Uneven density across a strip

Lifted from the sensitometer enclosure and exposure stage, calibrating the sensitometer and the step-wedge exposure series, whose preparation section already contains the uniformity map this entry is the failure of.

What you see

A step that is not one density. Read the same step near one edge of the strip and near the other and the two readings differ. On a long strip the difference usually runs one way — a gradient — and on a badly made enclosure it can be a bright patch instead.

The consequence is worse than the appearance. Every curve plotted from that strip inherits the gradient, and because the steps run along the strip, a gradient along the strip is indistinguishable from a change of slope. A sensitometer with a bright end quietly reports a contrast index that the film never had.

Likely causes

1. A non-uniform exposing field. The commonest cause on a home-built instrument, and the reason Part XIV’s enclosure page maps the field before anything else.
2. Poor contact between the wedge and the emulsion. Light spreads under a lifted step and softens the boundary between steps as well as lifting the density.
3. Uneven development. A strip lying against a tank wall, insufficient agitation, or a crowded tank — the same causes as mottled development and surge marks, acting on a piece of film small enough that the pattern is mistaken for a reading.
4. The densitometer. An aperture placed differently on each step, on a strip whose steps are only half an inch wide, samples a different place each time.

What is happening: the chemistry and physics

Nothing photographic. This is a geometry fault wearing a sensitometric costume, and that is precisely why it is dangerous: it produces a plausible curve.

A sensitometric exposure is only interpretable if the light arriving at the film is the same everywhere the wedge is not, because every density on the plot is attributed to the wedge step above it. Kodak’s workbook states both conditions — a uniform field over the area of the strip, and intimate contact between the modulator and the emulsion. Break either and the density you read is the sum of the step and something else, with no way to tell them apart from a single strip.

The development half is the same argument in chemistry. Development is a diffusion-limited process at the emulsion surface; where fresh developer reaches the film more readily, more silver is reduced. Part IX’s solvent and agitation experiments are where the course measured that, and a control strip is simply a very small negative subject to it.

!The confirming test: rotate the strip

Expose a second strip with the wedge turned end for end, and process it in the same run.

• If the dense end of the film stays at the same end of the instrument, the fault is the instrument or the wedge.
• If the dense end follows the wedge, the wedge is the culprit.
• If neither is consistent and the pattern moves from run to run, the fault is development.

A third check separates exposure from processing completely: expose one strip and cut it in half, processing the halves in different tanks. Kodak’s own first instruction when a point falls outside a control limit is to check the measurement — re-read the strip, then process another — before concluding anything about the process.

Diagnostic questions

• Has the exposure field ever been mapped? Part XIV’s enclosure page calls this the step everybody skips. A field map is a one-off measurement that makes every later reading interpretable.
• Is the wedge in contact, and under what pressure? A glass plate, a spring back, or a vacuum are the three usual answers, and a wedge lying loose is none of them.
• Where in the tank was the strip, and what was the agitation? A strip taped to a wall develops differently from one hanging free.
• How large is the densitometer aperture relative to a step? A 21-step wedge is half an inch wide; an aperture near that size averages over a boundary.
• Does the same fault appear on a strip from a different film? If it does, it is not the film.

Corrective action

Discard the strip. A gradient cannot be corrected out of a curve, because the correction and the signal have the same shape.

Fix the field before you fix anything else. An instrument that is not uniform makes every subsequent diagnosis unreliable, including the ones you would use to check the instrument.

If the fault is development, standardise the agitation as a written script — Part IX’s experimental design lesson is where the course argues for scripts — and process the strip in the same tank position every time.

Prevention

Map the field once and record the map. The calibration records worksheet exists for exactly this kind of measurement.

Give the strip its own tank position and its own place in the agitation script, so that whatever unevenness remains is at least constant.

Read the same place on every step, and record which place. A convention that is written down is a convention you can check.