The story of drinking water systems through the centuries

Aug 12, 2016

The “50th” anniversary issue of the EADIPS®/ FGR® annual journal has been an occasion to look back over the history of 500 years of drinking water transport and distribution. This area of life is inseparably linked to the traditional material of “cast iron”: it has put its stamp on the development of drinking water supply for half a millennium but, with a constant influx of improvements, optimisations and innovations over this timespan, it has nevertheless remained ever young. The cast iron pipe industry, in collaboration with its users, has always managed to keep abreast of the latest state of the art with a modern and sustainable piping system comprising pipes, fittings and valves.

Figure 1: Cast iron pipe from Dillenburg Castle (1455). [Source: EADIPS®/FGR®]

1 Introduction

It is undisputed that having sufficient drinking water is the most important prerequisite for the development of civil society. There is an impressive account which tells of the technical expertise, the organisational skills and the setting of standards for the supply of drinking water in Ancient Rome [1]. Frontinus was the first Water Commissioner for Rome and he documented his experiences for posterity.

The water transport pipelines of the time were aqueducts, magnificent structures which carried water from distant sources to the main cities by gravity. But during this phase they did not simply construct the technical infrastructure of water supply; they also set up the legal and financial principles, technical standards and management methods without which a secure water supply cannot function. Later on, however, it was above all the royal houses of the Baroque period who, seeking to outperform each other with the fountains in their palace gardens, were responsible for the invention of pressure pipelines. Here are a few landmarks in this development process:

  • 1412: The first known water pipeline using cast iron pipes in Augsburg/Germany
  • 1455: Cast-iron water pipeline to Dillenburg Castle/Germany (Figure 1)
  • 1680: Cast-iron flanged pipes in the gardens of the Palace of Versailles/France
  • 1700: Fountains with artificial cascades in Kassel-Wilhelmshöhe/Germany, a UNESCO World Heritage site since 2013
  • 1840: 1.8 km long castiron pipeline from the pumping house on the River Havel to the Ruinenberg in Potsdam for the fountains in Sanssouci Park
2 The industrial production of cast-iron pipe systems

With the Industrial Revolution in the middle of the 18th century it became possible to generate energy by means of steam engines. At the same time this phase was characterised by dramatic scientific progress. Primarily, progress in medicine and general hygiene resulted in an exponential increase in population figures. A basic element of this progress was the expansion of the supply of drinking water to towns and cities.

The time-consuming production of individual drinking water pipes using the sand moulding technique was replaced by production in vertically arranged casting carousels which were capable of being mechanised and thus made it possible to produce the quantities required for the first time (Figure 2). A further acceleration in the production of cast iron pipes was introduced in 1923 with the development of the centrifugal casting process. Even today, almost all cast iron pipes worldwide are still produced using the de Lavaud process (Figure 3).

Figure 2: Production of cast iron pipes with casting carousels (around 1900). [Source: EADIPS®/FGR®]

Figure 3: Production of a centrifugally cast iron pipe using the de Lavaud process. [Source: EADIPS®/FGR®]

3 Development of joint technology

Just as the production technique for pipes had to adapt to increasing requirements over the centuries, joint technology also underwent continuous development. Ancient and mediaeval pipes in stone, vitrified clay, lead and finally also cast iron had socket-shaped ends and were sealed with putty. Flanged joints came along for higher pressures. Sockets stopped with tarred hempen rope and sealed with lead are still in operation today.

It was only with the invention of vulcanised rubber that the important step was taken for the modern pipe joint: it was now flexible and hence could be bent and longitudinally adjusted in order to adapt to earth movement and subsidence. The “union” screw-gland socket joint was a quantum leap forwards in the 1930’s for the construction of pressure pipelines for water and gas, replacing the decades-old practice of caulking packed sockets with lead.

Figure 4: TYTON® push-in joint. [Source: EADIPS®/FGR®]

In the fifties the assembly of joints was once again simplified by the introduction of the “TYTON®(Figure 4) and “Standard” type push-in joint. These constructions are designed in such a way that tightness is achieved by simply pushing the spigot end into the socket which is fitted with a gasket and this can be loaded up to the bursting pressure of the pipe while retaining its flexibility over a useful life of 100 years; so far, 60 years of this have actually been proved in practice.

We will look at the further development of this joint – the restrained socket joint - later in the description of trenchless installation techniques.

4 Material developments

The material which has accompanied mankind since the Iron Age is cast iron or, more precisely, grey cast iron or even lamellar graphite cast iron. It is primarily produced by the smelting of iron ore in furnaces, but what is more significant is its production from recycled scrap metal (steel scrap, cast iron scrap, old cars, etc.) in cupola furnaces. The mechanical properties of this material essentially depend on the type of structure and form of the graphite.

Figure 5: Ductility – practical implications in the example of longitudinal bending strength. [Source: EADIPS®/FGR®]

The discovery of spheroidal graphite cast iron and its application in the area of pressure pipelines in the nineteen fifties was epochmaking in character. With this material the elementary carbon is embedded into the matrix in the form of graphite nodules, giving it plasticity and deformability. When it is used for pipes, fittings and accessories the material is referred to as ductile cast iron in accordance with EN 545 [2]. When used in valve bodies then, as is customary in mechanical engineering, it is called spheroidal graphite cast iron, the properties of which are determined in standard EN 1563 [3]. Both standards are used by various technical standardisation committees.

The practical consequences for the loading capacity of pipes, e.g. for longitudinal bending strength, are shown in Figure 5. This clearly illustrates the fundamental importance of replacing the traditional, brittle grey cast iron pipe by the ductile iron pipe system for pipeline construction in the fifties.

5 Development of the protection system

In the presence of water and oxygen, the oxides and hydroxides of iron are more thermodynamically stable than elementary iron. It corrodes. The consequence of this, when attacked from the outside, is corrosion damage and even perforation; in case of contact with drinking water on the inside this has a negative effect including brown discoloration of the water. Corrosion protection is therefore one of the main tasks in the production of cast iron pipes. The following landmarks highlight developments in this area:

  • 1960: External protection with zinc and bituminous finishing layer
  • 1966: Tar inside and outside (draft DIN 28600 [4])
  • 1970: Cement mortar lining
  • 1995: External protection with zinc and synthetic resin finishing layer (epoxy or polyurethane)

The important feature of the finishing layer is its porosity; this ensures that the soil electrolyte comes into contact with the underlying zinc.

In these cases the metallic zinc reacts with the soil electrolyte and forms insoluble zinc salts which seal both the pores of the cover coating and points of damage. The electrical resistance between the metal, first the zinc and later the iron, and the soil electrolyte increases as the corrosion current decreases.

The indispensable basic condition for the process happening as described is the precipitation of the insoluble zinc reaction product onto the boundary surface between pipe and soil electrolyte. For this the pH value of the soil electrolyte must not be below pH = 6. In acid soils (bog, marshland, peat) the protection mechanism does not work. Equally, no lasting protection is possible in an alkaline medium (pH ≥ 8.5) because the zinc salts go into solution as zincate in alkaline electrolytes.

Figure 6: DN 600 ductile iron pipes with cement mortar coating. [Source: EADIPS®/FGR®]

A particularly successful development in this area is cement mortar coating in accordance with EN 15542 [5] (Figure 6). This has a certain degree of porosity through which the underlying zinc coating interacts with the soil electrolyte. With a coating thickness of 5 mm, the resistance of the coating increases as the result of a progressive hydration of the cement over time. The cement mortar coating is structurally reinforced by inert fibres and therefore resists enormous mechanical loads, whether during trenchless installation in difficult soils or when laying pipes in open trenches where large-grained or rocky excavation material may in fact be directly reused. Therefore cement mortar coating is extremely appropriate for use in soils of all kinds. Inspections of pipelines after more than 30 years of use in aggressive soils have shown an iron surface which looks as good as new without any attack [6].

Figure 7: Butterfly valve enamelled on the inside and coated with fusion bonded epoxy powder on the outside. [Source: EADIPS®/FGR®]

While in the past tar and bitumen paints were used for the protection of fittings and valves, the state of the art today is characterised by epoxy resin and enamel. Both types of protection are so-called barrier layers with a high specific coating resistance; they are applied in layer thicknesses which result in a largely pore-free surface.

The epoxy coating is applied by fusing an epoxy powder to the freshly blasted and heated surface of the fitting or body. In the liquid phase, polymerisation reactions take place which produce a highly resistant and closed protective layer. The minimum coating thickness is 250 μm; additional requirements and test methods are described in EN 14901 [7]. The epoxy coating can be used in soils of all kinds.

The modern method of complete enamelling of fittings and valves has developed with the help of recent knowledge about silicate technology from the old craft of enamelling cast-iron ovens. As an inorganic lining for fittings and valves it is very popular in the area of drinking water supply. Requirements and test methods can be found in DIN 51178 [8] (Figure 7).

6 Restrained push-in joints

The push-in joints described earlier do not absorb any longitudinal forces which occur in changes of direction or cross-section, junctions or dead ends. These forces must be transmitted into the subsoil via suitable thrust blocks, usually in concrete. With large nominal sizes and pipelines outside urban areas, this continues to be the practice on the basis of DVGW worksheet GW 310 [9].

An important alternative to this is offered by the use of restrained joints, but their technical differentiation goes beyond the scope of this historical summary. Apart from a reference to chapter 9 of the EADIPS®/FGR® E-Book [10] which gives comprehensive details, only the landmark features will be listed here.

6.1 Friction-locking push-in joints

The longitudinal forces are transmitted by sharp, hardened teeth on retaining elements in the surface of the spigot end.

  • 1972: Tyton SIT®
  • 1985: Novo SIT®
  • 1995: BRS® , TYTON SIT PLUS®

In addition to these constructions, similar solutions have been developed in the various regional markets, usually based on other sealing systems.

6.2 Positive-locking push-in joints

Figure 8: BLS®/VRS-T® ( DN 500) and BLS® ( DN 600) restrained push-in joint. [Source: EADIPS®/FGR®]

The forces are transmitted via formed elements such as welded beads on the spigot ends in combination with force-transmitting elements. The most important representatives of this class developed between 1975 and 1985 are

  • TIS-K,
  • BLS® / VRS® - T and BLS®  (Figure 8)

With the ever improving distribution of forces between socket and spigot end it has in fact become possible to load the joints up to the bursting pressure of the pipes. This has had effects in two major directions: on the one hand in the area of high-pressure applications, such as turbine pipelines for hydroelectric power stations and snow-making equipment in the mountains. However, the second direction triggered a major expansion in construction technology as from the 1990’s, namely that of trenchless installation and renewal techniques.

7 Trenchless installation and renewal techniques

Figure 9: DN 900 installation with the HDD technique in Valencia, 2007. [Source: EADIPS®/FGR®]

With the efficient positive-locking push-in joints described in Section 6, the conditions are right for installing and replacing ductile iron pipelines using trenchless techniques. An unprecedented development of trenchless pipe-laying processes began in the early nineties; full details of this can be found in chapter 22 of the EADIPS®/FGR® E-Book [11].

The following milestones in development are mentioned here:

  • 1993: Horizontal directional drilling technique (Figure 9)
  • 1990: Pipe pulling technique
  • 1999: Auxiliary pipe technique

The last two techniques mentioned are practised by Berliner Wasserbetriebe with new pipes in ductile cast iron. In Berlin alone around 10,000 m of pipelines in nominal sizes of DN 80 to DN 500 are replaced in this way each year. A further development allows a considerable increase in nominal size if the excess soil is removed in this process.

1999: Pulling-in with the rocket plough process

The process uses a height-adjustable plough with which a pre-assembled pipe string is pulled into a new pipeline route. The application is limited to rural areas where there are no underground structures to be found as yet. Ductile iron pipes with robust cement mortar coating have proved themselves in particular in soils with large, sharp-edged stones.

Figure 10: Pipe relining with DN 800 pipes. [Source: EADIPS®/FGR®]

 2003: Static burst lining

Here again the new pipe is pulled in along the same route but in this case the old pipeline remains in the soil, either in the form of fragments or as a cut and opened-up pipe string. During the pulling-in process the new pipes can be dragged along on the sharp-edged fragments which is why materials which are extremely robust and resistant to notching are needed for them. Ductile iron pipes with cement mortar coating to EN 15542 [5] have proved to be excellent in numerous types of applications.

2003: Pipe relining

With this technique, new pipes are pulled or pushed into the unaltered old pipeline, whereby the free cross-section of the pipeline is downsized; however, against the background of reduced water consumption levels, this is often welcome as it means that flow speeds can be increased again (Figure 10). The technique has become a very popular one and has proved itself in more than 10,000 m total length.

8 Development of wall thicknesses

When the first ductile iron pipes were produced in the mid-fifties, safety characteristics were at first still limited to thick-walled grey cast iron pipes. After some initial practical experiences, in 1961 the FGR established a standard for the association with Preliminary Conditions of Supply. Then, because the DVGW had conducted a scientific investigation (the “Wellinger study”) to establish assessment bases with load assumptions taken from pipe bedding conditions, one year later it was possible to develop the first DIN standards – DIN 28600 [4] and DIN 28610 [12] – which were published as white papers two years later.

Nominal wall thicknesses S0 were determined taking account of the permissible circumferential stresses in the pipe wall using the formula

S0 = 5 + 0.01 · DN [mm]

in wall thickness range K 10.

In 1979 international standard ISO 2531 [13] showed a way in which smaller wall thicknesses could be coherently represented in their own wall thickness classes:

s = K · (0.5 + 0.001 · DN) [mm]

where K is to be selected from a series of whole numbers … 8, 9, 10, 11, 12 … As early as 1976, because of the possibility of also standardising thinner-walled pipes, the DVGW expert committee on “pipes and pipe joints” asked the FGR to revise standard DIN 28600 [4]. The aim was to introduce wall thickness classes K 9 and K 8 in 1980. With the development of the first European product standard EN 545 in 1995, wall thickness classes K 8 to K 10 were adopted as standard wall thicknesses.

The accurate production of the smaller wall thicknesses in the centrifugal casting process has matured in the five decades since the advent of ductile iron pipes due to improved machine control and continuous process optimisation. So it was only logical that the cast iron pipe industry would contribute to the trend of adapting water supply components to the prevailing pressures, as is made clear in EN 14801 “Conditions for pressure classification of products for water and wastewater pipelines” [14]. By 2002, in its edition current at the time, EN 545 was listing pressure class C 40 in addition to the K classes still in existence and by 2010 standard EN 545 [2] now only contains pressure classes. A look at the development of minimum wall thicknesses during the last half of the century in Figure 11 shows that they have almost halved since the introduction of ductile iron pipes.

In parallel to these improvements in terms of production technology, under the pressure of economic demands the trenchless installation techniques were also developing, for which restrained joints almost exclusively are used. The positive locking versions with their welded bead on the spigot end put an end to further development towards thinner walls: for perfect penetration of the weld bead, the welding process requires a wall thickness which should be between 5 mm and 6 mm at least.

Also the multiaxial stress states in a pipe wall which, in addition to the circumferential stress from internal pressure, must also absorb additional axial stresses, bring about a clear reduction in the allowable operating pressure PFA as compared with the non-restrained design. Stating the pressure class C as a synonym for the allowable operating pressure PFA is thus no longer adequate for pipelines operating with restrained joints. Each manufacturer must therefore state the lower value for the PFA for his restrained joints. EADIPS®/FGR® has taken this requirement into account in that it has published its own marking standard, EADIPS®/ FGR®-STANDARD 75 [15].

9 Conclusion

With this article which we are presenting on the occasion of the 50th anniversary edition of the EADIPS®/FGR® annual journal we hope to show how “cast iron” as a traditional material which has had such a great influence on water supply systems for half a millennium has managed to stay forever young by means of a constant flow of improvements, optimisations and innovations. In so doing, and in collaboration with its users, the cast iron pipe industry has succeeded in always keeping a modern and sustainable pipe system, consisting of pipes, fittings and valves, up to the latest state of the art.


[1] Frontinusgesellschaft e. V., München 2013
[2] EN 545: 2010
[3] EN 1563: 2012
[4] Draft DIN 28600: 1966-04
[5] EN 15542: 2008
[6] Rink, W. DUCTILE IRON PIPE SYSTEMS Issue 45 (2011), p. 25 ff Download:
[7] EN 14901: 2014
[8] DIN 51178: 2009-10
[9] DVGW-worksheet GW 310: 2009-01
[10] EADIPS®/FGR®-E-Book, chapter 9 Download:
[11] EADIPS®/FGR®-E-Book, chapter 22 Download:
[12] Draft DIN 28610: 1966-04
[13] ISO 2531: 1979
[14] EN 14801: 2006
[15] EADIPS®/FGR®-Standard 75: 2013-06

(First publication in: GUSS-ROHRSYSTEME - Information of the European Association for Ductile Iron Pipe Systems • EADIPS®)

More News and Articles


European Association for Ductile Iron Pipe Systems · EADIPS®/ Fachgemeinschaft Guss-Rohrsysteme (FGR®) e.V.

Dr.-Ing. Jürgen Rammelsberg

Doncaster-Platz 5

45699 Herten



+49 (0) 2366 / 9943905


+49 (0) 2366 / 9943906



To website