Illumination engineering and lighting design

In short

Illumination engineering is the quantitative design of lighting - choosing lamps, luminaires, mounting heights and spacing so that a space gets the required illuminance with acceptable uniformity, colour quality, glare control and energy use. It rests on four photometric quantities (luminous flux, luminous intensity, illuminance, luminance), two laws of illumination, and design methods such as the lumen (or watt-per-square-metre) method. In EEE courses it sits in "Utilisation of Electrical Energy", carries steady marks in university exams and appears in state-utility and PSU recruitment papers (SSC JE, RRB JE, state DISCOM AE); it is outside the current GATE EE syllabus. On the job it is used by electrical consultants, EPC contractors, municipal street-lighting projects (EESL's SLNP), industrial plant electrical teams and luminaire makers such as Bajaj Electricals, Havells, Wipro Lighting, Syska and Signify India.

Explained

Photometric quantities and units

• Luminous flux (Φ), unit lumen (lm): total light energy radiated per second, weighted by the eye's response curve V(λ). The eye peaks at 555 nm (green), where 1 W of radiation = 683 lm - that is the absolute ceiling on efficacy.
• Luminous intensity (I), unit candela (cd): flux per unit solid angle in a given direction, I = Φ/ω. The candela is an SI base unit (defined from a 540 THz source of radiant intensity 1/683 W/sr).
• Solid angle (ω), unit steradian (sr): ω = area/radius². A full sphere = 4π sr, a hemisphere = 2π sr, a cone of half-angle α subtends ω = 2π(1 − cos α).
• Illuminance (E), unit lux (lx) = lm/m²: flux falling per unit area of a surface. This is what a lux meter reads and what codes specify. 1 phot = 10⁴ lux; 1 foot-candle ≈ 10.76 lux.
• Luminance (L), unit cd/m² (nit): intensity per unit projected area of a surface or lamp - what the eye actually perceives as brightness; glare and road-lighting quality are judged on luminance.
• Luminous efficacy, lm/W: flux per watt of input power. For luminaires, luminaire efficacy (LER) includes optical and driver losses.
• MSCP (mean spherical candle power) = Φ/4π; MHCP (mean hemispherical) = flux in one hemisphere/2π; MHCP (downward) used for ceiling fittings. Reduction factor = MSCP/MHCP.
• Colour quantities: correlated colour temperature (CCT) in kelvin - warm white 2,700-3,000 K, neutral 4,000 K, cool daylight 5,700-6,500 K; colour rendering index (CRI, Ra) out of 100 - above 80 for offices and shops, above 90 for colour-critical work.
• Glare: *disability glare* reduces visibility, *discomfort glare* annoys. Interior glare is limited by the Unified Glare Rating (UGR) - typically ≤ 19 for offices and classrooms, ≤ 22 for general industry, ≤ 16 for drawing/CAD work.

Laws of illumination and photometric measurement

• Inverse square law: E = I/d². Flux spreads over an area growing as d², so illuminance falls as 1/d². Valid when the source is small compared with d (a "point source" - practically d > 5 × largest source dimension).
• Lambert's cosine law: when the surface is tilted by θ to the direction of light, E = (I/d²)·cos θ, because the flux is spread over A/cos θ.
• Cos-cube form: for a horizontal plane under a lamp at height h, with θ measured from the vertical, d = h/cos θ, so E = I cos³θ/h² - the standard exam form.
• Superposition: illuminance from several sources simply adds, each computed separately (lux is scalar at a point on a given plane).
• Polar curve / distribution curve: intensity plotted against angle in the vertical (and horizontal) plane; symmetrical for a bare lamp, asymmetric for street-light optics. Rousseau's construction converts a polar curve into MSCP by computing the mean ordinate of an equivalent rectilinear diagram.
• Measurement: integrating (Ulbricht) sphere measures total flux of a lamp in one reading; photometer bench with a Lummer-Brodhun head or Bunsen grease-spot photometer compares two sources (I₁/I₂ = d₁²/d₂²); goniophotometer plots distribution; digital lux meter (cosine- and colour-corrected photodiode) for field checks; luminance meter for roads and screens.
• Lab experiments typically examined: verification of the inverse square and cosine laws, polar curve of a luminaire, measurement of lamp efficacy, lux survey of a room/road.

Electric lamps - operation, types and comparison

• Incandescent (GLS): tungsten filament at about 2,700 K in vacuum or inert gas (argon-nitrogen); coiled-coil filament reduces heat loss. Very poor efficacy, unity power factor, CRI 100, instant and fully dimmable. Commercially obsolete in India.
• Tungsten halogen: halogen regenerative cycle redeposits tungsten on the filament, allowing higher temperature and longer life in a small quartz envelope; used for studio, display and automotive lighting.
• Fluorescent lamp (FTL, T12/T8/T5): low-pressure mercury discharge emits UV at 253.7 nm; phosphor coating converts it to visible light. Needs a starter (glow-switch) to preheat cathodes and a ballast/choke to strike the arc and limit current (a discharge has negative resistance). CFL folds the same tube with integral gear.
• High-pressure mercury vapour (HPMV): bluish, poor CRI, 5-7 min warm-up; largely withdrawn. Mercury-containing lamps are being phased down worldwide under the Minamata Convention (India is a party); 2023 amendments set phase-out dates for CFLs and linear fluorescent lamps around 2025-2027 - national notification dates vary, so check the current rules.
• Metal halide (MH): mercury discharge with metal-halide additives - good efficacy with CRI 65-90; used in sports and high-bay lighting.
• Sodium vapour: LPSV is monochromatic yellow (589 nm) with the highest efficacy but no colour rendering; HPSV is golden-white with good efficacy and long life - the former workhorse of Indian streets.
• LED: electroluminescence in a forward-biased p-n junction of a direct-bandgap semiconductor (InGaN for blue/green, AlInGaP for red); photon wavelength λ(nm) ≈ 1240/Eg(eV). White light is made by a blue LED + YAG phosphor or by RGB mixing. Life is quoted as L70/L80 (hours to 70%/80% of initial lumens), and junction temperature governs both output and life - hence the heat sink.

Control gear, power quality and switching

• Electromagnetic (copper/aluminium) choke: cheap and robust, but 8-12 W loss on a 36 W tube and a poor lagging power factor (~0.5) corrected by a shunt capacitor (typically 3.2-4 µF for a 40 W lamp) to about 0.95.
• Electronic ballast (HF, 25-40 kHz): lower loss, no flicker or hum, instant start, higher lamp efficacy; must meet harmonic limits.
• LED driver: a constant-current SMPS (buck/flyback) with power-factor correction; typical requirements for retail LED lamps are PF ≥ 0.9 above 5 W (BIS/BEE specification - check the current version) and THD limits.
• Stroboscopic (flicker) effect: a discharge lamp on 50 Hz AC extinguishes 100 times per second, so rotating machinery can appear stationary or slow - a real safety hazard in workshops. Remedies: feed adjacent fittings from different phases, use lead-lag (twin-lamp) circuits, or use high-frequency electronic ballasts / LED drivers.
• Controls as a design element: occupancy/PIR sensors, daylight-linked dimming (photocell), timers and astronomical clocks, 0-10 V / DALI / wireless dimming, and group switching of alternate rows parallel to windows.
• Lighting circuits are wired per IS 732 (code of practice for electrical wiring installations): usually 6 A, 10 A final circuits with their own MCB and RCBO where required, maximum points per circuit as per installation practice.

Lighting schemes and interior design (lumen method)

• Direct (>90% flux downward): high utilisation, hard shadows, risk of glare - factories, warehouses.
• Semi-direct (60-90% down): some upward flux lightens the ceiling - workshops, offices.
• General diffusing / direct-indirect (~40-60% each way): opal globes, pleasant and shadow-free - showrooms, halls.
• Semi-indirect (60-90% up) and indirect (>90% up): soft, glare-free, low utilisation and costly - hotels, conference rooms, operation-theatre surrounds.
• Local / task lighting supplements general lighting at the work point; accent and decorative lighting are architectural.
• Lumen-method steps: (1) fix required illuminance E from IS 3646 Part 1 / NBC 2016 for the task; (2) choose lamp type, CCT, CRI and luminaire; (3) find utilisation factor UF (also called coefficient of utilisation) from the manufacturer's table using room index RI and surface reflectances (ceiling/wall/floor, e.g. 0.7/0.5/0.2); (4) choose maintenance factor MF (0.8 clean air-conditioned, 0.7 normal, 0.6 dirty industry); (5) compute total flux and number of luminaires; (6) lay them out so that the space-to-height ratio (SHR) is within limits (about 1:1 to 1.5:1 for direct fittings, wall spacing half the interior spacing); (7) check uniformity (E_min/E_avg ≥ 0.4 generally, ≥ 0.7 over the task area), UGR and lighting power density (LPD).
• MF is itself the product of lamp lumen maintenance, lamp survival, luminaire maintenance and room-surface maintenance factors; a cleaning schedule is part of the design.
• Daylighting: daylight factor = (indoor illuminance / simultaneous outdoor horizontal illuminance) × 100%; NBC 2016 Part 8 Section 1 covers daylight and ventilation, and Eco-Niwas Samhita sets window-to-wall and visible-light-transmittance provisions for residences.

Street, flood and exterior lighting

• Road lighting is designed on average luminance and uniformity (and illuminance in Indian practice per IS 1944, Parts 1-6, with road classes from important traffic routes down to local streets; typical averages range from about 30 lux on major city roads to 5-10 lux on minor roads - confirm against the current part and the client's specification).
• Pole arrangements: *single-side* when road width ≤ mounting height; *staggered* when width ≤ 1.5 × mounting height; *opposite (both sides)* when width > 1.5 × mounting height; *central/twin-arm* on dual carriageways with a median; *high-mast* for junctions and squares.
• Mounting height 8-12 m for main roads, overhang and 5-15° tilt set the beam; spacing-to-height ratio normally ≤ 3-4 for acceptable uniformity; UF for street luminaires is low (0.25-0.40) because much flux misses the carriageway.
• Flood lighting: narrow/medium/wide-beam projectors on masts, aimed with a beam factor (fraction of flux inside the useful beam) and a waste light factor (1.2 for regular areas, up to 1.5 for irregular ones). Sports lighting additionally needs vertical illuminance, CRI > 80 and flicker-free output for TV broadcast.
• Environment: light-trespass and sky-glow control through full-cut-off optics and BUG/upward-light-ratio limits; IP (IS/IEC 60529) and IK ratings for outdoor luminaires - typically IP65/IP66 and IK08 or better; surge protection of 10 kV class on street-light drivers.
• Energy and policy: ECSBC 2024 (BEE; successor to ECBC 2017; States adopt it in their own versions) caps LPD by space type and mandates controls; EESL's UJALA programme distributed LED bulbs at subsidised prices and SLNP has replaced over a crore of conventional street lights with LEDs; BEE star labelling covers LED lamps; SP 72 - National Lighting Code consolidates Indian lighting practice. Statutory backing comes from the Electricity Act 2003, CEA (Measures relating to Safety and Electric Supply) Regulations and the Energy Conservation Act 2001 as amended.

Key points

• Luminous flux - total visible light output in lumens; luminous intensity - lumens per steradian (candela).
• Illuminance - lux falling on a surface (designed and measured); luminance - cd/m², the perceived brightness.
• Inverse square law - E = I/d², valid only for point-like sources.
• Lambert's cosine law - E = I cos θ/d², giving E = I cos³θ/h² on a horizontal plane.
• MSCP - Φ/4π for a uniform source; reduction factor = MSCP/MHCP.
• Luminous efficacy - lm/W; maximum physically possible is 683 lm/W at 555 nm.
• Lumen method - N = E·A/(Φ·UF·MF); UF from room index and reflectances, MF from dirt and lumen depreciation.
• Room index - RI = L·W/((L+W)·Hm), with Hm measured above the working plane (usually 0.8 m).
• Space-to-height ratio - 1:1 to 1.5:1 for direct fittings; wall spacing = half the interior spacing.
• Uniformity - E_min/E_avg ≥ 0.4 for a space, ≥ 0.7 over the task.
• UGR - glare index; ≤ 19 for offices and classrooms, ≤ 16 for CAD/drawing work.
• Ballast - limits current in a negative-resistance discharge; electronic HF types cut losses and flicker.
• Stroboscopic effect - fixed by phase distribution, lead-lag circuits or HF/LED gear.
• Street lighting - single-side/staggered/opposite chosen by road width versus mounting height; S/H ≤ 3-4.
• Flood lighting - projector count uses beam factor and waste light factor besides UF and MF.
• L70 life - hours until an LED falls to 70% of initial lumens; governed by junction temperature.
• Indian codes - IS 3646 (interior), IS 1944 (public thoroughfares), SP 72 (National Lighting Code), NBC 2016 Part 8, ECSBC 2024 LPD limits, IS 732 wiring.
• LPD - W/m² of connected lighting load; the single number by which an energy audit judges a lighting design.

Formulas and examples

Example 1 - Laws of illumination. A lamp of uniform intensity 500 cd hangs 4 m above the floor.

• Directly beneath: E = 500/4² = 31.25 lux.
• At a point 3 m away horizontally: d² = 4² + 3² = 25, cos θ = 4/5 = 0.8 → E = (500/25) × 0.8 = 16 lux.
• Two such lamps of 300 cd each are 8 m apart at the same height; at the midpoint each is 4 m away horizontally: d² = 32, cos θ = 4/√32 = 0.7071 → E each = (300/32) × 0.7071 = 6.63 lux → total 13.3 lux.
• A 20 W LED bulb emitting 2,000 lm uniformly: MSCP = 2,000/(4π) = 159 cd, efficacy = 2,000/20 = 100 lm/W.

Example 2 - Interior design by the lumen method. Office 20 m × 10 m, ceiling 3.2 m, working plane 0.8 m, required 300 lux. LED panel: 36 W, 3,600 lm. UF = 0.6, MF = 0.8.

• Hm = 3.2 − 0.8 = 2.4 m; RI = (20 × 10)/((20 + 10) × 2.4) = 200/72 = 2.78.
• Total flux needed = (300 × 200)/(0.6 × 0.8) = 60,000/0.48 = 125,000 lm.
• N = 125,000/3,600 = 34.7 → adopt 36 luminaires in 4 rows × 9 columns.
• Spacing: 10/4 = 2.5 m across (1.25 m from walls), 20/9 = 2.22 m along. SHR = 2.5/2.4 = 1.04 - within 1.5, so uniformity is acceptable.
• Connected load = 36 × 36 = 1,296 W → LPD = 6.5 W/m², comfortably inside ECBC-type office limits (of the order of 10 W/m²; use the ECSBC or State code table in force for the exact figure).

Example 3 - Street lighting. 7.5 m carriageway, single-side poles, mounting height 9 m, 120 W LED street light of 15,000 lm, UF = 0.30, MF = 0.80, target average 15 lux.

• Φ × UF × MF = 15,000 × 0.30 × 0.80 = 3,600 lm effective per lamp.
• S = 3,600/(15 × 7.5) = 3,600/112.5 = 32 m. But S/H = 32/9 = 3.56, too high for good uniformity.
• Adopt S = 27 m (S/H = 3.0): E_avg = 3,600/(27 × 7.5) = 17.8 lux - target met with margin.
• Single-side arrangement is valid because road width (7.5 m) ≤ mounting height (9 m). Load = 1,000/27 ≈ 37 poles per km × 120 W ≈ 4.4 kW/km.

Example 4 - Flood lighting a playground. 100 m × 70 m to 100 lux average with 400 W metal-halide projectors of 36,000 lm; BF = 0.75, MF = 0.80, WLF = 1.25.

• Area = 7,000 m²; flux required = (100 × 7,000 × 1.25)/(0.75 × 0.80) = 875,000/0.60 = 1,458,333 lm.
• N = 1,458,333/36,000 = 40.5 → adopt 44 projectors, 11 on each of 4 corner masts.
• Connected load ≈ 44 × 440 W (lamp + ballast) ≈ 19.4 kW. An equivalent LED array at 130 lm/W would need about 1,458,333/130 ≈ 11.2 kW, a 42% saving.

Example 5 - LED retrofit, energy and payback. A plant has 200 twin-tube 2 × 36 W FTL fittings on copper chokes drawing 90 W each; each is replaced by a 36 W LED batten of equal delivered lumens. Operation 10 h/day × 300 days = 3,000 h/year, tariff ₹8/kWh, retrofit cost ₹1,000 per fitting.

• Existing load = 200 × 90 = 18.0 kW; new load = 200 × 36 = 7.2 kW; saving = 10.8 kW.
• Energy saving = 10.8 × 3,000 = 32,400 kWh/year; money saved = 32,400 × 8 = ₹2,59,200/year.
• Capital = 200 × 1,000 = ₹2,00,000 → payback = 2,00,000/2,59,200 = 0.77 year ≈ 9.3 months (demand-charge and air-conditioning-load savings make it shorter).

Where it is used

• Electrical consultancy and EPC design: lighting layouts, load schedules and DIALux/Relux calculations for offices, malls, airports, metro stations and hospitals; deliverables include lux contours, luminaire schedules and LPD compliance sheets.
• Municipal and highway lighting: street-light design and LED retrofits under EESL's SLNP, smart-pole and centralised-control projects in city smart-city programmes, NHAI toll-plaza and interchange lighting.
• Industrial plant electrical teams: high-bay lighting for steel, cement, auto and textile plants, flameproof luminaires in refineries and chemical plants (zone classification), emergency and escape lighting.
• Luminaire and lamp manufacturers: photometry labs (integrating sphere, goniophotometer), driver design, LM-79/LM-80 style testing, BIS certification and BEE star labelling.
• Energy auditing: lux surveys against IS 3646, LPD benchmarking, retrofit business cases and ESCO contracts - a standard deliverable of BEE-certified energy auditors.
• Specialised: sports stadium and broadcast lighting, horticultural LED lighting, studio and stage lighting, railway yard and airport apron lighting, solar street lighting with battery and LED driver.

Interview questions

• Q: What is the difference between lumen, candela and lux? A: Lumen is total luminous flux, candela is flux per steradian in a direction, and lux is lumens falling per square metre on a surface.
• Q: State the two laws of illumination. A: The inverse square law (E = I/d²) and Lambert's cosine law (E = I cos θ/d²), combining to E = I cos³θ/h² on a horizontal plane.
• Q: Why is illuminance specified for design but luminance used for road lighting? A: Illuminance is easy to measure and relate to flux, but the driver actually sees brightness, which depends on the road surface's reflectance - that is luminance.
• Q: What are the utilisation and maintenance factors? A: UF is the fraction of lamp flux reaching the working plane (depends on luminaire, room index and reflectances); MF accounts for dirt and lumen depreciation over the maintenance cycle.
• Q: Why does a fluorescent lamp need a ballast and a starter? A: A gas discharge has negative resistance, so the ballast limits current; the starter preheats the cathodes and provides the inductive voltage kick to strike the arc.
• Q: What is the stroboscopic effect and how is it avoided? A: Discharge lamps on 50 Hz pulse at 100 Hz, making rotating machinery look stationary; it is avoided with lamps on different phases, lead-lag circuits, or high-frequency electronic/LED gear.
• Q: Why does an LED lamp need a heat sink if it produces "no heat"? A: It produces little infrared in the beam, but 60-75% of the input is still dissipated as heat at the junction; a high junction temperature cuts both output and life.
• Q: Which lamp has the highest efficacy and why is it still not used in offices? A: Low-pressure sodium, because it emits monochromatic yellow at 589 nm near the eye's sensitive band - but with essentially zero colour rendering it is unusable where colours matter.
• Q: How do you decide between single-side, staggered and opposite street-light arrangements? A: By road width relative to mounting height - single-side up to 1× height, staggered up to 1.5×, opposite beyond that.
• Q: What is LPD and why do auditors use it? A: Lighting power density in W/m² - it compares an installation's connected lighting load against ECBC limits and industry benchmarks independently of room size.
• Q: A room designed for 300 lux measures 180 lux after three years. What would you check? A: Lamp lumen depreciation and failed lamps, dirty luminaires and room surfaces, reduced supply voltage, and whether the as-built layout matches the design.
• Q: What is CRI, and is a high CCT the same as a high CRI? A: CRI measures how faithfully colours are rendered compared with a reference source; CCT only describes warm or cool appearance - a 6,500 K lamp can still have poor CRI.

Common mistakes

• Using lumens and lux interchangeably, or quoting lamp lumens where luminaire (delivered) lumens are needed.
• Applying the inverse square law to a long tube or large panel close to the surface, where the point-source assumption fails.
• Forgetting the cos θ term when the surface is inclined, or confusing the angle from the vertical with the angle from the surface.
• Multiplying E × A by UF × MF instead of dividing by them - or omitting MF entirely and under-designing by 20-30%.
• Measuring the mounting height from the floor instead of from the working plane when computing room index and SHR.
• Choosing a luminaire count from the lumen calculation but laying it out in a grid that breaks the space-to-height limit, giving dark patches between bright pools.
• Judging a design only on average lux, ignoring uniformity, UGR glare, CRI and flicker - the usual cause of user complaints after commissioning.

Related topics

Energy audit and conservation · Electrical installation, wiring and estimation · Electrical safety and earthing · Electric heating and welding · Utilisation of electrical energy and traction · Power electronics converters and LED drivers