Coverage Rate and Spread

A rate is not a quantity

A tin does not tell you how much to buy. It tells you how far a litre goes, and the buying decision is a division somebody performs afterwards. Roughly twenty calculators here perform it, and everything on this page concerns what that division quietly assumes.

The rate arrives in two reciprocal forms, and the words for them collide. Area per unit of product — square metres per litre, square feet per US gallon — belongs in the denominator. Product per unit of area — litres per square metre, kilograms per hectare for seed and fertiliser — multiplies. THE NAMES ARE NOT A RELIABLE GUIDE TO WHICH YOU HAVE: coatings call the first one the spreading rate, and ISO 6504 defines it that way, which is why a tin of paint prints its spreading rate in m²/L; agriculture calls the second one the spread rate and means kg/ha. Read the units, never the noun. Coatings and sealers are usually quoted the first way, construction adhesives, sprayed bituminous coats and fertiliser almost always the second. Feeding one into the slot meant for the other does not produce a plausible wrong answer; it produces an answer wrong by the square of the rate, which is at least visible.

The imperial and metric conventions do not convert by anything memorable. One square foot per US gallon is 0.024542 square metres per litre, so a datasheet reading 350 ft²/gal is 8.59 m²/L. Rounding that down to a comfortable 8 shifts the order by seven percent before a single real uncertainty has been applied, and the real uncertainties are much larger than seven percent.

Where the number on the tin comes from

The published rate is not found by painting a wall and seeing how far the tin went. It is calculated, from two properties of the product: how much of the wet material remains once the solvent or water has left, and how thick the remaining film is meant to be.

Volume solids is the first. A product at fifty percent volume solids leaves half its wet volume behind as film and evaporates the other half. The specified dry film thickness is the second, and on protective work it is the actual requirement — ISO 12944-5 writes a protective paint system as a nominal dry film thickness for each coat, never as a quantity of paint. The coverage rate is a consequence of those two figures, not an independent measurement.

Two of the pages listed below run this arithmetic in the open rather than folding it into a constant. The intumescent fireproofing and cold-applied liquid membrane calculators ask for solids content and a target dry film thickness, because on that work the film thickness is the performance requirement and no generic coverage figure can be substituted for it. A product at 70 percent solids needs a wet film almost half again as thick as the dry film it is being specified to leave.

The word worth holding on to is theoretical. This figure assumes every drop lands on the surface, at exactly the intended thickness, over a plane. None of those three assumptions survives contact with a building.

Per coat, and the coats are not equal

A coverage rate is quoted per coat. That is the single most frequent misreading of a datasheet, and it fails in the expensive direction: a specification calling for two coats at 200 square feet per gallon delivers 100 square feet for every gallon bought. With a single exception, every calculator here that offers a coats input multiplies the area before dividing by the rate, which is why the breakdown line reads area across all coats rather than area of wall.

The one page that inverts this is worth knowing about because it looks like the same field and is not. A DIY epoxy floor kit is rated for a complete two-coat system — 250 square feet, or 23 square metres, for the pair — so its coats input scales against two rather than against one. Reading that kit rating as per-coat orders twice the epoxy required.

What the multiplication cannot capture is that the coats differ from one another. On a bare, absorbent substrate the first coat is partly taken into the material rather than left on it. The second goes onto a surface the first has already sealed, and travels considerably further. Applying one rate to both averages a hungry first coat against a thrifty second — near enough right on a repaint over sound existing paint, where nothing is absorbed, and wrong in both halves on new plaster, bare timber or fresh masonry.

This is the reason primer is estimated separately rather than counted as a coat of paint. The primer page carries 200 square feet per gallon where the exterior topcoat page carries 300, and only part of that gap is primer being the thicker, more heavily pigmented liquid. The rest of it is that primer is the coat doing the absorbing. Fold it into the topcoat count and both figures come out wrong at once.

Porosity and texture move it further than anything else

Take one product family — a pigmented liquid, brushed, rolled or squeegeed onto a more or less flat surface — and line up the working rates this site carries for it. Smooth interior board sits in the 8 to 14 m²/L band that the interior paint page treats as plausible. Exterior siding: 300 ft²/gal, 7.4 m²/L. Deck boards and paver sealer: 200 and 175, near 4.9 and 4.3. Rough-sawn fence pickets: 150, or 3.7. Squeegee-applied driveway sealcoat: 80, under 2.

That is a factor of four to seven between the top and bottom of a list whose arithmetic is identical throughout — four if the interior band is read at its low end, seven at its high one. The products differ, but not by anything like that. What differs is the substrate, and each step down the list is a surface holding more true area than its measured area and more open pore to fill before any film exists at all. A tape measure returns the plan figure. The liquid follows the real one.

Measurement standards for protective coatings confront the same effect from the opposite end. ISO 19840 covers dry film thickness measured on rough surfaces and requires the gauge readings to be corrected for the surface profile — a measurement standard conceding that a blast-cleaned steel face holds paint down in the profile where an uncorrected instrument counts it as film standing over the peaks. That material is bought and applied, and none of it counts toward the thickness the specification asked for.

SSPC-PA 2 makes the corresponding point about acceptance: a coating conforms on the basis of measured film thickness, not on how many tins were emptied over it. Consumption is evidence of thickness and never proof of it, which is the honest limit of every figure this method produces. The order these calculators generate is a purchasing quantity. It is not a statement that the applied film will meet a specification.

The same asymmetry runs through the sprayed bituminous coats. A prime coat is meant to soak into a granular base, so an open crushed-stone base drinks several times what a tight, well-compacted dense-graded base will take, and the rate has to be chosen for the base actually present rather than the base the specification imagined.

Those two pages carry a second trap that has nothing to do with the substrate. An emulsion rate can be written as applied — the diluted liquid leaving the distributor — or as residual, the binder left behind once the water has broken off, and the residual figure is roughly a third of the applied one. Both calculators use whatever rate they are given, so reading a specification the wrong way round is an error of about three to one in a field that looks identical either way.

Beads, bulk, and the material that never reaches the surface

Four of the pages listed here divide a length rather than an area, and their rate carries a shape assumption folded inside it. A cartridge yields a run, and the run it yields depends entirely on the cross-section of the bead being laid.

The caulk page uses 25 linear feet from a 10.1 fluid ounce cartridge, and that constant can be checked rather than trusted: 299 millilitres divided by a quarter-inch square bead is 7.4 metres, or a little over 24 feet. It is also more fragile than it looks. Take the bead to half an inch in both directions and the cross-section quadruples — the same cartridge now yields about six feet. A linear rate quoted for one bead size is not a rate for sealant, and the same caution applies to a striping rate quoted for a four-inch line and a crack-filler rate quoted for a crack no wider than a quarter inch.

Bulk materials sit at the far opposite extreme, and their coverage figures deserve none of this suspicion. When a supplier says a cubic metre covers twenty square metres, no absorption, film formation or substrate texture is involved: that is a fifty-millimetre depth restated. The trade identity behind it — cubic yards equals square feet times inches divided by 324 — is nothing but 12 inches to the foot and 27 cubic feet to the yard. Figures of that kind are exact, and the uncertainty this page is about does not apply to them.

Finally, every rate discussed here describes material that arrives where it was aimed. None of them describe what does not: overspray past the edge of the work, tack coat lifted away on haul truck tyres, the film left in the tray and locked in the roller nap, the residue clinging to the drum. Those are losses, and they belong to a waste factor rather than to the rate. Inflating a coverage rate to cover site losses buries the assumption in a number that also has to serve as a physical property, and the next person to check the working has no way to see what was done. Keeping the two apart has a cost worth stating plainly. A few of the pages listed here do add something for loss: caulk, duct mastic and joint compound each bake in a flat ten percent, and the render-and-plaster and tile-adhesive pages expose a waste factor for you to set. On most of the rest the loss is real, the allowance is absent, and adding it is yours to do — which is what the overspray and transfer-efficiency page listed here exists for, and why it takes the quantity that must land on the surface as its input rather than recomputing an area.