temperatures and equipment assumptions remain explicit scenario inputs.

02 Inputs — mode toggles & locked CHW basis

The page has no free-form calculator inputs. Instead it exposes mode toggles and reads a locked CHW basis from conv-engine.js. All displayed numbers descend deterministically from these inputs — no random values, no per-page hardcoded constants that contradict the basis.

Input / ToggleValue / RangeTypeNotes
CHWS temperature19.4 °CAdopted current basisWarm-water primary chilled-water supply from snapshot.cooling.chws_c.
CHWR temperature27.0 °CAdopted current basisWarm-water primary chilled-water return from snapshot.cooling.chwr_c.
Current site IT load30,000 kWAdopted current basisIT sensible-load reference-flow term from snapshot.site.it_load_kw.
Current evaporator duty31,250 kWDerived current basisIT plus UPS loss; the plant-duty reference-flow numerator, not condenser/tower heat rejection.
Condenser/tower rejection36,403.4 kWDerived current basisEvaporator duty plus chiller electrical input.
Condenser water32 / 37 °CAdopted current basisDistinct condenser-water supply/return circuit; never relabelled as CHWS/CHWR.
Lead-lag modeAUTO / MANUALMode toggleAUTO: engine picks lead chiller by run-hours. MANUAL: operator selects.
View detail levelOVERVIEW / DETAIL / ENGINEERINGMode toggleControls label density on the P&ID. OVERVIEW shows only major status; ENGINEERING reveals all sensor annotations.
Chillers running / installed7 / 10Adopted current basisEach assumed at 5,000 kW thermal; central plant redundancy is N+1.
Loop identity rule: CHWS/CHWR are 19.4/27.0 °C. CDWS/CDWR are 32/37 °C. A page must fail closed to UNAVAILABLE when this complete current authority is missing; it must never revive retired temperature pairs or infer a secondary setpoint.

03 Calculation methodology

The page derives CHW delta-T, two explicitly scoped reference flows, chiller input power, and condenser/tower rejection. Actual header and branch flow remain unavailable without meters. Function names map one-to-one to js/conv-engine.js.

CHW delta-T

ΔT = CHWR − CHWS = 27.0 − 19.4 = 7.6 K In chwDeltaT(m). Source: snapshot.cooling.chwr_c − snapshot.cooling.chws_c. Anchors current flow and heat-transfer calculations.CONV_CALC v2.0.0

CHW reference flows

Q̇_IT-ref = IT_load / (c_p × ΔT) = 30,000 / (4.186 × 7.6) = 943.0 L/s Q̇_plant-duty-ref = evaporator_duty / (c_p × ΔT) = 31,250 / (4.186 × 7.6) = 982.3 L/s c_p = 4.186 kJ/(kg·K). The engine publishes 943.0 L/s as its IT sensible-load reference; the HMI also reconciles the 982.3 L/s current plant-duty reference. Both are calculations. Measured header flow is UNAVAILABLE.CONV_CALC v2.0.0 · chwFlowLps · cooling.heat_rejection_kw

Per-loop flow split

Flow_per_loop = Q̇_total / n_loops Displayed engine-reference split: 943.0 / 7 = 134.7 L/s per running chiller Plant-duty sizing reference: 982.3 / 7 = 140.3 L/s per running chiller Display-only derivations. Real plants balance by valve position and pump speed; actual branch flow is UNAVAILABLE without branch meters.doc-09 lines 86–93 · doc-04 §Cooling Calculation

Chiller input power (kW/ton sanity check)

P_chiller = evaporator duty × specific power = 31,250 × 0.16491 = 5,153.4 kW electrical COP = 31,250 / 5,153.4 = 6.06; plant efficiency = 0.58 kW/RT Condenser/tower rejection = 31,250 + 5,153.4 = 36,403.4 kW These are adopted engine-derived current KPIs. They are not universal ASHRAE setpoints and must be replaced by measured plant power and calibrated thermal load for live use.CONV_CALC v2.0.0 · chillerInputKwE

Lead-lag sequencing logic

Lead chiller: lowest cumulative run-hours (auto mode) Lag start: CHWS actual > CHWS SP + 0.5 K for t > 120 s Lag stop: Q̇_demand < 0.60 × Q̇_lead_rated for t > 300 s These thresholds are illustrative display-engine constants — the page animates the state machine. A real chiller plant controller would implement these in the PLC/DCS ladder; the HMI only reads back the RUN/STANDBY status bits.doc-04 §Lead-Lag · ASHRAE TC9.9 Table 3

N+1 cooling headroom

Available capacity = chillers_running × Q̇_rated_per_chiller N+1 margin = available − Q̇_demand Current running capacity = 7 × 5,000 = 35,000 kW thermal Post-one-unit-loss installed capacity = 9 × 5,000 = 45,000 kW thermal N+1 margin = 45,000 − 31,250 = 13,750 kW thermal The 45 MW figure is capacity with one of ten installed units unavailable. Current running capacity and post-failure installed capacity are different denominators and are labelled separately.CONV_CALC v2.0.0 · cooling capacity fields

04 ISA P&ID symbol conventions

The HMI follows ISA 5.1 tagging and symbol conventions. Equipment tags visible on the page map to the table below. Alarm colour hierarchy is strictly enforced — see the note on colour discipline.

ISA Tag PrefixInstrument TypeDisplay LocationExample Tag
TTTemperature TransmitterCHWS/CHWR headers, loop supply/returnTT-101 CHWS header
FTFlow TransmitterPrimary header, per-loop risersFT-201 primary header
PTPressure TransmitterPump discharge/suction, differential across coilPT-301 pump-A discharge
DPDifferential PressureAcross strainer, across chiller bundleDPT-401 strainer
MVMotor-operated ValveChiller isolation, bypassMV-501 chiller-1 isolation
CVControl ValveBypass, secondary-loop modulatingCV-601 DP bypass
PPumpPrimary CHW pumps, condenser water pumpsP-701A/B duty/standby
CHChillerChiller train blocksCH-1/CH-2/CH-3 lead/lag/standby
Alarm colour hierarchy (non-negotiable): Normal flow animation uses muted teal (#5da095 / --chwr). Standby equipment uses neutral grey (--ok). Warning conditions (N+1 margin low, pump near-trip) use signal amber (--warn). Fault/trip/alarm conditions use fault-red (--alarm). Instrument-cyan (--mn-accent) is reserved for page chrome only and must NEVER appear on a flow line or equipment symbol that can carry an alarm state — it would erase the alarm's visual signal. This is the same rule that governs all 34 manual pages.

05 Primary / secondary loop architecture

The chiller plant uses a decoupled primary/secondary (P/S) design. The primary loop circulates through the chiller bundles at a constant low flow; the secondary loop modulates to match the data-hall cooling demand. A hydraulic decoupling bridge (bypass header) absorbs any flow mismatch.

ParameterPrimary LoopSecondary Loop
Supply temperature19.4 °C (CHWS)19.4 °C at the facility distribution supply header
Return temperature27.0 °C (CHWR)27.0 °C at the facility distribution return header
Flow regimeConstant-flow (primary pumps run at fixed speed)Variable-flow (VFD pumps track DP setpoint)
DP setpointN/A (constant speed)Secondary pump DP SP (operator-configurable)
Bypass / decouplerShared — absorbs primary/secondary flow mismatchShared — bypass flow direction reverses under low-load
Pump tagsP-701A/B (duty/standby)P-801A/B (VFD, duty/standby)
HMI animationMuted teal flow arrows, constant velocityFlow arrow velocity scales with VFD speed display
Hydraulic scope: the HMI draws four representative machine modules for readability, while the governed plant contains ten chillers with seven running. The module mimic is not a four-machine total. Per-machine flow, pump head and pipe diameter require the as-built hydraulic design; absent values fail closed rather than being invented.

06 Worked example — current 30 MW site

This example reproduces the governed current plant fields. Every scope is explicit: site IT reference, evaporator-duty reference, condenser/tower rejection, running-machine display split, current running capacity, and post-one-unit-loss installed capacity.

  1. CHW delta-T: 27.0 − 19.4 = 7.6 K.
  2. Cooling-flow term: current site IT 30,000 kW.
  3. IT sensible-load reference flow: 30,000 / (4.186 × 7.6) = 943.0 L/s.
  4. Evaporator-duty reference flow: 31,250 / (4.186 × 7.6) = 982.3 L/s.
  5. Displayed engine-reference split: 943.0 / 7 = 134.7 L/s per running machine; plant-duty sizing reference is 140.3 L/s; actual balancing requires measured branch flow.
  6. Evaporator duty: 30,000 + 1,250 = 31,250 kW.
  7. Current running capacity: 7 × 5,000 = 35,000 kW thermal.
  8. Post-one-unit-loss capacity: 9 × 5,000 = 45,000 kW; N+1 margin = 13,750 kW.
  9. Chiller input: 31,250 × 0.16491 = 5,153.4 kW electrical; COP 6.06; 0.58 kW/RT.
  10. Condenser/tower rejection: 31,250 + 5,153.4 = 36,403.4 kW.
Honesty note: Steps 4–7 are display benchmarks derived from the locked basis, not measured or simulated from independent hydraulic/thermodynamic models. The page's purpose is SCADA HMI visualisation — the genuine engineering rigour is in the CHW basis derivation (steps 1–3) and the N+1 headroom logic. Equipment-level annotations (pump speed, valve %, VFD current) are display-only status fields.

07 References & standards

08 Assumptions & limitations

The HMI assumes a well-balanced, decoupled primary/secondary plant at the adopted steady-state operating point. Transient effects (pull-down, pump inrush, glycol correction, tower approach and dynamic staging) are not modelled. The displayed COP 6.06 and 0.58 kW/RT are engine-derived scenario values; live use requires measured electrical power, calibrated flow and temperature sensors.

Flow-velocity animation on the P&ID is a visual cue proportional to the displayed flow value; it is not derived from a hydraulic resistance network. Pipe diameters and friction factors are not modelled. Equipment-level display values (valve %, VFD speed %, run hours) are status-field annotations consistent with the engineering basis — they are not back-calculated from independent models and must be treated as illustrative display parameters in any off-page engineering analysis.

This page and its methodology are an engineering education and visualisation aid. Final plant design, setpoint commissioning, and N+1 validation must be performed by qualified mechanical/controls engineers against the actual installed equipment and as-built P&IDs.

▶ Open the live Chiller Plant HMI