Rendering and Plastering a Wall

What the Background Will Drink

Suction is the first measurement on any rendering job, and it gets taken with a hose and a hand rather than a gauge. Splash a litre of water across a square metre of wall and watch it. If the sheen vanishes in seconds and the surface goes matt again before you have turned round, the background is thirsty enough to pull water out of your first coat before the binder has anything left to hydrate with. If the water beads or runs off, the surface is dense, sealed, or contaminated, and no amount of pressure through the trowel will make render stick to it.

Common backgrounds sort into three families. High suction: aircrete block, soft handmade brick, old lime-bedded rubble stone. Moderate: most common clay brick and dense concrete block. Low or none: engineering brick, in-situ concrete cast against steel forms, painted or previously rendered faces, and anything still carrying curing compound or shutter oil. Each family needs a different opening move, and the move is either wetting down, a bonding treatment, or a mechanical key such as spatterdash or metal lath.

Damping down is control, not saturation. What you want is a surface that has gone dark but is not glistening, with no free water standing in the bed joints. On a hot dry wall in summer that can mean two passes an hour apart. Overdo it and the first coat floats on a film of water, cures soft, and lets go in sheets a season later. A wall rained on all night is the same problem from the other direction, too wet to take material, which is why the ambient condition clauses in BS EN 13914-1 Design, preparation and application of external rendering and internal plastering exist at all.

Contamination gets checked before anything else. Salt bloom, algae, old bitumen patching, gypsum residue on a wall about to receive cement render. All of these are bond breakers, and gypsum sitting against portland cement render in a wall that wets and dries is a well-documented route to expansion and spalling. Cut it out, do not cover it.

1. Wet a test metre of each different background on the elevation and time how long the sheen takes to disappear.
2. Sort the wall into suction zones and mark the boundaries; mixed suction on one elevation needs different preparation, not one compromise.
3. Remove coatings, salts, organic growth and loose material back to sound background before any wetting.
4. Damp high-suction zones in passes until the surface stays dark for several minutes without free water standing.

Key Before Coat: Dash, Bond and Lath

Mechanical grip outlasts chemical adhesion on any wall that gets wet. A bonding treatment brushed onto a dense concrete panel does work, but it is a single thin film with nothing behind it. A spatterdash coat gives the render several thousand small undercuts to hook into, and it does not much care whether the wall is warm, slightly dusty in the corners, or damp.

Spatterdash goes on wet and coarse, a sloppy cement and sharp sand slurry thrown from the trowel or flicked from a brush and left rough, never trowelled flat. Cover the wall completely. A dash coat with bald patches transfers the suction variation straight through into the scratch coat and prints as shading in the finish months later. Let it set hard and cure damp before the scratch coat follows, which is usually a day or more depending on temperature.

Lath takes over wherever a background cannot be keyed at all: timber or steel frame, sheathing board, insulation, and any wall that crosses from one substrate to another. ASTM C1063 Standard Specification for Installation of Lathing and Furring to Receive Interior and Exterior Portland Cement-Based Plaster covers fastener spacing, lap arrangement, furring, and the requirement that the lath stand off its backing far enough for mortar to key fully behind the wire. Lath pinned flat to sheathing with no furring produces a plaster slab with wire stuck on its back face instead of buried in it, and that wall cracks along every stud line.

Two lath details get skipped routinely and both cost real money. Run the sheets so the cups face upward, so the surface catches and holds mortar rather than shedding it. And carry a continuous water-resistive barrier behind the lath, lapped shingle fashion, with the number of layers and the flashing details set by the building code in force locally rather than by whatever the last job used.

Scratch Coat: Full Embedment, Combed Not Smoothed

Pressure, not thickness, defines a scratch coat. Mortar has to be pushed through the lath and squeezed hard against the backing so every wire is buried and the material behind the lath is continuous. A soft mix applied with a light hand leaves voids behind the wire that no later coat can ever reach. That wall sounds hollow when tapped and eventually parts company with the lath as one sheet, taking the brown and finish coats with it.

Nominal thickness for a scratch coat over metal lath in a three-coat portland cement system is tabulated in ASTM C926 Standard Specification for Application of Portland Cement-Based Plaster, and it is measured from the face of the lath rather than the face of the backing. That single definition explains why an unfurred lath job reads thin even when the correct volume of material has gone onto the wall. Under the European route the same decision arrives from the strength side instead, through BS EN 13914-1 and the mortar designations of BS EN 998-1 Specification for mortar for masonry, rendering and plastering mortar.

Comb the surface horizontally while the mortar is still green, deep enough that the scratch lines are unmistakable but not so deep that the tines reach the wire. Horizontal scoring gives the brown coat a shelf to sit on rather than a smear to slide off. Then leave it alone. A scratch coat that has been floated smooth because someone had spare time on their hands has lost most of its key and all of its point.

Estimating a scratch coat from face area alone under-orders on every lathed job, because the volume that disappears into the wire and the backing is real material that never appears on the finished face.

Brown Coat: Buying Flatness, Leaner Than What It Sits On

Flatness is bought in the brown coat and nowhere else. The finish is far too thin to correct a hollow, and grinding a high spot out of a cured brown coat is slow dusty work that usually breaks the bond around whatever area you attack.

Screeds go on first: narrow bands of mortar run to a level or a laser at internal corners, at openings, and at intervals a rod length apart across the wall, left to firm up before the bays between them are filled. Fill, rule off with a straight edge worked in two directions, then float once the surface has taken up. Floating too early drags fines and water to the face and leaves a weak crazed skin; floating too late tears the surface open and you are back to filling.

Strength steps down at this point and never back up. The brown coat should be no stronger than the scratch coat under it, equal or leaner. A rich brown coat over a leaner scratch coat shrinks harder as it dries, and that shrinkage stress has to be resolved somewhere. It resolves at the bond line, and it takes the scratch coat off the lath.

Suction management repeats itself here in full. A scratch coat left in the sun for a week will pull water out of the brown coat within minutes of it landing, stalling hydration right at the interface and leaving a dusty friable plane you can scrape open with a fingernail. Damp the scratch coat down, let it go matt, then start.

External Stucco: The Same Ladder, Worse Weather

Outside, every rule above still holds and the consequences arrive far sooner. Rain, freeze-thaw cycling, thermal movement across a south elevation and ultraviolet exposure all test the coat ladder inside the first season rather than over a decade, and they test it at the weakest interface rather than in the middle of a panel.

Movement joints do the work that mix design cannot. Large unjointed panels of cement stucco crack, and the crack runs where the wall wants it to run, usually diagonally out of a window corner. Set control joints to the panel areas and length-to-width ratios given in ASTM C1063, carry them through the lath rather than through the plaster alone, and keep them clear of the flashing line.

Casing beads, weep screeds and the terminations at grade and at roof-to-wall junctions decide whether the wall can drain. Stucco dead-ended into a slab with no screed wicks ground water upward, and the salt damage that follows looks like a mix fault while being purely a detailing one.

Three-coat over lath, two-coat direct to solid masonry, and proprietary one-coat systems all carry different thickness rules, different cure requirements and different tolerances, so settle which system is specified before a single tonne of sand is ordered.

Finish Coat: Thinnest, Weakest, Least Forgiving

Everything a client will ever look at lives in the last few millimetres, and that layer holds the least strength in the system by design. Gypsum finish plaster goes on in two thin passes and is trowelled up; cement finish coats are floated, dashed or textured. Either way the coat is a skin stretched over a flat background, not a device for making the background flat.

Variable suction across the brown coat is the fastest-moving problem on the wall at this stage. Patches that drink harder go off sooner, which forces the trowel, burnishes some areas and leaves others soft, and the difference shows up under raking light long after everyone has left. Even suction across a whole elevation is worth an extra half-day of damping down before the gang starts.

Texture consistency depends on gang discipline more than on individual skill. One hand mixing, one hand applying, a fixed water measure per bag, and no changing sand pile halfway up an elevation. A wall finished by two people working to slightly different consistencies reads as two walls forever, and no coating hides it.

Interior gypsum work runs under its own document, ASTM C842 Standard Specification for Application of Interior Gypsum Plaster, and carries its own hard limit: gypsum plaster has no business on a permanently damp wall, below grade, or in contact with cement render in a wet location.

Timing, Curing and the Weather Window

Time between coats is a moisture decision rather than a calendar one. Cement-based coats need the previous coat to have gained enough strength to take the next without being so dry that it strips water out of it, and the intervals and moist-cure periods are set out in ASTM C926 alongside the temperature conditions they assume. Read the specification for the job rather than repeating the interval from the last one, because the interval that worked in October will be wrong in July.

Cold weather stops the work rather than slowing it. Below the minimum application temperature named in the specification, cement plaster gains strength so slowly that a night frost can disrupt the surface before it has any, and a frosted finish coat will dust and craze no matter how well it was applied. Heated enclosures shift the problem instead of solving it if they also dry the wall out.

Hot dry wind does more damage than heat alone. Plastic shrinkage cracking on a finish coat comes from evaporation outrunning the mix, not from a bad mix, and the fix is shade, wind breaks, and fog spraying rather than extra water in the gauge. Water added at the board to keep material workable in the heat is water that lowers the strength of the very coat that is supposed to be weakest and most exposed.

Protection after application matters as much as the application. Cover new external render against driving rain for as long as the specification requires, keep it damp through the cure without soaking it, and do not strike scaffold sheeting on a hot afternoon just because the wall looks finished.

Reading a Failure Back Down the Coats

Map crazing over a whole elevation points at the finish coat: too rich, too wet, or dried too fast. Cracks in straight lines that follow framing or run out of an opening corner point at movement and jointing, not at the mortar. Cracks that appear at a change of background, brick to block or masonry to lath, point at the lath detailing over the transition.

Hollow-sounding areas that ring under a tapped knuckle are a bond failure, and the coat that failed is identifiable by where the fracture surface sits. A clean release at the background face means suction or contamination beat you before the first coat set. A fracture inside the scratch coat with mortar left on both faces means the coat itself is weak, which is usually water added on the board.

Salt bloom returning after every wash is not a plaster fault. It is a water path, and it will keep returning until the leak, the missing weep screed, the bridged cavity or the ground contact that feeds it is dealt with. Rendering over active efflorescence traps the salt behind a new face and moves the eventual blowout up the wall.

Diagnosis before demolition saves the whole elevation. Sound the wall, cut a small square out to the background, look at which face the fracture chose, and only then decide whether you are patching a bay, re-rendering an elevation, or re-lathing it. Cutting first and reasoning afterwards is how a two-day repair becomes a full scaffold.