Fifty lux and ninety seconds

Works on paper are displayed at fifty lux. That is not a guideline anyone in the sector treats as negotiable, and rightly — light damage to pigment and paper is cumulative and irreversible.

Fifty lux is roughly the light level of a restaurant at dinner. Sit in it for an hour and it reads as perfectly normal, warm even. Walk into it directly from a foyer at four hundred lux and you will see essentially nothing for the first ninety seconds.

Ninety seconds is a long time to stand in a room being unable to see the thing you came for.

The usual solutions

The standard responses are all lighting responses. Raise the ambient in the gallery to the limit and hope. Put a bright feature wall just inside the door as an anchor. Increase the contrast on the labels so at least the text resolves early.

None of them work well, because none of them address the actual mechanism. Dark adaptation is a property of the eye, not of the room. The eye takes the time it takes, and the only variable you control is what it is adapting from.

Adapting on the way in

At Whitlock Gallery we put the adaptation into the route. Between the entrance hall and the first gallery there is a top-lit ramp, twenty-two metres long, descending about a metre and a half.

Along that length the roof lights step down in area — generous at the top, almost nothing at the bottom. The light level falls from around three hundred and fifty lux at the entrance to about sixty at the gallery door. The gradient is shallow enough that at no point does it read as getting darker. It reads as a long, calm, naturally lit ramp.

By the time you reach the door your eye has already done most of the work. The gallery, at fifty lux, looks lit.

Nobody notices, which is the point

Visitors do not report that the ramp is dark. In two years of front-of-house feedback the ramp has been mentioned as pleasant, as long, and once as an accessibility improvement over the stair it replaced. It has never once been described as dim.

This is the quality we were after. A dark lobby with a sign explaining conservation lighting is a building announcing its constraint to the visitor and asking them to accommodate it. A ramp that happens to get gradually darker is a building absorbing the constraint on the visitor's behalf.

The second one is more work and considerably more expensive. It is also the difference between a gallery that feels considered and one that feels apologetic.

Where the section does the work

The wider point is that lighting problems in galleries are very often section problems wearing a lighting costume.

Glare on a glazed frame is a question of where the aperture sits relative to the eye, not of what fitting you specify. Uneven wall illumination is a question of ceiling height and throw distance. The perceived brightness of a room is substantially a question of what room you were in immediately before it.

All of those are decided at concept stage, in section, usually by someone who is not the lighting designer and often before the lighting designer is appointed. By the time there is a lighting package to specify, most of the outcome is already fixed.

What we would do earlier

If we ran Whitlock again, the one change would be to model the adaptation sequence at Stage 2 rather than Stage 3. We got the gradient right, but we got it right by iteration, and two of those iterations moved a structural line.

The tool for this is not complicated — a simple luminance model along the route, sampled at eye height every couple of metres. It takes an afternoon. We now run it on every project where a visitor moves between significantly different light levels, which turns out to be most cultural projects and quite a few civic ones.