Technical Manual · Cockpit Basis
Data Hall SCADA Console — Methodology & Formulas
Every basis parameter, derivation, constant, KPI output, and reference behind the conventional SCADA cockpit. Current operation derives from the deep-frozen CONV_CALC snapshot; the separate Hall A–D capacity context derives from RZConvDesignBasis.STUDY. Neither authority is mutated by the selector.
01 Purpose & engineering basis
The Data Hall SCADA Console is a governed-basis cockpit simulator for a four-hall conventional data centre. It does not accept arbitrary operating inputs. Every KPI derives from the deep-frozen CONV_CALC v2.0.0 snapshot shared by the Conventional BMS suite. Current operation is 30 MW IT across four equal 7.5 MW halls; 4 × 10 MW is the governed design boundary, not a second live operating value.
The basis is documented in Documents/screenshot bms rz/conv/review/09-engineering-basis-and-calculations.md and codified in js/conv-engine.js. The governing accuracy rules (ACCURACY_VALIDATION 6 rules) are:
| Rule | Description |
|---|---|
R1 | Single source of truth — every page reads from CONV_CALC; no page duplicates or hardcodes a value that conv-engine already computes. |
R2 | No Math.random() on basis KPIs — all KPIs are deterministic; values must be stable on reload. |
R3 | Explicit denominator on every metric — PUE = facility/IT; WUE denominator = IT kWh; carbon denominator = facility kWh (not IT). |
R4 | Design ≠ operation — 40,000 kW design and 30,000 kW current IT are separate fields; never relabel one as the other. |
R5 | Terminology must match engineering basis — CHWS/CHWR labels, UPS loss formula, heat rejection path must match doc-09 exactly. |
R6 | Basis chip on every critical KPI — every displayed derived number carries a source annotation traceable to a line in the review docs. |
No user inputs: this is a SCADA display cockpit, not a parametric calculator. The scenario is fixed to a real-world-representative conventional data hall. If you need to explore alternate PUE or CHW parameters, use the DC Conventional Calculator.
Current versus design: each hall carries 7,500 kW current IT over 500 installed rack positions (15.0 kW per installed position; approximately 16.2 kW per active rack in the current deterministic field). Design is 10,000 kW per hall at 20 kW/rack average, with a selected study peak of 30 kW/rack. Hall selection changes inspection context only. The 25.4 °C rack-inlet target remains normal inside the ASHRAE A-class 18–27 °C recommended envelope.
02 Locked basis parameters
These are the only authored inputs in CONV_CALC.model. Everything displayed on the cockpit is derived from this table — no other numbers are directly set.
| Parameter | Key in CONV_MODEL | Value | Unit | Source (doc-09 line) |
|---|---|---|---|---|
| Site IT design capacity | site.it_design_kw | 40,000 | kW | Governed four-hall design boundary |
| Current site IT load | site.it_load_kw | 30,000 | kW | Adopted current operating snapshot |
| PUE | site.pue | 1.45 | — | 09 line 11 ("PUE 1.45") |
| WUE | environment.wue_l_per_kwh | 1.20 | L/kWh | 09 line 13 ("WUE 1.20 L/kWh") |
| Grid carbon intensity | environment.carbon_kg_per_facility_kwh | 0.42 | kgCO₂/kWh | 09 line 14 (denominator = facility kWh) |
| Rack-inlet target | cooling.rack_inlet_target_c | 25.4 | °C | Project target within the ASHRAE A-class 18–27 °C recommended envelope |
| Relative humidity | environment.avg_rh_pct | 48 | %RH | Project operating basis. ASHRAE equipment guidance is class/dew-point dependent; 40–60% is not presented as a universal standard. |
| CHW supply temperature | cooling.chws_c | 19.4 | °C | Adopted warm-water conventional loop |
| CHW return temperature | cooling.chwr_c | 27.0 | °C | Adopted warm-water conventional loop |
| Chillers running / installed | cooling.chillers_running/total | 7 / 10 | — | Central water-cooled centrifugal plant |
| UPS efficiency | electrical.ups_efficiency | 0.96 | fraction | 09 line 39 ("UPS losses @ 96% efficiency 77 kW") |
| UPS modules (2N) | electrical.ups_module_count | 64 | — | Two 32-module systems; 2N topology |
| Metering tolerance | electrical.metering_tolerance_pct | ±2 | % | 12-qa-acceptance-criteria.md line 9 ("within 2%") |
| Fuel tank capacity | fuel.tank_capacity_l | 972,737 | L | Scaled current site authority; nameplate evidence class ASSUMED |
| Usable fuel fraction | fuel.usable_fraction | 0.90 | fraction | 09 line 141 ("Usable fraction = 90%") |
| Fuel level (current) | fuel.level_pct | 85 | % | 09 lines 16, 126 ("Fuel level 85%") |
| Generator consumption | fuel.generator_consumption_lph | 15,503 | L/hr | Current facility-load basis; fixed-rate simulation |
03 Calculation methodology
Each derivation below maps one-to-one to a named function in js/conv-engine.js. The function name appears in the source tag; the formula is transcribed exactly from the engine. No constant is hard-coded inside these functions — all literals come from CONV_MODEL.
Facility Load
Facility Load [kW] = IT Load × PUE
= 30,000 × 1.45 = 43,500 kW
In
facilityLoadKw(m). Total power at the utility meter, including IT + all overhead (cooling, UPS losses, lighting, BMS). PUE is the input constant; facility load is derived.CONV_CALC · facilityLoadKw · doc-09 lines 25–28
Non-IT Overhead Load
Non-IT Load [kW] = Facility Load − IT Load
= 43,500 − 30,000 = 13,500 kW
In
nonItLoadKw(m). The overhead power consumed by cooling, electrical distribution, and support systems.CONV_CALC · nonItLoadKw · doc-09 lines 32–34
UPS Losses
UPS Loss [kW] = IT Load × (1 / η − 1)
= 30,000 × (1/0.96 − 1) = 1,250 kW
In
upsLossKw(m). Input power above output at the adopted UPS efficiency η = 0.96. This loss is a component of non-IT overhead, not the complete PUE overhead.CONV_CALC v2.0.0 · upsLossKw
CHW Delta-T
ΔT [°C] = CHWR − CHWS = 27.0 − 19.4 = 7.6 °C
In
chwDeltaT(m). This is the adopted warm-water CHW temperature rise across the CRAH coil; it is distinct from the condenser-water circuit at 32/37 °C.CONV_CALC v2.0.0 · chwDeltaT
CHW reference flow
Flow [L/s] = IT Load / (4.186 × ΔT)
= 30,000 / (4.186 × 7.6) = 943.0 L/s
Evaporator duty = IT + UPS loss = 30,000 + 1,250 = 31,250 kW
Plant-duty CHW reference = 31,250 / (4.186 × 7.6) = 982.3 L/s
Condenser/tower rejection = 31,250 + 5,153.4 = 36,403.4 kW
In
chwFlowLps(m). This is the IT sensible-load reference, not a measured header flow. Current evaporator duty is 31,250 kW and its same-ΔT reference is 982.3 L/s; condenser/tower rejection is 36,403.4 kW.CONV_CALC v2.0.0 · chwFlowLps
Water Flow / WUE
Water Flow [L/min] = (WUE × IT Load) / 60
= (1.20 × 30,000) / 60 = 600 L/min
WUE check = (600 × 60) / 30,000 = 1.20 L/kWh ✓
In
waterFlowLpmForWue(m) and wueFromFlowLpm(m, lpm). This is the site-wide steady-state cooling-makeup equivalent. WUE denominator is IT energy, never facility energy or a selected Hall view.CONV_CALC v2.0.0 · waterFlowLpmForWue
Carbon Emissions
Carbon [kgCO₂/hr] = Facility Load × Grid Factor
= 43,500 × 0.42 = 18,270 kgCO₂/hr
In
carbonKgPerHr(m). Denominator is facility kWh per Rule R3 (carbon is a facility-level metric, not IT-level). Grid factor 0.42 kgCO₂/kWh sourced from doc-09 line 14.CONV_CALC · carbonKgPerHr · doc-09 lines 155–158, 167
Fuel Autonomy
Usable Fuel [L] = Tank × Usable Fraction × Level%
= 972,737 × 0.90 × 0.85 = 744,144 L (rounded)
Autonomy [hr] = Usable Fuel / Consumption = 744,144 / 15,503 = 48.0 hr
In
fuelUsableL(m) and fuelAutonomyHr(m). The current tank and consumption are scaled site assumptions; the fuel cockpit labels their nameplate evidence class separately.CONV_CALC v2.0.0 · fuelAutonomyHr
04 Constants & data sources
Every constant in CONV_MODEL carries a // source: comment referencing a review-doc file and line number. The primary authority document is conv/review/09-engineering-basis-and-calculations.md. Secondary references for individual values:
| Constant | Value | Authority | Notes |
|---|---|---|---|
| Specific heat of water | 4.186 kJ/(kg·K) | ASHRAE Fundamentals 2021 §Thermodynamics | Used in CHW flow and heat rejection calculations. |
| CHW supply / return | 19.4 / 27.0 °C | Adopted project scenario | Warm-water CHW circuit; condenser water is separately 32/37 °C. |
| UPS efficiency η | 0.96 (96%) | Project assumption; IEC 62040-3 reporting context | Gives 1,250 kW conversion loss at 30,000 kW IT. |
| Grid carbon factor | 0.42 kgCO₂/kWh | doc-09 line 14 | Facility-energy denominator per R3. Facility = IT × PUE. |
| WUE | 1.20 L/kWh | doc-09 line 13; ISO/IEC 30134-9:2022 reporting context | Project/model value with IT-energy denominator; no universal “good” grade is asserted. |
| Metering tolerance | ±2% | doc-12 line 9 ("within 2%") | Applies to EPMS total kW display; nominal = facility load without offset. |
| Generator consumption | 15,503 L/hr | CONV_CALC v2.0.0 adopted scenario | At 43.5 MW facility load. Used for the deterministic autonomy study, not a manufacturer fuel curve. |
| Tank usable fraction | 90% | doc-09 line 141; typical heel volume for above-ground diesel storage | Bottom 10% of tank volume is typically inaccessible sludge / unusable heel. |
05 KPI strip outputs
The cockpit renders the following nine KPIs from CONV_CALC.snapshot. Snapshot values are computed once at load and deep-frozen — they never change during a session.
| KPI | Snapshot path | Value at basis | Unit | Derivation function |
|---|---|---|---|---|
| IT Load | snapshot.site.it_load_kw | 30,000 | kW | Adopted current operating input |
| Facility Load | snapshot.site.facility_load_kw | 43,500 | kW | facilityLoadKw — IT × PUE |
| PUE | snapshot.site.pue | 1.45 | — | Authored constant (basis input) |
| IT sensible CHW reference | snapshot.cooling.flow_lps | 943.0 | L/s | chwFlowLps — IT/(4.186×ΔT); not measured flow |
| Evaporator duty | snapshot.cooling.heat_rejection_kw | 31,250 | kW | Historical compatibility key; IT plus UPS loss |
| Plant-duty CHW reference | snapshot.cooling.heat_rejection_kw/(4.186×ΔT) | 982.3 | L/s | Current evaporator-duty reference; not measured flow |
| Condenser/tower rejection | snapshot.cooling.tower_rejection_kw_th | 36,403.4 | kW | Evaporator duty plus chiller electrical input |
| WUE | snapshot.environment.wue_l_per_kwh | 1.20 | L/kWh | Authored constant; verified by wueFromFlowLpm |
| Water Flow | snapshot.water.flow_lpm_for_wue | 600 | L/min | waterFlowLpmForWue — (WUE×IT)/60 |
| Carbon Rate | snapshot.environment.carbon_kg_per_hr | 18,270 | kgCO₂/hr | carbonKgPerHr — Facility×GridFactor |
| Fuel Autonomy | snapshot.fuel.autonomy_hr | 48.0 | hr | fuelAutonomyHr — UsableL/ConsumptionLph |
| UPS Loss | snapshot.electrical.ups_loss_kw | 1,250 | kW | upsLossKw — IT×(1/η−1) |
EPMS cross-check:
snapshot.electrical.epms_total_kw = 43,500 kW, equal to site facility load. Hall-level EPMS reconciliation is UNAVAILABLE until a hall submeter authority exists; an equal 7,500 kW share is planning context only and must never render as a measured balance.06 Worked example — current 30 MW site
The exact scenario displayed on the live cockpit. Each number is reproduced from CONV_CALC.snapshot and must match the live page within ±0.1 (one decimal place rounding).
- Current site IT: 30,000 kW; design boundary: 40,000 kW.
- Facility Load:
30,000 × 1.45 =43,500 kW. - Non-IT Overhead:
43,500 − 30,000 =13,500 kW. - UPS Losses:
30,000 × (1/0.96 − 1) =1,250 kW. - CHW delta-T:
27.0 − 19.4 =7.6 °C. - IT sensible CHW reference:
30,000 / (4.186 × 7.6) =943.0 L/s; plant-duty reference: 982.3 L/s; actual header flow is unavailable. - Water Flow:
(1.20 × 30,000) / 60 =600 L/min. - WUE check:
(600 × 60) / 30,000 =1.20 L/kWh ✓. - Carbon rate:
43,500 × 0.42 =18,270 kgCO₂/hr. - Per hall:
30,000 / 4 =7,500 kW IT over 500 positions. - Installed-position density:
7,500 / 500 =15.0 kW/rack; design average is 20 kW/rack. - Campus rack inventory:
4 × 500 =2,000 racks. - Usable fuel study:
972,737 × 0.90 × 0.85 =744,144 L. - Fuel autonomy study:
744,144 / 15,503 =48.0 hr.
Engineering reading: PUE 1.45, WUE 1.20 L/kWh and 48-hour fuel autonomy are declared project/model values. This cockpit does not assign them a universal “good” grade or claim that a particular fuel duration proves an Uptime Tier. The owner’s risk policy, utility reliability, delivery contract, load-dependent generator curve and certification authority govern any operational target.
07 References & standards
- ASHRAE Handbook — Data Centers and Telecommunication Facilities — rack-inlet recommended dry-bulb envelope (18–27 °C for the applicable A classes) and class/dew-point-dependent humidity guidance. The project 25.4 °C target and 48% RH basis remain site decisions.
- Uptime Institute Tier Standard — topology and concurrently-maintainable intent. No certification, universal fuel-duration target or WUE grading band is inferred by this cockpit.
- ISO/IEC 30134-2 — Data centre key performance indicators — Part 2: Power Usage Effectiveness (PUE). Defines PUE = total facility power / IT equipment power.
- ISO/IEC 30134-3 — Data centre key performance indicators — Part 3: Renewable Energy Factor (REF). Referenced for carbon intensity context.
- IEC 62040-3 — UPS performance and test methods: double-conversion (VFI) typical efficiency 94–97 % at full load; basis uses 96 %.
- ASHRAE Fundamentals 2021 §Thermodynamics — Specific heat of water 4.186 kJ/(kg·K) at 10 °C (CHW regime).
- conv/review/09-engineering-basis-and-calculations.md — Primary authority for all basis constants (IT load, PUE, CHW temperatures, WUE, carbon factor, fuel system parameters).
- conv/review/00-overview-audit.md — Six data-quality rules and cross-page consistency contract for the conventional BMS suite.
- Engine source —
js/conv-engine.js(CONV_CALC v2.0.0) — deep-frozen current-operation and design-boundary snapshot. - Accuracy gates —
tools/test-conv-calc.mjs,tools/test-conv-coverage.mjs, andtools/test-conv-document-parity.mjs.
08 Assumptions & limitations
The cockpit models a four-hall, steady-state conventional data centre at the adopted load point. Hall A-D are equal operating views; central plant and site utilities remain site authorities.
| Assumption | Basis | Impact if violated |
|---|---|---|
| IT load is stable at 30,000 kW | Adopted simulated operating point; four halls at 7,500 kW each. | PUE, CHW flow, WUE, carbon and fuel assumptions must all be regenerated if the current authority changes. |
| CHW loop is 19.4/27.0 °C | Adopted warm-water CHW basis; condenser water is a distinct 32/37 °C circuit. | Changing either temperature changes ΔT and site CHW flow; all duplicate consumers must update together. |
| PUE is an adopted simulated input (not live telemetry) | The governed model adopts PUE = 1.45 as its current design-point ratio; it is separate from a marketing target and is not claimed as a measured value. Rule R4: target ≠ adopted model input ≠ field measurement. | Substituting a lower target PUE would understate modeled facility load and carbon by ~3–5 %. |
| Carbon factor is grid-average (not marginal) | 0.42 kgCO₂/kWh is an annual grid-average intensity. Marginal intensity varies hour-to-hour. | Carbon rate should be interpreted as an annual average indicator, not a real-time dispatch signal. |
| Generator consumption is fixed-rate | 15,503 L/hr at the adopted 43.5 MW facility basis. Actual manufacturer curves vary with loading and ambient conditions. | The 48-hour result is a planning scenario, not a procurement guarantee. |
| Water flow is instant-equivalent | waterFlowLpmForWue converts the annual WUE to an instantaneous flow rate. Actual cooling-tower makeup is intermittent. | Displayed as a steady-state equivalent; suitable for capacity sizing, not for variable-flow metering design. |
| UPS is transformer-free double-conversion (VFI) | η = 0.96 corresponds to modern transformer-free double-conversion UPS at full load per IEC 62040-3. | Legacy transformer-based UPS (η ≈ 0.90–0.94) would increase UPS losses by 35–75 kW and raise PUE slightly. |
Disclaimer: This cockpit and its methodology are an engineering education and pre-design reference. All values are derived from a representative simulated scenario, not from live telemetry. Final facility design, commissioning, and operational decisions must be validated by a qualified data-center engineering team against actual site measurements and authority-having-jurisdiction requirements.
09 Cockpit workflow & telemetry-point map
The console is a deterministic teaching SCADA, not a live BMS endpoint. Read it top-down: alarm strip, engineering roll-up, balance band, selected heat-map layer, equipment inspector, then event log. Every value below maps to a concrete DOM binding or CONV_CALC.snapshot field.
| Signal / control | ID or binding | Unit / state | Normal interpretation | Source binding |
|---|---|---|---|---|
| Hall state | dh-hall-state | NORMAL / WARNING / CRITICAL | Worst active alarm severity. | Deterministic alarms[] aggregation in datahall.html. |
| Hall IT load | dh-rack-load, bb-itload | MW | 7.500 MW for the selected equal-share Hall view. | CONV_CALC.snapshot.campus.halls[n].it_load_kw. |
| Rack count / average | bb-racks, bb-avg | count, kW/rack | 500 installed positions; 15.0 kW per installed position and approximately 16.2 kW per active rack. | Selected hall authority plus deterministic occupancy. |
| Power density | dh-pd | kW/rack | Same denominator as average rack load for the selected hall. | CONV_CALC.getHallSnapshot(selectedHall).it_load_kw / .racks. |
| PUE | dh-pue | ratio | 1.45 adopted simulated design-point ratio; neither live telemetry nor a target. | CONV_CALC.snapshot.site.pue. |
| Cooling margin | dh-cooling-margin, bb-cool | %, kW | Available CRAH capacity minus selected Hall IT sensible demand. | 58 required × 130 kW = 7,540 kW; 40 kW headroom. |
| Cooling N+1 | bb-np1 | PASS / FAIL | The installed standby unit replaces one failed running unit and preserves 7,540 kW available. | 59 installed; 58 required/running at the current hall load. |
| CHW plant | chws, chwr, chdt, chflow | state, °C, L/s | Central plant 7/10; 943.0 L/s is the governed IT sensible-load reference, not a measured header-flow value. | CONV_CALC.snapshot.cooling. |
| Alarm roll-up | as-crit, as-warn, as-maint | count | Rack-inlet and rack-density threshold breaches; maintenance is explicit equipment state. | alarms[] and current CRAH model. |
| Data health | as-comms, as-update, as-dq, as-scn | OK / timestamp / GOOD / Simulated | Declares freshness and provenance; does not imply a field connection. | Page clock plus static simulated-mode declaration. |
| Heat-map modes | mode-btn[data-mode] | power / temp / cooling-margin / alarms / space | One semantic layer at a time; selection is mirrored to the BMS layer toolbar. | setMode() and rackFill(). |
| Rack / CRAH inspector | sel-detail, crahPop | tagged detail | Hover/click exposes source, utilization, zone, temperatures and unit state. | Stable rack hash, zone aggregates and crahUnits[]. |
10 Spatial model, cooling balance & alarm logic
The rack field deliberately separates deterministic basis data from the temporary excursion simulator. Reloading produces the same occupancy, rack loads and normal cold-aisle temperatures; only an explicit excursion may introduce bounded short-lived variation.
rackTotal = Σ occupiedRackWeight × (7,500 kW / Σ occupiedRackWeight) = 7,500 kW per hall
installedPositionAverage = 7,500 / 500 = 15.0 kW/rack
currentActiveAverage ≈ 7,500 / 463 = 16.2 kW/active rack
coolingAvailable = 58 × 130 = 7,540 kW
coolingHeadroom = 7,540 − 7,500 = 40 kW
N+1 replacement = 59 installed − 1 failed = 58 available → 7,540 kW ≥ 7,500 kW → PASS
The fixed integer hash chooses roughly 8% spare cells and bounded occupied weights, then a normalization and residual correction force the exact EPMS reconciliation total. datahall.html rack model · CONV_CALC.snapshot
| Condition | Normal | Warning | Critical / alarm | Visual meaning |
|---|---|---|---|---|
| Rack utilization | <70% | 70–85%; 85–95% elevated tiers | >95% rating | Green → amber/orange → fault red; spare is muted. |
| Rack-inlet temperature | 18–27 °C recommended envelope; project target 25.4 °C | Outside the recommended envelope | >30 °C project critical threshold | 18–27 °C remains green; below 18 °C is amber, 27–30 °C is elevated, and above 30 °C is red. |
| CRAH availability | 58 required/running + 1 available standby | Loss of standby redundancy | Available capacity below 7,500 kW | N+1 chip reports the capacity comparison, not a decorative state. |
| Communications / quality | OK / GOOD | Stale or degraded when implemented | Unavailable when implemented | Current page is simulated and does not poll a live transport. |
Excursion boundary:
fireExcursion() may use bounded variation only for an explicitly simulated short training excursion. It does not alter the governed 30,000 kW site IT, 7,500 kW selected-hall IT, PUE, plant authority, or normal deterministic layout. Evidence must distinguish this training transient from measured telemetry.11 Glossary, operating sequence & limitations
CRAH, PUE, WUE, N+1 and EPMS terms are exposed to RZExplain. These definitions describe the cockpit model and must not be mistaken for site commissioning set-points.
| Term | Meaning in this cockpit | Operator action |
|---|---|---|
| EPMS reconciliation | Rack-field sum compared with UPS-output IT kW from the same frozen engine snapshot. | Investigate any non-zero balance before trusting downstream density or cooling calculations. |
| Rack inlet / cold aisle | Rack inlet-side air zone; project target is 25.4 °C inside the 18–27 °C recommended A-class envelope. | Select Rack-Inlet Temp, locate the hottest zone, then inspect assigned CRAHs and rack load. |
| Cooling margin | Running sensible cooling capacity minus modelled hall IT heat. | Confirm N+1 remains positive; margin alone is not proof of correct airflow distribution. |
| Standby / maintenance | At least one available unit supports the current 58-unit requirement. | Verify the standby can auto-start before removing a running unit. |
| Simulated data | Local deterministic model plus explicit training excursions; no BMS/SCADA transport. | Use for education and pre-design review only; validate against field sensors and approved sequences. |
- Read the alarm strip and data-quality/scenario chips before interpreting numbers.
- Confirm rack-field total balances to EPMS IT load and cooling N+1 passes.
- Select one semantic mode, locate the affected zone, then inspect its rack or CRAH.
- Use the event log as a training trace; use an actual alarm historian for operational diagnosis.