Break/Fix: The Exhausted Fixer

The bath you spent in the capacity experiment is the raw material for this session, so do not pour it away when that afternoon ends. Keep the last hundred millilitres, labelled and dated, and this page turns it into four samples and a diagnosis.

The situation

Four envelopes, marked A, B, C and D, each holding one strip of processed film and one small print. They were made three weeks ago and the key is in a fifth envelope you have not opened. One set was processed properly. The other three were broken, one way each, and the breaks were chosen because they look alike at first glance and behave completely differently when you push on them.

Your job is to rank them by how badly they were fixed, name the mechanism in each, say which can be rescued and which cannot, and name the one test that would have prevented all of it. Then you open the fifth envelope.

📓Making the four sets, which happens three weeks before the session

This is the preparation and it is half the teaching. Work from the mixing lab’s three baths and the loaded bath the capacity experiment left you, and measure the clearing time of each before you use it, because those numbers are the answers in the sealed envelope.

Set A, the control. Fresh rapid fixer at 1+4. Fix for twice its measured clearing time. Rinse and wash as you normally would.

Set B, mildly under-fixed. The bath at its crossing point — the one whose clearing time had just reached twice the fresh figure. Fix for twice the fresh clearing time, which is the mistake everybody makes: using the time that was right when the bath was new. That is about half the time this bath now needs.

Set C, grossly under-fixed. The same bath, worked well past the crossing, with the material given half the fresh time. This is the sample that will look wrong immediately, and its job is to calibrate your eye for B.

Set D, the dichroic route. Take material straight from the developer into the exhausted bath with no rinse and no stop bath, at 24 °C or above if you can manage it, and fix it for the full time. Kodak’s 1928 primer names exactly those conditions as the ones under which dichroic fog appears — a bath that is old, exhausted and carrying dissolved silver, with unrinsed developer going into it — and names the opposite conditions as the ones in which it never does.

The course cannot guarantee that D works. Wall’s 1912 dictionary already called dichroic fog “comparatively rare” and no source in the corpus gives a reliable recipe for provoking it. If your D comes out as merely another under-fixed sample, that is a result: report it, treat D as a second B, and read the diagnosis section for what you would have seen.

Label the envelopes in a random order somebody else assigns, or assign it yourself and seal the key. Then leave the four sets three weeks in an ordinary room, not in a dark drawer, because one of the symptoms takes time.

Before you start

Level A on the course rubric. On the day of the diagnosis there is almost no chemistry at all: a light box, a loupe, and one beaker of fresh fixer for the rescue. The preparation session three weeks earlier handles the same baths as the capacity experiment, at the same level and with the same controls — gloves, eye protection, an apron, and general ventilation at HSE’s standard of more than five air changes an hour with a through draught.

What is not a hazard here, and why. The samples are dry processed film and paper. Whatever is wrong with them chemically — retained halide, retained silver complexes, finely divided silver — is present in milligrams, locked in a hardened gelatin layer a few micrometres thick, and is not going anywhere. A badly fixed negative is a permanence problem and not a health one, and confusing the two is a habit worth breaking early: the same silver in the fixer bottle is a waste stream that must not reach a drain, and the same thiosulfate in a bottle beside an acid is a sulfur dioxide problem. In a dry negative in a sleeve it is neither.

The one test this page does not run. The iodide hypo check and the sodium sulphide residual-silver test both need reagents that put a session at Level B, and the second is Part XII’s laboratory. This page reaches its diagnosis with a light box and a stopwatch, which is the point: the diagnosis does not need the reagents, and the confirmation does.

Alternative route

The print half of every set needs a darkroom to make. If you have none yet — the printing parts are a long way ahead — run the session on film alone. Everything in the symptoms, the evidence and the diagnosis works on a strip of 35 mm, and film shows the milkiness more clearly than paper does because you can look through it. The print half exists because two of the four faults are worse on paper than on film, and the page says which and why, so you lose an illustration rather than an argument.

The symptoms

Read these before you look at the samples, so that you are looking for named things rather than for “something wrong”.

On film, looked at against a light box.

• Milkiness, opalescence, a faint cloudy veil across the clear areas. This is undissolved silver halide. Kodak’s 1928 primer defines the endpoint of fixing in exactly these words — the time for the milkiness or opalescence of the unreduced silver salts to disappear — so its presence is the definition of under-fixed. It is easiest to see in the rebates and the clear film between frames, which ILFORD’s own guide says should be “almost clear” in a good negative.
• A magenta or pink stain, even across the frame. Kodak names this on every T-Max sheet: a fixer near exhaustion or a time too short. Note the qualification the sheet attaches — a slight stain does not affect image stability, contrast or printing time, and can be removed with a hypo clearing agent; a pronounced and irregular stain is the one that means refix.
• A two-coloured veil: greenish or yellowish looking at the film, reddish-pink or violet looking through it. That is dichroic fog, and the two-colour behaviour is where the name comes from. Kodak’s 1928 primer and Wall’s 1912 dictionary describe it identically, thirteen years and an ocean apart.

On paper.

• A brown or yellow-brown general stain, present from the start. This is oxidised developer, carried into a bath with too little sulfite or too little acid left to deal with it. Kodak’s primer is explicit that this is what the sulfite in a fixer is there to prevent.
• A yellowing that was not there when the print dried, strongest in the borders and the highlights. This is the slow one, it is the reason for the three-week wait, and it is the most important symptom on the page.
• Bronzing — a metallic sheen in the deepest blacks, seen at a glancing angle.

⚗Why borders and highlights, and why later rather than sooner

The white parts of a print are where development reduced almost nothing, so they are where the most undeveloped halide was, so they are where the most silver-thiosulfate complex was formed. If the bath was spent, that complex formed as the sparingly soluble step-one species, in the paper, where — Kodak’s toning manual says it flatly — washing cannot remove it.

It is invisible when it forms. What makes it visible is sulfide, from the air over months, or from a toner in one minute:

2 [Ag(S₂O₃)₂]³⁻ + S²⁻ → Ag₂S + 4 S₂O₃²⁻

The retained complex becoming a stain

So the geography of the stain is a map of where the halide was, and the delay is the time it takes for something to come and find it. A print that looks perfect on the drying rack and yellow at its edges two years later was already faulty on the rack, and the fault was invisible by construction.

Evidence to collect

Work through all four sets before forming a view on any of them. The order matters, because the cheap observations constrain the expensive ones.

1. Transmitted light, film, rebates first. Light box, loupe, and the clear film between the frames. Rank the four for milkiness. Compare against a piece of the same film that was never exposed and never processed, which is the maximum milkiness, and against set A, which should be the minimum.
2. Reflected light, film, at a glancing angle over a dark background. You are looking for a greenish or yellowish sheen that was not visible by transmission. Then look through the same area again. A veil that changes colour between the two views is dichroic and nothing else is.
3. Reflected light, prints, in daylight. Borders and highlights. Compare each print with set A’s laid beside it — a stain is very hard to see alone and obvious in a pair.
4. Glancing light, prints, on the blacks. Bronzing.
5. The bath, if you kept it. Measure the clearing time of each bath you used, on a fresh scrap, and express it as a ratio to the fresh figure you recorded three weeks ago. This is the measurement that turns your ranking into a diagnosis.
6. The log. Look up how much work each bath had done, in films or 8 × 10 equivalents, from the capacity sheet. If there is no log, that is itself the finding.

✕The three-week wait is doing something, and shortcutting it costs the point of the page

It is tempting to make the samples on the morning of the session. Do not. Two of the symptoms — the yellowing of the borders and highlights, and the bronzing — are slow, and a page about a fault whose whole character is that it announces itself late cannot demonstrate that in an afternoon.

If you did make them the same day, run the session anyway and record honestly that the late symptoms could not be assessed. Then keep the samples and look at them again in three months, which is a better experiment than the one you meant to do.

Hints

Take these in order and stop as soon as you have it.

Hint 1. Three of the four faults are the same fault at different severities. One is not. Find the odd one before you rank the rest.

Hint 2. The odd one out is identified by a single observation that takes ten seconds and needs no equipment beyond a lamp: look at the film, then look through it.

Hint 3. Milkiness in the rebates is undissolved halide, and undissolved halide is a quantity. Rank it, do not classify it.

Hint 4. For the print samples, one kind of stain was there when the print was dry and one was not. If you photographed the prints when you made them — and you should have — the two are trivially separable. If you did not, ask which stain is strongest in the whitest parts.

Hint 5. Wall’s dictionary contains the discriminator that settles the odd one out completely, and it is a statement about what does not work: the plate “looks foggy and suggests that it has not been properly fixed, but prolonged immersion in hypo solution does not mend matters”.

Hint 6. Every one of these faults is a fixer that had done too much work, except in one case where it is a fixer that had the wrong thing put into it. The capacity log distinguishes them, and so does the clearing time of the bath.

The diagnosis

From symptom to mechanism to whether it can be rescued

1. 1Milky clear areas? — undissolved silver halide — the definition of under-fixed
2. 2Does the veil change colour? — greenish-yellow by reflection, reddish-pink by transmission = dichroic fog, and nothing else is
3. 3Dichroic fog — no rescue — finely divided silver metal. Refixing does not touch it; Wall says so in 1912
4. 4Under-fixed — rescuable — refix in fresh fixer for twice its clearing time, then wash in full
5. 5Print stain, from day one — oxidised developer in a bath with no reserve left. Not removable
6. 6Print stain, appearing later — retained step-one complex sulfiding. Not removable, and it was invisible when it formed
7. 7Prevention, which covers all four — a log, the clearing-time test on a schedule, two baths, and a rinse before the fixer

The tree's one genuinely diagnostic branch is the colour question at node 2: it separates a fault that answers to fresh fixer from one that does not, and it costs ten seconds.

Sets B and C: under-fixing, in two doses

What you see. Milkiness in the rebates and the clear film, worse on C than on B; possibly a magenta or pink cast; on the prints, no immediate stain but, after three weeks, a yellowing that starts in the borders.

The mechanism, in two parts. The first is trivial: silver halide that was never dissolved is still silver halide, it is still light-sensitive, and it scatters light, which is what the milkiness is. The second is the one that matters, and it is Kodak’s 1924 statement working exactly as advertised. As a bath loses fixing power, silver leaves the grain as the step-one mono complex more and more often, because the free thiosulfate that would carry it to step two is not there. That species is “almost insoluble in water”, it is invisible, and a material moved to the wash while it is present keeps it.

What separates B from C is only how much, and that is worth saying because it is why B is dangerous and C is not. C announces itself. B looks like a slightly flat negative and prints perfectly well, and its fault declares itself in the third year.

Set D: dichroic fog, if you got it

What you see. A veil that is greenish or yellowish by reflected light and reddish-pink or violet by transmitted light. Both Kodak in 1928 and Wall in 1912 describe it in those terms, and the two-colour behaviour is the whole diagnosis.

The mechanism. Not undissolved halide at all: finely divided metallic silver, deposited in the emulsion. Kodak’s primer says so directly when discussing its removal — dichroic fog “consists of very finely divided silver”. Wall reports Lumière and Seyewetz’s conclusion that it is due to the presence of silver and can arise in development or in fixing, that its formation is favoured by a silver-bromide solvent in the developer and by developer carried into the fixing bath, and draws the practical inference: hence the desirability of washing the plate between developing and fixing.

Put those together and the picture is coherent. A fixer dissolves silver halide into a complex. Carried-over developer is a reducing agent. Where the two meet inside the gelatin, some of the dissolved silver is reduced back to metal before it can diffuse out — physical development, happening in the fixer, in the wrong place. An old bath already carrying dissolved silver has far more to work with, which is why Kodak’s condition is a bath that is “old and exhausted and contains an excess of dissolved silver salts”, and why the prevention is a rinse or a stop bath.

The discriminator, and it is decisive. Wall: the plate “looks foggy and suggests that it has not been properly fixed, but prolonged immersion in hypo solution does not mend matters”. Of course it does not. Metallic silver is not silver halide, and a fixer has nothing to say to it.

!What the course does not give you, and why

Kodak’s 1928 primer states that dichroic fog can be removed with a plain permanganate solution of about 0.25 per cent, which attacks the finely divided fog silver and has “no appreciable action on the silver of the image”. Wall gives Lumière and Seyewetz’s ammonium persulphate route, acidulated with sulphuric acid and followed by a sulphite bath, and calls it of special value where the fog formed in the fixing bath.

The course records both and prescribes neither. Potassium permanganate has no entry in the chemical encyclopaedia and no hazard record anywhere in the corpus, and the persulphate route requires a concentrated mineral acid, which is Level B work at best. Rule 5 puts safety above historical authenticity, and a reducer whose reagents the course has not assessed is not something it will hand a reader as a procedure. The honest position is the one above: the fault is identifiable, the historical treatments exist, and the course’s answer to dichroic fog is prevention, which costs a thirty-second rinse.

The prints, and the two stains that share a word

A brown general stain present from the first day is oxidised developer. Kodak’s primer explains the whole chain in one paragraph: material goes from developer to fixer with little rinsing, the developer oxidises in the bath, the bath turns brown, and it stains negatives and prints. The sulfite is what prevents it, and an exhausted bath is one whose sulfite has been used up and whose acid has been neutralised.

A yellowing that appears later, in the borders and highlights, is the retained complex sulfiding. Kodak’s G-23 gives both halves: an exhausted bath leaves insoluble silver compounds that remain in prints and cannot be removed by washing, and when they meet a toner they form a dark yellow stain “especially noticeable in print borders and highlights”. A toner does in a minute what the air does in years.

They share the word “stain” and they are not the same fault: one is a developer problem in the fixer and the other is a fixer problem in the paper. The timing separates them, and so does the geography — oxidised developer stains generally, retained silver stains where the halide was.

The fix

Under-fixed film: rescuable, and worth doing today. Kodak’s instruction, standing on both the T-Max sheet and the current processing chart, is exactly this: where the stain is pronounced and irregular, refix the film in fresh fixer. ILFORD’s version for prints is more explicit about the sequence — soak in water for five minutes, then repeat the recommended fixing and washing sequence in fresh fixer.

Refixing, in full:

1. Measure the clearing time of the fresh bath first, on a scrap of the same film. You are about to use it as an instrument and you need its zero.
2. Soak the material in plain water for five minutes. It has been dry; the gelatin has to swell before anything can diffuse into it.
3. Fix for twice the fresh clearing time, agitating properly.
4. Wash in full, as though it had never been washed — because as far as the retained compounds are concerned it has not.
5. Dry, and look again by transmitted light. The milkiness should be gone.

What refixing does and does not repair. It dissolves the halide that was never dissolved, and it converts the retained step-one complex into the step-two complex, which the wash can then remove. Those are the two faults it was made for. It does nothing whatever for silver that has already been reduced to metal (dichroic fog), for a stain already formed in a print, or for a negative that has been sitting in a sleeve for two years while the retained compound sulfided — by then the silver has moved and there is nothing to redissolve.

Under-fixed prints: rescuable, with a caution. ILFORD’s procedure works, and the caution is Kodak’s from the other direction — a fibre-base print has an upper bound on total fixing time as well as a lower one, because prolonged fixing drives solution into the paper base. A rescued print has now been fixed twice. Wash it as though its life depended on it, because in the literal sense it does.

Dichroic fog: not rescuable in this course. See above.

Both print stains: not rescuable. They are not deposits on the surface; they are in the paper.

📷The judgement call, which is the point of a break/fix page

Set B will print. That is the difficulty. A mildly under-fixed negative has no visible defect on a work print and no measurable loss of contrast, and the fault it carries is a promise about the future rather than a statement about the present.

So the decision is not technical. It is: is this negative worth refixing today, on the strength of an argument about what it will look like in twenty years, when it looks fine now? For a test frame, no. For anything you would be sorry to lose, yes, and the cost is fifteen minutes and a litre of fresh fixer. That asymmetry — cheap now, impossible later — is the whole economics of permanence, and it recurs in Part XII at every step.

What to record

• Your ranking of the four sets before you opened the key, with the observation that produced each placement.
• The key, and where your ranking was wrong. Write down the wrong answer, not just the right one; a diagnosis you got wrong is worth more in the notebook than four you got right.
• For each set: the bath used, its clearing time when the samples were made, its ratio to the fresh figure, and the fixing time actually given.
• Which sets showed the late symptoms, and how long after processing you first saw them.
• Whether set D produced dichroic fog, and if not, what it produced instead.
• The refixing record for anything you rescued: fresh clearing time, time given, wash, and the appearance before and after.
• The one line that matters for the rest of the course: the capacity figure your own log now supports for your own fixer, and the date.

Prevention, which is four habits and no equipment

Keep the log. The running total is the early warning; the clearing time is the confirmation. Neither works alone, because the curve is flat for two thirds of a bath’s life and the log is the only thing that tells you where on it you are.

Test on a schedule, not on suspicion. Measure the clearing time every quarter of the published capacity. By the time something looks wrong, set B has already happened.

Use two baths for anything you care about. Kodak’s 1924 reasoning is the one to remember: the rotation exists to ensure that nothing leaves the fixer while the invisible insoluble compound is still in it.

Rinse before the fixer. A stop bath is better, and Part X costed it, but even plain water removes most of the developer that causes dichroic fog and most of the alkali that spends the fixer’s reserve. Kodak’s 1928 primer names an unrinsed film as one of the two conditions under which dichroic fog appears, and a rinsed one as a condition under which it never does.

Check your understanding

Question 1. A negative shows a veil that is greenish-yellow when you look at it and reddish-pink when you look through it. You refix it in fresh fixer for twice the clearing time and nothing changes. What is it, and why did refixing fail?

• Under-fixing; the fresh bath was itself exhausted
• Dichroic fog; the veil is finely divided metallic silver rather than silver halide, so a fixer has nothing to act on — Wall recorded in 1912 that prolonged immersion in hypo does not mend it
• Residual thiosulfate; it needs a longer wash rather than more fixing
• A sensitising dye stain; it will come out with a hypo clearing agent

Show the answer and why

Answer: Dichroic fog; the veil is finely divided metallic silver rather than silver halide, so a fixer has nothing to act on — Wall recorded in 1912 that prolonged immersion in hypo does not mend it

The two-colour behaviour is diagnostic and takes ten seconds to check, which is why it is the first branch of the decision tree. The mechanism is a silver solvent and a reducing agent meeting inside the gelatin: carried-over developer reduces some of the dissolved silver back to metal before it can diffuse out — physical development happening in the fixer. Kodak's primer names the conditions, an old bath carrying dissolved silver with unrinsed developer going into it, and names the prevention, a rinse and a fresh acid bath at a controlled temperature. The failure of refixing is not a disappointment; it is the confirmation of the diagnosis.

Question 2. Why does a print fixed in an exhausted bath look perfect on the drying rack and yellow in its borders two years later?

• The paper base yellows with age regardless of processing
• Because the fault at the moment of fixing is a sparingly soluble, colourless, step-one silver thiosulfate complex retained in the paper, which washing cannot remove and which becomes visible only when sulfide converts it to silver sulfide
• Because residual thiosulfate slowly oxidises to sulfate, which is yellow
• Because the developer continues to work in the paper base

Show the answer and why

Answer: Because the fault at the moment of fixing is a sparingly soluble, colourless, step-one silver thiosulfate complex retained in the paper, which washing cannot remove and which becomes visible only when sulfide converts it to silver sulfide

Kodak's 1924 primer gives the pair of compounds — one almost insoluble, one very soluble, and only the soluble one formed while the bath still has fixing power — and adds the sentence that makes it a permanence problem rather than an appearance one: the first insoluble compound is invisible. Kodak's toning manual gives the consequence: it remains in prints, cannot be removed completely by washing, and forms a dark yellow stain in borders and highlights when it meets a toner. The borders and highlights are where the most undeveloped halide was, so the geography of the stain is a map of where the fault was worst. Toning a scrap is the accelerated version of the same test.

Question 3. Set B was fixed in a bath at its crossing point, for twice the clearing time the bath had when it was fresh. Why is that the most instructive of the four faults?

• Because it is the most severe
• Because it is invisible — the negative prints normally and shows no measurable loss of contrast — so it is the fault a darkroom actually commits, and the only defence against it is a measurement made on a schedule rather than in response to a symptom
• Because it is the only one that can be rescued
• Because it is caused by the operator rather than by the chemistry

Show the answer and why

Answer: Because it is invisible — the negative prints normally and shows no measurable loss of contrast — so it is the fault a darkroom actually commits, and the only defence against it is a measurement made on a schedule rather than in response to a symptom

Set C teaches you what the symptom looks like; set B teaches you that the symptom is usually absent. Using the fixing time that was right when the bath was new is the specific error the doubling rule exists to prevent, and it is easy to make because nothing about the bath announces that its clearing time has changed — the exhaustion curve is flat for two thirds of its length. Everything else on this page follows: the log, the scheduled test, the two-bath rotation. Note also that B is rescuable today and not in two years, once the retained complex has sulfided and there is nothing left to redissolve.

Question 4. A print shows a brown general stain and it was there on the day it dried. Which fault is that, and can refixing help?

• Retained silver complexes; refixing helps if done within a week
• Oxidised developer carried into a bath whose sulfite and acid reserve were spent, which stains generally rather than in the borders; refixing does not help, because the stain is already in the paper
• Dichroic fog on paper; refixing helps
• Under-fixing; refixing removes it completely

Show the answer and why

Answer: Oxidised developer carried into a bath whose sulfite and acid reserve were spent, which stains generally rather than in the borders; refixing does not help, because the stain is already in the paper

Two stains share a word and are separated by timing and by geography. Kodak's primer sets out the chain for this one: material carried from developer to fixer with little rinsing, developer oxidising in the bath, the bath going brown, and the brown staining what passes through it — which is exactly what the sulfite is in a fixer to prevent, and why an exhausted bath is one whose sulfite has been consumed and whose acid has been neutralised. It stains generally, where retained silver stains where the halide was. Neither is removable once in the paper, which is why the prevention section is longer than the fix section.

Question 5. You have no capacity log, and you want to know whether your fixer is fit for a set of negatives you care about. What is the cheapest sufficient test, and what does it cost?

• A silver estimator strip; it costs pennies and reads down to the level that matters
• The clearing-time test — a scrap of undeveloped film, a drop of the bath, a stopwatch and the fresh figure you recorded when you mixed it; it costs a few square centimetres of scrap and two minutes, and it is the only one of the tests on this page that stays at Level A
• The sodium sulphide residual-silver test; it is the definitive answer
• There is no test; a bath older than a month should simply be discarded

Show the answer and why

Answer: The clearing-time test — a scrap of undeveloped film, a drop of the bath, a stopwatch and the fresh figure you recorded when you mixed it; it costs a few square centimetres of scrap and two minutes, and it is the only one of the tests on this page that stays at Level A

The clearing-time test needs no reagent, works in room light, and answers the question the doubling criterion is phrased in. Its one requirement is that you recorded the fresh figure when you mixed the bath, which is why the mixing lab makes that the last step. The silver estimator strip cannot help here: Kodak's own publication puts its useful floor at 1 g/L, above the 0.5 g/L that matters for fibre-base permanence. The sodium sulphide test is definitive and is a Level B laboratory in Part XII, because the reagent releases hydrogen sulphide on contact with acid. Discarding by the calendar is not wrong, but it throws away good fixer and does not catch a bath that was overworked in a single heavy session.

Check my answersClear