Lighting energy audit: advanced diagnostic methodology

Contrary to the dominant reflex in the profession, which prioritises LED relamping as the first response, a photometric audit alone produces inaccurate investment plans. The hierarchy observed across our audits reverses the conventional diagnostic priorities: usage behaviours are the leading source of the gap between theoretical and actual consumption, ahead of control system failures, with the intrinsic performance of luminaires coming only last. This breakdown calls for an integrated methodology rather than an isolated photometric audit.

A high energy class LED luminaire controlled by a manual switch can consume more than a T5 fluorescent unit regulated by presence detection and a luminosity sensor. Mapping the interfaces between equipment, control and usage becomes the core of the diagnostic.

Three concurrent lines of analysis for the lighting designer:

• Technical: photometric survey in accordance with NF EN 12464-1 (500 lux maintained in office areas, UGR less than or equal to 19), inventory of sources, measurement of power consumption circuit by circuit. This is the line natively mastered by the interior lighting architect, but it is not enough.
• Behavioural: observation of occupancy cycles, interviews with occupants, cross-referencing with the commonly used commercial occupancy ratios (7 to 12 m² per workstation in open space, 12 to 18 m² in enclosed offices). This is the dimension missing from competitions and conventional specifications.
• Control systems: audit of presence detectors, luminosity sensors, time programmers, DALI or KNX scenarios. Often split between electrical and ELV (extra-low voltage) lots, with no coordination.

  The absence of any single one of these lines invalidates the reliability of the quantified recommendations presented to the investment committee.

Three recurring pitfalls in adopted external audits

The diagnostics provided by clients reveal recurring blind spots that compromise the reliability of the reduction plans presented to the energy committee.

1. Measurements taken only during normal periods. The lack of readings during off-peak hours, weekends and closure periods masks lighting running continuously due to programming faults. This parasitic consumption can represent a significant share of the annual bill in buildings audited without complete temporal monitoring.
2. Omission of safety and signage lighting. Self-contained emergency lighting units (BAES), illuminated signage and wayfinding fixtures frequently remain oversized relative to the requirements of the Labour Code (articles R4227-13 et seq.). Their cumulative consumption can represent a non-negligible share of the lighting category, often underestimated due to the absence of readings on a dedicated circuit.
3. Unverified automation settings. Detectors with a time delay set to 30 minutes instead of 5, deactivated luminosity sensors, DALI scenarios locked at maximum values: these misconfigurations generate overconsumption that is invisible in an equipment inventory.

Best practice consists of taking readings over complete cycles of at least 15 days, with an exhaustive audit of automation systems and documented readjustment. This approach makes it possible to release significant savings with no equipment replacement at all.

When an in-depth lighting energy audit is NOT justified. Below 300 m² of usable area or with an annual lighting bill below €4,000, the ROI of a full assignment (15 days of measurements, 4 phases) exceeds 6 years: a targeted visual audit and a fixed-price readjustment are sufficient. Likewise, for a building delivered less than 3 years ago with a BMS calibrated at handover, the residual potential is generally low and does not justify deploying a 15-day protocol. Finally, for premises with continuous 24/7 occupancy (data centres, control rooms), priority goes to direct LED relamping rather than automation optimisation, whose leverage becomes marginal.

For the design architect: integrate the audit upstream of the project, not as after-sales support

The profession’s received wisdom treats the energy audit as a post-handover assignment, triggered by a regulatory obligation or a CSR committee. Our reading differs: the audit is part of the architectural brief. Across all the sites audited, the largest overconsumption gaps consistently concern buildings where lighting design was dissociated from automation design, with an ELV lot delivered after handover of the lighting lot.

For the design architect or lighting designer, there are three operational consequences:

• Photometric calculation in software (DIALux, Relux) projects a theoretical consumption far below the actual figure. Announcing in a competition a ratio of 4 W/m² calculated at 100% simultaneous occupancy is technically honest and functionally false. The actual ratio measured on our projects generally stands between 4.5 and 7 W/m² installed, the gap with the theoretical calculation being explained by real usage and the quality of the control system.
• Specifying the DALI or KNX protocol in the technical specifications must include the configuration scenarios, not just the equipment. A correctly specified detector delivered with a factory time delay of 30 minutes cancels out most of the expected gain.
• The architectural integration of the sensor (probe, detector) determines its relevance. A luminosity sensor recessed in the wrong place on the ceiling, hidden by an acoustic bulkhead or oriented towards a non-representative area, transmits false data to the system. This coordination is a matter of dialogue between architect, lighting designer and ELV, not the electrical lot alone.

The lighting energy audit methodology thus becomes a design tool: the feedback from the audits carried out informs the programming choices for new-build and refurbishment projects.