A third full, and slower than the chart says
Take the defaults: a 3.83 inch inside diameter line, n of 0.013, a quarter inch of fall per foot, and 25 gallons a minute going down it. The line runs 1.24 inches deep, which is 32 percent of the diameter, and the water in it is moving at 2.48 feet a second. Fill the same pipe and it would carry 109.8 GPM at 3.06 feet a second — so the real velocity is 81 percent of the full-bore figure, and the real flow is 22.8 percent of the capacity.
Sixty feet of that run takes the surge 24.2 seconds to cross. Running full it would take 19.6.
The pipe carries most when it is not full
The page computes two landmarks rather than asserting them, and they are worth knowing. Velocity in a circular pipe peaks at about 81 percent of the diameter, and capacity peaks at about 94 percent — 118.1 GPM at the default slope, against 109.8 GPM brim full. Past that point the last strip of perimeter around the top of the pipe adds friction faster than it adds cross-section, so a pipe running dead full is carrying less than the same pipe running an inch below the crown.
This is also why the depth answer is found by searching below the 94 percent point. Above it the relationship reverses and two depths pass the same flow, which is not a helpful thing for a page to be ambiguous about.
Bigger pipe, same flow, slower water — usually
Add two inches of diameter at the default flow and slope and the line runs 1.08 inches deep at 2.37 feet a second, against 1.24 inches at 2.48. Slower, and by only four percent. That is the whole of the classic argument about oversized drain lines, and the honest version of it is that the effect is real but small at this end of the range: the wider channel spreads the same flow into a shallower stream, which costs hydraulic radius, while the bigger pipe gives some back.
Which of those two wins is not fixed, which is why the page computes it rather than printing a rule. Sweeping the shipped function across pipe sizes from 2 to 12 inches, slopes from a sixteenth to two inches per foot and flows over three orders of magnitude, the bigger pipe comes out slower in about 95 percent of cases — and every single exception has the smaller pipe running more than half full. Below half full there is not one. Between 60 and 70 percent of the diameter about a quarter of cases flip, above 80 percent all of them do, because by then the smaller pipe is losing so much to its own wetted perimeter that a wider channel is a net gain. At household surge flows in a 3 or 4 inch line the depth is nowhere near that, and the bigger pipe is slower.
Slope buys less than people expect
Going from a quarter inch per foot to five sixteenths — a 25 percent increase in slope — takes the velocity from 2.48 to 2.68 feet a second, up 8.1 percent, and the depth from 1.24 to 1.17 inches. A quarter more slope bought eight percent more speed. Velocity follows the square root of the slope at a fixed depth, and the depth drops at the same time, so the two effects partly cancel. This is the arithmetic behind the observation that very steep drain lines do not scour dramatically better than moderate ones, though what slope you are permitted to build is a code question and not one this page answers.
A surge is not a steady flow
Everything above is uniform steady flow. A toilet emptying in four seconds is a wave: deeper at its front, stretching as it runs, arriving at the tank spread over rather longer than it left. Treat the depth as roughly what the middle of the surge reaches and the travel time as the time the front of it takes. Over a sixty foot run at these numbers the difference is a few seconds; over three hundred feet it is not.
Questions people ask
How full does a sewer line run?
Far less full than people picture. A 3.83 inch line at a quarter inch per foot running 25 gallons a minute is 1.24 inches deep, about a third of the diameter. It would take 109.8 GPM to fill it, and nothing in a house produces that except several fixtures dumping at once. The line spends nearly all its life empty and the rest of it a third full.
What velocity does a drain line need?
That is a code and design question, set by whatever your jurisdiction has adopted, and this page does not state a figure. What it does is tell you the velocity you actually have at a flow you specify — including the fact that a part-full pipe runs slower than the full-bore number on a capacity chart, 81 percent of it in the default case.
Does a bigger pipe really flow slower?
At a fixed flow and slope, usually yes. Two inches more diameter on the default case takes the velocity from 2.48 to 2.37 feet a second, because the same water spreads into a shallower stream. It is not universal — where the smaller pipe is already running around half full or more it is losing to its own wetted perimeter and the bigger pipe can come out ahead — so the page computes it for your numbers instead of asserting a rule.
Why does the pipe carry more at 94 percent full than at 100 percent?
Because the top of the circle adds a lot of wetted perimeter and very little area. Friction goes with perimeter and flow goes with area, so the last few percent of depth costs more than it earns. The same geometry puts the peak velocity lower still, at about 81 percent of the diameter. Both figures are computed on the page rather than quoted.
What flow should I use for a house?
A surge rather than a daily average. Switch the form to the discharge mode and enter what a fixture dumps and how long it takes: 1.6 gallons in 4 seconds is 24 gallons a minute, which is a rate no average will ever show you. The daily total belongs on the flow calculator; the pipe only cares about the peak.