Lux vs lumens: how to read a night light's brightness

Lumens are how much light a lamp gives out. Lux is how much of it reaches a surface such as a pillow or a page, and that depends on distance: for a small lamp, halve the distance and the lux is four times higher. So a lumen figure tells you little without your distance from the lamp, and a lux figure means little without the distance it was measured at. Kelvin is a separate measure, the tone of the light rather than the amount; our guide to warm white vs cool white covers it.

What lux and lumens each measure

A lumen (lm) is the unit of luminous flux: the total light a source puts out, in every direction. A lux (lx) is the unit of illuminance: one lumen spread over one square metre of surface, so 1 lx = 1 lm per square metre. Lumens belong to the lamp. Lux belongs to a place, so one lamp gives a different lux on a shelf, on a pillow and across the room. Light boxes and daylight lamps are a different job, covered in our guide to sunrise alarms and SAD lamps.

From lumens to lux at the bed

For a light that spreads evenly in every direction, the lux on a surface facing it equals the lumens divided by 4 × π × distance squared, with the distance in metres: E = Φ / (4πd²). This is the inverse-square law, which Stari and colleagues set out along with the cases where it fails.

Calculated, not measured: lux at a surface facing the light, from that formula. The assumptions are a small source spreading light evenly in every direction, no shade, and no light bouncing off walls or ceiling. The lumen values are illustrative inputs, not any product’s output.

Light output	Lux at 0.5 m	Lux at 1 m	Lux at 2 m
5 lm	1.6	0.40	0.10
10 lm	3.2	0.80	0.20
20 lm	6.4	1.6	0.40
50 lm	16	4.0	1.0

For example, a 10-lumen light gives about 0.80 lux at 1 m and 0.20 lux at 2 m. We did not measure anything for this page, and a real night light has a shade and a shape, so treat the figures as a rough guide.

When the estimate is wrong

• Beams and spots. A narrow beam puts the same lumens on a smaller patch, so lux straight down the beam is higher than the formula gives, and lower off to the side.
• LED strips. Stari and colleagues found that light from a strip falls off roughly in inverse proportion to distance, not distance squared.
• Shades, flat lights and wall mounts. A shade redirects light, and a flat or wall-mounted light throws more of it forward, so a meter pointed straight at it can read higher than the estimate. Walls and ceilings bounce some light back as well.
• Too close. The formula treats the lamp as a point. NIST notes the law holds poorly up close for a large lamp, and a trade article gives a rule of thumb of five times the lamp’s widest dimension (some experts say ten). For a lamp 10 cm across that is about 50 cm, which is why the table starts at 0.5 m.
• Projectors. A projector puts its light on a wall or ceiling, so lux at the bed is a different question. Our guide to baby projectors and night lights looks at when each suits a child’s room.

What has been measured in real night lights

Glickman and colleagues measured 25 children’s night lights, 79 settings in all, picked for popularity and consumer ratings on major online retailers between 2022 and 2024 (npj Biological Timing and Sleep, 2026). In a simulated bedroom they read the light at pillow height and reported melanopic EDI: a lux-style figure weighted towards the light-sensing cells in the eye that feed the body clock, rather than towards how bright the light looks.

Attributed to that study, the median melanopic EDI in the bedroom was 4.33 lx for handheld lights at 15 cm, 1.70 lx for tabletop lamps at 92 cm, 0.27 lx for wall plug-ins at 274 cm and 0.22 lx for projectors, read from the ceiling 180 cm away. Across all settings the median was 0.72 lx, and readings ran from 0 (below what the instrument could detect) to about 908 lx. The differences between the four types were not statistically significant, so read the medians as a description of this sample, not a ranking. With the meter 2 mm from the emitter, the median across all settings was 155.49 lx. The authors name distance and placement as critical.

That is a different metric from the Calculated table, from different lights in a different room, so the two cannot be lined up. It is one lab and one simulated room, with products chosen for popularity rather than at random. The paper gives no lumen figures and measured light, not sleep. The authors also note that manufacturers rarely report melanopic or photopic figures on packaging.

Which lux is that figure?

A lux meter or phone app normally reads photopic lux, which weights light by how bright it looks to the eye. Sleep researchers often report melanopic EDI at the eye instead. For many common domestic warm-white LEDs (2,700 to 3,000 K), Brown et al. give a melanopic efficacy ratio below 0.35: their melanopic EDI comes to less than 0.35 times their photopic lux. The two figures cannot be swapped.

Meter position varies too. Akacem et al. measured about 1,000 lux roughly 5 cm from children’s faces, and Lee et al. about 300 lux vertical at eye height. Neither tested a night light or babies; our guide to the best colour night light for babies covers both studies. The 2022 adult consensus from Brown et al. is stated in melanopic EDI at the eye: a recommended maximum of 1 lux for a sleep environment, or 10 lux where vision is needed at night. It covers healthy adults aged 18 to 55. The authors say children may need different values, and this page gives no figure for a child’s room.

Reading a night-light listing

As of 19 September 2026, the listings in our night-lights collection describe brightness steps, watts and runtime:

• The current listing describes GentleLume, a touch bedside lamp, as having three brightness levels, a 2W LED light source and three light temperatures. Three steps give the order of brightness, not the amount; our guide to choosing a touch lamp looks at fixed-step dimming.
• The current listing describes CapiZen with a 0.5W integrated LED, two light settings and runtime estimates of up to approximately 16 hours on the brighter setting and 56 on the lower. Runtime speaks to the battery, not the light.
• The current listing for Elephia gives brightness levels of 30%, 60% and 100%.

A watt is power drawn, not light given out, and converting one into the other needs a lumens-per-watt figure. Comparing 0.5W with 2W says which lamp draws more power; it hints at which is brighter only if the light source and shade are alike.

Sensor night lights use both units. Our guide to auto-on night lights notes lighting-industry guides putting professional photocell switch-on at roughly 10 to 30 lux, a threshold for the room, and quotes another retailer’s plug-in listing that gives 7 lumens, the lamp’s output. Neither tells you the lux at the bed.

What to ask of a brightness figure

For a lumen figure, ask how far away you will be; for a lux figure, ask which lux it is and where the meter was. This page does not say what brightness suits a child, and the estimates here cannot answer it.

Separately from brightness, follow a night light’s user guide: keep it on a stable surface outside the cot or bed, never under bedding. If you want to compare listings with these questions in hand, our night lights collection is one place to start.

References

Units and the distance law

1. BIPM, The International System of Units (SI), 9th edition, 2019, version 4.01 of June 2026. Supports: the SI definitions of the lumen and the lux.
2. CIE, International Lighting Vocabulary, entry 17-21-060, illuminance, from CIE S 017:2020. Supports: illuminance as incident luminous flux per unit area, in lux.
3. Stari, Abreu, Monteiro and Martí, Light intensity does not always decay with the inverse of the square of the distance: an open-inquiry laboratory, arXiv:2501.00622v2, 17 September 2025 (the PDF’s own title says “illuminance”). A preprint; we did not check its peer-review status. Supports: E = Φ / (4πr²), the quartering on doubling the distance, and the LED-strip and close-range limits.
4. Ohno, NIST Measurement Services: Photometric Calibrations, NIST Special Publication 250-37, January 1997. Supports: the inverse-square law not holding well at close distances for a large lamp.
5. Klages, Some Observations of the Inverse Square Law of Illumination, TV Technology, 2 January 2013. A trade article. Supports: the five-times rule of thumb (and the more cautious ten times), stated by the article rather than derived.

Measured night lights and light research

6. Glickman et al., Quantifying the biological impacts of nightlights: implications for sleep and circadian health in children, npj Biological Timing and Sleep 3:14, 2 April 2026, open access. Supports: the product selection and design, the medians by product type and distance, the source-level and overall medians, the finding that the type differences were not significant, and the packaging statement. All figures are melanopic EDI.
7. Brown et al., Recommendations for daytime, evening, and nighttime indoor light exposure to best support physiology, sleep, and wakefulness in healthy adults, PLOS Biology 20(3):e3001571, 17 March 2022. Supports: melanopic EDI versus photopic lux, the warm-white LED ratio below 0.35, and the adult maxima at the eye.
8. Akacem, Wright and LeBourgeois, Sensitivity of the circadian system to evening bright light in preschool-age children, Physiological Reports 6(5):e13617, 2018. Supports: about 1,000 lux measured about 5 cm from children’s faces.
9. Lee et al., Melatonin suppression and sleepiness in children exposed to blue-enriched white LED lighting at night, Physiological Reports 6(24):e13942, 2018. Supports: about 300 lux measured vertically at eye height.