Skip to content
Sensmat

Light and UV

Fading is spent permanently. How to measure it, how to filter it, and how to budget a year of display rather than dimming by feel.

Updated 14 August 2026

Light damage has a property that sets it apart from everything else in preventive conservation: it does not recover. Humidity that swings and comes back has done some damage but the object is still in the room. A dye that has faded is gone. Every lux-hour an object receives is spent permanently, and the total is what matters, not the peak.

This makes light the one agent where a budget is the natural way to think, and where a light meter pays for itself faster than almost any other instrument.

What actually causes fading

Photochemical damage scales with the energy of the photons and the sensitivity of the material. Ultraviolet radiation carries more energy per photon than visible light and is correspondingly more damaging per unit received, which is why UV filtering is the standard first move. But UV is a small fraction of the energy in most lighting, and visible light does the bulk of the fading in most real installations simply because there is so much more of it.

Infrared does something different again. It does not fade, it heats, and a hot object in a cool room sits at a lower local relative humidity than its surroundings. A halogen spot at close range can create a genuine microclimate problem on the surface of a painting.

Sensitivity varies enormously between materials, and the standard grouping used in practice splits collections into broad responsivity categories: materials with no appreciable response such as stone, metal and glass; low responsivity such as oil and tempera paintings and undyed leather; medium responsivity such as most textiles, watercolours, prints and dyed leather; and high responsivity such as many early photographic processes and objects with fugitive dyes. The recommended illuminance for each group differs by a factor of several, and the materials section covers individual cases.

Measuring it properly

A lux meter measures visible illuminance, weighted to the response of the human eye. A UV meter measures ultraviolet content, usually expressed in microwatts per lumen so that it scales with the visible light present. You want both numbers, and a combined instrument is the usual way to get them. Lux and UV meters covers what separates a usable meter from a phone app.

Two mistakes recur. The first is measuring at the wrong place: the number that matters is the illuminance at the object’s surface, held in the plane of the surface and facing the light, not the ambient reading in the middle of the room. The second is measuring once. Daylight varies by orders of magnitude between a January afternoon and a June noon, and a single reading taken on a grey day proves nothing about a case beside a south-facing window.

From readings to a budget

The number that governs fading is cumulative exposure, illuminance multiplied by time, usually expressed in lux-hours. This is what makes budgeting possible: a fixed annual allowance can be spent as a long display at low light or a short display at higher light, and the choice becomes a curatorial one rather than a technical one.

In practice this means a sensitive object might be shown at 50 lux for a limited number of hours per year, and the museum then knows exactly how many years of display it has before reaching the total it decided was acceptable. It also makes rotation legible: an object on display for three months and in the dark for nine has spent a quarter of the exposure of the same object left out permanently. Light exposure budgets works through the arithmetic.

What to change first

Filter the UV. UV filter film on windows and glazing, and UV-filtering acrylic in frames, is cheap relative to almost everything else and removes a disproportionate share of the damage. Films age and lose effectiveness, so they need a check and a replacement date rather than being fitted and forgotten.

Control daylight before electric light. Daylight is the largest and least controllable source in most historic buildings. Blinds, sun curtains and film do more than any amount of careful lamp selection if a case sits in direct sun.

Then choose lamps. Museum LED lighting covers colour rendering, colour temperature and dimming, which affect how a display looks far more than they affect how fast it fades.

Then rotate. For the most sensitive material, no lighting technology solves the problem. Limiting the hours does.

Frequently asked questions

Is removing UV enough?
No, and this is the most persistent misconception in the field. UV is disproportionately damaging per unit of energy, so filtering it is worthwhile and cheap. But visible light fades pigments and dyes on its own, and a UV-filtered showcase at 500 lux will still fade a watercolour.
Are LEDs safe for collections?
Safer than the alternatives in one specific respect: a good LED emits essentially no UV and much less infrared than a halogen lamp, so it heats objects less. It is not damage-free, because the visible light it does emit fades things exactly as any other visible light does.
How do I know how much light an object has already had?
You usually do not, and that is the honest answer. Unless someone kept a display record, previous exposure is unknown. This argues for starting a record now rather than trying to reconstruct one.

In this section

Related reading