Repair conference talk

Document Sample
Repair conference talk Powered By Docstoc
					       The Composite Overwrap Option for Pipework Repair
  Simon Frost, AEA Technology PLC, Culham, Abingdon, Oxfordshire OX14 3ED
                                March 2002


This paper provides a description of work carried out by an industry forum concerned with the
development of documentation for the repair of pipework and pipeline systems using composite
materials. The membership of the group includes both users and suppliers of the various repair
options. The primary objective of the workgroup is to establish a framework within which
composite repairs can be specified and implemented with confidence. The paper addresses issues
such as material qualification, design, installation and inspection, with the aim being to
demonstrate that composite overwrap repairs are an engineered solution for pipework and pipeline.

1. Introduction

Design codes and standards for pressurised equipment provide rules for the design,
fabrication, inspection and testing of new piping and pipeline systems. ISO 15649,
ASME B31.3 and BS 8010 are examples(1)-(3). These codes do not address the fact that
equipment degrades in service or may require to be up-rated due to a change in duty,
nor do they consider options for remedial actions should such events occur. There are
standards, such as API 579 and BS7910(4)(5) that address fitness for service of a degraded
part, but, again, they do not specify or give guidance on possible repair options.

Ageing assets have created a market for repair methodologies that can be used as an
alternative to equipment replacement. This is particularly the case for pipework and
related components as degradation is often limited to isolated areas and a local repair can
offer re-instatement rather than replacement of a significant part of the system. A
further benefit of having repair schemes available is that it potentially allows immediate
action to be taken thereby minimising any system downtime. As a result there is a
considerable demand for documentation that covers the repair situation.

Repair methods using composite technology have been available for some time and
have both tempted and intrigued operators. There are many examples of successful
application. However, success has not been universal and there have been
circumstances where repairs have not met expectations. In some cases this may have
been because expectations were unrealistic in the first place, whilst in others the
behaviour of the repair material may not have been properly understood leading to
inadequate design or less than adequate specification and control of application
procedures. It was in this context that the Composites Workgroup was established with
the objective of delivering a documentation framework that would take the composites
option forward so that it could realise its full potential, i.e. an engineered solution.

2. Documentation Requirements

Composite repair systems that are currently available fall into 2 generic groups. In the
first, material often in pre-packed form can be held in stock and applied by maintenance
personnel on the facility. For these products it is envisaged that the suppliers would
produce technical data detailing the situations where their material may be used. This

would be made available as a part of the documentation accompanying the material
packs. For the second type repairs are specified and designed on a project basis and
usually applied by specialist contractors.

For documentation to be useful, therefore, it must be performance based so that it can
cover all commercially available products. A prescriptive approach that, for example,
gives specific information regarding constituent materials or laminate lay-up would
unlikely to be sufficiently inclusive and would almost certainly hinder future product
development. The features of an effective set of documentation that covers the repair
situation are:

-   it should provide guidance on the testing that suppliers need to carry out in order to
    demonstrate fitness for purpose and to derive design allowables;
-   it should identify how operators should specify the repair to be carried out;
-   it should provide design methodologies that allow operator and third party
    verification of proposed repair solutions;
-   it should advise key issues for control during installation;
-   it should define respective responsibilities to ensure that repairs are carried out safely
    and with due regard to the environment;
-   it should advise operators on issues associated with ongoing operations and NDT

The document suite that has been prepared in response to these requirements is shown
in Figure 1(6)-(8) . At each stage data is required for input and it is also a role of the
documents to define who is responsible for its provision. Together the documents
provide a systematic approach for the execution of an effective repair.

                              Operator        Operator                    Operator
                                data            data                    responsibility

             Qualification     Design        Installation        Inspection

               Vendor          Vendor          Vendor                      Vendor
                data            data            data                    responsibility

                                Figure 1: Repair documentation

3. Repair Design

Figure 2 shows the layout of the design process. The design document provides details
on the information that is necessary to specify the repair situation in the first place and
this is used as input to the calculation. Four design options are defined depending on
the defect type and available design data. A point that is emphasised is that the load

conditions should be carefully considered. For example, specifying a design pressure
well in excess of operation could prejudice the viability of composites as an option.

                                 Pipe allowable stress

                                 Composite allowable
              Design                  strain                    Design details
             allowable:                                       Minimum thickness
             4 options                                          Overlay length
                                  Requirements for
                                    leaking pipes

                                Long term performance
                                   based allowable

            Qualification               Material
               data                      data

                                   Figure 2: Design process

The route for design varies according to the type of damage being repaired. Where
there is no leak the repair material will be largely subjected to membrane forces and the
calculation reduces to one of load share between the repair laminate and the underlying
substrate. The required laminate thickness may be determined from:

                           E 
                tmin = D ⋅  s  ⋅ ( P − Ps )                                (1)
                       2 s  Ec 
                            

                             E   2F          
                t min = D ⋅  s  ⋅       − Ps                             (2)
                        2 s  E a   πD 2
                                              
If the contribution of the substrate pipe is to be ignored Equations (1) and (2) reduce to
the following:

                t min =      PD                                              (3)
                            2E c ε c

                t min =       F                                              (4)
                            πDE a ε a

where;      D is the external pipe diameter
            s is the allowable tensile stress for the steel
            Ec and Ea are the tensile modulus for the composite laminate in the
                       circumferential and axial directions

            Es is the tensile modulus for steel
            P is the internal pressure
            F is the sum of axial tensile loads due to pressure, bending and axial thrust
            εc is the allowable circumferential strain
            εa is the allowable axial strain
            Ps is the remaining maximum working pressure for the steel pipe.

The maximum allowable working pressure of the substrate pipe (MAWP) can be
determined using API 579 or other similar fitness for service codes. The complexity of
this is dependent on the nature of the defect, but often it would be a simple assessment
based on minimum remaining wall thickness. The design allowable strains used in
Equations (3) and (4) are obtained from tabular data given in the design document. The
values given vary with the design conditions and if the repair is intended for temporary
or permanent duty. The values given are broadly commensurate with those used for
the design of composite process equipment (9) .

When the pipe section is leaking a further set of calculations is required. This is because
in these circumstances the repair material is exposed directly to radial pressure forces and
to the process media. The combined action of these factors will tend to cause a
delamination along the interface between the laminate and the underlying steel. The
design method that is used involves use of a fracture energy term that characterises the
adhesion between the composite and the substrate. Figure 3 shows the situation.

                           Figure 3: Delamination of repair laminate

It can be shown that the derivation for the delamination pressure is based on an energy
balance between that stored under the deformed laminate and that required to cause
delamination, i.e.;

                        1   ∂V
                γc =      P
                       4πa ∂a
where;      a is the radius of the defect
            V is the volume under the delamination.

Using this approach the Equation for a circular defect can be shown to be:

                                          
                                          
                            γc            
      P= f                                                                  (5)
            (1 − υ 2 )  3  4  1    3   2
                            d + d+
            E 512t 3
                              π  64Gt  
where;      d is the diameter (or diameter of the equivalent circle) of the leaking region
            G is the shear modulus for the composite laminate
            E is the bending modulus for the composite laminate
            ν is the Poisson's ratio for the composite laminate
            γ is a toughness parameter or energy release rate for the composite steel
            f is the service (or safety) factor
            P is the required internal design pressure

For defects of small diameter the first term of Equation (5), which represents
deformation due to simple bending, becomes negligible. The details of Equation (5) are
strongly dependent on the boundary conditions assumed for the shear deformation at
the edge of the hole(10) .

Where the defect is of long aspect ratio, i.e. a slot, for example leaking from a weld
region a similar Equation can be derived, i.e.;

                                             
                                             
                                             
                          γc                 
   P= f                                                                           (6)
                                 4 υ       
                2                  + 
         (1 − υ )  1    π    3 5 2 2 
                    3 W + W +
         E  24t                           W 
                         4  16Gt (1 + υ )   
where W is the width of the slot.

The design or service factor, f, to be used in Equations (5) and (6) depends on the details
of the application and, as with the allowable design strains necessary for Equations (3)
and (4), can be obtained from tables within the document or determined using
performance testing. Using these analyses it is possible to develop design curves for
different defect types. Figure 4 shows a schematic example.

                         Short term pressure limit                      Design envelope

                                                               Curve scales with energy
          500                                                      release rate, γ

                                                    Service factor = 3


                   0             5             10          15                  20         25
                                                 Defect size

                         Figure 4: Delamination pressure for different defect widths

4. Product Qualification

The objectives of product qualification are threefold:

-   to demonstrate fitness for purpose for the required operating conditions;
-   to obtain quantitative data for use in design calculations;
-   to define those parameters and the limits to those parameters that need to be
    controlled during repair.

It is a fundamental premise behind performance based design methods that the materials
and processes that are used to produce test samples for qualification are identical to those
to be used in the delivered product or service. It is also important that testing replicates
the conditions to be seen in service. Of necessity, qualification always represents a
compromise between testing rigour and practical limitations. If the testing requirement
is too limited its value will be minimal and the uncertainty that this would represent
would need to be catered for through the imposition of large safety factors. Too
extensive a test programme would cause the document to fall into disrepute – it would
not be used, hinder the uptake of the product and inhibit development.

In the repair documentation procedures for product qualification are defined. These

-   specification of tests for basic material properties of the repair laminate, e.g. modulus
    and strength values. Wherever possible existing test methods are specified;
-   description of simple tests to demonstrate a minimum level of durability. For
    example, there is a requirement to carry out simple lap shear testing after a period of
    exposure and achieve a minimum level of residual strength;
-   description of tests to determine the toughness design parameter, energy release rate
    γ, needed for leaking repairs, Equations (4) and (5). Essentially this entails pressure

    testing a series of specimens with flaws of different geometry and then determining γ
    empirically by reference to the relevant Equations.

A key point when considering qualification data is that the achievement of a high γ
should not necessarily be seen as an objective. What is more important is that the value
that is used can be replicated with confidence under site conditions. Arguably a lower
value that has with it an associated reduced degree of scatter could be a preferred
arrangement. It would be rare for the additional material costs that this would imply to
be important in the competitiveness of the product. Figure 5 demonstrates schematically
the data fit procedure. Essentially the 95% lower confidence bound as the data fit curve
and the design curve is this curve divided by the service factor, currently defined as 3.

                                 Data          Data fit - mean               Data fit - 95% LCL

      Pressure (bar)



                             0          5        10          15         20         25         30          35
                                                    Defect diameter (mm)

                       Figure 5: Delamination pressure for different defect widths - data fit procedure

Work is ongoing in assessing qualification testing. A specific interest at the present time
is the merit of introducing simpler tests in order to reduce potential test burdens. This
need arises due to the fact that there is interest in using composites for a range of
substrates. In principle this would require the testing to be repeated for each variant. It
is important to appreciate that when considering a repair arrangement it is the
composite/surface/substrate that is the qualified unit. It is not possible to do tests on
one arrangement and then infer data for a second substrate or, indeed, any other change.

The test method under consideration is the wedge test(11) . Figure 6 shows the
arrangement. This has considerable merit as the loads applied during the test are
broadly similar to those for a repair in service, data can be obtained in a short test period
and it is possible to derive an estimate for the energy release rate. Even if there is doubt
regarding the absolute levels of toughness data it may be possible to derive scaling
factors so that the effect of substrate, change of service temperature or other variable can
be assessed quickly, scaled accordingly from a pre-defined datum.

                                                                    Composite repair
                     Load, Pfinal
                                            Measured crack
                                              length, a

                                                  Pfinal2 ∂C δ 2 ∂C
                                      γ wedge =             =
                                                     2 ∂a 2C 2 ∂a
                             where C is the compliance of the repair laminate

                                    Figure 6: Wedge test specimen

5. Installation

Inevitably each repair product will have its own installation requirements. The repair
documentation(6) provides guidance for each step of the process and advises on what
should be included within an installation manual. Figure 7 shows the layout of the
document. The fundamental issue is that the site installation should mirror those
processes that were applied in the preparation of samples for qualification testing. This
is especially the case for surface preparation as this is the single most important task to be
performed. It is also likely that failure to execute this operation correctly is the root
cause of many of the examples of disappointing performance. In many respects the
challenges of achieving adequate surface preparation are similar to those encountered
during painting so they are not unduly onerous.

                       Risk                Installer
                    assessment           qualifications

                      Method                                        Testing and
                     statement                                       inspection

                     HSE input                                     Documentation

                                  Figure 7: Installation process

To assist in the achievement of the necessary level of control the installation document
advises on the contents of method statements including the definition of hold points.
These include simple on-site tests that are helpful in checking surface preparation.
What the document does not specify is what the surface preparation should be or how
to achieve it. There is guidance on good practice for given circumstances, but the detail
is for the supplier to state following the qualification process.

Installer qualification also includes guidance on what constitutes a minimum level of
training. This applies where the repair is being executed by local maintenance
personnel as, regardless who is doing the repair; they need to be trained. Also the
qualification and training for supervision of the repair application is currently under

6. Repair Inspection

The Inspection/NDT document is written as a guidance note as opposed to
specification form in recognition of the fact that this area is still in development (8) . In
essence there are 3 inspection challenges for composite repairs:

-   inspection of the repair laminate;
-   inspection of the interface between the repair and the substrate;
-   inspection of the underlying pipe in service after repair.

Of these, the third is of most concern to the operators, especially if the pipe is subject to
internal corrosion. Indeed, this issue is perhaps the most significant with regard to the
potential use of composites for the more demanding applications. For hydrocarbon
service the ability to inspect the status of the pipe post repair is probably a pre-requisite.

The document provides information on both aspects of the inspection issue; tabulated
allowable defects followed by the inspection methods that can be used. In reality
inspection of the repair laminate itself is probably of limited value and the main means

of assuring quality, i.e. to employ effective process control during application. This is
similar to the practice adopted for the construction of composite process equipment.

For detection of defects within the substrate, electromagnetic techniques are currently
used. They are able to see through both carbon and glass based composites and
therefore the repair does not need to be removed. Both low frequency and pulsed eddy
current techniques have been applied with success.

For the detection of delaminations within the interface, currently no techniques are
used. However, laser shearography is one potential technique which measures the
surface deflection of the laminate may offer some chance of success

For the detection of defects within the composite laminate visual inspection is the best
available technique are present.

Further work in assessing the effectiveness of NDT techniques for delamination
detection and monitoring the growth of defects within the substrate is continuing.

7. Current Work

In addition to further work on the present sequence of documents and development of
the qualification test procedures, the Workgroup is expanding its scope to include the
following areas:

− other pipework components, e.g. fittings and flanges;
− long term performance testing;
− repairs to vessels and tanks;
− audit of repair suppliers to advise operators on compliance with the intent of the
− assessment of NDT techniques.

8. Summary

This paper has described a suite of documentation that covers the design, installation
and inspection of composite repair methods for pipework systems and pipelines. The
benefits of the work are that it provides a framework that allows operators to select the
composite option with confidence. In addition the establishment of an accepted
approach to material qualification gives suppliers a firm basis on which to invest in
material testing and product development programmes. Together these points
represent the necessary next steps in taking composite repair products forward so that
they can realise their potential in the marketplace.

9. References

1.   ISO/DIS 15649, Petroleum and natural gas industries - piping.
2.   ASME B31.3, Chemical plant and petroleum refinery piping.
3.   BS 8010, Code of practice for pipelines.
4.   API 579, Recommended practice for fitness for service.
5.   BS 7910, Guide on methods for assessing the acceptability of flaws in fusion welded

6. AEAT - 57711/1, Design of composite repairs for pipework, March 2000
7. AEAT - 57756/1, Installation procedures for composite repairs, Nov 2000
8. AEAT - 57756/2, Review of NDT techniques for the inspection of composite
    repairs, Nov 2000
9. PrEN 13121, GRP tanks and vessels for use above ground.
10. Timoshenko, SP and Gere, JM, Mechanics of Mateials, Van Nostrand
11. ASTM D3762, Standard test for adhesively bonded surface durability.

9. Acknowledgements

The support and guidance of the Composite Repair Workgroup is gratefully
acknowledged. The Workgroup members consist of users (Shell, BP, Saudi Aramco,
Amerada Hess, Petrobras and BG-Hydrocarbon Resources Limited) and material
suppliers (Walker Technical Resources, Devonport Management Ltd, Clockspring and
Industrial Maintenance Group).

Organisations wishing to participate in the Workgroup should contact Simon Frost or
Geoff Eckold at AEA Technology.


Shared By:
Description: Repair conference talk