Build an Emulsion Coating Station

Purpose

To build the five pieces of apparatus every make in this part depends on — a levelled coating bed, a chilled setting slab, a water bath you can hold at a stated temperature, a drying box that excludes dust while still moving air, and a set of coating tools — and then to prove each of them with a test before a gram of silver nitrate is weighed.

The building is quick. Most of it is glass, tape, a plastic box and a pan of water, and the whole station can be made without cutting a single piece of wood. The part that decides whether the rest of this part works is the last half hour, where you level the bed by watching water, log the bath for thirty minutes, coat a sheet with plain gelatin and count what fell on it, and run a safelight fog test. Four measurements, four numbers in the notebook, and from then on every fault you meet has one fewer place to hide.

That is the whole argument for building the station first. When a coated sheet comes out wedge-shaped, speckled or streaked, the question is always emulsion or apparatus? — and a station with four dated test results attached to it answers half of that question before you ask it.

Learning objectives

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

• lay out a coating station in a darkenable room so that wet work, dry work and the safelight are in fixed positions and a mains cable never crosses a wet surface;
• level a coating bed by the water-puddle method, say why a bubble level is not sufficient on its own, and name the test that finally settles it;
• hold a water bath inside a stated band for half an hour, and produce the log that shows it;
• build a drying enclosure that keeps dust off a wet gelatin layer without stopping the air movement that dries it, and explain why those two requirements pull against each other;
• choose a safelight class for an undyed emulsion, state what changes when a sensitising dye is added, and run the fog test that decides whether yours passes;
• coat a sheet of plain gelatin and read it as evidence about the station rather than about the emulsion;
• list every substitution you made for cost or space and state what each one might cost you in coating quality.

Prerequisites

Coating, drying and hardening, which is where the coating temperatures, the coating-weight arithmetic and the hardening ruling come from. This page builds the apparatus that lesson describes; it does not re-argue it.

Laboratory layout and workflow from Part II, for the wet-side and dry-side discipline the station plan below applies to a new job.

The latent image made visible, whose Station 4 is the safelight fog test this page runs again on the finished station. You need its method and its verdict language, not a summary of them.

Safety classification

Level A. The hazards here are a hot water bath, glass with cut edges, a knife, and a mains appliance standing near water. Every one is controlled by ordinary domestic care plus eye protection, and none of them needs a control that the safety classification reserves for a higher level.

What is not a hazard here, and why. This page handles no photographic chemistry. The only substance on the bench is gelatin in warm water, for the validation coat: no silver salt, no developer, no fixer, no acid, no hardener. There is therefore no splash of a corrosive liquid to protect against, no vapour that would call for extraction, no glove-selection problem, and no silver-bearing waste stream — the first of those arrives with Project 1 and not before. Saying so plainly matters, because after four parts of chemical discipline the reflex is to import all of it wholesale, and importing controls that do not apply spends attention the real hazards need. Warm gelatin in a beaker behaves like warm gravy; the same gelatin becomes part of a Level B procedure the moment silver nitrate joins it, and that is a property of the procedure, not of the gelatin.

Hazards

Hot water, and hot glassware. The bath runs at up to about 55 °C for later sessions and a jar lifted out of it holds that heat. Lift with a cloth, put it down on a board rather than on cold glass, and keep the pan handle turned inwards. For a scald, NHS first-aid guidance is cool running water for 20 minutes, as soon as possible, and Part II’s laboratory safety and PPE carries the full sequence and the criteria for seeking help.

Glass edges. A cut sheet of plate glass has an edge that behaves like a blade until it is dressed. Buy the glass with polished or arrised edges if you can; if you cannot, take every edge back with a 220-grit diamond hand pad under running water before the sheet goes anywhere near a bench, and tape the edges of anything that will be handled often. Handle sheets by the faces, not the edges, and carry a large sheet vertically.

The knife, for cutting plastic sheet, tape and card. HSE’s principles transfer from catering directly: a sharp blade, because a blunt one needs force and force is what slips; a stable surface; the tool put away the moment it is set down; and never an attempt to catch a falling one.

A mains appliance beside water. The bath is the one electrical item in the station, whether it is a slow cooker, a heat mat or an immersion circulator. Plug it into a socket protected by a residual current device; route the cable so that it runs away from the bench edge and never lies across a wet surface or under a tray; dry your hands before you touch a switch; and keep the appliance’s own instructions with the station, because they are the authority for how that particular product may be used near water.

!One control on this page has no source behind it, and the page says so

This course has not obtained a published guidance document on the use of portable mains appliances near water in the launch market, and it does not have manufacturer stability specifications for consumer immersion circulators either. Both were on this page’s research list and neither was closed.

So the electrical advice above is the general practice a domestic darkroom already follows, plus the appliance’s own instructions, which are a manufacturer source and therefore the strongest thing available; and the bath’s accuracy is not asserted at all — you measure it, in the thirty-minute log below, and the number you get is the number your station has. Mains work of any kind, including switched outputs and timers, belongs to Part XVII, which owns the subject and the sourcing.

Required PPE

Eye protection while glass is being cut, dressed or drilled, and while the knife is in use.

Something between your hands and a cut glass edge — a folded cloth, a leather glove or the factory tape still on the sheet — until every edge has been dressed. This is cut protection, not chemical protection, and the two are different jobs: nitrile splash gloves would be the wrong control here and are not needed until Project 1 brings silver nitrate to the bench.

Dry hands at any switch, which is protective equipment in the sense that matters even though you cannot buy it.

A dust mask is not among the controls, because nothing on this page is a dust: gelatin arrives as sheets or granules and is dissolved in water, and no powder is weighed. The dust that this page cares about is the dust falling onto your work, which is an image-quality problem and is dealt with by the drying box.

Parts

Quantities make one station sized for an 11 × 15 inch sheet, which is the paper size the chloride and chlorobromide projects use. Scale the bed and the box if you intend to work larger.

Item	Quantity	Notes
Float glass, 4–6 mm, edges dressed	1 sheet, about 450 × 600 mm	The coating bed. It must be bigger than your largest sheet by at least 50 mm all round
Float glass or a marble tile, 10 mm or more	1 piece, about 300 × 400 mm	The chilled setting slab; mass is the point, so thicker is better
Rigid base board, 18 mm ply or a kitchen chopping board	1, larger than the bed glass	Carries the three feet
Levelling feet: M6 bolts with wing nuts and washers, or screw-in furniture levellers	3	Three, not four — see the build
Small bubble level, 150–200 mm	1	For the coarse setting only
Plastic storage box with a close-fitting lid, 60–80 litres	1	The drying box, laid on its side
Wire cooling rack or plastic-coated shelf	1–2	Drying shelves inside the box
Fine open-weave fabric: floating row cover, net curtain or nylon organza	about 0.5 m²	The filter over the box’s air openings
Hygrometer with a thermometer, digital	1	Reads the box, not the room
Freezer blocks	2	Under the setting slab
Slow cooker, seedling heat mat, mug warmer, deep pan, or immersion circulator	1	The bath; choose from the three budget paths below
Thermometer reading to 0.5 °C, with a stem that reaches into a jar	2	One lives in the bath, one goes in the emulsion
Borosilicate jars or beakers, 250 mL and 600 mL	2 of each	Dedicated to emulsion; never shared with food preparation
Glass rod, 8–10 mm diameter, 250 mm long	1	Becomes the puddle pusher
Epoxy putty and PTFE or electrical tape	small quantities	The rod’s end plugs
Hake brush, 50 mm, and foam brushes, 25 and 50 mm	1 of each	The brush coating route
Stiff plastic card or a squeegee blade	1	The card coating blade
Plastic wrap (cling film), plus masking tape	1 roll each	The bed surface and the dam bars
Syringes, 10 mL and 50 mL, without needles	1 of each	Measuring the coating volume per sheet
Red LED string light, or an amber or red safelight with a rated filter	1	See the safelight stage
Kitchen scales reading to 1 g, and photographic gelatin	1 set, 20 g	For the validation coat only

Tools

A steel rule; a craft knife and a cutting mat; scissors; a 220-grit diamond hand pad or wet-and-dry paper for glass edges; a drill and a bit to suit the levelling feet; a screwdriver; a marker; a torch; a watch or timer; and a notebook page ruled for the temperature log.

Not needed: a saw, a router, a workbench or any woodworking at all if you take the tray route below. The reference station has exactly three drilled holes in it.

Estimated cost

Cost band ££, and the spread inside that band is wide because the bath is the item that decides it. Three paths reach a working station:

The tested minimum. A pan of hot water on a stove or a hotplate with a nested jar standing in it, a thermometer, a sheet of glass on three bolts, a plastic box, and a rod made from a length of glass. The practitioner whose paper recipes this part uses coats with exactly this arrangement; the whole of the temperature control is you, a thermometer and a kettle.

The middle path. A slow cooker or a seedling heat mat with a plug-in thermostat gives a bath that holds a band without attention, which matters most during the thirty to seventy-five minutes of ripening that Projects 1 and 2 ask for. A second-hand slow cooker is the single best value item in the station.

The top path. An immersion circulator in a deep container, which holds a temperature closely and stirs the bath while it does it. It is also the only item here that this course cannot quote a specification for, so buy on the maker’s figures and then measure it yourself with the log below.

This page quotes no prices, because it cannot verify a current one. Dated figures for the launch market live in the laboratory planner, where they can be corrected without rewriting the text.

Estimated consumables cost

A station is built once and then it is equipment, so almost everything in the Parts table above is outside this section. What the build itself uses up is the plastic wrap, the tape, the putty, the fabric and the gelatin for the validation coat.

Consumed	This session	Sourced price	Cost this session
Plastic wrap (cling film) and masking tape	one roll of each; the wrap is renewed every session	None. tape-and-adhesives carries a cost band and no dated figure	—
Epoxy putty and PTFE or electrical tape	a few grams, for the rod end plugs	None. tape-and-adhesives carries a cost band and no dated figure	—
Fine open-weave fabric for the air filter	about 0.5 m²	None. card-and-paper-stock carries a cost band and no dated figure	—
Gelatin, photographic inert	20 g, validation coat only	£17.45 per 100 g, inert photographic gelatin (£0.17 a g)	£3.49
Abrasive: 220-grit diamond pad or wet-and-dry	one pad or sheet, for dressing glass edges	None. A named price gap: pinhole-making stock — shim, wet-and-dry abrasive paper, a pin vice and a bought laser-drilled pinhole	—

The priced rows come to £3.49 for one run of this session, at the retail ranges read on 5 September 2026 and recorded in the laboratory planner. That is a floor, not a total: 4 of the 5 rows carry no dated price, so they are counted as nothing here and are certainly not free. A priced entry is a dated range to plan against, never a quotation.

The float glass, the setting slab, the base board, the levelling feet, the storage box, the shelves, the hygrometer, the freezer blocks, the bath, the thermometers, the jars, the glass rod, the brushes, the syringes and the safelight are all equipment that the rest of Part V and Part XXI reuse — and the planner records the whole of that list as one of its named price gaps (coating-station hardware), so the band above is the only cost statement the course can make about them. The plastic wrap is the row worth noticing: it is renewed for every coating session, not once for the build.

Alternative route

No room that can be blacked out. The build itself needs no darkness at all; only the safelight test and the later coating do. Build the station in daylight, and run the fog test in a windowless bathroom at night, which is the route Part IV’s overview already assumes.

No space to leave a station standing. Everything here is designed to be picked up. The bed lives on its three feet on a board that slides under a bed; the drying box is a storage box; the bath is a slow cooker. Mark the feet’s positions on the board with a pencil so that levelling is a two-minute job rather than a fresh one each time.

No woodworking, no drill. Substitute a shallow plastic tray for the base board and levelling feet entirely: put the bed glass in the tray and level it with folded card shims, checking with the water puddle exactly as below. It is slower to adjust and it is not less level when you have finished. The shims are the thing to mark and keep.

Build

Station plan: where each piece stands, and why

1. 1Drying box — opening away from the wet side; lid off only while loading
2. 2Levelled coating bed — three feet, on a board, water-puddle levelled
3. 3Chilled setting slab — next to the bed, because a coated plate must travel a hand-span, not a room
4. 4Notebook, syringes, timer — the dry island; nothing wet is ever set down here
5. 5Water bath and emulsion jar — thermometer in the emulsion, second thermometer in the bath
6. 6Rinse pan and sink — the wet end of the bench and the end the cable never reaches
7. 7Safelight, bounced off the ceiling — behind the shoulder, so the wet sheet can be viewed at a low angle

Drawn as an arrangement, not to a measured plan. The two things worth copying exactly are the short trip from bed to slab, and the cable route.

Stage 1 — Set out the station (15 minutes)

Decide the two sides first and write them on tape stuck to the bench. Dry side: the bed, the setting slab, the drying box, the notebook, the dry tools. Wet side: the bath, the rinse pan, the sink. The darkroom-design principle is the one Kodak set out for enlarging rooms and it transfers unchanged — the work moves in one direction and hands do not travel back against it.

Two placements are specific to coating and worth getting right now. The setting slab goes beside the bed, because a freshly coated plate has to reach it before the layer stops being able to flow, and a walk across a room is long enough for a wedge to form. And the safelight goes behind you, bounced off a pale ceiling rather than aimed at the bench, so that you can crouch and look across a wet sheet at a shallow angle. Almost every coating fault is visible in that raking view while there is still time to do something about it, and invisible from directly above.

Finally, run the bath’s cable before anything else stands on the bench, and route it along the back and down to the socket. A cable decided at the end is a cable that ends up where it can be caught.

Stage 2 — The coating bed (25 minutes)

Drill three holes in the base board — two along one long edge, one centred on the opposite edge — and fit the levelling feet. Three, not four. Three points define a plane, so a three-footed bed has exactly one level position and reaches it by adjustment; a four-footed one rocks and can be level in two different ways at once. Set the glass on the feet, sitting on three small pads of self-adhesive felt so it cannot slide.

Now level it, and do not trust the bubble. Set the bubble level across the glass in two directions and get it roughly right; that is what a bubble level is for here. Then pour a small puddle of water onto the middle of the glass and watch it. If it creeps, the surface falls that way, and the foot that raises it is the one you turn. Add more water to the puddle and watch again. When you can keep adding water and the puddle grows round rather than pulling to one side, the bed is level enough to coat on. That method is the tested one and it is more sensitive than the instrument.

!There is no published level tolerance, and this page will not invent one

A number in millimetres per metre would be a comforting thing to print, and no source in this course’s research corpus gives one for hand coating. What the sources give instead is an operational test — the water puddle — and an acceptance test that comes later, in the fixer: if one edge or one corner of a coated sheet clears noticeably later than the rest, the layer is thicker there and the bed was not level. That is a real, repeatable criterion, and it has the advantage of measuring the thing you actually care about rather than a proxy for it.

So the honest form is this. The puddle sets the bed; the fixer grades it; and if you want a tolerance in millimetres, it is a number you will have to establish for your own bed by coating at deliberate tilts and finding where the clearing time starts to differ. That would be an original measurement and worth publishing.

The levelling table: three feet, and which one to turn

1. 1Base board — rigid; level this with the bubble level before the glass goes on
2. 2Three levelling feet — bolt through a clearance hole, wing nut above and below; two on one long edge, one centred on the other
3. 3Three felt pads — three, never four — four makes the glass rock
4. 4Plate glass, 4–6 mm — bigger than the largest sheet by 50 mm all round
5. 5Working surface — water, plastic wrap, water, wetted paper — renewed each session
6. 6The puddle, and the rule — turn the foot nearest the way the water runs, a quarter turn at a time, and pour again

Drawn as an arrangement, not to scale. The only dimension that matters is that the glass sits on exactly three points.

Finish the bed by laying its working surface: wet the glass, lay plastic wrap over it and smooth it down — the water holds it almost as a vacuum seal — then wet the wrap. Paper is coated on that; plates sit directly on it with taped card dam bars around them. The wrap is renewed each session and is the reason the bed itself never has to be scrubbed clean of set gelatin.

Stage 3 — The chilled setting slab (10 minutes)

Lay two freezer blocks on the bench, put the thick glass or marble on top, and leave it while you build the rest. That is the whole construction.

The reason it matters is in the coating lesson’s account of setting: a coated layer flows until the gelatin sets, and everything it does while it is still flowing — creeping to a low corner, running to an edge, taking a bubble with it — becomes permanent when it stops. Chilling the support shortens that window. Duffin’s industrial version is chilled rollers or cold air immediately after the coating head; a cold slab is the domestic form of the same idea. Get the slab genuinely cold before the session, not during it, because a slab at room temperature does nothing.

Stage 4 — The water bath (20 minutes)

Assemble whichever bath you have chosen and get it running with plain water, before any gelatin is involved. The station has to hold two different bands: about 40 to 45 °C for coating, which is the window this part’s paper work uses, and about 50 to 55 °C for melting and ripening, which the projects reach for later. Industrial trough coating runs between 36 and 40 °C, chosen so that the emulsion is above its setting point but close enough to it to set almost as soon as it is laid, and the domestic windows sit a little above that.

Two rules that save more trouble than they look worth. Put a thermometer in the emulsion vessel, not just in the bath — the bath’s temperature is not the emulsion’s temperature, particularly in the first ten minutes after a cold jar goes in. And stand the vessel on something, a jam-jar lid or a folded cloth, rather than directly on the bottom of a heated pan, so that the heat arrives through water rather than through metal.

If your bath is a pan on a hob, work out now how far to turn the heat down once it is up to temperature, and mark the knob. If it is a slow cooker, find out which of its settings holds which band. Either way the answer comes from the log in the next section, not from the appliance’s markings.

Stage 5 — The drying box (25 minutes)

Lay the storage box on its long side so the lid becomes a door. Cut two openings, low at one end and high at the other, each about a hand’s width, and tape the fine fabric over both on the inside. Fit the wire rack across the middle on four bolts or four blocks of foam. Put the hygrometer at the back, where you can read it through the lid.

The drying box in section: where the air goes in, where it leaves, and what stops the dust

1. 1Low inlet, fabric taped inside — about a hand's width; the filter goes on the inside face so it cannot be knocked off
2. 2Rack on four blocks — sheets flat, not touching each other or the walls
3. 3Hygrometer at the back — read it through the lid; above about 80 per cent and staying there means not enough air is moving
4. 4High outlet, fabric taped inside — warm moist air leaves high and at the far end
5. 5Optional fan, drawing OUT — never blowing in; a fan at an unfiltered opening is a dust delivery system
6. 6Lid as a door — load it, shut it, leave it shut

Drawn as an arrangement, not to scale. The whole design is one idea: all the air the box breathes has been through fabric.

The design has one job with two halves that pull against each other, and the tension is worth naming because it explains every drying-box decision.

Air must move, or the water in the layer has nowhere to go. About 85 per cent of a freshly coated layer is water, and it has to come down to about 5 per cent before the sheet can be handled or stored.

Dust must not, because a wet gelatin layer catches and keeps everything that lands on it. Baker makes the drying cupboard the item on which success with experimental emulsions depends, and his reason is not only cosmetic: dust and bacteria are the plate maker’s enemies, mould spores travel on dust, and a speck landing on gelatin has found an ideal growing medium — especially if the plate is drying slowly in a warm, humid atmosphere.

The fabric over the openings is the compromise, and it is the tested one: the same trick as taping floating row cover over a room’s air vent, which cuts the dust without blocking the flow. A fan is optional and, if you fit one, it goes at the high opening, drawing air out, so that everything entering the box has come through fabric. A fan blowing in through an unfiltered opening is a dust delivery system.

✕Slow drying is better, and slow drying is also riskier — you have to choose knowingly

Duffin found drying rate to matter: as water leaves, the gelatin molecules already linked by the hydrogen bonds formed during setting are drawn into a closer-packed array, and the more slowly drying is carried out, the more orderly that array will be. Maximum order comes from drying at room temperature, which a factory cannot afford and you can.

Baker’s dust-and-mould warning points the other way: the longer a wet gelatin layer sits, the more time there is for something to land on it and start growing. Both statements are from Tier 1 sources and both are right; they are simply about different failure modes.

The practical resolution is not a compromise on time but a better box: dry slowly, and make the enclosure good enough that slow drying is not a risk. If your room is cool and damp and sheets will not dry at all, the tested answer is gentle background warmth — a small radiator at about 24 °C — rather than a heat gun or a hairdryer, either of which will dry the surface before the depth and can leave the layer permanently marked.

Stage 6 — The coating tools (15 minutes)

The puddle pusher. Take the glass rod and build a plug at each end: a turn or two of tape, then a pea of epoxy putty pressed on and smoothed, so that each end finishes in a small collar standing proud of the rod. Make the clear span between the collars about 25 mm wider than the widest sheet you will coat; a 9 inch rod against an 11 × 15 inch sheet is the tested combination. The collars are the whole of the tool’s design: they hold the rod off the sheet by a fixed amount, and they stop emulsion escaping past the ends and running down the sides, where it dries as a thick dark border.

The brushes. A hake brush for a soft edge, foam brushes for speed, a stiff card blade for a straight-drawn coat. Wash them in warm water immediately after use — gelatin left to set in a brush never fully comes out — and keep them in a lidded box, not standing in the open where they collect the dust the drying box exists to exclude.

The measuring gear. Two syringes, one small and one large, both without needles, are how the volume per sheet is measured; a coating volume you did not measure is a coating weight you cannot compute, and coating weight is the number the whole part turns on. Dedicate the beakers, the jars and the thermometer to emulsion work and store them apart from anything used for food.

A note on metals. Cheap stainless steel can corrode in contact with silver nitrate solutions, and rust must not go anywhere near an emulsion. Prefer glass, borosilicate and plastic for everything that touches a make, and if you use a stainless whisk, buy a good one and inspect it before every session.

Stage 7 — The safelight (10 minutes)

Fit the safelight where the plan puts it, bounced off the ceiling, and note three numbers on tape stuck to the lamp: filter or lamp type, bulb wattage, and distance to the bench. Those three are the geometry your fog test will certify, and changing any of them means running it again.

Which class of light you need depends on what the material can see, and this part changes that twice.

Material	What it sees	Safelight class
Silver chloride contact paper, Project 1	Ultraviolet and the violet edge	The most forgiving material in the part. The tested domestic practice is a red LED string or a low-wattage yellow lamp; the maker’s general recommendation for black-and-white papers is an SL1 or 902 filter with a 15 W bulb at not less than 1.2 m
Chlorobromide and bromide emulsions, Projects 2 and 3	Further into the blue as bromide rises, and further again when iodide is added	A red or dark brown filter of the class made for fast blue-sensitive material. The margin is narrower than for chloride and is not assumed — it is re-tested
Orthochromatic, Project 5	Blue and green; still blind to red	A deep red filter, of the class the manufacturers specify for orthochromatic material. Not darkness
Panchromatic	Everything	Not attempted on this course. It is studied and not performed

The manufacturers’ own tables put the dark red 906 filter against orthochromatic materials and the dark brown 904 against fast blue-sensitive ones, and the current technical sheet for an orthochromatic film says plainly that its blue and green sensitivity still allows handling in deep red safelight, with the 906 filter, a 15 W bulb and not less than 1.2 m. Kodak’s darkroom-illumination guidance says the same thing in one line: blue-sensitive and orthochromatic black-and-white films may be handled under red safelights.

The point to take from that table is not the filter numbers. It is that a safelight is safe for a material, not in general, and the material changes under you as this part progresses. Your undyed emulsions are the easiest thing you will ever handle in a darkroom; the moment a dye goes in, the room you proved is a different room.

Testing and calibration

Four tests. None takes long, all four go in the notebook with a date, and no silver is coated until they have been run.

The validation half-hour, in the order the tests actually fit together

Bath

Heat up, 15 min30-minute temperature log, readings every 5 minHold at temperature for the gelatin, 45 min

Bed

Puddle level, 10 minLay wrapGelatin test coat, 10 minDry in the box, then count specks in raking light

Safelight

Waits for the box, 40 minFog test on paper: 0, 1, 2, 4 min, in four masked areasProcess and read

030 min60 min90 min

Blocks are drawn to a common scale of one unit to five minutes, so their widths are proportional to the times named on them. The bath log runs while the bed is levelled and coated, which is why the session fits into two hours; only the drying is genuinely idle, and the safelight test fills it.

1. The level test

Method: the water puddle, as in Stage 2, followed later by the fixer reading of the first silver-coated sheet. Pass: the puddle grows round rather than running. Record: which feet you turned and in which direction, so that the bed can be reset without starting over.

2. The thirty-minute temperature log

Method: bring the bath to your coating band, put a jar of water in it with a thermometer in the water in the jar, and read both thermometers every five minutes for thirty minutes. Seven rows.

Pass: the jar stays inside a band you are willing to work in, and you can write down what that band is. This course does not set a figure for you, because the band a station can hold is a property of that station and this page has no manufacturer specification to hold it to. What it does say is that the band is a measurement, not a claim, and that you should know it before an emulsion depends on it. A swing of a couple of degrees in the jar is workable for a coating hold; a swing of ten is a bath that will need attention every few minutes during a seventy-five-minute ripening.

Record: appliance and setting, the two temperature columns, the room temperature, and the time the jar took to reach the band from cold. That last figure is the one you will use for the rest of the part when planning a session.

3. The plain gelatin test coat

Method: dissolve about 10 g of gelatin in 200 mL of water — sprinkle it on cold water, let it swell for half an hour, then melt it in the bath — and coat a sheet of paper or a glass plate with it on the levelled bed at coating temperature, using the rod. Set it on the chilled slab, then dry it in the box with the door shut.

Read it in raking light, with a torch held almost parallel to the surface. Count what you find in a marked 100 mm square: specks of dust, hairs, bubbles, rod marks, streaks and the line left by a stroke that stopped and restarted. Write the count down.

Pass: there is no published threshold and the course does not manufacture one. The count is a baseline — the number your station puts on a sheet today — and its value is comparative. Improve the box or the sequence, coat again, and see whether the count falls. When you later coat silver and find twenty specks, the question of whether they came from the emulsion or the room has an answer already in the notebook.

📓The one published acceptance number in this whole area, and what it is actually for

Baker gives an industrial fog criterion for experimental emulsions: on a trial coating on glass, fog should not exceed 0.02 density; and after a keeping test of ten days at 105 °F below 65 per cent relative humidity, incubated plates should show no more than 0.02 extra fog compared with controls.

Both of those are figures for a silver coating, read on a densitometer, and neither applies to a plain gelatin coat, which has nothing in it to fog. They are recorded here because this station’s real acceptance test is the first emulsion coat that comes off it in Project 1, and 0.02 is the number to judge that one against if you can measure density. If you cannot, the comparative version still works: coat a strip, develop and fix it without exposing it, and compare it against a strip of the same material fixed without development. A visible difference is fog worth chasing.

4. The safelight fog test

Method: exactly Station 4 of the latent image made visible, run in the station’s own geometry and using a commercial paper as the test material. Give the strip a light grey exposure first, carry it to the brightest safelit spot in the room, and mask off areas for 0, 1, 2 and 4 minutes; then expose the other end of the same strip after the safelight exposure, and process both against an unexposed control.

Pass: no density change out to four minutes. A small change — of the order of 0.04 in density — after only one minute means the conditions are inadequate. The working figure to carry away is the safe time, defined as half the time at which a change first becomes detectable; compare it with how long a sheet is actually out of its box during one of your coating sessions, and if the safe time is shorter, the room fails and the remedies in order of cost are: move the lamp further away, fit a lower bulb, replace an aged filter, or work with the lamp switched off.

Record: the filter or lamp, the wattage, the distance, the date, and the verdict. Then note the one thing this test cannot tell you: it certifies the geometry for the paper you tested with. Your own emulsions are not that paper, and Project 5’s dyed emulsion is a different material again.

Clean-up and waste

Wash the brushes, the rod and the jars in warm water and detergent before the gelatin sets in them, and dry them before they go back in the lidded box. Peel the plastic wrap off the bed and bin it. Wipe the bed and the slab with a damp cloth rather than dry-dusting them, because dry dusting moves dust around rather than removing it.

The only waste this session produces is plain gelatin and its rinse water, with no silver, no hardener and no developer in it. From Project 1 onward that changes completely: every rinse from a coating session carries silver and joins the collected stream that Part II’s chemical waste and silver waste sets up, and local regulation governs what may then be done with it, differing between authorities even within one country. Establishing the collecting container now, while the station is being built, is easier than establishing it mid-make with a jug of silver rinse in your hand.

Troubleshooting

What you find	Likely cause	What to do
The dried gelatin coat is noticeably thicker along one edge	The bed falls towards that edge	Turn the foot nearest that edge clockwise to raise it, a quarter turn at a time, re-running the puddle test after each. If the thick edge is a corner rather than a side, it is the single foot nearest that corner
The coat is thicker at both ends and thin in the middle	The glass is dished, or it is resting on four points instead of three	Check that only three pads are carrying the sheet. A 4 mm sheet over 600 mm will also sag under its own weight; go thicker rather than trying to adjust it out
The puddle stays put but the fixer clears one corner late	The bed moved after levelling, or the board itself is on an uneven bench	Level the board first with the bubble level, then the glass with the puddle. Mark the board’s position on the bench
A stroke line runs across the coat	The rod stopped and restarted, or the emulsion had begun to set	One pass, no hesitation, and check the emulsion temperature rather than blaming the rod
A thick dark border down both long edges	Emulsion escaped past the ends of the rod	The end plugs are missing, too small, or set too far in. Rebuild them
The bath overshoots the band and then undershoots	The vessel is standing on the heated surface, or the thermostat is cycling widely	Stand the vessel on a lid or a cloth; add water so the bath has more mass; take the reading from the jar, not the bath
The drying box’s hygrometer reads above about 80 per cent and stays there	Not enough air is moving, or the box is in a damp room	Enlarge the openings before you add a fan. Gelatin at high relative humidity is a growing medium, and mould on a stored plate starts here
Sheets in the box dry with a bloom or feel tacky for days	The room is cool and damp	Gentle background warmth at about 24 °C, not a heat gun. Warmth applied to the surface alone dries the skin before the depth
The speck count is high and rises towards the door of the box	Dust enters when the box is opened	Load the box, shut it, and leave it shut. Every opening is a fresh dose
The safelight test fails only on the strip exposed after the safelight	You are looking at latensification, not simple fog	The lamp is fogging paper that already carries a latent image, which is the harsher and more realistic condition. Treat it as a fail and move the lamp

Questions

1. Explain, in terms of what defines a plane, why this build uses three feet rather than four, and say what symptom a four-footed bed produces that a three-footed one cannot.
2. Your thirty-minute log shows the bath at 44 to 45 °C throughout and the jar inside it at 39 to 43 °C, still climbing at the end. What has the log actually told you, and what would you change about the session plan rather than about the apparatus?
3. The plain gelatin coat is clean but shows six bubbles in a line across one end. Give two explanations, one belonging to the tools and one to the technique, and say what evidence would separate them.
4. Baker says the drying cupboard is the thing on which success with test emulsions depends; Duffin says the slower the drying, the more orderly the gelatin array. Explain why those two statements together argue for a better enclosure rather than for a compromise on drying time.
5. You pass the safelight test with a commercial paper at 1.2 m, then coat and dry Project 1’s chloride paper. Are you entitled to assume the same lamp is safe for it? Say what you would do to find out, and what you would do differently again before Project 5.
6. List every substitution you made in this build for reasons of cost or space, and for each one state the coating fault it makes more likely and the test on this page that would catch it.

Further experiments

Find your own level tolerance. Coat four plain gelatin sheets on a deliberately tilted bed — level, and then with one foot raised by one, two and three turns — and measure the dried thickness at each corner by weighing the sheet before and after coating and by reading the clearing time when you later repeat it with a silver emulsion. Somewhere in that series is the tilt at which the fault becomes visible, and that number is a genuine original measurement this course does not have.

Test the fabric filter. Run the drying box for one drying cycle with the fabric fitted and one without it, coating a plain gelatin sheet each time, and count the specks in the same marked square. If the difference is small, your room is cleaner than the box; if it is large, you have just measured what the box is worth.

Compare the three coating tools on one emulsion. Rod, foam brush and card blade, three sheets, one melt, same temperature, same volume measured with the syringe. Dry them together and read them in raking light. This is the cheapest way to find out which tool your hands are good with before a batch of your own silver is riding on the answer.

Log the bath against the room. Take the bath’s log again on a cold morning and on a warm evening. A slow cooker’s behaviour is not independent of the room, and knowing by how much is the difference between planning a seventy-five-minute ripening and improvising one.

Recap

The station is five things and four numbers. The bed is levelled by watching water rather than a bubble, and finally graded by the fixer. The slab shortens the window in which a coated layer can still go wrong. The bath holds two bands, and the band it actually holds is something you measured rather than something the box it came in claimed. The drying box resolves a genuine conflict between two Tier 1 sources by being good enough that slow drying is not a risk. The tools are a rod with two plugs, some brushes and a syringe, and the syringe is the one that makes coating weight computable.

The four numbers — the level, the log, the speck count and the safe time — are worth more than the apparatus. They are what turns the next four sessions from a sequence of hopeful attempts into experiments with a known instrument.

Check your understanding

Question 1. Why does this page tell you to put a thermometer in the emulsion vessel rather than relying on the one in the water bath?

• Because bath water and emulsion have different boiling points
• Because the vessel lags behind the bath, so the two readings differ, and it is the emulsion's temperature that sets its viscosity
• Because a thermometer in the bath will be damaged by the heater
• Because gelatin changes the boiling point of the water around it

Show the answer and why

Answer: Because the vessel lags behind the bath, so the two readings differ, and it is the emulsion's temperature that sets its viscosity

Coating temperature matters because it sets viscosity, and viscosity decides how the layer flows and how soon it stops flowing. The jar takes time to reach the bath's temperature, so during the first minutes — exactly when a cold jar has just gone in — the bath reads one figure and the emulsion another. Measuring the thing you are controlling is the general rule; this is one case of it.

Question 2. A dried plain gelatin coat is even, but shows a thick dark border down both long edges. What has gone wrong, and where?

• The bed falls towards both edges, which is impossible, so the glass is dished
• The emulsion was too cool and would not flow to the middle
• Emulsion escaped past the ends of the coating rod and ran down the sides
• The plastic wrap was not wetted before the paper went down

Show the answer and why

Answer: Emulsion escaped past the ends of the coating rod and ran down the sides

The plugs on the ends of the rod exist to stop exactly this. Emulsion that gets past them is no longer being metered by the rod at all: it runs down the outside of the sheet and dries as a heavy border. The fix is in the tool, not the bed — rebuild the plugs so they stand proud of the rod and sit just outside the sheet width.

Question 3. You are told that a safelight passed its fog test at 1.2 m with a commercial paper. Which of these later changes most clearly requires the test to be run again?

• Coating a chloride paper instead of buying one
• Adding a sensitising dye to the emulsion so that it responds to green
• Changing the developer used to process the test strip
• Moving the drying box to the other side of the bench

Show the answer and why

Answer: Adding a sensitising dye to the emulsion so that it responds to green

A safelight is safe for a material, not in general. An undyed chloride emulsion sees less than the tested paper, so the geometry that passed is at worst unchanged; a dye that extends sensitivity into the green makes the material see part of what the lamp may be emitting, which is a new question. The manufacturers answer it with a filter change to deep red rather than with darkness — and then you re-run the test, because the test is what certifies the change.

Question 4. Duffin reports that slower drying gives a more orderly array of gelatin molecules; Baker warns that dust and mould spores are the plate maker's enemies and that slow drying in a warm humid atmosphere is when spores find their medium. What does this page do with that pair?

• Prefers Duffin, because he is the later source
• Prefers Baker, because his warning concerns a visible defect
• Splits the difference by drying at a moderate rate
• Treats both as correct about different failure modes, and improves the enclosure so that slow drying is not a risk

Show the answer and why

Answer: Treats both as correct about different failure modes, and improves the enclosure so that slow drying is not a risk

The two statements are not in conflict: one is about the physical order of the dried layer, the other about what can land on and grow in a layer that stays wet longer. Averaging them would forfeit the benefit of both. Building a box that genuinely excludes dust lets you take the slow drying Duffin recommends without paying Baker's penalty — which is the general shape of a good engineering answer to two true constraints.

Question 5. What does the plain gelatin test coat actually establish, given that this page sets no pass threshold for it?

• Nothing useful, since without a threshold there is no test
• A dated baseline speck count for your station, against which later coats and later improvements can be compared
• That the gelatin you bought is of photographic quality
• The maximum density the station will later be able to reach

Show the answer and why

Answer: A dated baseline speck count for your station, against which later coats and later improvements can be compared

A measurement with no published threshold is still a measurement, provided it is repeatable and dated. The count answers a question you will ask for the rest of the part: when a silver coat comes out speckled, did the specks come from the emulsion or from the room? Absolute standards are useful; a control you took yesterday is often more useful.

Check my answersClear