Project 4: Coating Glass Dry Plates and a Lantern Slide

Purpose

To put an emulsion on a support that absorbs nothing, forgives nothing and lasts indefinitely.

Paper hid three of your mistakes. It soaked up the by-products of an unwashed make, its fibres broke up an uneven coat into something that read as texture, and its own opacity concealed a thin patch. Glass does none of that. A plate shows its coating faults as faults, will not accept an emulsion that was not washed, and demands a bed that is level to a tolerance you cannot see by eye.

In exchange you get the dry plate the nineteenth century built an industry on — Maddox’s gelatino-bromide of 1871, made practical by Bennett’s long low-temperature digestion published in 1878, both of which Abney sets out in Photography with Emulsions — and two objects: a lantern slide, which is a positive transparency made to be looked through, and a stack of camera plates for the pinhole work in Part VI.

Learning objectives

By the end of the session you should be able to:

• specify and source glass for a plate, and say why thickness is a holder question before it is an optical one;
• level a coating bed by a test that is more sensitive than a spirit level, and name the second test that confirms it during processing;
• compute a coating weight for a plate and say why the figure for a lantern slide is not the figure for a camera plate;
• state both published positions on subbing, name the test that decides between them, and explain why the disagreement is not a contradiction;
• process a plate without frilling it, and choose between hardening in the emulsion and hardening in the bath;
• name three things about hand-coated plates that no source this course has read establishes.

Prerequisites

Project 3, which supplies the washed bromide emulsion for the camera plates, and either Project 1 or Project 2 for the lantern slide. The coating lesson owns the coating methods, the coating-weight arithmetic and the hardening ruling; this page performs them and does not re-derive them.

Safety classification

Level B, and the reasons are unusual for this part: the emulsions were made in earlier sessions, so the silver nitrate handling is behind you, and what is left is glass and, if you clean deeply, caustic soda.

What is not a hazard here, and why. Nothing on this page evolves a vapour, and nothing is heated above the 55 °C of a remelt. The chrome alum in the subbing dip, if you use it, is chromium(III): its notified classification is Warning with GHS07 — skin and eye irritation and possible respiratory irritation — with no sensitisation and no carcinogenicity statement notified, and a British exposure limit of 0.5 mg/m³ as chromium against 0.01 mg/m³ with the carcinogen and sensitiser notations for chromium(VI). The classification page holds that ruling once and this page does not restate it in its own words. The potassium alum in a hardening fixer carries no GHS classification at all: 43 of 43 notifiers report that it does not meet the criteria. The coated plates themselves, once dry, are silver halide locked in a gelatin layer, and the hazard they present is a cut edge rather than a chemical one.

Level D, named and not given. Baker’s first cleaning method is potassium dichromate with sulfuric acid added to it — chromic acid, which is chromium(VI). This course publishes no procedure for it at any level. Baker himself calls it very corrosive and offers the alternative in the very next sentence, which is why the exclusion costs nothing.

Hazards

Glass edges. A freshly cut edge is sharp enough to cut through a nitrile glove. Every plate is arrised — its edges dulled with a diamond hand pad — before it is handled again, offcuts go straight into a lidded tin and never into a bin bag, and the work happens over a tray so that a dropped plate breaks somewhere findable. This course has not found a published domestic glass-cutting guidance document for the launch market and therefore quotes none; the control it recommends instead is the one that removes the hazard, which is to buy the glass cut to size.

Hot two per cent caustic soda, if you deep-clean. Sodium hydroxide is corrosive to eyes, skin and respiratory tract, and its own safety card carries the instruction that matters here: never pour water into the substance — always add it slowly to the water. Make it up cold, warm it afterwards, and brush rather than pour. Splash goggles, not spectacles.

Chrome alum powder, once, while the 2 per cent dip is made. The hazard is dust: weigh over a tray in still air, gloves and eye protection while the jar is open, and make up the whole packet at once.

Wet, tender emulsion on a slippery plate. The commonest injury in this session is a plate slipping out of a wet glove into a sink. Work over a tray, and lift plates by two diagonal corners.

Required PPE

Splash goggles, nitrile gloves, apron and closed shoes. For handling clean dry glass before coating, add cotton gloves or handle the plate by its edges alone — this course has not tested whether one glove material transfers less than another, so the control it relies on is that nothing touches the face. A fingerprint on a plate is a coating defect you will not see until the fixer.

Ventilation

Ordinary room ventilation. The controls that matter here are about dust rather than vapour, and they run the other way from extraction: the drying box door stays shut, no fan blows across the bench while plates are wet, and the two powder weighings happen in still air. Baker is blunt about why — dust and bacteria are the plate maker’s enemies, mould spores travel on dust, and a speck on a gelatin-coated plate is an ideal medium for one.

Materials

• Glass, 1/16 inch (about 1.5 to 1.6 mm), bought cut to size. The tested source gets four 4 × 5 inch plates from an 8.5 × 11 inch sheet.
• A 220-mesh diamond hand pad for arrising the edges.
• Plastic food wrap, a sheet of plate glass for the bed, and four dam bars — strips of the same glass, or of acrylic, that fence the plates in.
• A round-bowled tablespoon, dedicated to emulsion.
• Distilled water, lint-free cloth, a lidded tin for offcuts, and a plate rack or an egg box to dry in.

Chemicals

Chemical	Quantity	Form
Your own emulsions	15 mL per 4 × 5 plate	from Projects 1 or 2, and Project 3
Sodium hydroxide	2 % w/v, about 200 mL	deep clean only; made cold, used warm
Chrome alum	2 % w/v, about 300 mL	subbing dip only, if you decide you need one
Potassium alum	15 g/L in Kodak F-5	in the fixing bath, if plates frill

!Three of this project's questions have no published answer, and the page says so rather than filling them

Varnishing finished plates. Nothing this course has read says whether to varnish a hand-coated plate or with what. The honest answer is deferred: plates here are stored in the dark and handled by the edges, not varnished.

The keeping fog rate of a hand-coated plate. Baker gives the industrial keeping test — ten days at 41 °C (105 °F) below 65 per cent relative humidity, with incubated plates showing no more than 0.02 extra fog against controls — and no figure exists for anything coated by hand. Your own dated plates are the only data there will be.

Whiting and alcohol. The manifest names it as a cleaning method, and it is a wet-collodion tradition, but this course has read no published formula or procedure for it. What it can publish is Baker’s caustic soda, and the observation that the tested modern practitioner cleans her glass and coats it directly without saying, in the tutorials this course read, exactly how. So the page gives the method it has and names the one it has not.

Equipment

The levelled bed and the drying box from the coating station, the water bath, a thermometer, trays and tongs, a 20 mL syringe, and the silver-waste container. Add a light box or a window for reading the finished slide by transmission.

Estimated cost

Cost band ££. The glass is the cost, and it is small: an 8.5 × 11 inch sheet of thin glass yields four plates and plates are reusable if they are not scratched. Chrome alum and caustic soda are bought once. This page quotes no prices; dated figures live in the laboratory planner.

Estimated consumables cost

This session consumes no silver of its own: the emulsion comes from Projects 1, 2 or 3 and is already paid for there, at about 15 mL a 4 × 5 plate. What it does consume is glass, plastic wrap and the two cleaning solutions — and plates are reusable if they are not scratched, which makes the glass closer to a slow consumable than to a purchase.

Consumed	This session	Sourced price	Cost this session
Your own emulsion, from Projects 1–3	15 mL a 4 × 5 plate	Costed on the project page that made it	—
Glass, 1.5–1.6 mm, cut to 4 × 5 in	four plates from one 8.5 × 11 in sheet	None. A named price gap: glass plates, 1.5 to 1.6 mm, cut to 4 × 5 in	—
Sodium hydroxide, 2 % w/v	about 200 mL, deep clean only	None. A named price gap: chrome alum, potassium alum and sodium hydroxide	—
Chrome alum, 2 % w/v	about 300 mL of subbing dip, only if needed	None. A named price gap: chrome alum, potassium alum and sodium hydroxide	—
Potassium alum	15 g/L in the hardening fixer, only if plates frill	None. A named price gap: chrome alum, potassium alum and sodium hydroxide	—
Plastic food wrap and dam bars	one bed per session	None. tape-and-adhesives carries a cost band and no dated figure	—
Distilled water and lint-free cloth	about 1 L	None. distilled-water carries a cost band and no dated figure	—

Not one row in that table has a sourced price, so this session has no consumables figure at all — only a list of what it consumes. 6 of the 7 rows carry no dated price at all. A priced entry is a dated range to plan against, never a quotation.

That is the honest state of it: glass plates and the three hardeners and alkalis are two of the planner’s named gaps, and the emulsion is costed on the page that makes it. The diamond hand pad, the bed glass, the tablespoon, the rack and the tin are equipment.

Waste streams

Silver-bearing rinse water and failed coats, as in every project here — including the emulsion cut from between the plates and the dam bars, which is a real quantity.

The caustic soda rinse, alkaline and dilute.

Chromium(III), only if you sub. A 300 mL dip at 2 per cent carries 6 g of chrome alum, which is far more chromium than an in-emulsion dose, and it is reused rather than discarded after each plate.

Broken glass, which is a sharps stream and not a chemical one.

Alternative route

Without silver nitrate, coat the bought liquid emulsion from the practice run onto glass. Everything this page teaches about levelling, coating, drying and processing a plate is unchanged, and the lantern slide works just as well.

Without a levelled bed, do not coat plates. Paper forgives a slope; glass does not, and the fault it produces looks like an emulsion fault.

Preparation

Choosing the glass, which is a holder question first

Thickness. The tested figure is 1/16 inch, about 1.5 to 1.6 mm — thinner than the 2 mm glass a picture framer sells. The reason is mechanical: most period plate holders were made for thin glass, and holders smaller than 4 × 5 inches almost always need it. 4 × 5 holders run either way, with the older ones usually thin; 5 × 7 and larger almost always take 2 mm or 2.5 mm, except the old book-form holders, which take thin glass only. The instruction that follows is practical rather than clever: take the holder to the glass merchant.

Size. Most holders take glass an eighth of an inch under the named format, so a “4 × 5” plate is actually 3⅞ × 4⅞ inches. Cut to the holder, not to the name.

Fit to the Part VI camera. The pinhole camera does not use a plate holder at all: its flat back holds whatever you coat against paper stops, and its own format table lists a student-coated plate from Part V as one of the four options. One consequence belongs in your register rather than in your coating notes. A plate is roughly five times the thickness of a sheet of paper, and whether that moves the emulsion surface depends on which face of the material the back’s stops locate — which is a question about your own back and not one this page can answer for you. So the film-plane register is measured with a plate in the back, and recorded separately from the figure measured with paper.

Reused glass carries a warning. Baker cites Clerc for the suggestion that ultra-microscopic particles, probably of metallic silver in solid solution in the glass, cause the ghost of a previous image to appear in a new picture on reused glass. Whether that survives modern scrutiny this course cannot say; what it can say is that a scratched plate is certainly no good and that the observation is a hundred years old and has not been retested here.

Preparing a plate, in the order the operations have to happen

1. 1Cut or buy cut, to the holder1/16 inch glass, an eighth under the named format; buying it cut removes the sharpest hazard on the page
2. 2Arris the edges220-mesh diamond pad, four edges, under running water; offcuts into the lidded tin
3. 3Cleandetergent and a thorough rinse for a new plate; hot 2 % caustic soda, brushed on and washed off under the tap, for a reused or greasy one
4. 4Final rinse in distilled watertap water leaves what is dissolved in it exactly where the emulsion is about to go
5. 5Sub, or do notthe two published positions are below; the fixing tray is the test that decides between them
6. 6Dry dust-free, and handle only by the edgescotton gloves; a fingerprint is a coating defect you will not see until the fixer

The order is not arbitrary: arrising before cleaning, because it makes glass dust; the distilled rinse last, because it is the only water whose residue you know; and the sub, if used, straight after that rinse with no further rinsing at all.

Levelling, by a test more sensitive than a spirit level

A spirit level gets you started and is not precise enough. The published test is a puddle of water: pour one onto the bed, watch which way it wants to go, adjust, and repeat. When you can pour more water into the puddle and it grows round rather than running, the bed is level.

The second test happens later and is the one that catches a slope you did not find. Fix a coated plate and watch it clear. If one edge or corner clears noticeably later than the rest, the emulsion is thicker there, and the bed was not level. That is a measurement disguised as a processing step, and it is worth doing on your first plate before you coat five more.

Procedure

Stage 1 — The bed (15 minutes)

Moisten the levelled plate glass, lay plastic wrap over it — it will practically vacuum-seal down — and moisten the wrap. Arrange the plates and the dam bars: a 1/16 to 1/8 inch gap between each plate and its neighbours, and the bars tight against each other around the outside. The water under the wrap is what holds everything in place.

Four or six plates at a time is the working maximum. Beyond that the first plate has set before the last is poured.

The coating bed, the stroke, and the cut that separates the plates

1. 1Wrap on wet glass — moisten, lay, moisten again; the water is what holds the wrap and the plates down
2. 2Dam bars, butted — four strips around the group, tight to each other, loose to the plates
3. 3The stroke — spoon held nearly flat, spreading outward from a poured puddle; bubbles chased to the edge
4. 4The cut — when dry, a blade along every gap, so plates are never pulled apart

Drawn from the published practice rather than measured. The single most consequential detail is the one that looks least important: the gap between the plates, which exists so that the dried layer can be cut instead of torn.

Stage 2 — Coat the camera plates (30 minutes)

Remelt the bromide emulsion in the bath, cool it to the coating window, and strain it. Under the safelight, pour 15 mL onto each plate from the syringe and spread it with the bottom of the round-bowled spoon, held nearly flat, working the puddle outward to all four edges and chasing bubbles out to where they can break.

Do not rework a plate. The emulsion sets in seconds at coating temperature, and a second pass across a setting layer leaves a ridge that survives drying, exposure and development.

Stage 3 — Coat the lantern slide (10 minutes)

The lantern slide is the same operation with a different emulsion and a very different target weight, and the arithmetic is worth doing before you pour.

🧮A camera plate and a lantern slide are not the same coat, and the gap is larger than it looks

A 4 × 5 inch plate is 20 in², which is 129 cm², which is 1.29 dm².

The camera plate. Project 3’s washed emulsion carries about 5.6 g of silver halide in a finished 150 to 180 mL, so 31 to 37 g per litre. At 15 mL per plate that is 465 to 555 mg over 1.29 dm² — about 360 to 430 mg of silver halide per square decimetre.

The lantern slide. Baker’s figure for chloride and chlorobromide plates is not more than 25 to 40 mg per square decimetre, that is 2.5 to 4.0 g per square metre. Project 1’s chloride emulsion is about 20 g of silver chloride per litre, so hitting Baker’s band on the same plate needs 1.6 to 2.5 mL — against fifteen.

Six to nine times lighter, for the same plate. That is not a rounding difference, and it is the honest finding of this page: no source this course has read gives a hand-coating method for a layer that thin. Pouring 2 mL onto a 4 × 5 plate and spreading it with a spoon is not a technique anyone has published. What you can do is coat as light as you can lay evenly, weigh the pot before and after, compute the weight you actually achieved, and expect a denser and slower slide than a commercial lantern plate.

And a check that puts the whole part in perspective. Fifteen millilitres over 129 cm² is a wet layer 1.2 mm thick. Duffin gives industrial coatings as 30 to 600 µm wet, drying to 2 to 40 µm — so a hand-coated plate at the tested rate is about twice the top of the industrial wet range, and at his own fifteen-to-one drying ratio it dries to roughly 80 µm, twice his dry ceiling. A paper coat is not like this: Project 3’s 15 mL on an 8 × 10 inch sheet is 290 µm wet and about 20 µm dry, comfortably inside his range. It is the plates that are heavy, and most of what goes wrong with them follows from that one number.

Stage 4 — Set, cut, dry (overnight)

Move the whole bed to the chilled slab, or set the plates on it individually, and let the layer gel. When the plates are dry or nearly dry, cut the emulsion along every gap with a sharp blade so that plates never have to be pulled apart, then prop each plate up to dry the smear on the underside.

Dry in the box with the door shut. Plates take longer than paper and the reason is not the layer’s thickness alone: paper loses water into its own fibres from the moment it is coated, and glass gives water only one way out. In a cool damp room the tested practice is a small radiator holding the drying space at about 24 °C (75 °F), with fabric taped over the air vent so that airflow continues and dust does not.

Stage 5 — Print the lantern slide, and expose the camera plates

The slide is a contact print made on a slow plate: negative and plate emulsion-to-emulsion in the glass-and-board sandwich under a weight, exposed at your fixed lamp and distance, and developed to the sequence Project 1 established. A transparency is read by transmitted light, so judge it on a light box or against a window rather than in the hand.

The camera plates go, wrapped and labelled, to Part VI. Wrap each in clean paper, emulsion side marked, with the version code, coating weight and date on the wrapper.

Expected observations

• The puddle test is startlingly sensitive. A bed that a spirit level calls flat will send a puddle in a definite direction.
• The emulsion sets faster on glass than on paper. Glass conducts heat away; paper does not.
• The first plate is worse than the fourth. Coating is a motor skill, and it is worth coating two plates you intend to throw away.
• Held up to the safelight, a wet chloride coat is white and a bromo-iodide coat is deep yellow.
• In the fixer, an uneven plate clears unevenly, and the last corner to clear is the thickest.
• Pepper, if you have it, may not be visible at all by eye. Wall’s instruction is explicit and worth obeying: test plates should always be examined with an eyepiece, especially in the parts protected from light.

What is happening chemically

Adhesion, which is the whole subbing question. Gelatin sticks to clean glass by hydrogen bonding and by mechanical keying into a surface that is microscopically rough. Both are defeated by grease, which is why cleaning is the operation that decides whether a plate frills. When the layer swells in the developer it grows in every direction and the glass does not, so the layer is in compression against a rigid surface; if the bond at the edges is weaker than that stress, the edge lifts. That is frilling, and it is a mechanical failure of adhesion rather than a chemical fault in the emulsion.

What a sub does. Chromium(III) cross-links gelatin by forming ionic links between carboxyl groups on different chains, and a chrome alum dip leaves a trace of chromium on the glass that ties the first gelatin it meets to the surface. Duffin notes the condition that comes with it: the action is pH-dependent, effective near the usual coating pH of about 6.0 and much less so higher.

Why plates dry differently from paper. Duffin’s numbers are the mechanism: a coating is about 85 per cent water when it goes down and about 5 per cent when it is dry, because the gelatin holds on to 10 to 12 per cent of its own weight and will not give it up. On paper part of that water leaves downward into the fibres. On glass all of it must leave upward, through a layer that is getting more concentrated and less permeable as it goes.

A coated plate in section, drawn twice: with a sub and without

Depths drawn readable rather than to scale; the real figures are in the labels, and the two interface options are alternatives at one position rather than two layers. The three absences matter as much as the layers that are present. Layer depths are drawn to be readable, not to scale: on real film the base is roughly a hundred times the emulsion, and drawn honestly the emulsion would vanish. Any thickness given in the labels is the real one.

⁇Does a hand-coated plate need subbing? Two sources, both credible, and a test that settles it for you

Baker, Tier 1, 1941, page 140. “Substratuming is done by dipping them into a dish containing a two per cent solution of chrome alum after the final rinse and racking them without further rinsing. They should be dried in a perfectly dust-free cupboard.”

The tested modern practitioner, Tier 2, from 2013. Clean 1/16 inch glass, no sub at all, coated directly on a moistened plastic-wrap bed. Her plates process and print.

This is less a contradiction than a difference in what is being asked of the plate. Baker is preparing plates for a commercial coating head and for keeping tests measured in weeks; she is coating by hand and processing soon afterwards. Neither is wrong about their own case.

One correction the course has to make in the open. Part V’s own manifest asked for a “chrome alum gelatin subbing solution”. No source this course has read gives one. Baker’s dip is plain chrome alum in water with no gelatin in it, and the course publishes what the source gives rather than inventing a compound layer to match a plan.

The test that decides it for your glass and your emulsion: fix a plate and watch the edges. If the layer lifts, rolls back or frills, you need either a sub or a hardening bath. If it does not, you do not. Run it on your first two plates before you commit a batch.

Data to record

The batch record fields from Project 3, plus, for each plate: glass thickness and supplier; cleaning method used; subbed or not, and with what; coating temperature, volume and computed coating weight; drying temperature, humidity and hours; the fixing observation — did the plate clear evenly, and if not, which corner was last; and the eyepiece check for pepper.

For the camera plates, add the film-plane register measured with a plate in the back, which belongs in the Part VI register rather than here, and an exposure log entry carrying the version code, so that a negative made in three weeks can be traced to the melt it came from.

Version code the plates BrI-01b if they came from the same melt as Project 3’s paper: a different coating of the same melt takes the next letter, and only a different formula takes a new number.

Analysis

1. What the fixing tray told you

Even clearing means a level bed and an even stroke. Uneven clearing gives you a map of your own coating: the last area to clear is the thickest, and where it sits tells you whether the fault was the bed (a consistent direction across every plate) or the stroke (different on each).

2. The lantern slide, read by transmission

Wall’s account of why these emulsions were used for slides is worth testing against what you are holding: the chloride and bromo-chloride emulsions give an extremely fine-grained image, “more like a stain than the normal silver image”, and warm tones are easy to reach on them. He also records that lantern plates of a pure bromide type were much in favour in England because almost any colour could be obtained by suitable variation of the exposure and developer — which is the same finding Project 2 carried, arriving from the other direction.

If your slide is dense and muddy, the coating weight is the first suspect, not the emulsion, and the arithmetic above says why.

3. Hardening: in the bath, not in the emulsion

The course’s default is no hardener in the emulsion, and the plates are the case that most tempts you to change it. Resist for one reason: a hand coat is thick and uneven, and Duffin states that excessive hardening interferes with developer penetration. A hardened thick patch fixes incompletely and bronzes weeks later, in room light, when the print or plate is finished and put away.

If plates frill, harden in the bath. The recommended route is a potassium-alum acid hardening fixer — Kodak’s F-5, or the F-53 stock made up as F-54a — because potassium alum carries no GHS classification at all and because a fixer is something you are buying anyway. Mix it in the stated order: hypo, then sulfite, then acid, then alum, or the hypo decomposes and the alum precipitates.

Chrome alum baths exist and this course does not reach for them first. Kodak’s SB-3 and SB-4 are 30 g of chrome alum per litre, one to two orders of magnitude more chromium than an in-emulsion dose. SB-4 does supply one thing worth knowing: a fresh bath is violet-blue and keeps indefinitely unused, while a partly used one turns yellow-green within a few days and should be replaced. That is a keeping test you can read with your eyes.

Troubleshooting

What you see	Likely cause	What to do
A ridge across the plate	The stroke stopped and restarted, or the plate was reworked after setting began	One pass, and do not go back. Coat two sacrificial plates first
The layer is thicker at one edge on every plate	The bed is not level	Re-run the puddle test; the fixing tray will confirm it
The layer lifts or rolls back in the fixer	Adhesion: grease, or no sub with a thick layer	Deep clean with hot 2 % caustic soda; fix in an acid hardening fixer; consider the chrome alum dip
Circular clear spots	Air bells trapped under the layer, or dust that repelled the emulsion	Chase bubbles to the edge during the stroke; keep the drying box shut
Black specks in unexposed areas	Pepper, or undissolved halide from the make	Examine with an eyepiece before blaming the coating; the three published remedies are in Project 3
Plates still tacky after a night	Too cool or too humid a drying space	Hold the box near 24 °C with airflow and a fabric dust screen. A plate that dries slowly is also the one that grows mould
The layer tears when plates are separated	They were pulled apart instead of cut	Cut along every gap with a blade when dry. This is why the gaps exist
Fog across the whole plate	A safelight adequate for a chloride paper and not for this emulsion, or an over-ripened make	Re-run the fog test on a coated plate of your own material rather than on a commercial paper

Clean-up

Peel the wrap off the bed with the waste emulsion still on it and put it in the silver-bearing solid waste. Wash the dam bars, the spoon and the syringe in warm water before the gelatin sets. Sweep for glass with a damp paper towel rather than a brush, and empty the offcut tin into a sharps route rather than a bin bag. Wash gloves before removing them.

Storage

Coated plates, dry, emulsion side marked, interleaved so that no two emulsion faces touch, in a lightproof box out of warm damp storage, with the version code and coating weight on the wrapper.

Uncoated glass, cleaned or not, stored dry and separated by paper so the faces do not scratch.

The chrome alum dip and the caustic soda, labelled with substance, strength and date, in closed bottles, stored apart from each other and from anything acid.

What nobody has measured: how long a hand-coated plate keeps. Coat two extra plates, date them, and develop one unexposed now and the other in a month. The difference in base density between them is your own keeping-fog figure, and it is more than the literature offers.

Disposal considerations

Silver-bearing rinse water, failed coats, the emulsion cut from the gaps, used fixer and first wash: all collected, none to a drain, because silver compounds are very toxic to aquatic life with long lasting effects.

Chromium(III), if you subbed, in the spent dip and in the rinse that follows. The waste classification method in the launch market treats compounds of chromium as transition-metal compounds where they are classified as hazardous, and chrome alum is classified — so the assessment is not automatically trivial. This course states the chemistry and names the method rather than computing a threshold, and local regulation governs what may then be done with the collected waste. The option that puts no chromium anywhere is the one the tested practitioner takes: no sub, and no hardener.

Broken and offcut glass is a sharps stream, boxed and labelled, not a chemical one.

Questions

1. A holder measures 3⅞ × 4⅞ inches inside and takes glass up to 1.7 mm. Say what you would order and why the second figure matters more than the first.
2. Compute the coating volume that puts Baker’s 25 to 40 mg/dm² on a 5 × 7 inch plate using an emulsion carrying 39 g of silver halide per litre. Then say what practical problem your answer creates.
3. A plate clears evenly in the fixer but its layer lifts at one corner. Which of the two faults does each observation rule out, and what do you change first?
4. Explain, from the water content of a coating, why a plate takes longer to dry than a sheet of paper coated at the same weight, and name one consequence of that for mould.
5. State both published positions on subbing, say what each source was preparing plates for, and describe the test that decides which applies to you.
6. Your camera plate at 15 mL puts down about 400 mg/dm². A commercial bromide paper was coated at 23 to 31. List three things you would expect to be different about your material, and say which of them you could measure with a step wedge alone.

Further experiments

The subbing pair. Coat four plates from one melt: two on clean glass, two on glass dipped in 2 per cent chrome alum and racked without rinsing. Process all four in the same fixer and record which edges lift. That is a one-variable test of a hundred-year-old disagreement, and it takes one evening.

The coating-weight series, on glass. Five plates from one melt at 4, 8, 15, 20 and 25 mL, each weight computed, all exposed identically. Maximum density against coating weight, measured on your own material, is a curve nobody has published for a hand-coated plate.

Baker’s keeping test, at home. Seal four coated plates in a box with a hygrometer, hold them at about 41 °C for ten days below 65 per cent relative humidity, and develop them unexposed against four controls kept cold. Baker’s industrial criterion is no more than 0.02 extra fog. You will not have a densitometer, but a side-by-side comparison against a control still answers the question that matters: is the fog visible?

Recap

• Glass is chosen by thickness first, because it is a holder question: 1/16 inch, an eighth of an inch under the named format, and the Part VI camera’s flat back takes whatever you coat.
• Level the bed with a puddle of water, and confirm it in the fixing tray by watching where the plate clears last.
• 15 mL on a 4 × 5 plate is about 360 to 430 mg of silver halide per square decimetre, a wet layer of about 1.2 mm — roughly twice the industrial range at both the wet and the dry end. A lantern slide at Baker’s 25 to 40 mg/dm² needs 1.6 to 2.5 mL, and no source gives a hand method for a layer that thin.
• The two sources disagree about subbing and both are right about their own case; the fixing tray decides yours. There is no chrome alum gelatin sub in any source this course has read.
• Harden in the bath, not in the emulsion, with a potassium-alum acid hardening fixer, because a hardened thick patch fixes incompletely and bronzes later.
• Not established, and said so: varnishing, the keeping fog rate of a hand-coated plate, and any published method for whiting-and-alcohol cleaning.

Check your understanding

Question 1. You have an emulsion carrying 20 g of silver chloride per litre and a 4 × 5 inch plate. What volume puts Baker's stated coating weight for a chloride plate onto it, and what does the answer tell you?

• About 15 mL, which is the standard plate volume
• About 1.6 to 2.5 mL, which is six to nine times less than the tested plate rate and is thinner than any published hand method can lay
• About 8 mL, half the plate rate
• It cannot be computed without knowing the gelatin concentration

Show the answer and why

Answer: About 1.6 to 2.5 mL, which is six to nine times less than the tested plate rate and is thinner than any published hand method can lay

A 4 × 5 inch plate is 20 square inches, which is 129 square centimetres or 1.29 square decimetres. Baker's band for a chloride or chlorobromide plate is 25 to 40 milligrams of silver halide per square decimetre, so the plate wants 32 to 52 milligrams, and at 20 milligrams per millilitre that is 1.6 to 2.5 millilitres. The tested hand rate is fifteen. The gap is the honest finding of the page: a transparency wants a very thin film, hand coating lays a heavy one, and nobody has published a hand method for the light coat. What you do about it is measure what you actually achieved and expect a denser slide, not pretend the numbers agree.

Question 2. A plate frills at one edge in the fixer, and the same plate cleared perfectly evenly. What does the second observation rule out?

• Nothing; the two are the same fault
• An unlevel bed and an uneven stroke, because both would have produced a thickness variation that showed as uneven clearing — so the fault is adhesion at the interface, not thickness
• A dirty plate, since dirt would show in clearing
• The emulsion, which must be at fault

Show the answer and why

Answer: An unlevel bed and an uneven stroke, because both would have produced a thickness variation that showed as uneven clearing — so the fault is adhesion at the interface, not thickness

Even clearing is a statement about thickness: the fixer works through a thick layer more slowly, so a plate that clears evenly was coated evenly. That leaves the interface. Frilling is a mechanical failure of adhesion — the layer swells in every direction against a rigid support that does not, and lifts where the bond is weakest — so the things to change are the clean, the sub, or the hardening, in that order of cost. This pair of observations is a good example of what a break-fix page calls a discriminating test: one symptom narrows the field and the other closes it.

Question 3. Why does this course recommend hardening a frilling plate in the fixing bath rather than adding chrome alum to the emulsion?

• Because chrome alum does not work in an emulsion
• Because a hand coat is thick and uneven, and hardening interferes with developer and fixer penetration, so a hardened thick patch fixes incompletely and bronzes later; and because potassium alum carries no GHS classification at all while chrome alum carries three irritation statements
• Because chrome alum is a chromium(VI) compound
• Because the emulsion is already hardened by the gelatin

Show the answer and why

Answer: Because a hand coat is thick and uneven, and hardening interferes with developer and fixer penetration, so a hardened thick patch fixes incompletely and bronzes later; and because potassium alum carries no GHS classification at all while chrome alum carries three irritation statements

Two reasons, and the first is photographic. Duffin states that excessive hardening interferes with developer penetration, and hand coating produces exactly the thick uneven patches that make that bite. The second is a hazard comparison: potassium alum is reported as not meeting GHS criteria by every notifier, where chrome alum carries skin, eye and respiratory irritation statements — and you are buying a fixer regardless. Note the third option's error, which is the one the course's classification page exists to prevent: chrome alum is chromium(III), a different substance from the chromium(VI) of dichromate, with a fifty-fold more permissive exposure limit and neither a carcinogenicity nor a sensitisation notation.

Question 4. The manifest for this project asked for a "chrome alum gelatin subbing solution". What does the page do?

• Gives a formula compounded from the chrome alum dip and a gelatin solution
• Publishes the plain 2 per cent chrome alum dip that the source actually gives, states that no chrome-alum-and-gelatin sub was found in any source read, and says that both the industrial dip and a tested practice of no sub at all exist
• Omits subbing entirely
• Recommends a commercial subbing product

Show the answer and why

Answer: Publishes the plain 2 per cent chrome alum dip that the source actually gives, states that no chrome-alum-and-gelatin sub was found in any source read, and says that both the industrial dip and a tested practice of no sub at all exist

Rule 6 does not allow a formula to be assembled from parts to match a plan. Baker's page 140 gives a dip into a two per cent solution of chrome alum after the final rinse, racked without further rinsing; there is no gelatin in it. Compound subbing layers carrying gelatin certainly exist in industry — that is what a polyester film base needs — but this course has read no formula for one, so it prints none and says so. The page is better for the gap being visible: the reader now knows there are two published positions and a test that decides between them, rather than one invented recipe.

Question 5. Fifteen millilitres of emulsion on a 4 × 5 inch plate is a wet layer of about 1.2 mm. Why is that number worth carrying?

• It is not; only the dry thickness matters
• Because it is about twice the top of the industrial wet range of 30 to 600 micrometres, and most of what goes wrong with hand-coated plates — long drying, thick patches, incomplete fixing, bronzing after hardening — follows from the layer being that heavy
• Because it tells you the exposure time
• Because it sets the safelight requirement

Show the answer and why

Answer: Because it is about twice the top of the industrial wet range of 30 to 600 micrometres, and most of what goes wrong with hand-coated plates — long drying, thick patches, incomplete fixing, bronzing after hardening — follows from the layer being that heavy

Divide 15 cubic centimetres by 129 square centimetres and you get 0.116 cm, which is 1.16 millimetres of liquid standing on the plate — held there by the dam bars and by surface tension. Duffin's industrial figures are 30 to 600 micrometres wet drying to 2 to 40 dry, so the hand coat is off the top of his scale at both ends. That single number explains the long drying, the tendency to thick patches, the incomplete fixing of a hardened plate and the higher maximum density a hand-coated plate can reach. It is also the sharpest contrast in the part: a paper coat at 290 micrometres wet sits comfortably inside the industrial range, and it is the plates that are heavy.

Check my answersClear