Will there be pressure left?
Fixture units total up, demand comes off the curve in your book, and then the question that matters: after lift, meter, fittings and friction, is there enough left at the top?
Water supply: fixture units, then a pressure budget
Fixture units total up. Demand comes off Hunter's curve in your book. Then the real question is whether there is pressure left when the water arrives.
Per-fixture WSFU values come from your code's fixture table — and note whether the row is for a flush tank or a flushometer valve, because they differ sharply. The demand in GPM then comes off Hunter's curve for the total. That relationship is a probability curve, not a formula: it flattens hard, so scaling it linearly overshoots badly on big systems. Look it up; this page will not fake it.
The fixtures
Demand, off the curve
The pressure budget
Everything the water spends on the way to the highest fixture comes out of the street pressure. Whatever is left has to meet the fixture's minimum.
How this is calculated
total WSFU = Σ (count × WSFU per fixture), kept split between flushometer and tank fixtures.
velocity = GPM ÷ (2.448 × d²) where d is the inside diameter in inches.
available = street pressure − static lift − meter − fittings − friction, with static lift = height × 0.433 psi/ft at 60°F, corrected for water density at other temperatures.
No fixture-unit-to-GPM conversion is offered. Hunter's curve is a probability model, printed as a graph or table because it has no closed form. Any formula this page gave you would be a fabrication, and it would be quoted back as though it were the code.
Why flushometers are kept separate
A flush tank refills gently over a minute or so. A flushometer valve takes its whole volume in a few seconds at a very high rate. The fixture unit values reflect that, and the demand curves for the two are genuinely different curves — not the same curve read differently.
This matters most on small systems. One flushometer in a building of tank fixtures can dominate the peak demand entirely, because the probability of it being open at the same moment as anything else is what the curve is modelling. Total everything as though it were tank-fed and the main comes out too small, and the symptom is a fixture that works fine until someone flushes.
Sizing is really a pressure budget
Supply sizing is usually taught as "pick a pipe size", which hides what is actually going on. There is a fixed amount of pressure at the street, and four things spend it: lifting the water to height, pushing it through the meter, pushing it through fittings, and overcoming friction in the pipe. Whatever survives has to satisfy the fixture.
Seeing it as a subtraction tells you something a pipe-size table cannot: only friction responds to a larger pipe. The static lift is pure geometry — 0.433 psi for every foot of height, and no pipe size changes it. If your budget is short and friction is already a small part of it, upsizing will not rescue the run, and the honest answer is a booster pump or a different service pressure. Plenty of money gets spent on larger pipe for buildings whose real problem was height.
Velocity is a separate constraint
A pipe can satisfy the pressure budget and still be the wrong size. Water moving too fast erodes copper at the fittings and makes noise that carries through the structure; the usual ceiling is around 8 ft/s cold and lower on hot lines, where erosion is faster. This is why the velocity check sits alongside the pressure check rather than inside it — passing one does not excuse the other.