The dual pressure procedure in polymer butt-welding, where the pressure is reduced after a few seconds, is usually used for large-scale pipes and is also standardized in regulations like ISO 21307. This work addresses the impact of using dual-pressure welding procedures on small-scale pipes made of PE 100 and PE100RC materials. Special attention is given to mechanical and also fracture-mechanical short- And long- Term properties of the welded areas and their surroundings.

Figures - uploaded by Florian Arbeiter

Author content

All figure content in this area was uploaded by Florian Arbeiter

Content may be subject to copyright.

ResearchGate Logo

Discover the world's research

  • 20+ million members
  • 135+ million publications
  • 700k+ research projects

Join for free

IMPACT OF SINGLE AND DUAL PRESSURE BUTT-WELDING PROCEDURES ON

THE RELIABILITY OF PE 100 PIPE WELDS

Florian J. Arbeiter and Gerald Pinter, Montanuniversitaet Leoben, Leoben, Austria

Andreas Frank, Polymer Competence Center Leoben, Leoben

Abstract

The dual pressure procedure in polymer butt-welding,

where the pressure is reduced after a few seconds, is

usually used for large-scale pipes and is also standardized

in regulations like ISO 21307. This work addresses the

impact of using dual-pressure welding procedures on

small-scale pipes made of PE 100 and PE 100 RC

materials. Special attention is given to mechanical and

also fracture-mechanical short- and long-term properties

of the welded areas and their surroundings.

Introduction

Pipes made from high density Polyethylene (PE-HD)

are used for gas and water distribution for well over 50

years now [1–3] . Modern pipe grades are estimated to last

over 100 years under normal operating conditions [3–5] .

Therefore it is vital to be able to characterize and test

these materials with accelerated test methods. In the last

years various methods like the Full Notched Creep Test,

Notched Pipe tests etc. have been developed and are well

established now and can be used to determine long-term

properties of pipe materials (ISO 16770, ISO 13479, etc.).

Pipe welds on the other hand are usually only

characterized by standardized short term tests

(DVS 2203). These tests can be used to differentiate

between good and bad welds, however do not necessarily

describe the long-term properties and fracture mechanism

of welds in actual application. Therefore it is important to

test short- and long-term properties of welds to give a

clear view of their performance.

Various authors have shown in the past that cracks in

pipe welds are most likely to initiate at notches, created

from the beads which form during the welding process.

The cracks however tend to grow alongside the weld in

the pipe material itself and not the actual weld [6–8] .

Therefore the behavior of the bulk material is important

when considering long term failure of pipe weld.

Imperfections created during the welding process however

can lead to actual crack growth inside the pipe weld.

Hence it is important to also examine the short- and long-

term properties of the weld material itself.

To characterize material and short term properties

Charpy-Impact, tensile- and differential scanning

calorimetry tests were performed on specimen, directly

cut from pipe welds. The long term properties of the

welds were tested by means of the recently standardized

Cracked Round Bar (CRB) test (ONR 25194). To

simulate imperfections in the welds, Charpy and CRB

specimens were notched with razor blades inside the

actual weld.

The goal of this paper was to use the upper mentioned

tests to characterize PE 100 and PE 100 RC pipe welds.

To emphasize the influence of welding parameters on

crack growth, heat soak time and the applied pressure

profile have been varied. Focus has been given to the

effect of using the dual pressure procedure, which is

usually used for large scale pipes, on small scale pipes.

Butt welding

There are various standards which are used for butt-

welding PE 100 pipes. They are usually optimized for the

different demands of the application site. Nevertheless

most standards have similar procedures. First there have

to be clean level surfaces, produced by milling and a first

contact pressure on the heating plate. Afterwards material

is melted to a certain degree during the heat soak time.

Then the hot plate is removed and the two pipes are

pressed together, forming the weld.

In the course of this work the welding procedure

according to the German standard DVS 2207-1 and the

ISO 21307 standard for the dual pressure procedure were

used (figure 1). After leveling the two pipes by milling

and cleaning the surfaces, both pipes are pressed against

the heating plate with a pressure of 0.15 MPa until an

initial bead is formed. When the initial bead is formed the

pressure is reduced to ≤ 0.01 MPa. Afterwards follows the

heat soak time. According to DVS 2207-1 it is 10 times

the wall thickness (e.g. 160 s for 16 mm thickness). The

next step is to remove the heating plate as fast as possible

to minimize cooling of the surfaces. The last step is to

press the two pipe ends together with 0.15 MPa. The

difference in the dual-pressure procedure is that the

pressure is reduced to 1/6 of the initial value after 30

seconds. This procedure, according to ISO 21307 is only

used for large scale pipes with a wall thickness of at least

22 mm.

0,00

0,05

0,10

0,15

Contact Pressure [MPa]

Time

levelling surfaces

initial bead build up

cooling - single-pressure

cooling - dual-pressure

heat soak time

Figure 1. Butt welding procedure according to

DVS 2207-1 and ISO 21307

Failure mechanisms in pressurized pipes

Failure of pressurized PE pipes can be divided into

three regions [2, 9]. The first region is where pipes

fracture after relatively short times due to ductile

deformation and rupture. The second region is governed

by crack initiation and slow crack growth (SCG). The

third region is failure due to global material ageing after

long periods of time. Failure in the first region often

occurs when the applied pressure is too high for the used

wall thickness. Failure in the third region can be reduced

by using long term stabilizers. Failure in the second

region is mainly governed by the molecular structure and

molecular weight distribution of the used polymer [10].

Failure times in this region should be greater than 50

years to ensure operational reliability of pipe systems. It is

necessary to use testing methods which can describe

failure due to SCG of PE pipe systems in shortened

testing times. A possibility poses the use of linear elastic

fracture mechanics (LEFM) which can be used to describe

said crack growth phenomena. By using LEFM the

distribution of applied stress in the vicinity of a crack tip

can be described by the stress intensity factor KI (I stands

for mode I which describes a crack opening load normal

to the crack plane). This factor is a function of applied

stress (), crack length (a) and a geometric factor (Y) that

depends on the used specimen (eq. 1) [11]. Laboratory

tests according to LEFM can be used to shorten testing

times significantly.

(1)

By applying cyclic loading instead of static loads testing

times can be further decreased, but the results are still in

good accordance to long term tests like the internal

pressure test. Recent work has shown that the use of

fatigue testing, within the limitations of LEFM and the

usage of CRB specimen is a valid tool to compare

different PE pipe grades in regard to SCG [12–16].

Therefore this testing method, which is also described in

the Austrian standard ONR 25194, was chosen to compare

different welds with respect to their resistance against

SCG.

Materials

The materials used for the experiments were three

different PE-HD types. PE 100 class materials have a

"minimum required strength " (MRS) of 10 MPa at 20°C

over a time period of 50 years. All used materials are of

this class and described in Table 1.

Table 1. Used Materials for welding.

bimodal PE -HD ,

standard PE 100

bimodal PE -HD

modified for high crack resistance

Experimental

To produce weld samples with varied parameters

pipes in the size DN160 were chosen. These pipes have a

diameter of 160 mm and a wall thickness of 14.6 mm

which requires single-pressure welding procedures

according to DVS 2207-1 and ISO 21307. Heat soak time

and pressure profile were varied during the welding

procedures. The exact variation of parameters can be seen

in Table 2.

Table 2. Welding parameters.

Differential Scanning Calorimetry

Differential Scanning Calorimetry (DSC) was used to

determine changes in cristallinity due to welding.

Therefore small samples of about 5 mg were cut from the

weld near the outer surface, the inner surface and the

middle of the pipe wall and also in respect to the distance

from the weld. The result for the weld, which was

prepared according to DVS 2207-1, is shown in figure 2.

Inner Surface Middle of Pipe Wall Outer Surface

0

1

2

3

4

45

50

55

60

65

PE 100 Pipe Weld according to DVS 2207-1

DSC: Heating Rate 10 K/min

Sample Size: 5 mg

Weld-Material

Pipe-Material (Distance from weld 20 mm)

Pipe-Material (Distance from weld >100 mm)

Degree of Cristallinity [%]

Sample Area

Figure 2. Degree of cristallinity in reference to sample

position in the pipe/weld

It can be seen, that the welding procedure increases

the degree of cristallinity of material near the weld

compared to the untreated pipe. Higher degrees of

cristallinity usually lead to higher stiffness of the material.

A schematic drawing of this gradient can be seen in

figure 3. Mechanical properties behave accordingly.

Figure 3. Gradient of degree of cristallinity in pipe weld

The variations in degree of cristallinity between the

different welding parameters were between 58 % and

61 %. Differences of this small magnitude are within the

uncertainty of the measuring apparatus and no further

statement in regard to welding parameters could be made.

Tensile Testing

To test the mechanical properties of the welds, thin

samples (200 µm) were cut from the welds at the same

positions as for the DSC tests. A schematic drawing of the

samples can be seen in figure 4.

Figure 4. Specimens used for tensile testing of welds

Not only tensile properties were observed, but also

the location where the first neck-in during the tests

occurred. The majority of samples showed neck-ins

outside of the welds. This strengthens the proposition of

the variation of mechanical properties according to the

degree of cristallinity. The Young's modulus showed no

significant change in respect to welding parameters as to

be suspected when considering results from DSC

measurements and the fact, that the tensile tests were

performed on specimens which consisted of bulk- and

weld material alike. Using the dual-pressure procedure

however increased yield stress and decreased yield strain,

respectively. The influence on yield stress for the PE 100

welds can be seen in figure 5. A similar trend was found

for tensile strength. The tests at the middle of the pipe

wall showed little to no variation and since crack initiation

is most likely to start at the surface areas, only outer and

inner samples are compared.

I II III IV V

0

5

10

15

20

25

Yield Stress [MPa]

Welding Parameter

PE100 - Yield Stress

Testing Speed: 50 mm/min

Testing Conditions: 23°C

Yield Stress - Inside Surface

Yield Stress - Outside Surface

I Single short soak time III Dual short soak time

II Single long soak time IV Dual long soak time

V according to DVS 2207-1

Figure 5. Variation of yield stress depending on welding

parameter

Impact Testing

To further characterize mechanical behavior of the

welded material impact testing was performed. The tests

were performed based on ISO 179 in Charpy setup. To

ensure similar testing conditions and to ensure crack

initiation inside the weld all specimens (10x10x50 mm)

were additionally notched with a sharp razor blade to

further sharpen the milled V-notch. All tests were

performed using a 2 Joule Hammer at 23°C. After the

tests the fracture surfaces were measured using a

microscope to get more precise areas. In figure 6 the

results of the tests are shown, normalized by the fracture

energy of the unwelded pipe material.

III III IV V I II III IV V

0,00

0,25

0,50

0,75

1,00

normalised Impact Energy [/]

Welding Parameter

Verhältnis geschweißter zu ungeschweißter Pendelschlagprobe

Impact Testing

Impact Hammer: 2J

Testing Conditions: 23°C

PE100 RC Welds PE100 Welds

I Single short soak time III Dual short soak time

II Single long soak time IV Dual long soak time

V according to DVS 2207-1

Figure 6. Normalized Impact Energy in regard to

welding parameters of PE 100 and PE 100 RC

According to these results it seems the Dual-pressure

procedure improves the impact behavior of welds which

can hint at better resistance against rapid crack

propagation.

Fatigue Testing

To properly validate the influence of welding

parameters on long-term properties of welds fatigue tests

were performed. The testing conditions were a frequency

of 10 Hz , an R-ratio of 0.1 and a testing temperature of

23°C. Hysteretic heating was monitored via IR-sensors

and showed no increase in temperature more than 5°C. All

tests were performed on CRB-specimens directly lathed

from the pipe walls. The specimens were

circumferentially notched with a sharp razor blade

precisely in the welded area. Therefore again the

properties of the weld itself, and not the pipe material

around were tested. Specimens without welds were tested

as a comparison. To be able to differentiate between

Single and Dual-pressure procedure more clearly, the two

extreme cases, Single-pressure with elongated and Dual-

pressure with shortened heat soaking time, were tested.

The results can be seen in figure 7.

100000 1000000

0,50

0,60

0,70

0,80

Fracture in PE 100 welds

Testing Conditions: 23°C

Test Frequency: 10 Hz

Loading Ratio: 0.1

PE100 - Dual-pressure, short soak time

PE100 - Single-pressure, long soak time

PE100 - According to DVS 2207-1

PE100 - Pipe without a weld

KImax [MPam0.5]

Cycles

Figure 7. Cycles until failure of PE 100 pipe welds and

pipe material as a function of the maximum

applied KI .

Similar to the impact testing an improvement of the

fracture behavior can be observed in the dual-pressure

welds compared to the single-pressure welds. The pipe

material itself shows significantly better resistance against

SCG than the welds.

Discussion

As expected butt welded pipes show good mechanical

overall behavior. Mechanical properties such as Young's

Modulus of the weld are of similar or even higher

magnitude than the bulk material of the pipe itself. The

development of welding beads, however, leads to the

creation of notches at the interface between weld and bulk

material of the pipe, creating local stress concentrations.

Initiating from these notches cracks can start to grow into

the pipe wall. Studies showed that cracks tend to grow

alongside the welded area and not in the welded area

itself. Thus the crack resistance of the bulk material

adjacent to the weld is very important when considering

failure mechanism of pipe welds.

Another matter altogether is when imperfections,

such as voids, cavities or degraded material due to

excessive heating during the welding process, are within

the weld itself. Imperfections can lead to crack initiation

and crack growth within the weld and were simulated by

notching with razor blades. This can shorten the lifetime

of the weld significantly due to lower resistance against

crack growth of the welded material compared to the pipe

material. This decrease of resistance against SCG could be

explained by molecular orientation of polymer chains in

the welded area. In pipes polymer chains are most likely

oriented parallel to the direction of extrusion, possibly

acting as an improvement against crack growth. Polymer

chains in welds are oriented more arbitrary or even

perpendicular to the direction of extrusion due to the

welding process [17], thus decreasing the resistance

against crack growth.

Using the Dual-pressure procedure on small scale

pipes, crack growth resistance could be improved. Both,

impact and fatigue testing provided results where Dual-

pressure welds performed better than Single-pressure

welds.

An explanation to this phenomenon could be, that the

decrease in pressure during cooling, allows the polymer

chains to move more freely than under constant high

pressure. Hence they can abandon the positions

perpendicular to the extrusion line which are forced upon

them by the welding process and improve the crack

resistance of the weld.

Conclusions

Higher degrees of cristallinity due to welding can

benefit the mechanical properties of welds. As cracks tend

to grow according to the principle of least constraint, a

crack path around the weld more or less perpendicular to

the point of initiation seems plausible, when considering

stress state, higher stiffness of the weld and the concave

shape of the weld itself. Flaws in the actual weld however

can decrease the resistance against crack growth

significantly compared to the bulk material, therefore

posing a risk in regard to total service lifetime of pipe

systems.

The use of Dual-pressure procedures for small scale

pipe welding has proven to increase fatigue and short-

term impact behavior of pipe welds. Nevertheless the

comparison to the fatigue behavior of the normal pipe

material shows how critical imperfections in the pipe

weld, which act as crack initiation points, can be. Whether

the use of dual-pressure increases the overall risk of

imperfections such as voids etc. due to the decrease of

pressure has to be researched more thorough.

Acknowledgements

The research work of this paper was performed at the

Polymer Competence Center Leoben GmbH (PCCL,

Austria) within the framework of the COMET-Project

"Comprehensive lifetime assessment of pressurized PE

pipe systems by an accelerated fracture mechanics based

methodology" (Project Nr. IV-3.01) of the Austrian

Ministry of Traffic, Innovation and Technology with

contributions by the University of Leoben, AGRU

Kunststofftechnik GmbH (Austria), DOW Europe GmbH

and Österreichische Vereinigung für das Gas und

Wasserfach (Austria). The PCCL is funded by the

Austrian Government and the State Governments of Styria

and UpperAustria.

References

1. Richard, K., Gaube, E., Diedrich, G.:

Trinkwasserrohre aus Niederdruckpolyäthylen.

Kunststoffe 1959(49(10)), 516–525

2. Lang, R.W., Stern, A., Dörner, G.F.: Applicability

and limitations of current lifetime prediction models

for thermoplastics pipes under internal pressure. Die

Angewandte Makromolekulare Chemie(247), 131 –

145 (1997)

3. Brömstrup, H. (ed.): 50 Jahre Rohrsysteme aus

Polyethylen für Gas, Wasser und Abwasser. Vulkan-

Verl, Essen (2006)

4. Hessel, J.: 100 Jahre Nutzungsdauer von Rohren aus

Polyethylen. 3R International(46), 242–246 (2007)

5. Frank, A., Pinter, G., Lang, R.W.: Fracture

MEchanics Lifetime Prediction of PE 80 and PE 100

Pipes under Complex Loading Conditions. In

Proceedings Plastics Pipes XV Vancouver (Cdn)

(2010)

6. Fleissner, M.: Experience with a full notch creep test

in determining the stress crack performance of

polyethylenes. Polymer Engineering and Science(38),

330– 340 (1998)

7. Frank, T.: Schweißen von Rohren mit großem

Durchmesser. Joining Plastics(4), 272–284 (2008)

8. Hessel, J.: Langzeitverhalten von

Schweißverbindungen an Großrohren aus

Polyethylen. 3R International(4-5), 314–317 (2011)

9. Gaube, E., Gebler, H., Müller, W., Gondro, C.:

Zeitstandfestigkeit und Alterung von Rohren aus

HDPE. Kunststoffe 1985(75(7)), 412–415

10. Brown, N., Lu, X., Huang, Y.: The fundamental

material parameters that govern slow crack growth in

linear polyethylene. Plastics, Rubber and Composites

Processing and Applications(17), 255– 258 (1992)

11. Hertzberg, R.W.: Deformation and fracture

mechanics of engineering materials, 4th edn. J. Wiley

& Sons, New York (1996)

12. Barker, M.B., Bowman, J., Bevis, M.: The

performance and causes of failure of polyethylene

pipes subjected to constant and fluctuating internal

pressure loadings. J Mater Sci 18(4), 1095 – 1118

(1983)

13. Parsons, M., Stepanov, E.V., Hiltner, A., Baer, E.:

Correlation of fatigue and creep slow crack growth in

a medium density polyethylene pipe material. Journal

of Materials Science 35(11), 2659–2674 (2000).

14. Haager, M.: Bruchmechanische Methoden zur

beschleunigten Charakterisierung des langsamen

Risswachstums von Polyethylen-Rohrwerkstoffen.

Dissertation, Montanuniversität Leoben (2006)

15. Lang, R.W., Balika, W., Pinter, G.: Applicability of

linear elastic fracture mechanics to fatigue in

amorphous and semi-crystalline polymers. The

Application of Fracture Mechanics to Polymers,

Adhesives and Composites(33), 83–92 (2004)

16. Pinter, G., Haager, M., Balika, W., Lang, R.W.:

Cyclic crack growth tests with CRB specimens for

the evaluation of the long-term performance of PE

pipe grades. Polymer Testing 26(2), 180–188 (2007).

17. Potente, H.: Fügen von Kunststoffen. Grundlagen,

Verfahren, Anwendung ; mit 31 Tabellen. Hanser,

München ;, Wien (2004)

... To properly describe the long term properties the fatigue tests using the CRB specimens directly lathed from the pipe wall were performed as documented in the paper [10]. The testing conditions (frequency of 10 , R-ratio of 0.1 and temperature of 23°) were applied. ...

... The testing conditions (frequency of 10 , R-ratio of 0.1 and temperature of 23°) were applied. It was found, that the pipe material itself shows significantly better resistance against the fatigue loading than the welded area itself [10]. The results of the CRB tests were confirmed by our own PENT tests using the same material. ...

The main aim of the paper is to study the influence of both material inhomogeneity and weld bead geometry on crack propagation in welded polyolefin pipes. Lifetime of three pipes welded by different welding procedures is numerically estimated. Experimentally observed shapes of weld bead and change of material properties inside the welded region (the change of Young's modulus) is implemented into the numerical model of welded pipes. Circumferential crack is of interest during the crack propagation through the pipe wall and the stress intensity factor is evaluated. It is shown that the deformation of welded region plays an important role whatever it is caused by, either the inhomogeneous distribution of Young's modulus or the amount of material in that region. The change of weld bead notch radius is not proved to be important for slow crack growth. It is shown that non-optimal welds can significantly decrease lifetime of pipe systems. The results of this research can be used for lifetime estimation and prediction of creep crack growth and further optimisation of welding conditions and butt weld technology.

... As the result of higher degree of crystallinity inside the welded joint both the resistance against the stable crack initiation (J 0.2 ) and stable crack propagation (T j ) are higher inside the welded joint rather than in the basic pipe material [14], as shown in Fig. 6. However, these short-term testing, which may be useful to differentiate between good and bad welds, does not necessarily describe the long-term properties of welds in actual applications [20]. Long-term properties of the welded pipes are in a better way described by PENT (Pennsylvania Notch Test) or CRB (Cracked Round Bar) tests. ...

... Comparison between the optimal weld and basic material is shown in Fig. 7. It can be seen, that the pipe material itself shows significantly better resistance against fatigue loading than the welded area [20]. ...

The main aim of the paper is to study the influence of both material inhomogeneity and weld bead geometry on crack propagation in welded polyolefin pipes. Axially and circumferentially oriented cracks are studied and the stress intensity factors are compared considering different positions of the cracks. Two cases of the welded pipe system are compared, one considering the optimal weld bead geometry and the other one considering the geometry after removing the weld bead once the welding process is finished. In both cases the inhomogeneous distribution of material properties inside the welded region is considered. The results show that the weld might have a negative effect on the lifetime especially when the weld bead is removed. Though the weld bead increases the stress concentration near the notches due to which a circumferential crack may appear the resulting lifetime is still comparable to that of an axial crack propagating in a homogeneous pipe.

... Schweißparameter; In Anlehnung an [13] Die CT-Prüfkörper aus dem Großrohr wurden gemäß Bild 5 entnommen. Zusätzlich wurden die CT-Prüfkörper mit Seitenkerben versehen, um einen überwiegend ebenen Dehnungszustand zu erzwingen. ...

... Aufgrund der hohen lokalen Behinderung wuchsen die Risse in diesem Fall innerhalb des geschweißten Materials. Dabei zeigte sich, dass dieses Material, aufgrund eventueller morphologischer Orientierung die durch den Schweißprozess hervorgerufen werden[13,15], einen geringeren Widerstand gegen langsames Risswachstum zu haben scheinen, als das Grundmaterial selbst. Bei Versuchen mittels CT-Prüfkörpern zeigte sich, dass die Risse trotz Seitenkerben aus der Schweißzone hinaus in das Grundmaterial wachsen. ...

... To test H-NBR, specimen geometry had to be changed according to Fig. 1. Frequency was chosen at 10 Hz for PE, PVC and POM according to findings or experience in literature (PE [15], POM [16], PVC [17]). For PP [7], PB, PA12 and H-NBR [13] testing frequency was reduced to 5 Hz to avoid excessive hysteretic heating. ...

Abstract The aim of this work is to demonstrate the applicability of the cracked round bar test recently developed for PE-HD to other polymeric materials. The main advantage of this new test method are rather short testing times for PE-HD materials used in long-term applications such as piping. Therefore, this test is of high interest for other polymers used in similar applications. Five thermoplastic materials used for plumbing (PE-HD, PP-B, PB, PVC-U, PA12), a technical polymer (POM) and an elastomeric material (H-NBR) have been tested. Scanning electron microscopy has been applied to investigate fracture surfaces. Results show that the test method seems to be basically applicable to all tested materials. Most materials showed similar fracture behaviour as postulated in literature, despite the high acceleration factor of the cyclic CRB test.

A three-dimensional model of a pressured polymer pipe with a weld is created in order to estimate the stress intensity factor for a crack located inside the weld. A comparison with values obtained for an edge-cracked tension specimen, usually employed for an experimental determination of weld properties, is performed. The difference between the stress intensity factor in a homogeneous pipe and the results taking into account the changes in material properties inside the weld is presented. A conservative and easily applicable relationship for calculating the stress intensity factor in a pipe weld is proposed. Keywordsfracture mechanics–polyolefin pipe–polymer weld–stress intensity factor–graded structure

Lifetime prediction of pressurized pipes made of polyethylene (PE) under complex loading situations is of special interest. Methods of linear elastic fracture mechanics (LEFM) are suitable to characterize the relevant long-term failure mechanisms in such pipes, which is slow crack growth. However, these methods require the knowledge of the creep crack growth (CCG) kinetics and material specific fracture mechanics parameters at application near tem-peratures. As testing of CCG with common test methods is not possible in feasible times, an accelerated extrapolation procedure based on fatigue tests with cracked round bar (CRB) specimens was developed previously. Within the present work this test procedure was used to characterize the CCG behavior of two commercial available PE pipe grades. Lifetime predictions of these materials are correlated to real failure times of internal pressure tests. Furthermore, failure times of more complex loading conditions with external soil load and external point load were predicted with fracture mechanics tools.

  • N. Brown
  • X. Lu
  • Y. Huang
  • N. Ishikawa

The effect of stress, notch depth, and temperature on slow crack growth are generally the same for all linear polyethylenes. The great difference in the rate of slow crack growth resides in the material parameters such as molecular weight distribution and branch density. Branch sequence and the placing of branches on the high molecular weight molecules make a contribution. The effects of varying the morphology by varying the thermal history are presented. The importance of the density of tie molecules is emphasized. The density of the tie molecules depends on the molecular weight and the spacing of the branches. Suggestions are made for producing a resin with a maximum resistance to slow crack growth.

  • M. Fleissner

A laboratory method to measure the stress crack resistance of polyethylenes was developed and has since been applied in our laboratory for more than twelve years. The experience gathered since our first paper is herewith reported. The creep rupture test of circumferentially notched specimens cut from plaques or pipes has proven to be a rapid and reliable method to evaluate the stress crack performance. Surfactant-assisted stress cracking was employed to accelerate testing. The stress crack resistance of several polyethylene samples was studied with respect to its dependence on stress, temperature, and environment. The creep rupture behavior at different temperatures could be superposed by horizontal shifting when the stresses were normalized in proportion to the respective bulk yield stresses. The notch tip radius turned out not to be very crucial, and variation of the nominal concentration of the surfactants, nonylphenolpolyglycolethers, scarcely affected slow crack growth. Acceleration of testing by surfactants equalized property differences to a noticeable extent but did not influence the ranking of the materials. The activation energy of crack growth was in the expected range. Defects introduced into the line by butt joint welding were precisely modeled by the full notch creep test.

  • R. W. Lang
  • A. Stern
  • G. Doerner

Engineering thermoplastics, in particular polyolefins such as special grades of poly(ethylene), are gaining importance in pipe applications such as gas and water supply systems. To ensure proper performance of such pipes over the required lifetime, polymer physics and mechanics concepts are needed to adequately account for the effects of time, temperature, and environmental conditions as well as the occurrence of pipe defects and imperfections on relevant polymer properties and pipe performance. This article provides a critical overview of the scientific background of current methodologies to describe the long-term behavior of thermoplastic pressure pipes. In particular, the merits and limitations of two different approaches-namely, the standard extrapolation method (SEM) described in ISO/TR 9080 and the linear elastic fracture mechanics (LEFM) approach-are compared. Special attention is given to effects associated with material ageing and degradation.

  • Michael Bradley Barker Michael Bradley Barker
  • J. Bowman
  • M. Bevis

Three different pipe-grade polyethylenes, in the form of one large and three small diameter pipe systems, have been tested at elevated temperatures, using constant and fluctuating internal pressure loadings that resulted in brittle fractures. The behaviour under fatigue of two of the three types of small diameter polyethylene pipes was substantially described by a cumulative damage model, whilst the third exhibited a fatigue weakness, an observation not previously reported. The performance of the large diameter pipes under fatigue was dominated by the presence of large voids in the pipe wall that arose from incorrect processing and resulted in premature failure. The sites of crack initiation in one material grade of the small diameter systems were examined in detail. In particular the size, position and composition of particles initiating fracture were determined. The maximum particle size on the fracture surface of the pipe was found to correlate reasonably well with a measure of pipe lifetime, as predicted by a fracture mechanics approach, and indicated that the lifetime of this one type of polyethylene pipe was dependent on the size of the inclusions initiating fracture.

  • M. Parsons
  • E. V. Stepanov
  • A. Hiltner
  • E. Baer

The relationship between slow crack propagation in creep and fatigue in a medium density polyethylene pipe material was studied by increasing the R-ratio (defined as the ratio of minimum to maximum stress in the fatigue loading cycle) from 0.1 to 1.0 (creep). The study included characterization of the effects of R-ratio and temperature (21 to 80C) on the mechanism and kinetics of slow crack propagation. With increasing R-ratio and decreasing temperature, the fracture mode changed from stepwise crack propagation, i.e. crack growth by the sequential formation and breakdown of a craze zone, to a quasi-continuous mode of crack growth through the preexisting craze. Despite the change in fracture mode, the damage zone, as characterized by the length of the main craze, shear crazes, and crack tip opening displacement, followed the same dependence on loading parameters, and crack growth rate followed the same kinetics. Crack growth rate (da/dt) was related to the maximum stress intensity factor KI, max and R-ratio by a power law relationship (da/dt) = BK4 I, max(1 + R)–6. Alternatively, crack growth rate was expressed as (da/dt) = BK I 4 (t)T() with a creep contribution B‹K I 4 (t)›T, calculated by averaging the known dependence of creep crack growth rate on stress intensity factor KI over the period T of the sinusoidal loading curve, and a fatigue acceleration factor () that depended on strain rate only. The correlation in crack growth kinetics allowed for extrapolation to creep fracture from short-term fatigue testing. The temperature dependence of crack growth rate was contained in the prefactors B and B. A change in slope of the Arrhenius plot of B at 55C indicated that at least two mechanisms contributed to crack propagation, each dominating in a different temperature region. This implied that a simple extrapolation to ambient temperature creep fracture from elevated temperature tests might not be reliable.