Lighting a mineral display cabinet
Set a cabinet to 50 lux and the fading clock slows about threefold against 150 lux. A UV-free LED does not make amethyst or green fluorite in a collection safe.
Light a mineral cabinet to about 50 lux at the specimen, which the Canadian Conservation Institute gives as the benchmark for full colour vision, and treat exposure as a dose in lux hours rather than a setting. At 150 lux for 4 hours a day a shelf of crystal specimens collects 0.219 million lux hours a year; at 50 lux, 0.073.
In short
- Fading is driven by visible light, not only ultraviolet. The CCI puts the photochemistry of colourant fading between about 2 and 3 electron volts — the range we see in. A zero-UV LED reduces the problem and does not remove it.
- 50 lux is the museum benchmark, adopted because colour science established it is enough for the eye to be working in full colour vision. It looks dim for about thirty seconds and then looks normal.
- Exposure is a dose: lux multiplied by hours. Two hours at 100 lux and four hours at 50 lux do the same damage. That is the lever most people never touch, because they think about brightness and not about the switch.
- Heat is the fixture problem nobody measures. The CCI's field test needs no instrument: put your hand where the specimen sits and shade it with a card. If you can feel the light warming your palm, it is too close or too hot.
- Three groups should not be lit at any level. Realgar and the arsenic sulphides are destroyed irreversibly. Amethyst and green fluorite fade and do not come back. Hydrated copper sulphates want humidity, and a warm lamp dries their air.
| Control | What to aim for | Where the figure comes from | What it costs you |
|---|---|---|---|
| Illuminance at the specimen | 50 lux; 150 lux as a compromise | CCI benchmark for full colour vision | At 150 lux a highly sensitive colourant shows just-noticeable fade in roughly half a year to 7 years; at 50 lux, 1.5 to 20 years |
| Hours lit per day | As few as you will accept | Dose is illuminance times time | Nothing. This is the free lever and it is the one people ignore |
| Annual dose | Calculate it: lux x hours x days | Museum practice uses millions of lux hours (Mlx h) | Nothing but the arithmetic — and it turns an argument into a number |
| Ultraviolet content of the source | Below about 75 microwatts per lumen | CCI cites this as the level of daylight through a good UV filter | A filter, or an LED source; most white LEDs are already well under it |
| Heat at the specimen | No perceptible warming on your palm | CCI's hand-and-card field test | Distance, or a cooler fixture. Halogen inside a sealed case is the common failure |
The arithmetic, with real inputs, so you can substitute your own
Exposure is not a setting, it is a dose, and the dose is illuminance multiplied by time. Museums count it in millions of lux hours, abbreviated Mlx h. Here is the calculation for an ordinary domestic cabinet.
Cabinet lit to 150 lux at the shelf, switched on 4 hours a day. 150 x 4 = 600 lux hours a day. Over a year: 600 x 365 = 219,000 lux hours, or 0.219 Mlx h a year.
Same cabinet at 50 lux, same hours. 50 x 4 x 365 = 73,000 lux hours, or 0.073 Mlx h a year. One third of the dose, for a change nobody notices after the first minute.
Same cabinet at 150 lux, switched on 1 hour a day instead of 4. 0.055 Mlx h a year — better than dropping the brightness, and it costs nothing at all.
The Canadian Conservation Institute's note on light, ultraviolet and infrared publishes the fade times that make those doses mean something. For its highest-sensitivity class of colourants, just-noticeable fade takes roughly half a year to 7 years at 150 lux, and 1.5 to 20 years at 50 lux. Dropping to a third of the illuminance buys roughly three times the life, which is exactly what you would expect from a dose model and is worth knowing is actually measured rather than assumed. Our page on displaying a collection works the same problem from the lit-hours end.
Why a zero-UV LED does not make amethyst safe
The commonest piece of display advice is to fit LED strips because they emit no ultraviolet. The advice is sound as far as it goes and it does not go as far as people think.
The CCI states the mechanism plainly: the photochemistry that produces yellowing and disintegration typical of ultraviolet needs photon energies above about 3 electron volts, but the photochemistry typical of colourant fading happens between about 2 and 3 electron volts — which is the visible range, the range our own retinas work in. Their phrasing is worth keeping verbatim: We are fated, in fact, to see in the same band as that which causes sensitive colourants to fade, given the related photochemical phenomena.
So removing the ultraviolet removes one mechanism and leaves the other running. Amethyst and green fluorite fade under visible light. An LED-lit case at 300 lux is not a safe place for them because the LED is UV-free; it is simply a less bad place than a daylit windowsill. The species that genuinely must not be lit — realgar above all — must not be lit by anything, and the NatSCA and Icon leaflet says so: susceptible species should be stored in light-proof containers.
Where the LED genuinely wins is heat. An incandescent or halogen lamp inside a closed case is a heat source in a sealed box, and that dries the air as well as warming the specimens — which matters to chalcanthite, langite and the other hydrated species, for whom dry air is the failure mode. Our note on which minerals fade in light lists the species involved.
Fitting it: where the light goes, and the test that needs no meter
Three placement decisions do most of the work, and none of them needs equipment.
Light from above and slightly in front, not from behind the glass at specimen height. A source in the viewer's line of sight puts glare on the glazing and kills the contrast you are lighting for. Crystal faces are specular: they show their lustre when the light comes in at an angle and the eye is somewhere else.
Keep the fixture out of the sealed volume if you can. A strip mounted under the shelf above, shining down, is outside the air the specimens sit in. A lamp inside a closed case heats that air.
Then run the CCI's field test, which is the most useful thing in their note. Put your hand where a specimen sits. Alternately illuminate and shade your palm with a piece of card. If you can feel the light warming your hand, the fixture is too hot or too close, and anything in that case that is sensitive to temperature is at risk. It takes ten seconds and it replaces an instrument most collectors will never own.
For illuminance itself, a lux meter is inexpensive and there are usable ones in phones. Measure at the shelf where the specimens are, not at the glass, and measure with the room lights in their normal state. If you are not going to measure, assume you are over: domestic cabinet lighting is almost always set brighter than museum practice, because the person setting it is looking at it in a lit room.
What to put under the light, and what to keep out of the case entirely
Lighting decisions get much easier once you accept that a display cabinet is not where the whole collection lives. Some species should be in a drawer, and putting them there is not a compromise.
Excellent under light, at any sensible level: quartz, hematite, cassiterite, pyrite, galena, the silicates. These are chemically stable and their appeal is lustre and form, which is exactly what directional light rewards.
Fine in a glazed case away from daylight, at 50 lux: most carbonates, most of the copper secondaries, pyromorphite, the zeolites.
Not on display: realgar and orpiment, which the conservation literature puts at the top of the light-damage list; amethyst and green fluorite, which fade; and the hydrated sulphates, which need stable humidity more than they need an audience.
And the one that catches people out: fluorescent specimens are a separate problem. Ultraviolet display lighting inside a case sounds attractive and it works badly — the inverse-square penalty at case distances is severe, and the shortwave lamps that produce the good effect are not things to leave running where people stand. Light the case for daylight viewing and do the ultraviolet work by hand with a proper filtered lamp. If a particular specimen has to do both jobs — display in daylight and respond under a lamp — that is a specific brief, and the sort of thing worth stating on a wanted list rather than hoping to find by browsing.