Rainwater First Flush Diverter Volume Calculator

Introduction to first flush diversion for roof catchments

Rainwater harvesting is only as clean as the roof it starts on. Between storms a catchment surface accumulates dust, pollen, soot, tyre and brake particles, leaf litter, insect fragments and bird and rodent droppings. The moment rain begins, that material is mobilised in a short, highly concentrated pulse commonly called the first flush. Monitoring studies summarised in the Virginia Rainwater Harvesting Manual show that sediment, metals, bacteria and even pesticides all peak in the opening minutes of an event and then fall back towards a much lower steady-state concentration for the rest of the same storm. Keeping that opening pulse out of the cistern is one of the cheapest water quality interventions available to a harvesting system.

A first flush diverter is the device that does it. The simplest version, described in the Texas Water Development Board manual, is a vertical PVC standpipe hung off the downpipe: it fills with the dirtiest water first, backs up, and only then lets the remaining flow continue to the tank. A widely used refinement adds a floating ball that rises with the water and seals against a seat once the chamber is full, so the trapped pre-wash water cannot re-mix with the clean flow behind it. Other systems use tipping buckets, vortex pre-filters, or wetted-mesh filters that divert until the screen is wet. Whatever the hardware, the design question is identical: how much water must the chamber hold?

This calculator answers that question from the roof area draining to the device and the depth of rainfall you intend to sacrifice, then carries the answer through to a physical standpipe length and a water-cost check against a design storm. It reports results in litres and gallons at the same time, and it plots your chosen diversion rate against the published benchmarks so you can see immediately whether you are inside, below or above the range that recognised manuals recommend.

How to use the diverter sizing tool, field by field

1. Unit system. Switch between metric (m², mm, litres, millimetre pipe bore) and US customary (ft², inches, gallons, inch pipe bore). Every numeric field you have already filled in is converted for you, so you can enter a roof in square feet and read the chamber in litres, or the reverse.
2. Roof catchment area. Use the horizontal projected area of the part of the roof that drains to this diverter, not the sloped surface area and not the whole building. If four downspouts share a roof, divide by four (or by the true split) and size a diverter for each one.
3. Diversion guidance preset. Pick one of the published benchmarks and the depth field is filled in for you: the Texas Water Development Board minimum of 10 gallons per 1,000 ft², its upper recommendation of 2 gallons per 100 ft², the 25 gallons per 1,000 ft² (1 mm) used in the Virginia manual, or the 49 gallons per 1,000 ft² upper bound of the vendor study the Texas manual quotes. Choose Custom to type your own.
4. First flush diversion depth. The depth of rainfall you intend to discard, in millimetres or inches. This is a rainfall depth, not a runoff depth, which is how every source in the reference list expresses it.
5. What-if depth slider. Drag it, or focus it and use the arrow keys, to sweep the diversion depth from 0.10 mm to 3.00 mm. The depth field, the result panel and the benchmark scale all update live once you have calculated once, so you can watch the chamber volume and the share of the storm you sacrifice move together.
6. Diverter capture efficiency. The fraction of the intended depth the device really retains, expressed as a percentage. The chamber volume is the target volume divided by this figure, which is an explicit oversize allowance for wetting losses, leakage past the ball seat and slow-release drains that bleed water out while the storm is still running. 90 % is a reasonable default; drop it if your drain orifice is generous or the chamber is shallow and mixes badly.
7. Standpipe internal diameter. Choose a preset with the published actual bore — ARCSA lists 4.046 in for nominal 4 in PVC and 6.065 in for nominal 6 in — or enter your own bore for a tank, a barrel or a metric pipe. Pick Not a pipe if you plan to use a moulded tank and only want the volume.
8. Design storm depth and collection efficiency. These drive the water-cost check. The storm depth is a representative event for your site; the collection efficiency is the runoff coefficient of the catchment and conveyance together. The Texas manual reports that most installers assume 75 % to 90 %, and the ARCSA standard uses roof coefficients of approximately 0.90.
9. Press Calculate. Use Reset to return every field to its default, and Download summary to save a plain-text record of the inputs, the results and the sources for your design file.

The first flush sizing formula and its unit conversions

The metric form is the reason first flush sizing is usually done in millimetres: one millimetre of rain falling on one square metre is exactly one litre, because 1 m² × 0.001 m = 0.001 m³ = 1 L. The target, or ideal, first flush volume is therefore the roof area multiplied by the diversion depth with no conversion factor at all.

The ideal first flush volume in litres equals the catchment area in square metres multiplied by the diversion depth in millimetres.

In US customary units the same volume needs the depth-to-volume factor that the ARCSA design standard states explicitly: one inch of water covering one square foot is 0.623 gallons, because 7.48 gallons per cubic foot divided by 12 inches per foot gives 0.6234. The Texas manual rounds the same number to 0.62 gallons per square foot per inch.

The ideal first flush volume in gallons equals 0.623 multiplied by the catchment area in square feet multiplied by the diversion depth in inches.

The chamber you actually build has to be larger than the target, because no diverter retains every drop it is aimed at. Dividing by the capture efficiency expressed as a fraction gives the installed capacity.

The chamber volume equals the ideal first flush volume divided by the capture efficiency expressed as a fraction.

Because the published guidance is written as a volume per unit of catchment area rather than as a depth, the calculator also reports the diversion rate in gallons per 1,000 ft². That is what makes your design directly comparable with the Texas and Virginia figures. One millimetre of diversion depth is 24.54 gallons per 1,000 ft².

The diversion rate equals one thousand multiplied by the ideal volume divided by the catchment area.

Turning a volume into a standpipe means dividing by the internal cross-sectional area of the pipe. For a circular bore of internal diameter the volume held per unit length is the area of the circle, and the required length follows directly.

The volume held per unit length of pipe equals pi times the internal diameter squared divided by four.

The required standpipe length equals four times the chamber volume divided by pi times the internal diameter squared.

Finally, the water-cost check uses the standard catchment yield relation. The runoff a storm of depth delivers to the tank inlet is the area multiplied by the depth multiplied by the collection efficiency (runoff coefficient) . In US units the same expression carries the 0.623 factor.

The storm runoff equals the catchment area multiplied by the storm rainfall depth multiplied by the collection efficiency.

The fraction of storm runoff discarded equals the diversion depth divided by the product of the storm depth and the collection efficiency.

• V_ideal – target first flush volume (litres or gallons)
• V_chamber – installed diverter capacity after the efficiency allowance
• A – horizontal projected catchment area draining to this diverter (m² or ft²)
• d – first flush diversion depth (mm or in)
• η – diverter capture efficiency as a fraction (90 % → 0.90)
• r – diversion rate, reported in gallons per 1,000 ft² for comparison with published guidance
• D – internal diameter of the standpipe (mm or in)
• L – required standpipe length (m or ft)
• P, C, f – design storm depth, collection efficiency, and the fraction of that storm's runoff discarded

Published diversion benchmarks compared side by side

Every figure below is converted from the source's own units using 1 US gal = 3.785411784 L and 1,000 ft² = 92.903 m². Use it as a sanity check on whatever the calculator returns: a design far outside this table is not necessarily wrong, but it should be a deliberate choice.

From litres to a standpipe: pipe volume per unit length

The two independent sources agree once you allow for the bore each of them assumed. ARCSA tabulates storage volume against length for real PVC bores; the Texas manual expresses the reciprocal, quoting the length of pipe needed per gallon. The right-hand columns below are recomputed from and reproduce both published sets, which is a useful check that the geometry in this calculator is the geometry in the manuals.

Worked example: a 150 m² metal roof on four downspouts

A rural house has a standing-seam metal roof with a 150 m² footprint, drained by four downspouts that split the roof roughly evenly. The owner follows the Virginia manual and targets 1.0 mm of diversion, allows 90 % capture efficiency, and plans to build the chambers from nominal 6 in PVC with the 6.065 in (154.05 mm) bore ARCSA lists. A 10 mm design storm and a collection efficiency of 85 % — the value the Texas manual uses in its own worked monthly water balance — give the water-cost check.

1. Target first flush volume for the whole roof: 150 m² × 1.0 mm = 150 L (39.6 gal).
2. Installed chamber capacity: 150 ÷ 0.90 = 166.7 L (44.0 gal). The efficiency allowance adds 16.7 L.
3. Diversion rate for comparison: 1.0 mm = 24.5 gal per 1,000 ft², which sits just inside the 13–49 band and just below the Virginia figure of 25.
4. Volume per metre of 154.05 mm bore: π × 0.15405 m² / 4 = 0.018639 m³ per metre = 18.64 L/m.
5. Standpipe for the whole roof: 166.7 ÷ 18.64 = 8.94 m, which is obviously unbuildable under a single eave.
6. Split across four downspouts: 41.7 L each, so 41.7 ÷ 18.64 = 2.24 m of 6 in pipe per downspout. That is a normal, buildable standpipe.
7. Water cost: the 10 mm storm yields 150 × 10 × 0.85 = 1,275 L of runoff, of which 150 L is diverted, leaving 1,125 L for the tank. The diverter discards 11.8 % of that storm.

Step 5 is the single most useful line in the example, and it is the reason the calculator carries the volume all the way through to a length. A number in litres looks harmless; nine metres of vertical 6 inch pipe does not. Step 6 is exactly the response the Texas manual anticipates when it notes that roofs producing quantities of runoff which require multiple downspouts will need a diversion device on each of them.

Reading your result: chamber, benchmark position and water cost

The headline figure is the installed chamber capacity in litres and gallons. Round it up to a real component: a moulded tank's nominal capacity, or a pipe length cut to a convenient dimension. Rounding down defeats the purpose, because the last part of the diverted volume is the part that determines whether the ball seals before or after the dirty pulse has passed.

The benchmark scale under the headline places your diversion rate against the published figures. Landing below the 10 gal per 1,000 ft² mark means you are diverting less than the minimum rule of thumb in the Texas manual, which is defensible only on a clean, steeply pitched, well-screened roof in an open setting. Landing above 49 means you are past the upper bound of every source in the reference list, which is defensible after a long drought or under heavy tree cover but expensive in a dry climate.

The water-cost line converts that trade-off into litres. If your design discards more than roughly a fifth of a typical storm you are paying a real price in yield, and the better lever is usually pre-filtration — leaf screens, gutter guards, a vortex filter — rather than a bigger diverter. If the design storm is smaller than the diversion depth divided by the collection efficiency, the calculator says so explicitly: in that case the chamber never fills, the ball never seals, and nothing at all reaches the tank during that event.

Maintenance and drain-down between rain events

A diverter that does not empty is a diverter that does not work: the next storm arrives, finds the chamber already full, and passes straight through. ARCSA/ASPE 63 makes this a requirement rather than a suggestion, stating that first flush diverters and pre-filters shall be provided with an automatic means of self draining between rain events, that the drained water shall be piped away from the storage tank to a location that will not cause damage or erosion, and that roof washers shall be readily accessible for regular maintenance. The same section requires an inlet debris screen with openings no larger than 0.5 in and no smaller than 0.25 in unless the pre-filter is self-cleaning.

• Slow-release drain. A drilled orifice or a slightly open hose bibb at the base, as illustrated in the Texas manual, empties the chamber over a few hours. It satisfies the automatic self-draining requirement, but note that whatever leaks out during the storm is capacity you never had — that loss belongs in the capture efficiency input.
• Manual drain valve. Simpler and fully effective, and it lets you route the diverted water to an irrigation bed instead of wasting it, but it depends on someone opening the valve after every event.
• Clean-out. The Texas manual notes standpipes usually have a cleanout fitting at the bottom and must be emptied and cleaned after each rainfall event. Sediment accumulates fastest where the diverted water is dirtiest, which is precisely the systems that need the diverter most.

After a long dry spell the first storm of the season carries a disproportionate load. Some operators temporarily raise the diversion depth for that event and drop it back afterwards; the what-if slider is a quick way to see what that costs in chamber volume and in discarded yield.

Limitations and assumptions behind a fixed-volume model

• Fixed volume, variable first flush. The model assumes a fixed diverted volume. The Virginia manual points out that the depth of rain needed to rinse a roof depends on rainfall intensity and on the type and condition of the roof, so a wetted-mesh filter that diverts in proportion to intensity may match the true first flush better than any fixed chamber.
• Rainfall depth, not runoff depth. The diversion depth is applied to the rainfall, matching how every cited source states its guidance. Wetting losses on the roof mean slightly less than that depth actually reaches the diverter in a light event; the collection efficiency input is used for the storm yield check but deliberately not applied to the diversion target itself.
• Uniform catchment. Wind-driven rain, valleys, dormers, partial roof wetting and unequal downspout splits all move water around in ways a single area figure cannot capture. Size per downspout and be conservative about the split.
• Capture efficiency is a design margin. No standard publishes a diverter capture efficiency. It is an installer's allowance, and if you can measure your own chamber's retained volume you should replace the assumed figure with the measurement.
• Geometry is idealised. The pipe calculation assumes a plain cylinder of the stated internal bore, full to the top. Real chambers lose capacity to the ball assembly, the seat, tee fittings, the drain and any sediment left from the previous storm; add a margin or round up a size.
• Not a treatment step. Diversion reduces the load reaching the cistern. It does not deliver potable water, and it is not a substitute for the filtration and disinfection that ARCSA/ASPE/ANSI 63 requires for potable end uses. Local regulations govern.

Common sizing questions about first flush diverters

Sources for the diversion depths, factors and pipe volumes