Session exposure drift

Owned by Break/Fix — the timer that lies, which says that this entry owns the general shape of the fault and the reason a heatsink is a photometric component, and that the page itself adds the discrimination: nine candidate causes, a test for each, and a four-sheet decomposition that separates the exposure side from the developer’s decline for good. It was written before that page, from low-voltage electronics for the darkroom, whose heatsink section is the mechanism, and calibrating the sensitometer, whose Part 3 measures it.

What you see

The last print of the evening does not match the first, at the same nominal exposure, the same filter and the same negative. Usually the drift is one-directional and slow enough that no single pair of prints looks wrong — which is why it is found by re-printing the first negative at the end of a session rather than by noticing.

Likely causes

1. A source still warming. Its output falls as it heats, and it may still be falling several minutes after switch-on.
2. Supply variation. A lamp sharing a circuit with something that cycles.
3. The developer exhausting. A print developer works on far more silver per unit area than a film developer, so it exhausts on the work rather than on the air, and its performance falls faster and more predictably.
4. The timer. A repeatability worse than the drift being chased makes every measurement of the drift noise.

What is happening: the chemistry and physics

Light out falls as temperature rises. Cree plot relative flux against junction temperature from 25 °C to 150 °C for the XP-E2, and Part XIV reads the direction off that chart and deliberately not the ordinates, because a value taken from a printed graph is a guess wearing a decimal point. The direction is enough for a diagnosis: an emitter bolted to a star board and nothing else will still be climbing towards its final temperature — and falling towards its final brightness — several minutes after switch-on.

That is what makes a heatsink a photometric component rather than a mechanical one. It is not there to stop the device dying. It is there to hold the junction at a temperature that stops moving, so that the light output stops moving.

The spectrum moves too. Cree publish relative chromaticity against current and temperature, which establishes that the mixture of a phosphor white emitter is not a fixed thing — and for variable-contrast paper a shifting spectrum is a shifting grade as well as a shifting exposure.

The developer half is chemistry rather than physics, and it is the faster of the two on paper. Every print consumes developing agent from a bath measured in hundreds of millilitres; ILFORD publish the capacity in 8 × 10 inch prints per litre for each dilution, and a session that exceeds it is a session whose later prints were made in a different solution.

!The confirming test: print the first negative again, last

Keep the first negative in the carrier or go back to it, and make an identical print at the end of the session. Two prints, one variable — time.

Then separate the causes with two more measurements, neither of which needs an instrument:

• Repeat the last print in fresh developer. If it matches the first print, the drift was chemical. If it still does not, the drift was optical or electrical. Part XVII’s four-sheet test is the complete version of that comparison, and it is worth the extra two sheets: one pair exposed at the start and one at the end, each pair split between the working tray and developer mixed a minute earlier, every sheet developed immediately so that latent-image keeping never enters the arithmetic. Its three differences isolate the exposure side, the developer’s decline and the total drift, and the first two should add up to the third — an identity that catches a variable you have not controlled.
• Repeat the first print after a long warm-up rather than from cold. If the difference disappears, the source was warming.

Part XIV’s own drift arm does the instrument version of this properly: expose from cold at 0, 10, 30 and 60 minutes, log a monitor reading with each, and develop all four together. And it makes the point that matters here — the monitor watches the lamp; the film sees whatever arrives at the film plane. A flat monitor trace with a drifting film series means something between them is moving.

Diagnostic questions

• Is there a warm-up rule, and was it followed? A professional bench instrument specifies two minutes; a home-built head in a plywood box is probably longer, and the only way to know is to measure it.
• Which way does it drift? Falling exposure through a session points at the lamp; darkening prints at the same exposure point at the developer having weakened and the printer having compensated.
• How many prints through this dish? Against the published capacity, not against a feeling.
• Does the enlarger share a circuit with a kettle, a heater or a dryer? Anything that cycles is a supply variation with a schedule.
• How repeatable is the timer? Ten nominally identical exposures onto one sheet, read as densities, gives the answer in an evening.

Corrective action

Establish and obey a warm-up rule derived from your own measurement rather than from any published figure. This is the cheapest fix and it removes the largest term for most people.

Replenish or replace the developer part-way through a long session, and note the change on the print record so that later prints are known to have been made in fresh chemistry.

Re-make the reference print after any change, and keep it. A physical reference print is worth more than a remembered exposure.

Prevention

Heatsink the source properly if you built it, and give it the time it needs if you did not.

Work within the developer’s published capacity, and mix enough for the session at the start rather than topping up unrecorded.

Keep a print record with exposure, filter, dish and time, so that a drift is visible as data rather than as a suspicion — the exposure log worksheet is the course’s sheet for it.