Root resistance of ductile iron pipe joints

Sep 24, 2018

Root resistance of push-in joints.

1 Introduction

“Root penetration into sewers and drains” is recognized as an obstruction to flow in the context of the camera inspections regularly performed on the inside of pipes. In private drains, root penetration may only be noticed at a late stage when blockages and backing up occur, along with the resulting consequences. In public sewer systems, root penetration is one of the most frequent causes of damage [1]. In both private drains and public sewer systems, these sections of pipeline are considered not to be tight.

2 Root growth and root pressure

The reasons why tree roots tend to grow into drains and sewers has been researched over the last 15 years. An important result of the “Root penetration into sewers and drains” research project [2] has been the development of models which can describe the penetration of roots in the area of sewers and pipelines. Based on results from numerous excavations, the density trap model and the oxygen model have been developed and described; they are useful for describing the growth of roots in the underground space. In the context of a further research project [3], root pressures of primary roots and contact pressures of push-in joints were determined on different DN 150 nominal size pipe systems made of vitrified clay, plastics and ductile cast iron. No choice could be made of the pipe joints – e.g. in the form of the maximum socket gap – in which the minimum contact pressure could be expected, as the range of contact pressures needed to be identified. This fact led the Association to have the contact pressures in the TYTON® push-in joint system determined. With the help of the following criteria such as

  • the density trap and oxygen models,
  • the investigation results on the root pressure of the primary roots of different trees,
  • the contact pressure investigations on push-in pipe joints in the context of [3] and
  • measurements on the TYTON® push-in joint system with a maximum socket gap

it was possible to estimate the root resistance of the TYTON® push-in joint system.

3 The density trap model

The entire environment of buildings and their infrastructure is a ground space which has been produced anthropogenically. In contrast to natural ground it often has a lower compaction level and larger pore spaces. The differences in density affect the direction of growth of the root tip. The elasticity of the calyptra (root tip) means that the roots grow in the direction of the substrate which is easiest to penetrate (Figures 1 and 2). The growth of roots back into an area of high compaction or poorer penetration conditions can be excluded as a rule. The roots are “trapped” in areas of soil with greater penetration potential. The annular gap or space before the sealing element can also, depending on the pipe joint, be an area which the roots can easily access. They can grow there for several years before they finally penetrate inside a pipeline. To do this, they have to overcome the contact pressure of the sealing agent.

Figure 1:
Tough roots in the utility trench of a telecommunications line
[Source: EADIPS/FGR]

Figure 2:
Fine roots in the utility trench of a gas pipeline
[Source: Heidger, C.]

4 The oxygen model

The availability of oxygen in the soil has a great influence on root propagation. All plant organisms need oxygen to sustain their metabolic processes. The sealing of urban soils has the consequence that oxygen is seriously restricted from entering the soil. Sewers are usually operated as gravity pipelines and are sufficiently ventilated via maintenance and inspection openings (manholes). The major proportion of the pipeline is filled with air. With cast seals, cracks can occur in the casting material due to shrinkage. In this way, the oxygen contained in the air can get into the environment of pipes and pipe joints in the soil. When planning sewer systems, no consideration is given to the gas permeability of pipes and pipe joints. However, it can also influence the growth of roots with newly installed sewage pipes. With pipe materials which are not gastight, oxygen can leak out even if the pipelines are intact. According to the oxygen model, roots grow towards the source of oxygen and so find the pipe joint.

This model (oxytropism) is supported by the results of [4]. On the roots of pea seedlings (Pisum sativum L.) it was able to be demonstrated that these followed an oxygen gradient in the direction of the higher O2 concentration. Based on this, it was the aim of the “Root penetration into sewers and drains – supplementary project” research project [3] to prove the causes of root growth into pipelines scientifically and to describe the mechanisms behind the penetration of a root into the pipeline and the interactions between pipe properties and root penetration. In addition, proposals were to be developed for testing procedures to illustrate the mechanical and biological processes involved in root penetration realistically so that pipe joints can be evaluated in terms of their resistance to root penetration. This included a description of the differences in the growth behaviour of various tree species as characteristics of different root systems. One particular point of focus was the interaction of roots with different DN 150 pipe joints and mechanical trials to investigate root resistance. The results of these investigations are summarised in Section 9, Investigations and results.

5 Practical examples

The influence of the pore space on root growth can be seen in Figures 3 to 6. For the tests at the Botanical Gardens of Ruhr University Bochum shown in Figures 3 to 5, an artificial layering of two substrates with different porosity levels was set up in planters. Compost of the type used for container cultivation in the Botanical Gardens of Ruhr University Bochum was selected as the substrate with a high pore proportion and good root penetration characteristics. Bentonite was selected for the reproduction of soil areas with a low pore proportion and poor root penetration characteristics. The material used consists of clay minerals with a mean particle size of 0.063 mm. The small size of the particles has the result that the spaces (soil pores) needed for root growth with a size from 100 μm upwards are not available [5]. The substrates were applied alternately in horizontal layers around a central column of one of the substrates. The layer structure can be seen in the schematic longitudinal section represented in Figure 3. Figure 4 shows a planter opened up on one side.

Figure 3:
Schematic longitudinal section through a planter
[Source: Schmiedener, H,]

Figure 4:
Planter opened down the side - no roots are to be seen in the layers of bentonite
[Source: Stützel, T.]

A simple glance at this picture gives a hint that no roots have grown into the bentonite with low pore space, which was confirmed after cutting through the planter including the soil it contained and flushing off the roots with water. The roots had spread through the compost down to the bottom of the planter (Figure 5). Between the layers of bentonite, they had in fact reached almost as far as the vertical edge of the planter [2].

This laboratory-simulated behaviour of root growth was confirmed in practice. The root system shown in Figure 6 was found at a depth of 7 m. The roots had spread along a service pipeline to a house, connected via the fitting shown to the sewer system. They continued to grow into the main sewer pipe trench and into the space around the sewer. At this depth, the growing root system needs a sufficient supply of oxygen. There is reason to believe that this oxygen came from the non-gastight socket joints of the house connection pipeline and the main sewer, along the lines of the oxygen model.

Figure 5:
Planter cut open - roots flushed with water. The roots have not grown into the bentonite.
[Source: EADIPS/FGR]

Figure 6:
Roots at a depth of 7 m, which have grown down to this depth because of the good oxygen supply
[Source: Schmiedener, H.]

6 Characteristics of different root systems

The possible effects of differences in the anatomical structure of roots were investigated with the help of root pressure measurements on primary roots. Basically, a lower root pressure was measured with gymnosperm roots (coniferous trees) than with angiosperm roots (broad-leafed trees). The peak values for the root pressure of gymnosperms varied between 4.0 bars for araucaria roots (Araucaria araucana) and 8.8 bars for pine roots (Pinus pinea). The root pressures of angiosperms (broad-leafed trees) varied as well. As the lower peak value, 8.8 bars were measured for robinia roots and as the upper limit, a brief value of 11.9 bars was recorded for oak tree roots. On average root pressures are between 4 bars and 8 bars. The results of the root pressure measurements are summarised in Table 1.

Root pressure under ideal growth conditions for the seedling roots used – at every point there was a continuous supply of water and oxygen for the roots [Source: EADIPS/FGR]

Species Measurement
period [h]
Maximum pressures
Average values
Pisum sativum L./
1 62.5 4.9 4.07
2 62.5 5.9
3 62.5 2.5
Quercus robus L./
Common oak
1 50.0 1.2 8.42
2 50.0 5.9
3 64.0 11.9
4 46.0 10.8
5 46.0 12.3
Robinia pseudoacacia L./
Lack locust tree
1 58.0 8.8 6.43
2 58.5 8.4
3 48.0 6.7
4 48.0 6.5
5 25.0 3.7
6 22.0 4.5 6.28
Pinus pinea L./
Pine/Italian Stone Pine
1 700.0 3.6
2 700.0 8.8
3 670.0 9.8
4 670.0 2.9
Araucaria araucana
Monkey puzzle tree
1 530.0 4.0 4.0

7 Pipe joints

Figure 7:
Push-in joint for ductile iron pipe system TYTON® system
[Source: EADIPS/FGR]

Sewage pipes are most often connected together by means of push-in joints with elastomer as the gasket. As compared with other systems they offer the advantage that they are comparatively easy to assemble even under difficult site conditions. In recent years, the development of these joints has been advanced and optimised in terms of constructional criteria. The most commonly used socket joint for ductile iron pipes is the TYTON® push-in joint (Figure 7). This is standardised in the range DN 80 to DN 1400. Since its introduction onto the German market in 1957 it has proved itself a million times over in drinking water, raw water and wastewater pipelines. The essential dimensions of this joint are determined in DIN 28603 [6] for nominal sizes DN 80 to DN 1400. The sealing function of the TYTON® push-in joint is performed by a profiled gasket consisting of a mixture of softer rubber (sealing part) and a harder rubber (retaining part) [7].

The STANDARD – push-in joint (Figure 8) is comparable with the TYTON® push-in joint in its constructional concept and its function. In Germany, its joint dimensions are determined in DIN 28603 [6], (Form C), from DN 80 to DN 2000. The gasket consists of rubber with a single hardness grade.

8 Interaction between pipe joints and roots

Figure 8:
Push-in joint for ductile iron pipe system STANDARD system
[Source: EADIPS/FGR]

In order to test pipe joints, load situations such as those which may occur in the pipe trench are simulated in laboratory tests so that pipe and pipe joint quality are ensured. Roots which grow into pipe joints and result in leakages represent an as yet undefined burden for pipe joints. The load case represented by root growth was described for the first time in the context of [2]. According to this, roots do not only penetrate pipe joints which lack tightness. Indeed, according to the generally accepted rules of technology, they can overcome “root-proof” pipe joints, so the assumption that “watertight pipe joints mean root-tight pipe joints” does not apply in all cases. Results of investigations in Sweden [8] and Australia [9] confirm these observations. The strategies by which roots can vanquish a pipe joint have already been described in [2]. Particular importance is placed on the pipe material used, the geometric design of the pipe joint and the elastomer gasket applied. The sum of the properties of these individual components influences the growth behaviour of the roots in the area of pipe joints. Additional factors, such as the supply of the roots with oxygen via the piping system (diffusion tightness of pipe materials and push-in joints), are to be taken into account here.

9 Investigations and results

Figure 9:
Measuring the contact pressure using pressure measurements film. Pipe joint of a DN 150 ductile pipe with TYTON® push-in joint without shear load and with shear load to EN 598 [10].
Substantial contact pressure in the area of the pipe bottom:
Measurement 1) without shear load effect 24.8 bar, with shear load effect 21.2 bar,
Measurement 2) without shear load effect 23.9 bar, with shear load effect 20.4 bar,
Measurement 3) without shear load effect 17.8 bar, with shear load effect 10.8 bar.
Mean values for contact pressure without shear load effect 22.2 bar, with shear load effect 17.5 bar ([3], Fig. 35)
[Source: EADIPS/FGR]

In the context of [3], the contact pressures were determined on a total of eleven different pushin joints of nominal size DN 150 and OD 160. This included three push-in joints for vitrified clay pipes, four for PVC pipes, three for PP pipes and one push-in joint for ductile iron pipes. The results of the contact pressure tests on the DN 150 TYTON® push-in joint, without shear load and under the influence of a shear load, are shown in Figure 9 (Table 18 in [3] gives all the results). As an average contact pressure without the effect of a shear load, a value of 22.2 bars was determined for the TYTON® push-in joint system. Under a shear load an average contact pressure of 17.5 bars was produced on the non-loaded side of the TYTON® push-in joint system.

10 Penetration risk and root resistance in the light of the Technical Rules

The root resistance of pipe joints was considered to be proven in Germany in accordance with DIN 4060 [11] if the pipe joint under shear load had withstood tightness testing under positive and negative pressure (Figure 10). This was under the assumption that roots can only penetrate pipe joints which are not tight.

This changed on the basis of the research results described in [2] and [3] and has been taken into account with the publication of the interdisciplinary rules on “Trees, underground pipelines and sewers”, appearing with the same text as DWA-M 162 [12], DVGW data sheet GW 125 and FGSV 939.

As an important change, in the two chapters Tightness and root resistance ([12], chapter 5.5) and Pipe joints ([12], chapter 5.6) a risk of penetration for tight pipe joints is described for the first time ([12], chapter 5.5): “Roots can grow not only into leaky pipes and pipe joints but also tight pipe joints which do not offer sufficient resistance to the roots.”

It goes on to state: “With new constructions and the correct manufacture of pipe joints (e.g. to DIN EN 1610/ DWA-A 139 for sewerage) it can be assumed that the danger of root penetration into the pipeline is slight. In order to increase resistance to root penetration, additional constructional safety measures can be adopted ([12], chapter 7)”.

By way of additional constructional safety measures, [12], chapter 7.2.2 describes measures which are taken directly in the area of underground pipelines and trenches: so-called passive protective measures. Among the passive protective measures are:

Figure 10:
Shear load testing for pipe joints in sewers and wastewater pipelines with elastomer gaskets as per [11]

  • the use of low pore space filling material in the pipe- and pipeline trench,
  • the installation of casing pipes (protective conduits) around the pipeline,
  • the installation of plates or sheetings in the pipe trench,
  • the choice of root resistant pipe joints, other installations.

Therefore, from [12], chapter 5.5 of the data sheet “Trees, underground pipelines and sewers” it is possible to deduce the risk of penetration into push-in joints, even where the pipe joint has been assembled correctly. A push-in joint with the lowest risk of penetration, or none at all, would then be classified as root resistant according to [12], chapter 7.

11 Measuring the contact pressure on pipe joints

The geometry of the sealing agent and the resulting influence on the contact pressure produced remained unconsidered in shear load testing [11]. It is only the short-term strain on the sealing agent which has been demonstrated in tests. No attention was paid to changes in the properties of the elastomer such as the reduction of contact pressure under more enduring shear load.

In EN 14741 [13] a testing process is described for determining the long-term sealing behaviour of elastomer gaskets by extrapolation and estimation of the sealing pressure after 100 years. The process reads as follows in EN 14741 [13]: “The sealing pressure in a joint is estimated by measuring the pressure required to lift the gasket in each of the three PTFE tubes which are distributed evenly around the circumference of a joint between the rubber gasket and the outside wall of the spigot end or, where applicable, the socket (Figure 11). In a temperature-controlled environment and at increasing time intervals, nitrogen or air is forced through three flexible PTFE tubes at a constant flow velocity of 120 ml/min. The nitrogen or air pressure p required to achieve this flow velocity is measured. The pressure pt is measured within a period at increasing intervals of time. The extrapolated regression lines for pt are used in order to calculate the estimated values of px after 100 years and py after 24 hours.”

PTFE (polytetrafluoroethylene) tube:
stretched tube which is normally used as shrink tubing. The original diameter and the original wall thickness after shrinking are determined normally. It should be noted that the dimensions in the stretched state are not normally determined. The wall thickness determined and the diameter determined are to be checked carefully. The limit deviations stated should be considered as a guideline for suppliers.

the pressure measured [bars] in the PTFE tube with a flow velocity of 120 ml/min during time t [hrs]

extrapolated pressure after 100 years [bars]

calculated pressure after 24 hours [bars]

The test set-up is shown in Figure 11. The investigations are carried out according to EN 14741 [13] without the influence of shear loads. Figure 12 shows a typical procedure for contact pressure measurements and subsequent extrapolation by way of example.

Figure 11:
Measurement of contact pressure, taken from [2], published in [14]
[Source: EADIPS]

Figure 12:
Typical course of contact pressure measurement with subsequent extrapolation
[Source: EADIPS/FGR]

12 Evidence of the root resistance of TYTON® push-in joints

The test below shows a way in which the root resistance of push-in joints can be demonstrated based on EN 14741 [13] with a modified test for determining the long-term sealing behaviour of elastomer gaskets by estimating the sealing pressure. The following conditions must be met for the application and subsequent interpretation of the results:

Figure 13:
Schematic representation of the test set-up based on EN 14741 [13]: push-in joint with a maximum joint annulus. Circumference of the spigot end reduced by the amount of the measurement device (PTFE tubes). Measurement with four PTFE tubes which are inserted into the shear-load-relieved area of the pipe joint (drawing EN 14741 [13] supplemented by J. Rammelsberg)
[Source: EADIPS/FGR]

  • The pipe materials and push-in joints used in the context of the tests are demonstrably diffusion-tight [15], so that there is no oxygen supply to the roots in the pipeline trench via the pipe system.
  • The minimum possible contact pressures are determined for the push-in joint to be tested. For this purpose, the boundary conditions for the type testing of ductile iron pipe joints, for example, are to be observed. Push-in joints with a maximum joint annulus to EN 598 [10] are used, loaded with a shear load in the aligned position.
  • Even the wall thickness of the means of measurement used (PTFE tube) can influence the contact pressure measurement. In order to take this into account, before the test the diameter of the spigot end is to be reduced by a corresponding amount (e.g. by “turning”).
  • The tests are carried out representatively for the DN groups described in EN 598 [10] on DN 200, DN 400 and DN 800 push-in joints.

The test set-up is shown in Figure 13.

Notwithstanding the specifications in EN 14741 [13] four, and not three, PTFE tubes are positioned on the non-loaded side of the coaxially-mounted and shear-force loaded joint represented in Figure 13 between the seal and the surface of the spigot end according to Figure 14 at positions 0°, 45°, 90° and 300°. After this, the pressure required to lift the seal is measured according to EN 14741 [13] at the intervals stated.

Figure 14:
Arrangement of the test tubes
[Source: EADIPS/FGR]

The root resistance of the push-in joint tested is considered proven if the contact pressure between the elastomer seal and the spigot end of a push-in joint is greater than the average pressure of the root tips and therefore the width of the sealing surface can also be assumed to be sufficiently large to cut the root tips of from the oxygen available in the pore space of the soil (diffusion tightness).

This is the case with the TYTON® push-in joint system and the STANDARD system if the pressure after 100 years, extrapolated according to the process described here based on EN 14741 [13], determined on push-in joints with a maximum joint annulus as per EN 598 [10], is on average greater than 7.0 bars.

With the process described here, tests were carried out to estimate the sealing pressure on push-in joints from the TYTON® system, nominal sizes DN 200 and DN 400 [16]. The real test set-up is shown in Figures 15 and 16.

For the TYTON® push-in joint DN 200, a mean value of 7.67 bars after 100 years was estimated and for push-in joint system DN 400 the estimated mean value was 7.76 bars after 100 years. In the result, the push-in joints tested were classified as root resistant.

Figure 15:
Test set-up for a DN 200 TYTON® push-in joint based on EN 14741 [13] at the IRO Institut für Rohrleitungsbau an der Fachhochschule Oldenburg e. V.
[Source: EADIPS/FGR]

Figure 16:
Test set-up for a DN 400 TYTON® push-in joint based on EN 14741 [13] at the IRO Institut für Rohrleitungsbau an der Fachhochschule Oldenburg e. V.
[Source: EADIPS/FGR]

The details of the measurements, results and evaluation are included in test report no. G 32 980 dated 14.02.2013 – Determination of the longterm sealing behaviour of TYTON® push-in joints with elastomer gaskets by estimating the sealing pressure based on EN 14741 [13] – Iro GmbH Oldenburg [16]. It can be viewed and/or downloaded at

13 Summary and outlook

The reasons for the penetration of tree roots into wastewater pipelines and sewers have been researched in the last 15 years. The following should be highlighted as significant results:

  • density trap model,
  • oxygen model,
  • possibility of root penetration into watertight joints,
  • influence of joint construction and gasket geometry,
  • diffusion tightness of the material for pipe and gasket,
  • consequences for the Technical Rules.

Ductile iron pipe systems, including their push-in joints, are demonstrably diffusion-tight, meaning that a supply of oxygen to the root system in the pipe trench can be excluded and hence a significant stimulus for root growth is lacking. Furthermore, in the context of [2] it has been shown that the contact pressures and contact pressure surfaces of ductile iron pipe joints are way above the average root pressures which were also determined experimentally.

In the product standards for ductile iron pipe systems [10], the usual function tests for the tightness of movable pipe joints are carried out with a maximum joint annulus under the simultaneous effect of a shear load. In the revision of product standard EN 598 [10] the requirement for and testing of root resistance is included as an additional element in the standard in the form of a long-term test.

On the basis of these results, there are applications in the underground space which require a root-resistant push-in joint. One of these applications is described in the article “The sponge city principle – from pipe-soil-systems to soilpipe- systems – solutions with ductile iron pipe systems”.

[1] Stein, D. and Kaufmann, O.: Damage analysis on concrete and vitrified clay pipe sewage systems in the Federal Republic of Germany – West Korrespondenz Abwasser, 1993-02
[2] Stützel, T. et al: Root penetration into sewer systems and drains, Ruhr University Bochum in cooperation with IKT – Institut für Unterirdische Infrastruktur gGmbH, Research report July 2004
[3] Stützel, T. u. a.: Root penetration into sewer systems and drains – Supplementary project, Ruhr University Bochum in cooperation with IKT – Institut für Unterirdische Infrastruktur gGmbH, Final research report June 2007
Download: f0160langbericht.pdf
[4] Porterfield, D. M. and Musgrave, M. E. (1998): The tropic response of plant roots to oxygen - Oxytropism in Pisum sativum L. Planta 206 (1): Quoted in Streckenbach, M.: Interactions between roots and technical infrastructure underground – Principles and strategies for avoiding problems, Dissertation 2009-06, Download:
[5] Kuntze, H., Roeschmann, G. and Schwertfeger, G.: Soil science, Thieme-Verlag, 5th edition 1994
[6] DIN 28603: 2002-05
[7] E-Book 10.2015, Chapter 8, Download:
[8] Ridger, D. u. a.: Evaluation of testing of concrete and PVC pipes, Published in: Final scientific report of COST Action C 15 “Improving relations between technical infrastructure and vegetation”, 2008
[9] Whittle, A.: The resistance of elastomeric seal pipe joints to tree root penetrationm, 2003-07
[10] EN 598: 2009-10
[11] DIN 4060: 1998-02
[12] DWA-M 162: 2013-02
[13] EN 14741: 2006-05
[14] Scharwächter, D.: Long Term tightness of sealing joints in non-pressure plastic pipe systems, Plastic Pipes XI, 2001
[15] Wolf, W.: fgr volume 10 (1975), p. 55 et seq
[16] Rolwers, S.: Test report no.: G 32 980 – Determining the long-term tightness behaviour of TYTON® push-in joints with elastomer gaskets by estimating the sealing pressure based on EN 14741 [13], Iro GmbH Oldenburg, 2013-02-14
Dipl.-Ing. Christoph Bennerscheidt
EADIPS®/FGR® European Association for Ductile Iron Pipe Systems/
Fachgemeinschaft Guss-Rohrsysteme e. V.
Doncaster-Platz 5
45699 Herten/Germany
Telephone: +49 (0)2366/9943905


More News and Articles


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

Dipl.-Ing. Christoph Bennerscheidt

General Manager

Doncaster-Platz 5

45699 Herten



+49 2366 9943905


+49 2366 9943906



To website