Conservation standards are a number and a tolerance, and the building spends a fortune holding them. With the air handling off for a fortnight, every gallery drifted with the weather. Gallery 4 sat at fifty two percent, plus or minus one, for the whole fortnight.
I am a building services technician at a city museum and I cover nights and weekends.
The numbers, because conservation is the reason the building exists in its present form.
Relative humidity is how much water the air is holding compared with how much it could hold at that temperature. Objects take up and give off moisture until they are in balance with it, and the damage is done by the changes rather than by the level. Wood and ivory and panel paintings crack when it swings.
So the standard is a target and a tolerance. We run fifty percent plus or minus five, and twenty degrees plus or minus two, and a deviation generates a report that goes to the conservators.
Every gallery has a logger. Ten minute intervals, calibrated annually against a reference hygrometer, and the data goes back to 2009.
Holding that costs about a hundred and forty thousand pounds a year in plant and energy, which I know because I have sat in the meeting where it was argued about.
In the summer of 2022 we shut the air handling down for a fortnight for a chiller replacement.
The collection was protected: the sensitive material went into a controlled store and the conservators signed off the risk.
What the galleries did in that fortnight is what any masonry building does. They drifted with the weather: up into the sixties in the humid week, down to the low forties on the two hot dry days, each room at its own pace depending on its volume and its windows.
Gallery 4 sat at fifty two percent.
Fourteen days. Two thousand and sixteen readings. The highest is fifty three point one and the lowest is fifty one point two.
With no plant running, no humidifier, no dehumidifier, nothing in the ductwork moving, and the dampers shut.
The first thing anybody says is that the logger failed and froze on its last value, and that is exactly what a failed logger looks like.
A frozen logger repeats the same value to the decimal place. This one varies by a few tenths in a pattern with a daily shape to it, the same daily shape the other galleries have, just flattened to almost nothing.
And there are three loggers in that gallery now, two of them put there afterwards by me, different makes, and all three agree.
What gallery 4 is.
A single room, about twelve metres by nine, stone walls, a blocked fireplace, no exterior windows since a 1960s alteration, two doors onto the corridor.
It is the Roman gallery. Stone, pottery, two mosaics, a lot of metalwork, and the collection in there is the least humidity sensitive in the building, which is why it was chosen for that position in the first place.
What has been measured since, because I could not leave it and because a conservator became interested.
The moisture buffering of the fabric. Lime plaster and stone do buffer humidity, and that is a real and well documented effect: a heavy masonry room with a lot of exposed hygroscopic surface is more stable than a plasterboard one. It is the reason old buildings are kinder to collections than new ones.
Buffering slows a change. It does not cancel one. The conservator's own written note says that the degree of stability observed in gallery 4 is an order of magnitude beyond what the fabric can account for and that she has not seen it in twenty years of collection care.
Air exchange. We smoke tested it and put a tracer gas test through it in 2023, paid for out of the capital project. The room changes its air about a third of a time an hour with the doors shut, which is a normal leaky old room.
So the air in there is being replaced, with corridor air at whatever the corridor is doing, which in that fortnight was between forty one and sixty four percent.
And it comes out at fifty two.
What happens officially.
Gallery 4 is now exempted from the humidity control zone, which saves a genuine amount of money, and the conservation policy records it as a naturally stable environment attributable to the building fabric.
I helped write that sentence. The words attributable to the building fabric are mine and I knew they were not true when I typed them, and the alternative was a document that said we do not know, which the funders would have read as a risk.
What I do on a night shift.
The round is every two hours and gallery 4 is on it, and the job is to look at the panel, look at the room and move on.
The change I made is that I take the reading from the doorway.
The display for the original logger is on the wall by the fireplace, which means walking into the middle of the room, and there is a handheld that reads the new ones from the door, and I bought the handheld myself.
The reason is a night in March 2023. I was at the logger by the fireplace writing the figure on the sheet, and the room at that point was inside its normal band and so was the corridor, and there was nothing wrong with anything.
A room with no ventilation running is a still room. You notice your own breathing in it.
While I was standing there the numbers on the display did what they always do, which is nothing, and I became aware that the air on the back of my neck was moving, steadily, the way it does under a vent, and the vent in that room was sealed in 2022 and I sealed it.