Painting Woodwork and Trim

One Door, Two Whites

The hall door was painted eight years ago and it is now two colours. The face looking into the hall is still recognisably white. The face that shuts into the cupboard is the colour of milky tea, and the top edge — the one nobody has looked at since it was hung — is worse than either. Same tin, same brush, same afternoon. Nothing leaked, nothing has been smoked near it, and washing it does nothing at all. The customer wants to know what went wrong with the paint, and the honest answer is that nothing did: it behaved exactly as that kind of paint behaves, and it did so fastest wherever the light could not reach.

That one observation sorts most of the arguments about woodwork. Trim gets a hard, closed, high-binder film rather than the open pigment-heavy one a wall gets, and the binder that gives it the hardness is the binder that changes colour in the dark. The rest of the job follows from which trade-off you buy — and then from the fact that a moulding is not a rectangle and a skirting is not sold in the units it is painted in.

Why a White Alkyd Goes Ivory Behind the Door

An alkyd is a polyester resin carried on drying oils, and it does not dry so much as react. Oxygen attacks the unsaturated points along the fatty acid chains, hydroperoxides form and break down, the fragments cross-link, and over days and then months a liquid becomes a solid — metal driers pushing it along. The same reaction also leaves a set of coloured compounds sitting in the film. That is the ivory. Not dirt, not a stain from behind, not a bad batch: part of the chemistry that made the film hard in the first place.

What surprises people is that daylight undoes a good deal of it, because those chromophores bleach under ultraviolet. An alkyd white in a south-facing room holds its colour for years while the identical paint in a north hall, inside a wardrobe, on the closet side of a door or behind a headboard drifts warm within a winter. It is why the top edge of a door is always the worst place on the door, and why the skirting behind a sofa that has not moved is a different white from the skirting either side of it. It is also why the cure is sometimes free: take the wardrobe doors off for a fortnight of summer light before agreeing to repaint anything, and a fair share of the yellowing walks back on its own.

Oil chemistry decides how badly. Oils high in linolenic acid — linseed above all — yellow hardest, which is why traditional oil-bound whites did it so visibly and why products sold as non-yellowing are built on soya, safflower or tall-oil fatty acids, or on urethane and acrylic modification. Waterborne acrylics carry no drying oil and do not do this at all; waterborne alkyds and hybrids keep some of the mechanism along with some of the flow, so read the class rather than the word on the front of the tin. Where it has to be argued rather than eyeballed, ASTM E313 turns measured colour coordinates into a yellowness index and ASTM D2244 computes the difference between two of them — which matters, because the accelerated tests routinely quoted for coatings are light exposure tests and this failure happens in the dark.

Before blaming the binder, rule out the three things that imitate it. A thin waterborne film over an old nicotine-loaded gloss lets tar through as a warm cast that worsens with each coat rather than improving. An over-generous shellac knotting patch grins through as an orange-brown bloom the shape of the patch. And oak, mahogany or cedar under a waterborne primer pushes tannin up as a brown ring. Look at the shape of it: the binder's own chemistry is even and follows the light, and everything else is patchy and follows the timber.

Waterborne or Solvent-borne, Settled on the Job Rather Than in the Van

The trade has spent twenty years treating this as allegiance and it is not. Each system fails at a different point, and which failure you can live with is a property of the room rather than of the painter — which is what the table below is for.

Two published tests are worth knowing by name, because they describe exactly the two things people argue about. ASTM D1640 separates drying into stages — set to touch, dry to touch, dry hard, dry through — and the gap between the first and the last is where every waterborne complaint lives: a skirting dry to a fingertip within the hour may be weeks off a film that can be leaned on. ASTM D4946 measures blocking resistance, the property that decides whether a painted door edge fuses to a painted stop, and it is measured on a fresh film precisely because that is when it matters. ASTM D3363 gives the pencil hardness a data sheet will quote if you ask. None of those numbers reach a retail tin; all of them are available from a technical department, and on a large job they are worth the phone call.

Regulation narrows the choice before preference gets a say, and it does it by product category rather than by brand. Woodwork has its own two subcategories in Directive 2004/42/EC — (d) for interior and exterior trim and cladding paints on wood and metal, and (e) for trim varnishes and woodstains — each carrying a separate ceiling for waterborne and solvent-borne products, so the category a gloss falls into decides how much solvent it is allowed before anyone chooses a colour. The equivalent in the United States is a federal architectural coatings rule with stricter regional ones over it. The consequence is unglamorous: a high-solids alkyd gloss stocked and legal on one job may be neither on the next, and the morning of the second coat is an expensive time to find out.

Three Different Bleeds, and Only One of Them Is a Knot

A live knot is a resin reservoir with a working temperature. Softwood knots hold colophony and terpenes that stay put at room temperature and mobilise when the timber warms, which is why a knot that behaved perfectly through a February refit pushes an amber weep through the paint on the first hot afternoon in a bay window, and why a skirting above a run of pipework does the same in October. The mechanism is thermal rather than chemical, so nothing laid on top of a knot solves it by being thicker. Sealing means putting down something the resin will not dissolve and cannot pass, and the traditional answer is a shellac knotting solution: it dries in minutes, is unaffected by resin, and carries no water. Two thin applications over the knot and a small margin round it — not one flood coat, which stays soft in the middle and grins through the finish as an orange patch of its own.

Dead knots are a carpentry repair rather than a coating one. A dead or loose knot is not bonded to the timber around it; it will move, drop out, or open a ring crack no film bridges. Cut it out, plug or fill it, then seal — sealing a knot that is going to fall out only buys the time until it does. Resin pockets deserve the same look and are easier to miss on a sanded face than a knot is.

Tannin is the third bleed, and it catches anyone who has moved from softwood to hardwood without changing anything else. Oak, sweet chestnut, western red cedar and mahogany carry water-soluble extractives, and a waterborne primer redissolves them and carries them into the film as brown staining along the grain and around fixings. The answer is not a heavier waterborne coat but a primer that introduces no water at all — shellac-based or solvent-borne — after which the waterborne system goes on top perfectly happily. The same logic covers any ferrous fixing left in painted joinery, which bleeds rust from below where no topcoat reaches.

The preparation of woodwork for painting, sealing resinous knots included, is set out in BS 6150, the code of practice for painting of buildings — the document to reach for when a specification is being argued rather than a habit defended. Moisture content belongs in the same conversation, since paint on timber that is still too wet blisters and never develops the film the sheet describes; ASTM D4442 covers how that content is actually measured, and a pin meter costs far less than coming back. MDF has neither knots nor tannin but has its own two problems, a machined face and cut ends that drink primer, and an irreversible swell if it gets properly wet — the cut-end sealing and board grades are covered in the skirting and architrave guide rather than repeated here.

What an Eleven-Coat Skirting Will Take

Old trim carries its history as film thickness, which is why a Victorian ogee in a hallway reads as a soft blur rather than a profile: every repaint has laid another few tens of microns into the quirks and against the arrises where a brush cannot help but load, and the moulding has been filled in from the inside out. Two more coats onto a profile that has already lost its detail is a decision rather than a default — strip it, or take the build out at the quirks alone, and price that as its own line instead of burying it in a preparation rate.

Where the film is sound, the work is grip, and trim collects exactly the contaminants that defeat it — hand oils at handle height, aerosol grease near a kitchen, polish on anything a cleaner has been near. Degrease and rinse before an abrasive is opened, because sanding a greasy surface drives the contamination into the scratch pattern instead of taking it off. The quirks and arrises then have to be keyed by hand, since that is where no machine reaches and where the next film gets worked hardest. The general case for abrading a glossy film before overcoating it belongs to the interior walls guide; the test a British specification is likelier to name is ISO 2409, the cross-cut test.

Chipped edges are the defect specific to painted trim, and they get skimped. A chip has a hard shoulder where the old film ends, and filling the hollow without taking that shoulder back leaves a raised ring that reads under gloss from three metres — feather it out until a fingertip cannot find it, then fill, sand flush and spot-prime the bare timber and a hand's width beyond. On anything built before the late 1970s, treat the old film as leaded until shown otherwise: dry sanding and hot-air stripping are the two operations that put lead into the air or into fume, both are controlled work under the Control of Lead at Work Regulations in Great Britain and under the United States Environmental Protection Agency's renovation, repair and painting rule in pre-1978 housing, and trim is where the old lead gloss actually lives.

1. Establish whether the existing film is leaded before sanding or heating anything in pre-war or pre-1978 stock.
2. Degrease and rinse every surface before an abrasive is opened.
3. Key the sound film, working the quirks and arrises by hand where no machine reaches.
4. Feather back the shoulder around every chip until it cannot be felt, then fill and sand flush.
5. Cut out dead knots and resin pockets, plug or fill them, and seal the live knots with two thin applications of knotting.
6. Spot-prime every bare patch and a hand's width beyond, and let it flash off fully before anything else lands.

Brush Marks, Sags, and the Door That Sticks in August

Waterborne trim paint punishes a slow worker and solvent-borne punishes a heavy one. A 4.8 metre skirting run in a warm room gives a waterborne product a matter of minutes before the edge sets, and a brush going back into a set edge leaves a rope no later coat disguises — so work between natural breaks, a corner or an architrave or a chimney breast, rather than trying to hold a wet edge the length of a wall. Solvent-borne gloss fails the other way, hanging on a stile or in the top of a panel long enough to sag, so it goes on lighter than instinct wants, spread out and laid off in one direction with an unloaded brush. Match the filament to the product, synthetic for waterborne and natural bristle for solvent-borne, and keep the gloss brush for gloss.

The stuck door is the commonest callback on painted joinery, and it is a blocking failure rather than a fitting one. Waterborne trim paint is dry to a fingertip within the hour and nowhere near its final hardness for weeks, so a leaf painted on the Friday and closed onto a freshly painted stop over the weekend fuses to it, and the tear takes film off both. Wedge doors and casements open, paint the leaf and the stop on different days where the programme allows, and where a door genuinely has to close, put greaseproof paper or a low-tack tape between the two surfaces rather than trusting the touch test.

One condition matters more on trim than it does on walls, and it is substrate temperature rather than air temperature: a skirting on a cold outside wall takes and holds a film differently from the identical skirting on a partition in the same warm room, and the through-dry that blocking depends on is the thing that suffers. Ventilation is the other, and on this trade it is a control measure rather than a comfort: knotting solution is carried on alcohol and gloss on white spirit, so the room's air change rate decides where the vapour goes. In Great Britain that sits under the Control of Substances Hazardous to Health Regulations, and the product's own safety data sheet is the document that says what is in the tin.

1. Work the woodwork after the ceilings and walls are finished, and after the room has been vacuumed rather than swept.
2. Take handles, escutcheons and finger plates off rather than cutting around them.
3. Prime and undercoat the leaf's top and bottom edges while it is off its hinges, since neither is reachable once it is hung.
4. Break a long run at a corner or an architrave rather than trying to hold a wet edge the length of it.
5. Lay off in one direction with an unloaded brush, and stop touching it once it starts to set.
6. Wedge every painted door and casement open, and never close one onto a stop painted the same week.

What to Look At After the First Heating Season

Come back once the heating has run, because that is when woodwork tells the truth. Look first at the knots, particularly any in a bay, above a radiator or behind a south-facing window board — those are the ones that reached a temperature the February refit never tested. An amber weep at a knot is a sealing failure, cheap to correct in one small area and expensive once it has softened the film around it.

Then run a hand along the bottom arris of the skirting and the leading edge of every door, where the film is always thinnest and the first wear shows: that is the case for two thin coats rather than one heavy one, made visible. Finally, open every cupboard, look at the closet face of every door and pull the sofa out, then compare those whites with the ones the light has been on all year. If they have parted company, the binder did what the binder does, and the conversation is about which system goes on next time rather than about who applied this one.