CEA cheat sheet

Audit levels per ASHRAE/CEA convention: Level 1 = walk-through with utility bill benchmarking (rough payback ranges); Level 2 = energy survey with end-use breakdown, energy balance, and costed ECMs; Level 3 = detailed/focused analysis including calibrated simulation and hourly modeling for major capital measures.

Scope-setting inputs: audit objective (compliance, capital planning, ISO 50001 baseline), facility boundary, baseline period selection (12 consecutive months minimum, weather-normalized), and required accuracy of savings estimates (typically ±10% for Level 3).

Baseline period rule: choose the most recent full year with representative occupancy and no major process changes; document adjustments for square footage changes, shutdowns, fuel switching, and degree-day anomalies.

Sampling: 90/10 rule (measurement covers ≥90% of energy use, with ≥10% of the remaining use sampled) is a common screening criterion for Level 2 end-use disaggregation.

Deliverable structure: executive summary, facility description, baseline energy and cost, end-use breakdown, ECM list with capital cost, savings, simple payback, and implementation priority ranking.

Instrumentation plan: power logger (kW and power factor), ultrasonic or clamp-on flow meter, portable temperature/RH loggers, combustion analyzer, light meter, and infrared camera - calibrate and record serial numbers and calibration dates.

Measurement and verification planning: specify IPMVP Option (A/B/C/D) at the planning stage because it determines the metering and baseline data that must be collected during the audit.

Safety and access constraints: arc-flash boundaries, confined space entry for air handlers, hot-work permits, and building owner approval for data logging and equipment shutdowns.

Energy balance: total purchased energy = end-use sum + losses. Convert all fuels to a common unit (kBtu, GJ, or kWh) using site or source energy factors; 1 kWh = 3,412 Btu; 1 therm = 100,000 Btu.

Degree-day method: annual heating energy ≈ (UA) × HDD × 24 / efficiency. Balance-point temperature (typically 55–65°F) must be used rather than 65°F for internal-gain-dominated buildings.

Weather normalization: regress monthly energy against HDD/CDD; the slope gives weather sensitivity (kBtu/degree-day) and the intercept gives weather-independent baseload. High R² (>0.75) indicates a reliable model.

Load factor and load duration curves: plot hourly load sorted descending to identify peak demand charges versus baseload; a low load factor (annual kWh / (peak kW × 8,760)) signals demand-management opportunity.

End-use disaggregation methods: submetering, temporary logging, nameplate with run-hour estimates, and calibrated simulation; cross-check totals against utility bills within 5–10%.

Energy intensity benchmarking: EUI = annual site energy (kBtu) / conditioned floor area (ft²); compare to CBECS/ENERGY STAR Portfolio Manager medians for the same building type and climate zone.

Peak demand analysis: identify coincidence of loads (chiller + AHU + process) and check utility ratchet clauses (demand ratchets can lock in 12-month peak for billing).

Loss accounting: envelope conduction, infiltration (ACH × volume × 1.08 × ΔT for sensible Btu/h), duct leakage, steam trap losses, and standby/idle equipment loads.

Utility bill data: collect 24–36 months of electric, gas, oil, and water bills including demand (kW), power factor, ratchets, and riders; verify meter multipliers and estimated vs. actual reads.

Spot vs. interval data: spot readings (5–15 min) for steady loads; 15-minute interval data for load profiles, duty cycling, and demand analysis. Logging duration should cover at least one full operating cycle (typically 1–2 weeks).

Power measurement: P (W) = V × I × PF for single phase; three-phase P = √3 × V × I × PF. Measure true RMS with a power meter, not a clamp meter alone, for non-linear loads.

Flow measurement: ultrasonic transit-time for closed-loop water; pitot traverse for duct air (velocity pressure → velocity: V = 1,096 × √VP with VP in in. w.c. and V in ft/min).

Airflow and psychrometrics: cfm = Btu/h / (1.08 × ΔT) for sensible; total heat = 4.5 × cfm × Δh. Use measured wet-bulb/dry-bulb to compute enthalpy.

Lighting measurement: illuminance in foot-candles with a calibrated meter at the work plane; spot-read several points and average; compare to IES recommended levels.

Data quality checks: instrument accuracy vs. required precision, sample rate vs. Nyquist concerns for cycling loads, missing intervals, and time-synchronization between meters and BAS trends.

Statistical analysis: normalize by production or occupancy, use regression to separate weather and non-weather effects, and report uncertainty ranges rather than single-point savings.

Simple payback = initial cost / annual savings. Acceptable thresholds commonly 1–3 years for operational measures and 5–10 years for capital measures; payback ignores time value and measure life.

Life-cycle cost (LCC) = present value of capital + PV of energy, maintenance, and replacement costs − PV of salvage. Use PV = FV / (1 + d)^n and the uniform series factor for annual costs.

Time value formulas: present worth factor for a recurring annual amount over n years at discount rate d: PWF = [1 − (1 + d)^−n] / d. Capital recovery factor = d / [1 − (1 + d)^−n].

Internal rate of return (IRR) is the discount rate at which NPV = 0; accept when IRR exceeds the owner's hurdle rate. Simple payback of n years corresponds roughly to IRR above 1/n.

Levelized cost of conserved energy (CCE) = annualized capital cost / annual energy saved; compare CCE to the marginal cost of purchased energy to rank measures independent of fuel type.

Utility rate structures: demand charges ($/kW), energy blocks, time-of-use periods, ratchets, power factor penalties, and riders; savings must be computed on the marginal rate, not the average rate.

Incentives and tax treatment: rebates reduce first cost, tax credits/deductions reduce after-tax cost, and depreciation schedules affect cash flow; model both with and without incentives.

Uncertainty and sensitivity: run NPV at ±20% on savings and ±2 points on discount rate; report measure life, escalation rate for energy prices (typically 2–4%/yr), and non-energy benefits separately.

Luminous efficacy (lm/W): incandescent ~15, halogen ~20, T8 fluorescent ~90–100 (system), T5 ~95–105, LED ~100–150+; compare system efficacy including ballast/driver losses.

Lumen method: maintained illuminance E = (N × lamps × lumens × CU × LLD) / area; CU = coefficient of utilization, LLD = light loss factor (lamp lumen depreciation × dirt depreciation).

Lighting power density (LPD): W/ft² = total connected lighting watts / floor area; compare to ASHRAE 90.1 or IECC allowances for the space type; the lower of space-by-space or whole-building method governs.

Controls savings: occupancy sensors 20–30% in intermittent spaces; daylight dimming 20–60% near windows; scheduled/bi-level switching 10–30%; tuneable LED systems allow additional task tuning savings.

Ballast and driver factors: fluorescent ballast factor (BF) typically 0.85–1.10; input wattage = lamp wattage × number of lamps × BF / ballast efficiency. LED driver losses typically 5–10%.

Retrofit decision rules: relamp/reflector retrofits only when fixture condition is good and labor is low; full fixture replacement when lamp life, color quality, or controls integration matters; T12 magnetic → T8 electronic is a standard high-payback measure.

Color metrics: CRI ≥ 80 for general spaces, ≥ 90 for retail/art; CCT 3000K–4000K typical; check flicker (percent flicker) and glare (UGR) for occupant acceptance.

Daylighting and controls commissioning: photocell setpoints, time delays, and dimming curves must be commissioned; measure actual savings with post-retrofit logging for M&V.

Cooling capacity: 1 ton = 12,000 Btu/h = 3.517 kW. Chiller efficiency in kW/ton; COP = 3.517 / (kW/ton); EER = Btu/h per watt; SEER/IPLV are seasonal metrics.

Sensible and latent load: total heat = 4.5 × cfm × Δh; sensible = 1.08 × cfm × ΔT; latent = 0.68 × cfm × Δgrains (or 4.5 × cfm × ΔW in lb/lb).

Chiller plant optimization: condenser water reset (lower approach), chilled water supply temperature reset, variable primary flow, and sequencing to run chillers near their best efficiency point (typically 60–90% load).

Boiler efficiency: combustion efficiency from flue O2/CO2 and stack temperature; typical 80% (conventional) to 90–95% (condensing, requires return water below ~130°F). 1 boiler hp = 33,475 Btu/h.

Air-side measures: economizer (dry-bulb or enthalpy), VAV supply air temperature reset, static pressure reset, demand-controlled ventilation (CO2-based), and duct sealing/insulation.

Fan and pump laws: flow varies linearly with speed, pressure with the square, power with the cube. VFDs on variable-torque loads yield large savings at part load; affinity laws are the primary justification.

Equipment part-load: check staging, minimum turndown, short cycling, and simultaneous heating/cooling; simultaneous operation is a common fault found in BAS trend review.

Refrigerant and maintenance: refrigerant charge, economizer damper linkage, coil fouling, belt tension, and filter pressure drop all affect actual versus rated efficiency.

Hot water energy: Q = 8.33 lb/gal × gallons × ΔT (°F) = Btu; annual energy = gallons/day × 365 × 8.33 × ΔT / (system efficiency × 100,000) in therms.

Storage and standby losses: tank standby loss typically 0.5–2%/hr of stored energy; R-value of tank insulation and pipe insulation are high-payback, low-cost measures.

Distribution losses: recirculation systems lose heat continuously; controls such as demand recirculation, timers, and thermostatic balancing valves reduce losses; insulate all hot water lines and the first 3–6 ft of cold lines at the tank.

Fixtures: low-flow showerheads (2.5 → 1.5 gpm), aerators (2.2 → 0.5 gpm), and faucet controls; each gpm reduction saves both water and the energy to heat it.

System types: storage tank, tankless/instantaneous, heat pump water heater (COP 2–3, needs conditioned air and condensate drainage), solar thermal (needs backup), and condensing gas (efficiency > 90% with low return temperature).

Temperature setpoints: 120°F storage for scald safety, 140°F with mixing valve where Legionella control is required; Legionella risk rises in lukewarm 68–122°F ranges and in oversized/dead-leg piping.

Boiler-connected DHW: indirect tanks and tankless coils suffer from boiler short cycling in summer; check whether a separate high-efficiency water heater is more economic than summer boiler operation.

M&V: measure hot water flow and inlet/outlet temperatures; for savings verification, use run-time loggers on recirculation pumps and pre/post fuel bills weather-normalized for boiler-connected systems.

Motor input power: kW = hp × 0.746 / efficiency; motor efficiency varies with size and load - NEMA Premium motors typically 90–96% for 5–200 hp; rewinding vs. replacement decision considers efficiency loss and cost.

Affinity laws: for fans and centrifugal pumps, flow ∝ speed, head ∝ speed², power ∝ speed³. Throttling or bypass control wastes energy; VFD control is preferred for variable loads.

Motor loading: measure with a power meter; a lightly loaded motor (<50%) runs at poor power factor and efficiency; consider right-sizing or replacing with a smaller premium motor.

Compressed air is the most expensive utility (often $0.02–0.05 per 1,000 scfm·min in energy); typical system efficiency is 10–15% from input power to useful work at the tool.

Compressed air rules: 1 hp ≈ 4–5 cfm at 100 psig; each 2 psi of pressure reduction saves ~1% of compressor energy; leak loads often 20–30% of compressor output.

Compressor controls: load/unload, modulation, and VSD. VSD compressors are efficient at part load but should not run below ~30% capacity; sequencing multiple trim compressors is often better than one large unit.

Leak detection and management: ultrasonic leak surveys, tagging and repair, and pressure/flow monitoring during non-production hours to quantify leak load (cfm = leak flow at no-load run time).

Air system measures: reduce pressure setpoint, use storage receivers, eliminate inappropriate uses (open blowing, vacuum generation), heat recovery from compressors, and dryers with dew-point control.

Heat transfer: Q = U × A × ΔT; U-value = 1/R. Overall envelope UA is the sum of opaque, window, and door areas times their U-values; infiltration adds 1.08 × cfm × ΔT.

Insulation metrics: R-value in ft²·°F·h/Btu; typical roof R-20 to R-40, walls R-13 to R-20, and windows U-0.30 to U-0.50 for efficient assemblies; thermal bridging reduces effective R-value.

Infiltration: blower door test measures cfm50; natural infiltration ≈ cfm50 / N (N typically 10–20). ASHRAE 62.1 ventilation requirements are separate from uncontrolled leakage.

Windows: low-e coatings, argon fill, and thermally broken frames; SHGC controls solar gain (lower SHGC in cooling-dominated climates, higher in heating-dominated for passive solar).

Roof measures: cool roofs (high solar reflectance and emittance) reduce cooling load; roof insulation and radiant barriers are common retrofits; check structural capacity before adding insulation.

Moisture and durability: vapor retarders on the warm side, air sealing before insulation, and dew-point analysis prevent condensation; envelope changes can alter pressure relationships and combustion air supply.

Envelope economics: envelope measures are long-lived but capital-intensive; they are most cost-effective when bundled with roof replacement or major renovation, and when air sealing is done before insulation.

Commissioning checks: thermal imaging for missing insulation and air leakage, door/window weatherstripping, and verification that envelope changes do not degrade ventilation rates below code.

Definitions: EMCS = energy management and control system; BAS = building automation system (HVAC, lighting, access); PAS = process automation system (industrial process control). Integration via BACnet, Modbus, or LonWorks gateways.

Control strategies: scheduling/optimal start-stop, night setback, economizer control, static pressure reset, supply air temperature reset, chiller/boiler sequencing, and demand limiting.

Sensors and calibration: temperature sensors ±1°F, RH ±5%, CO2 ±50 ppm; calibration drift is a leading cause of wasted energy; verify with calibrated reference instruments.

Trend data: log 15-minute interval trends for major equipment (kW, temperatures, valve/damper positions, run status) for at least 2 weeks; trend review is the highest-value BAS audit activity.

Fault detection: simultaneous heating/cooling, valves stuck open, schedules overridden, sensors failed high/low, and economizer dampers stuck; these often account for 5–15% of HVAC energy.

Demand limiting and load shedding: BAS can shed non-critical loads at a kW setpoint; verify that shed loads actually reduce demand and that setpoints recover without occupant complaints.

Cybersecurity and network: separate BAS network from IT, use managed switches and VLANs, change default passwords, and document points list and sequence of operations.

M&V with BAS: use BAS trends as the primary data source for IPMVP Option B; export data to spreadsheets and normalize with degree days for savings verification.

Solar PV: capacity factor typically 15–25%; annual kWh ≈ kW × 1,400–1,800 kWh/kW in good US locations; performance ratio accounts for inverter, soiling, and temperature losses.

PV economics: installed cost $2–4/W commercial; incentives (ITC, state rebates, SRECs) and net metering materially change payback; model with and without net metering.

Solar thermal: flat-plate or evacuated tube for DHW; annual output ~50–70% of incident solar; requires freeze protection and backup heating; SRCC ratings give collector efficiency curves.

Wind: power in wind = 0.5 × ρ × A × v³; capacity factor 25–40% for good sites; requires wind resource assessment (atlas data, met tower) and utility interconnection approval.

Combined heat and power (CHP): overall efficiency 70–85% when heat is recovered; sized to thermal baseload for best economics; check utility standby and exit fees.

Battery storage: round-trip efficiency 85–95% (lithium-ion); sizing considers peak shaving, demand charge reduction, backup duration, and degradation; levelized cost of storage includes cycle life.

Grid interaction: net metering, demand response, and time-of-use arbitrage; verify interconnection rules, inverter settings (IEEE 1547), and whether storage can discharge to the grid.

Screening: compare levelized cost of energy (LCOE) of the alternative to the marginal retail rate; LCOE = (capital + PV of O&M) / PV of lifetime kWh.

Fleet energy: fuel consumption (gal or kWh) per mile or per ton-mile; compare to baseline by vehicle class; idling can consume 0.5–1.0 gal/hr for heavy trucks.

Vehicle efficiency measures: route optimization, driver training (5–15% fuel savings), tire pressure and low-rolling-resistance tires, aerodynamic fairings, and weight reduction.

Electrification: EV efficiency ~0.3 kWh/mile for light duty; charging infrastructure sizing (Level 2 ~7 kW, DC fast 50–350 kW) and demand charges are key cost drivers.

Alternative fuels: CNG, LNG, propane, biodiesel, and hydrogen; compare on energy content (diesel 128,000 Btu/gal, gasoline 114,000 Btu/gal, CNG ~38,000 Btu per 100 scf) and infrastructure cost.

Modal shift and logistics: consolidate shipments, shift to rail or intermodal where feasible, and use backhauling; these often yield larger savings than vehicle technology alone.

Telematics and data: GPS/telematics provides idle time, speeding, and route data; use it to build a baseline and verify savings from driver behavior programs.

Electrification economics: total cost of ownership includes fuel, maintenance (EVs lower), infrastructure, and residual value; compare per-mile cost, not sticker price.

Emissions accounting: convert fuel to CO2 using EPA factors (e.g., gasoline 8.89 kg CO2/gal, diesel 10.18 kg CO2/gal); report scope 1 (owned fleet) and scope 3 (contracted transport) separately.