Where 96.3 comes from, and why it is not a fudge factor
The constant is pure unit conversion, and it is worth seeing once because it tells you exactly what the formula assumes. A US gallon is 231 cubic inches by definition. A zone running at Q gallons per minute therefore delivers 231 × 60 = 13,860 cubic inches of water per hour for every gallon per minute. Spread that over an area measured in square feet — 144 square inches each — and the depth in inches per hour is:
PR = (231 × 60 ÷ 144) × Q ÷ (S × L) = 96.25 × Q ÷ (S × L)
PR (in/hr) = 96.3 × zone GPM ÷ (S × L in feet)
96.25 rounded to three figures is 96.3, which is the number printed in every irrigation design manual. There is no hidden efficiency term and no empirical fitting in it: it is gallons and feet turned into inches, nothing more. That also means it answers a narrow question honestly — the average depth over the area, assuming the heads between them cover that area. It says nothing about whether the water landed evenly, which is what distribution uniformity is for and why this tool asks for that separately.
The area a head is responsible for
The spacing numbers are not the head's throw radius — they are the footprint each head has to cover. On a square or rectangular grid that footprint is simply S × L: each head sits at a corner of the rectangle and contributes a quarter of itself to each of the four around it, which adds back up to one whole head per rectangle. Triangular (staggered) spacing is the same accounting with different geometry. Offset each row by half a space and the rows can sit closer together for the same throw — specifically at 0.866 of the in-row spacing, because that factor is sin 60°, the height of the equilateral triangle the three nearest heads form.
L = 0.866 × S ⇒ PR = 96.3 × Q ÷ (0.866 × S²)
The consequence surprises people: for the same heads and the same in-row spacing, triangular spacing packs the rows about 13% closer, so the same flow lands on about 13% less ground and the precipitation rate goes up by the same proportion. Staggering the rows buys better uniformity, not a slower rate. If you re-space a zone from square to triangular and leave the controller alone, you have quietly started over-watering it.
Matched precipitation rate, and the zone that cannot be scheduled
A run time is a single number and a zone gets one. That only works if every head in the zone applies water at the same rate. A full-circle head has to cover four times the ground of a quarter-circle head at the same spacing, so it has to flow four times as much water to match it — which is exactly what manufacturers build matched-precipitation-rate nozzle sets to do. Mix a rotor and a spray head on one valve and you are not out by a few percent; sprays typically apply water several times faster than rotors, so whichever one you schedule for, the other is badly wrong. The tool computes each nozzle group's own rate and flags the spread.
The convention this page follows is that the design run time serves the lowest rate present in the zone, because under-watering shows up as dead grass and over-watering shows up as a water bill. That is a defensible default and not a law — a mismatched zone's real fix is a valve, not a compromise.
Why the number the soil accepts is a different number
Water enters soil quickly at first and then progressively slower as the profile wets and the larger pores fill. The rate approaches a roughly steady value — the basic or steady infiltration rate — which is what a long sprinkler run actually has to live within. Coarse sand takes water far faster than any sprinkler applies it. Clay does not: its steady rate is a small fraction of an inch per hour, well under what a spray zone puts down. When the precipitation rate exceeds the intake rate the surplus has nowhere to go, and it does what water does — it moves downhill, off the turf, onto the drive.
trunoff = allowable surface storage ÷ (PR − intake) × 60
cycles = ⌈ total run time ÷ tcycle ⌉
Splitting the run is the fix that does not require re-plumbing anything. Run the zone until the surface is about to shed water, stop, let what has already been applied drain out of the top inch, and run again. Repeat until the total adds up to the depth you wanted. The soak gap is not arbitrary — it has to be long enough for the surface layer to drain, which is why this tool derives it from the same intake rate rather than defaulting to a round number.
Distribution uniformity, and the run time it forces
A catch-can test measures the thing the formula cannot: whether the water arrived evenly. Set out cans on a grid, run the zone, and record what each caught. The average tells you the precipitation rate directly. The spread tells you something more useful — the low-quarter distribution uniformity, the mean of the driest quarter of the cans divided by the mean of all of them:
DUlq = mean of the lowest 25% of catches ÷ mean of all catches
PR = mean catch depth ÷ test minutes × 60
A DUlq of 1.00 would be perfectly even and does not happen. The reason it matters for scheduling is blunt: if you water until the driest quarter has had enough, everywhere else has had more than enough, and the worse the uniformity the larger that surplus. Fixing a low DU pays back faster than any scheduling change, because scheduling can only choose which parts of the lawn you over-water.
What this tool will not print