Water is patient. It finds the low spot on a deck, settles in, and goes to work on the concrete, the waterproofing, the reinforcement, and the bearings below. Most bridge engineers would say they know this. Yet walk beneath almost any older urban viaduct and you will find rust runs down the fascia, staining across the pier caps, and a scupper that has been discharging onto the same patch of concrete for thirty years.

None of that is mysterious. It is the predictable result of a handful of drainage mistakes that keep getting made, largely because drainage is the last thing detailed and the first thing trimmed when the budget tightens. Here are the ones that do the most damage, and what to do instead.

Ponding Starts Before the Puddle Appears

Ponding is the most common complaint and the least useful way to describe the problem. By the time water is standing on the deck, the geometry has already lost. What you want to find at design stage is where the water will want to go, and whether anything is there to receive it.

Start with the vertical profile. On a sag curve the low point is not always where the drawing suggests, especially on a skewed or widening deck where the cross slope rotates through a transition. Plot finished surface grades along the gutter line, station by station, and mark the true low points. Then check them again once the superelevation transition is set — the low side of a curve switches, and drains laid out on an even spacing pattern will cheerfully sit on the high side of a curve while the water runs past them on the other edge.

Then there are tolerances. A deck placed to nominally correct grades can still hold a puddle three metres across where the screed dipped. A local depression of 15 mm is enough. This is why sensible details call for a shallow sump around each inlet — often in the region of 20–25 mm — and why the inlet frame should sit slightly below the surrounding surface rather than proud of it. A frame standing 10 mm high becomes a small dam, and the pavement upstream of it stays wet.

Inlets Placed by Convenience, Not by Hydrology

Equal spacing looks tidy on a plan. Water does not care about tidy. Contributing area changes along a bridge: a deck that widens at a ramp gore may triple its catchment in a few metres, and a grade that flattens pushes the gutter spread out towards the traffic lane.

The better habit is to compute the spread at the design flow and place inlets where the spread exceeds what you are prepared to accept — normally the edge of the shoulder or the edge of the running lane. Spacing then falls out of the calculation instead of the other way round.

Grate selection matters more than it looks. Generous open area with narrow slots clogs fast with leaves and windblown litter. Fewer, wider openings pass debris but are a hazard for cyclists, so slot orientation relative to the direction of travel deserves a moment's thought rather than a copied note. There is no universal grate; there is only the right grate for that site and that debris load.

The Pipe Run Nobody Checked

The pipe is the part of the system most likely to be drawn once and never examined again. A 150 mm pipe at 1% is generous. The same pipe with two 90-degree bends, a 40 m run beneath a wide deck, and a shallow fall is a different proposition entirely — friction losses accumulate, velocities drop, and sediment settles out at exactly the points you cannot reach.

Draw the run in profile, with every bend, every penetration through a diaphragm, and every change of material. Watch for the sag that appears when a hanger settles or a support is relocated during construction. Any low point in a drain line is a future blockage with a date already set.

Then follow each outlet to where the water actually lands. Discharge onto a pier cap, into abutment backfill, onto an unprotected slope, or straight down onto the road below are all common and all damaging. The consequences range from staining and freeze–thaw scaling to ice on the carriageway beneath the structure in winter. A drain that works and discharges in the wrong place is still a drain that is slowly damaging the bridge.

Maintenance Access Is a Design Decision

A drainage system that cannot be cleaned is not a drainage system. It is a sediment trap with a pipe attached. Access has to be drawn, dimensioned and defended like any other structural detail:

• A rodding eye at every change of direction, not only at the two ends of the run.
• Hatches in the soffit sized for the equipment crews actually use, not for a hand and a torch.
• Clear space around the drain body so a coupling can be undone with a wrench.
• Outlets placed where someone can see them from a platform or an inspection vehicle.
• Drain locations marked on the as-built drawings, so nobody has to hunt for them in the dark.

The classic failure is a drain body fully embedded in a concrete diaphragm with the pipe connection cast in place. When it blocks and the rod will not pass, the remedy is demolition and reconstruction of that section. A cast-in sleeve or a mechanical connection costs very little at construction time and saves an unreasonable amount later.

Details That Fall Between Disciplines

Roadway drainage plans tend to stop at the bridge joint. Bridge drainage plans tend to start at the deck. The few metres where the two meet are sometimes nobody's responsibility, and that is where water arriving at the joint finds nowhere to go except into the joint — and from there onto the bearings.

Intercept before the joint. Put the last inlet upstream of it, give the approach pavement a positive fall towards that inlet, and do not run a drainage connection across a moving joint unless it is genuinely designed for the movement. Scuppers discharging over the edge are cheap and easy to maintain, but they wet the fascia, the bearings and anyone working below; where they are used, shape the outlet to throw water clear of the face and protect the ground or slope beneath.

Materials, Corrosion and Replaceability

A galvanised drain body cast into chloride-exposed concrete is a corrosion cell waiting for a reason to start. Specify the material for the exposure it will see, avoid direct contact between dissimilar metals, and think about how the component will be replaced rather than only how it will be installed. Vibration loosens fixings that were never intended to be re-tightened, and a drain that cannot be unbolted will eventually be broken out.

A Review That Pays for Itself

Before the drawings leave the office, run through a short check. It takes an afternoon and it changes projects.

1. Plot finished surface grades at the gutter and mark the true low points, including sag curves and superelevation transitions.
2. Check inlet spacing against gutter spread, not against a standard note carried over from another job.
3. Draw every pipe in profile, with all bends, crossings and falls.
4. Follow every outlet to its discharge point and ask what the water will hit.
5. Explain, in words a maintenance crew would recognise, how each drain is reached and rodded.
6. Confirm who owns the transition from the roadway drainage plan to the bridge drainage plan — and get it in writing.

Bridge decks are judged by how they look at fifty years, not at opening. Drainage is one of the few things on the structure that can be got right on paper for almost no cost, and almost impossible to fix later without scaffolding, lane closures and a budget conversation nobody enjoys. Give it the same attention as the bearings, and the fascia will still be clean when the next generation inspects it.

Photo: Kranich17 / Pixabay