Control strip processing and charting

Purpose

To put one pre-exposed strip through every processing run, turn it into three numbers, and plot those numbers so that a change in the process is visible on the day it happens rather than on the negative you cared about.

When to use it

Every run, once the aim and the limits exist. Establishing them is a separate job — the mean of the first three runs at your standard condition — and it is done by the process-control experiment, not here.

Before you start

You should already have a batch of strips pre-exposed in one sitting from one cassette, numbered, and stored frozen; Kodak’s instruction for their own strips is to keep them frozen to maintain consistency. You should have the aim for each of the three numbers, the action limit at 2s and the control limit at 3s, where s came from your own instrument certificate rather than from anybody’s published table.

The chart is ruled up before the first run. A chart drawn after the fourth point is a chart whose aim was chosen to fit the data.

On the bench: the next strip in the batch, still frozen until you need it, and an unexposed strip alongside it; the densitometer; the chart, on paper; the notebook.

Procedure

1. Take one numbered strip from the batch and let it reach room temperature before it goes in the tank. Take an unexposed strip from the same batch with it.
2. Process both in the run, in the same tank as the work, at your standard condition — your dilution, 20.0 °C, your written agitation script, by film processing.
3. Read fog as D-min, from the masked tail of the strip.
4. Read speed as D(shadow step) − D-min, at the step nearest your own criterion point.
5. Read contrast as (D(high) − D(shadow)) ÷ Δlog H between those two steps, taking Δlog H from the wedge values you used to expose the batch.
6. Compute the deviation from aim for each of the three, and plot the deviation rather than the raw number.
7. Write beside the point, on the chart itself, what else was true that day: which bottle, how old, what temperature the bath actually held, whether the agitation was the usual script, and how many films had already been through that developer.
8. Rule a vertical line on the chart at every event — a fresh mix, a batch crossover, a new fixer, a day that was unusually warm. A movement with an event beside it is an explanation; a movement without one is a question.
9. Read the chart. Points scattered on both sides of aim, inside the limits, in no pattern, is a process in control: do not react to it.
10. Treat a point on or beyond a control limit as an outlier and check the measurement first — re-read the strip, then process another — before concluding anything about the process.
11. Treat four to eight consecutive points on one side of aim as a level shift, and four to eight climbing or falling in a row as a trend. Two or three of either is noise.
12. When the batch runs low, change to a new one by the crossover: one strip from each batch in each of three separate runs, average each set of three, take the difference, and apply that difference to the aims.

How you know it worked

• Every run this month has a point on the chart, including the runs that went well.
• Each point is a deviation from aim, and the aim has not moved since it was set, except at a crossover.
• Every point that moved has either an event rule beside it or a written question in the notebook.
• The three numbers were read the same way on every strip: same steps, same instrument, same operator judgement.
• The chart has caused you to do nothing at all on most days, which is what a process in control looks like.

If a step fails

Step 1, the batch has run out mid-month. Do not start a fresh batch without the crossover at step 12. A new batch with no crossover puts a step in the chart that will be read as a process change for the rest of its life.

Step 2, the run was not at the standard condition. Plot the point anyway and annotate it with what was different. A point you excluded because it was inconvenient is the one that would have explained the next four.

Steps 3 to 5, the strip cannot be read — it is fogged, blank or damaged. Read nothing from it. Process the next strip in the batch in the next run and note the gap; a guessed point is worse than a missing one.

Step 6, the deviation exceeds the action limit. Investigate before the next run: measurement first, then temperature, then the developer’s age and how much has gone through it, then agitation. Beyond the control limit, stop and find the cause before you process anything you care about.

Step 9, you are tempted to adjust after two points on one side. Do not. Over-controlling a process by reacting to random variation is the failure Kodak name first, and it manufactures the very instability the chart exists to detect.

Step 11, speed is drifting slowly downward. This is the ambiguous case and it has to be said plainly. A slowly fading latent image on a pre-exposed batch looks exactly like a developer being exhausted. Run the crossover: a trend that disappears at the crossover was your strips, not your developer. Until it is resolved, the method has detected a change and has not identified it.

Step 11, speed falls after a fresh mix and then settles. Some developer systems are formulated so that a fresh solution gives a higher speed than a properly seasoned one, and Kodak name Developer D-76 with Replenisher D-76R among them. Know which system you are running before you chase your own tail.

What to record

The point on the chart, and one line in the notebook beside it: the date; the strip number and its exposure date; the three readings and the three deviations; the bath, its dilution, its mixing date and its running total; the temperature the bath actually held; and the agitation used. At a crossover, record the three paired runs, the two averages, the difference, and the aims before and after it.