An aqueduct starts in a water collection system&action=edit&redlink=1 "Water collection systems (Rome) (not yet written)"). The water passes in a controlled manner to the pipeline from a header tank (caput aquae")), also known as castellum, from where it was distributed throughout the city.[1].
The construction of an aqueduct requires a thorough study of the terrain that will allow choosing the most economical route to allow a gentle and sustained slope without lengthening the route of the work too much.[15].
Channels (riui")) without pressure (circulating in a free layer) are used whenever possible and only on rare occasions is conduction under pressure used.[16].
In any case, whenever the water is intended for human consumption, the channel is covered by vaults, false vaults, stone plates or tegulas.[17].
The channel adapts to the terrain through different procedures. Whenever possible, it runs on the ground supported by a wall (substructio) in which culverts are made to facilitate the normal transit of surface waters. If the terrain rises, the canal is buried (riuus subterraneus")) and forms an underground gallery (specus")) excavated directly into the rock or built inside a trench. When a strong depression has to be overcome, the construction of complicated systems of arches (arcuationes) that support the channel and keep it at the appropriate level is resorted to.[18].
If a mountain that cannot be circumvented is in the way of the conduction route, the construction of a tunnel that pierces it is resorted to. This procedure is only used if it is unavoidable. Tunnels pose major technical problems. Normally they start at both ends, which requires great precision in the work so that the two branches meet at the planned point. The narrowness of the cutting areas requires that only one or two men work on each cut, so the work progresses very slowly.
Underground canal conduits are usually connected to the surface by means of wells (putei")) arranged at regular intervals. Through them you can access the aqueduct for cleaning and maintenance. In the case of tunnels, they were also used to extract debris and introduce materials during construction, as well as to ensure the correct layout and depth of the excavation.
The canals, unless they were directly excavated in impermeable rock, were lined with an impermeable mortar composed of lime and small fragments of crushed ceramic (opus signinum). The interior corners were protected with a convex molding (half-round) of the same material.
Although Roman technicians also used pressure conduction through lead pipes (fistulae) or ceramic (tubuli fictiles), they only rarely did so, since the poor technology available to them for the construction of pipes made them expensive and unsafe. The ceramic ones were cheap and easy to manufacture on site, but they were too fragile. The lead ones, apart from the cost of the material, required very laborious transportation, given their weight.
There were also rudimentary stone pipes, made up of large pierced ashlars that were joined together thanks to a tongue-and-groove joint that was sealed with lime mortar. This procedure was used to transport water through flat areas where it was impossible to maintain an adequate slope for driving through a free channel. Sometimes wooden tubes and channels also seem to have been used.
Siphons were used to overcome not very pronounced depressions, difficult to overcome by other systems. The water carried through the aqueduct channel was collected in a header tank from which it came out under pressure through the tubes of the descending branch, which sought the bottom of the depression. Behind it began the ascending branch that, once the depression was overcome, poured the water into an outlet tank, from which the water came out again through a free sheet channel. The angles formed by the tubes were ballasted with thick stones to prevent breakage in the joints due to the pressure that the water exerted on them. When the gap to be bridged was significant, the water pressure could burst the pipe, so it was more economical to bridge the ravine using an arch.
In some cases, the arch work of an aqueduct was used to also build a road bridge, as in the case of the Pont du Gard (hence it is known by the name of bridge, pont).
At different points in the pipes, devices were interspersed whose purpose was to calm the water flow to allow the solid waste to decant, water boxes or water arks. They all have in common that they consist of intermediate receptacles with the bottom at a lower level than the water inlet and outlet channels. The drags precipitate and fall to the bottom, from where they could be periodically extracted. The simplest type is a simple well made at the bottom of the canal. The largest specimens are small construction tanks (piscinae limariae")) covered with opus signinum.
When the pipeline reaches the city walls, its flow is collected in a terminal cistern, which serves to regulate the supply. From there, water is supplied to citizens through a distribution system&action=edit&redlink=1 "Water distribution systems (Rome) (not yet written)").
According to Frontinus, the Romans were satisfied for a long time with the water they extracted from the Tiber, wells and springs. Things were like this for the first 441 years since the founding of the city. Their first aqueduct was underground, the Aqua Apia"), which extended about 16 km, built on the initiative of the censor Appius Claudius the Censor in the year 312 BC. Later, they built the first one that carried water above the surface, the Aqua Marcia, in Rome, which ran about 90 km (144 BC). During the time of Frontinus, who wrote his treatise on the aqueducts of Rome in the 97 or very shortly after, the pipeline is underground for most of its route, but upon arrival near the city it has a short section that runs on the surface over a wall and arches.
The Old Anion pipeline, built in 273 BC. C., although it travels on the surface for a not very long stretch, it lacks arches, according to Frontino's description. All the other pipelines in Rome have significant sections of arches, which, in general, are longer the more modern they are. The aqueduct that fed Carthage, in present-day Tunisia, from the 1st century, ran a distance of 132 km from Zaguán, of which 17 km were in arches.
It is, therefore, a fact that the oldest aqueducts in the city tend to prefer underground conduction whenever possible. It is also known from Frontino that on occasions, with the passage of time, in some of the pipelines the detours required by the underground route were replaced by shorter routes over arches. The preference for underground routes in the oldest aqueducts is due, more than to technical limitations, to the interest in protecting the pipes from sabotage in periods of war or to the old Roman experience in the construction of drains and sewers.
But it is no less true that in the two aforementioned aqueducts the main conduction systems used in all periods are combined: underground conduction, in a tunnel, on a wall and on arches. As if this were not enough, the layout of the Anión Viejo aqueduct includes a siphon, which quite completes the repertoire of technical solutions.
Vitruvius, who wrote his De architectura libri decem probably shortly before 27 BC. C., already mentions all the conduction systems mentioned in this article, with the exception of stone pipes, which do not appear in any treatise, and wooden tubes and channels, which must have been used very late. In fact Faventino, who practically follows Vitruvius in everything, is the only theorist who mentions them, thereby departing from his source. And keep in mind that he most likely wrote his work at the end of the century.
Therefore, we must think that all the technical resources available to the Romans to convey water were at their fingertips from the first moments.
The aqueducts that were built from the 19th century onwards were far from the beautiful Roman works, many of which are still in operation today, such as those that supply water to the fountains of Rome.
Modern aqueducts are generally built underground, as extensive networks of iron, steel or concrete conduits. The Delaware aqueduct, which transports water from the Catskill Mountains to New York, has a length of 137 km and is the second longest continuous transport aqueduct to supply populations (although it is only 5 kilometers longer than the Roman one that fed Carthage).
The longest aqueduct in the world is called Acueducto Vizcaíno-Pacífico Norte in the municipality of Mulegé, Baja California Sur, Mexico. This work has a length of 315 kilometers and consumes 60 liters of water per second.