Why Sanitary and Storm Drainage Are Kept Separate

Why this matters

Sanitary flow and storm flow are two different animals wearing the same pipe. One is roughly steady, bounded by how many fixtures a building has and how many people can plausibly use them at once. The other is an event: nothing for weeks, then a peak that runs more than an order of magnitude higher for twenty minutes. They are separated because no single pipe and no treatment works can be economically sized for the second while doing a useful job on the first. The part a service shop meets is the consequence: one downspout tied into a sanitary lateral can deliver more flow than the entire building's sanitary peak, and it does it during a storm, at night, into a basement nobody is standing in.

When you arrive at a surcharge

• Sewage in an occupied space is a contact and inhalation hazard, not a mess. Keep occupants out of the affected area, wear nitrile or heavier gloves, splash goggles and boots, and ventilate the space to outside air before working in it for any length of time.
• Standing water in a basement or mechanical room around energized equipment is an electrical problem before it is a drainage problem. Open the disconnecting means for anything in or near the water, lock it, and verify dead at the point of work under 29 CFR 1910.333(b)(2) in general industry or 29 CFR 1926.417 on construction work, before anyone steps in it.
• Do not open a cleanout on a surcharged lateral while standing over it. The line is under head. Back the plug out a few turns from the side, in splash goggles and gauntlet gloves, and let it bleed.

What each system is sized for

Sanitary drainage is sized from drainage fixture units. A fixture unit is a probability-weighted allowance, not a flow rate, and the conversion from a fixture unit total to a probable peak flow is a curve that flattens as the count rises, precisely because a hundred fixtures never discharge together. That curve lives in the plumbing code your jurisdiction adopted and amended, and the number it produces is a peak that recurs daily and varies little across the year.

Storm drainage is sized from a rainfall event. The usual tool on a building-scale area is the rational method, Q equals C times i times A, with Q in cubic feet per second, i the rainfall intensity in inches per hour, and A the contributing area in acres. It is derived for small, reasonably uniform drainage areas with a short time of concentration, and it assumes the rainfall intensity is uniform over the area for that duration; carried onto a large watershed with mixed cover it stops being appropriate. C is the runoff coefficient, near 0.95 for a roof because a roof absorbs essentially nothing.

The two systems therefore answer different questions. One asks how many people are in the building. The other asks how hard it is raining.

Worked example: the two peaks on one building

A single-story building with a 10,000 square foot roof, offices, on a separated municipal system. The comparison below is what makes the separation non-negotiable.

Storm peak.

• Contributing area: 10,000 square feet divided by 43,560 equals 0.230 acres
• Runoff coefficient for a roof: 0.95
• Rainfall intensity: 4 inches per hour, taken from the local rainfall frequency data for the return period this jurisdiction requires. This value carries the arithmetic; it is not a national figure and yours will differ. Use the one your code official specifies.
• Q equals 0.95 times 4 times 0.230, which is 0.87 cubic feet per second
• Converted: 0.87 times 448.8 equals 392 gallons per minute

Sanitary peak. From this building's fixture unit total through the adopted code's conversion curve, the probable peak lands in the tens of gallons per minute. Call it 30 gallons per minute, from that table, and substitute your own jurisdiction's figure when you run this on a real building.

The ratio.

• 392 divided by 30 equals about 13

More than an order of magnitude, and the two peaks are not coincident: the sanitary peak occurs every weekday morning, the storm peak occurs for a matter of minutes, a handful of times a year.

What that ratio costs in pipe. For uniform flow in a circular pipe, capacity scales with diameter to the 8/3 power, so the diameter required scales with flow to the 3/8 power. That exponent holds when both cases are compared at the same slope and the same relative depth of flow, which is the condition it is derived under.

• 13 raised to the 0.375 power equals about 2.6
• A 4 inch sanitary building drain carrying the combined flow becomes roughly a 10 inch pipe

Why that is not simply an expensive answer. Because the pipe is the cheap half. Everything downstream of the building - the municipal collector, the interceptor, the pumping, the treatment works - would have to be sized for the same multiple, to treat a flow that for most of the year is not there. A treatment works sized for the storm peak is oversized by an order of magnitude for its normal duty, which degrades its performance on the flow it actually receives. Combined systems that do exist resolve this by overflowing untreated during large events, which is the entire reason separation is the modern practice and the reason separating an existing combined system is a generational public works project rather than a project.

The number that matters to a service shop. Run the same rational method on a single downspout serving 1,000 square feet of that roof, at the same intensity.

• 1,000 square feet divided by 43,560 equals 0.023 acres
• Q equals 0.95 times 4 times 0.023, which is 0.087 cubic feet per second
• Converted: 39 gallons per minute

One downspout delivers more than the entire building's sanitary peak of 30 gallons per minute. Not a share of it, more than all of it. That is why a single wrongly-connected leader is not a minor code deviation.

What a cross-connection does inside the building

Follow the 39 gallons per minute into a system that was sized for 30 and it stops being a hydraulics question and becomes the air-pressure question that runs through every drainage fault:

1. The lateral fills. Once it runs full, it is no longer a gravity drain, it is a pressurized pipe with the head of whatever is standing above it.
2. The stack above it fills from the bottom up. Air ahead of the water is compressed rather than vented, because the venting was sized for a system that does not do this.
3. Positive pressure at the base of the stack pushes trap seals upward, into the fixtures. The signature is bubbling into the lowest fixtures, water on the floor around them, and a smell that arrives with the weather rather than with occupancy. A sibling card covers back-pressure as one of the six mechanisms that empty a seal.
4. When the lateral is full, the lowest opening in the building overflows. That opening is the floor drain, because the floor drain is at the low point by definition. This is the situation backwater protection exists for.

Every one of those four steps happens during the storm and stops when the storm stops, which is why the complaint is so often reported as "the basement smells when it rains" and so rarely traced past that sentence.

Inflow versus infiltration, and why the distinction sets the fix

Water gets into a sanitary system two ways, and they need completely different answers.

Inflow is a direct connection: a downspout, a sump pump discharge, an areaway or yard drain, a foundation drain plumbed into the sanitary lateral, a manhole lid in a low spot. It arrives fast, peaks with the rain, and falls off almost as fast. On a flow chart it is a spike sitting on top of the rain event.

Infiltration is groundwater entering through defects: cracked pipe, failed joints, a bad lateral connection. It arrives slowly, rises over hours or days after the water table does, and takes days to fall back. On a flow chart it is a broad hump lagging the rain.

That difference is diagnostic before you dig anything. A sharp coincident peak means somebody connected something and the fix is a disconnection. A slow sustained rise means the pipe itself is leaking and the fix is rehabilitation, which is a different scope, a different price and a different contractor. Reading a peaky chart as infiltration sends a customer into a lining project that changes nothing, because the downspout is still connected.

Finding a cross-connection

Three methods, each with a hazard the method itself creates:

Dye tracing. Introduce a non-toxic tracer dye into the suspect storm inlet during dry weather and watch the sanitary cleanout downstream. Notify the sewer authority before you do it; coloured water arriving at a treatment works unannounced is an alarm condition and some authorities require prior notice as a condition of the sewer-use ordinance. Handle the dye per its safety data sheet, and keep it out of the potable system entirely.

Smoke testing. Blow tracer smoke into the sanitary lateral and watch for it emerging at roof leaders, yard drains and area drains. Two hazards, both created by the test rather than found by it. First, notify every occupant and, where the jurisdiction requires it, the fire department, before smoke is introduced, because smoke appearing inside a building reads as a fire and produces an evacuation, a call and occasionally an injury. Second, smoke entering an occupied space is itself a finding: it means a dry trap or a defect, and the trap gets refilled before anyone leaves the site. Follow the smoke product's safety data sheet for the respiratory control it requires in confined or poorly ventilated areas; a glove does nothing for an airborne product.

Flow monitoring. Meter the lateral through at least one rain event and compare against a dry-weather baseline. This is the only one of the three that separates inflow from infiltration on evidence rather than inference, and it is the one worth doing first on a building where nobody knows what is connected.

Where the protection goes when you cannot fix the source

A building on a system that surcharges for reasons outside the property line does not get to solve the problem at the source, and the answer is backwater protection at the fixtures below the surcharge elevation, sized, located and accessed per the code as adopted and amended by the jurisdiction. Two field points that the drawing rarely makes:

• A backwater valve protects what is below it. Fixtures above the surcharge elevation should not be routed through one, because a valve that closes during a storm also stops those fixtures draining, and now the building has an indoor problem it did not have.
• A valve without a maintainable access is a valve nobody services, and a backwater valve fouled open by a rag is a backwater valve in name only. Its access location belongs in the same conversation as its installation, not in a later one.

References

• The plumbing code and the storm drainage provisions as adopted and amended by the local jurisdiction, which set fixture unit conversions, storm sizing, design rainfall return period and backwater valve requirements
• The local sewer-use ordinance, which governs prohibited connections and any notification required before dye or smoke testing
• 29 CFR 1910.333(b)(2) for electrical lockout and verification in general industry; 29 CFR 1926.417 for the construction counterpart
• Safety data sheets for any tracer dye or smoke product introduced to a drainage system
• See related: What a Trap Primer Is Solving and When It Is the Wrong Answer; What a Floor Drain Needs That a Fixture Drain Does Not; Drainage Hydraulics Reference; What a Backflow Assembly Is Actually Protecting Against