Final Report Summary - COMPIPE (Composite Pipes and Fittings for Aero-Engines Dressing)
Executive Summary:
The COMPipe project is developing non-metallic dressings for use in civil aerospace gas turbine engines. The project is being undertaken by a consortium of three members; Rolls-Royce Plc., Sigma Precision Components Ltd. (“Sigma”) and TWI Ltd. This project is part of the Clean Sky programme. High level project targets and development activities are given in two documents: The original project Call for Proposal (CfP) and the project Description of Work (DoW) document.
Primary development activities are being performed by Sigma and TWI. Rolls-Royce is the topic manager and has design authority for the engines being considered.
The object of this project is to update the concept of aero-engine dressing to the most advanced concepts already used in commercial aircrafts. The existing design concepts and components utilised within the engine dressings have remained largely static for a long period of time; there is little difference in the components and materials used in the engine dressings on the latest Trent engines to the early RB211’s from the 1970’s. The purpose of the topic call was to challenge the existing technology, by utilising non-metallic alternatives to save weight with minimal impact upon cost and functionality.
COMPipe allows to design and manufacture an agreed set of non-metallic pipes and support systems to replace traditional metallic variants in engine dressings. It has identified the manufacturing processes that could be used to scale up the production requirements to meet future delivery needs.
The scientific and technical achievements reached from the project are:
i. Design and development of a new concept of non-metallic pipes and fittings suited for aero-engines, specifically those for drains, scavenge, sensor, vent lines and oil.
ii. Selected materials meeting the required environmental constrains, both in the range of temperatures and in the chemical resistance to different products present in the aero-engine, such as Skydrol hydraulic fluid, oil and aviation fuel.
iii. Developed manufacturing practices to improve existing methodologies driving to lower cost production. Notably these have been development of a thermoplastic and thermoset matrix composite pipe.
iv: Additionally the project has managed to develop a solution which incorporates the technology required to meet the requirements of fuel carrying systems.
v. The project has produced a thermoplastic composite product standard that has been shown capable to reach zero wall porosity and high pressure retaining capability.
vi. The project has proven the ability to replicate the 3D geometries compatible with existing metallic technology.
vii. In the case of thermoplastic pipes, the present manufacturing capability and future full automation of the process has been assessed and future developments identified.
viii. To meet the requirements of fire performance, the project has developed new standards from the existing standards of Rolls-Royce in order to achieve measurable performance in the composite product.
ix. The project has provided technical results that can be used to upgrade existing elements of metallic pipe systems.
x. The project has demonstrated the applicability of existing technologies for fire resistant composite materials.
xi. Through a series of rig tests, the project demonstrated that composite pipe assemblies are capable of meeting, or even exceeding the performance and environmental requirements for an aero engine
Project Context and Objectives:
The Clean Sky Joint Technology Initiative (JTI) is the European Union's largest ever aeronautics research programme, with a total value of €1.6bn supporting demonstration of a wide range of aeronautics technologies for reduced emissions of CO2, NOx and noise.
The objective of Clean Sky is to demonstrate technologies in six programmes, all contributing to the ACARE (Advisory Council for Aeronautic Research in Europe) goals of 50 per cent reduction in CO2, 80 per cent reduction in NOx and 50 per cent reduction in perceived noise emissions by 2020. A Technology Evaluator will model each of the ITD (Integrated Technology Demonstrator) technologies and assess the contribution of the technologies towards delivery of the ACARE goals.
The Sustainable and Green Engine (SAGE) ITD of Clean Sky demonstrates five engine technologies contributing towards the environmental targets. There are six engine projects contained in the programme. Each one targets specific technologies and market sectors, led by a member of the European engines industry.
Clean Sky SAGE3 project aims to develop and demonstrate a large 3-shaft bypass engine Demonstrator. The topic manager is the major engine manufacturer, Rolls-Royce. The COMPipe project will develop non-metallic aero engine pipes and support systems to replace traditional metallic variants for engine externals. The objective of the project is to develop the technology and demonstrate to Technology Readiness Level (TRL) 6.
Over the last 30 years there has been little change to the existing metallic technology for pipes and supporting systems. Common pipe base materials are Titanium, Stainless Steel and Nickel Alloys (for example Inconel). Pipe fittings and details utilise the same materials and are commonly welded or brazed. The design of the end fittings have also largely remained static during this period. Applying the advances in computational analysis like CFD (computational fluid dynamics) and FEA (finite element analysis) into the design process for new end fittings can demonstrate significant potential for weight optimisation. Component Designers 30 years ago would have focussed on design for ease of manufacture and economical costings using tooling and machining available at the time. The objective of this project is to develop new end fitting designs utilising modern technology to dramatically optimise the weight of each component. In high temperature and high pressure applications, OEMs often elect to use Titanium materials, however long term stability and cost of Titanium has a high level of associated risk due to fluctuations in supply and sustainability of Titanium ore sources to support a retrofit of stainless steel technology.
The purpose of the topic call was to challenge the existing technology, by utilising non-metallic alternatives to save weight with minimal impact upon cost and functionality.
Specific objective criteria extracted from the call for proposal are:
• Operating range capability of -85°C to 165°C, but materials capable of operating in excess of this range are preferred.
• Materials are to be Fire-resistant to JES314-1 (which conforms to ISO2685) as a minimum and the system capable of continued operation in an engine environment.
• Pipe diameters to range from 0.25” to 1.5”
• Pipe bend radii of 1.5 x OD (later amended by Rolls-Royce to 2.5 x OD)
• Pipe length up to 1.8metres
• Pressure capability of -10psig to 450psig working and 800psig maximum.
The systems capability requirements are stated as follows:
• Non-metallic solution to be used for all Aero engine Air and Oil pipework. Specifically drains, scavenge, sensor and vent lines.
• Demonstrated resistance to Skydrol hydraulic fluid, oil and aviation fuel.
• Material porosity should be as close as possible to zero.
• Demonstrate how the non-metallic solution will combat handling damage in service.
• Core material and design must enable complex 3D shapes to be created.
• Demonstrate how potential inter-weave leaks at pipe bend radii are combated.
• Demonstrate alternative pipe support systems for practical integration and assembly on an engine.
• Demonstrate new and novel end fittings.
Manufacturing automation was regarded throughout the project as an additional primary objective. Typical large turbo fan engines have in excess of 200-300 different pipe designs and therefore any technology proposed by the project must be capable of meeting the technical specification whilst also having a flexible manufacturing technique. Throughout the project this objective developed further from focussing on designing manufacturing processes for ease of manufacture and with high levels of automation (to reduce inherent labour costs) to producing manufacturing methods with high levels of process stability to demonstrate a repeatable output from manufacture. This objective was deemed necessary to ensure suitable maturity of parts for testing to demonstrate that both the technology objectives could be met and that it could be met in large scale manufacture with a level of confidence.
Although the rate of adoption of composite materials in the gas turbine engine has not met the expectations of a decade ago, their superior specific strength and stiffness still promise major performance improvements. Indeed, there is a significant competitive advantage to be gained if the European aero engine industry can successfully develop new technologies and commercialise them rapidly. Complete airframes can now be produced in PMCs, and they are essential to modern helicopters. Temperature requirements limit their use in aero-engines, but most of the nacelle of a modern aero gas turbine is PMC, a component that accounts for around 25% of the weight and 20% of the cost of the power plant. Other PMC parts in current engines include fan blades, outlet guide vanes, bypass ducts, nose cone spinners, core engine fairings, annulus fillers and variable guide vane rings.
Until recently, the two major, inter-related, drivers for the application of composites in engines have been weight reduction and performance improvement. For example, MMC (Metal Matrix Composites) compressor drums have the potential for 80% weight saving over a conventional disc and blade assembly and PMC components typically provide 20-30% weight saving. If COMPipe can reduce the weight of 1 kg pipes by 25%, this equates to 3325 kg of CO2 during the life of an engine. Given predicted engine deliveries of 137,000 engines over the next 20 years, this is a lifetime saving of over 450,000 tons CO2.
Project Results:
1. Development of Two Approaches: Thermoplastic and Thermoset Fibre Reinforced Composite
The purpose of the call was to develop new technologies or to adapt existing ones to produce lightweight pipes and fittings for aero-engine dressings. Initially a short technical review of both the use of composites in aero-engines and the specific requirements related to the dressing was completed. The study and empirical data provided by the topic manager, Rolls-Royce formed the basis for the technical requirements for which the pipes must meet. These included factors such as strength, fatigue life, electrical conductivity, resistance to damage, fire performance, porosity, pressure retention, maintenance and installation and operating temperatures.
Currently aero-engines dressing are made out of metals such as titanium, nickel alloys for instance Inconel and stainless steel. This technology has remained steady for the last 30 years. A requirements capture phase demonstrated that non-metallic replacement technology must provide equivalent performances as the metallic components. Focussing on the specific strength or force per unit area at failure (σ/ρ) of stainless steel is typically 0.254 unreinforced polymer (Polypropylene) is 0.037 and a carbon/epoxy (61% resin) is 1.08. The project identified that the majority of ceramic materials have very poor fatigue and impact performance therefore the project concluded that the non-metallic solution must be a composite material in order to meet the strength requirement.
To put this into context, a composite material is a material made from two or more constituent materials with significantly different physical or chemical properties that, when combined, produce a material with characteristics different from the individual components. Examples include carbon fibre, fibreglass and concrete. In this aerospace application each material would not satisfy the requirements individually. It is the combined performance of the two materials that form the composite that realises a pipe with sufficient strength and performance characteristics.
Pipes have been the main focus of the call as these were considered throughout to be the most challenging element. Pipe systems have been trialled with composites in many different ways. The most common small scale production method found in a study of existing composite pipe technology identified many companies producing hand lay-up thermoset pipes formed in rigid moulds. This is a labour intensive process and requires complex tooling for each individual pipe geometry. The project identified that the most appropriate, for high production rates and cost effective manufacture, were found to utilise the following technologies or processes:
Thermoplastic Pipe
- Braided carbon fibre reinforced thermoplastic polymer
- Straight moulding tools
- CNC post formed into complex 3D geometries
- Mechanical fitting connections
- Metallic and hybrid polymer fittings
- External fire protection
Thermoset Pipe
- Braided carbon fibre reinforced epoxy polymer
- Chemically modified fire resistant polymer
- Moulded into final shape
- Chemically bonded fitting connections
- Metallic and hybrid polymer fittings
During the period of trialling, the project has looked at alternatives to the above solutions including materials, different preforming methods, 3D forming processes, end fitting designs and methods of joining them to the pipe.
The trials creating pre-prototype parts have demonstrated that the two pipe approaches detailed above were the preferred solutions as only they demonstrated capability to meet the complete technical requirements of the project. The project team reviewed a complete set of technologies and merits at the Preliminary Design Review, chaired by Rolls-Royce, prior to embarking upon the two thermoset and thermoplastic fibre reinforced composite options.
2 Thermoset Pipe
One method for the manufacture of a composite reinforced polymer pipe is to utilise a thermoset resin for the polymer material. Such a material can only be moulded into the final shape of the pipe, thus requiring individual rigid tooling to create the 200-300 pipe geometries for an engine. Thermoset resin pipes were produced as part of the project activities and TWI developed a modified thermoset resin to provide the pipe with enhanced fire performance capability. This is a completely novel application, combining the TWI technologies for producing thermoset pipes and fire resistant resins into a singular product. TWI successfully developed a tooling and production process for producing thermoset fibre reinforced pipes to the example 3D geometries supplied by Rolls-Royce. These pipes were tested for pressure retaining capabilities but due to the difficulties in achieving pipes to meet the large number of geometric shapes without the need for many complex moulding tools, the thermoset fibre reinforced pipes were not developed to the same TRL status as thermoplastic pipes.
2.1 Thermoset Material Selection
The most challenging aspect of material selection is the desired maximum continuous working temperature (165°C defined in the call but preferably up to 350°C per Rolls-Royce requirements). Additionally, the pipe and fitting assemblies must demonstrate fire resistance according to RRES 90023 - Rolls-Royce Fire Test Specification. A ranking exercise of available off-the-shelf materials was performed; scoring is based primarily on temperature performance.
For the thermoset resins the available materials included
TRC prepregs (A&P Braided Sleeve)
Resin System (in prepreg)
UF3357 Tg : 182°C
UF3350 Tg : 191°C
Reinforcement
Braided Carbon / Biaxial Sleevings (GAMMASOX™)
Biaxial Carbon Fabric (BIMAX™)
Triaxial Fabric (TRIMAX™)
P2SI prepregs
Material Description Process Information (Tg TCure)
P2SI® 635LM High temperature structural thermosetting polyimide prepreg with low melt viscosity.
500°F – 550°F service temperature applications. Out-of-autoclave cure capable.
Autoclave curing @ 85 psi. 335°C 343°C
P2SI® 4600 High temperature aerospace-grade epoxy. Out-of-autoclave cure. Post-cure at 200°C.
260°C 177°C
P2SI® T410 High temperature epoxy with low temperature initial cure. Composite tooling applications. Out-of-autoclave cure. Post-cure at 200°C.
Lower initial cure temperatures possible.
210°C 80°C
o Liquid Moulding Systems - Taylor-made Reinforcement by Sigma
▪ Hexcel
• RTM6 Tg : 200°C (Epoxy/Infusion)
• RTM651 Tg : 285°C (BMI/RTM)
▪ UBE
• PETI-330 Tg : 330°C (BMI/RTM)
▪ Cytec
• PRISM™ EP2400 Tg : 200°C (Epoxy/Infusion)
▪ Huntsman
• Tactic 742 Tg : 310°C (Epoxy - solid)
This exercise considered a range of thermoset matrices but, since the fire test is an extreme requirement, it was not included in the initial ranking of materials.
2.2 Thermoset Resin Fire Resistance
Combining the capabilities of TWI and the nature of the thermoset resins, two different approaches for producing composite pipes with fire resistance capabilities inherently engineered into the material were adopted.
The first is applicable to both thermoset and thermoplastic matrix materials. In this approach it is assumed that the matrix itself cannot be modified and it is necessary to isolate the polymer from the flame. Intumescent paint as well as other inorganic coatings, both metallic and ceramic, have been tested during the project with limited results.
The second approach is applicable to the thermoset resin used in the resin transfer moulding (RTM) process (Vitolane®). In this case the matrix has been modified using two different methods, both showing very promising results. Elements of this technology may also be developed to provide a fire resistant coating or additional protective layer for use in thermoplastic matrix composites.
Vitolane®
To improve the fire resistance of the epoxy resin RTM6 (HEXCEL®), which is used in the manufacture of the composite pipes, TWI has developed a novel and innovative technology based in the functionalization of silica - Vitolane® technology. Within this new technology it is possible to obtain high organic-inorganic hybrid materials. The addition of a proportion of this inorganic compound in high levels (50%wt in loading, approximately), can provide an improvement in the fire resistance of the organic resin. If successful, the developed RTM6 resin will provide to the core structure of the composite fire protection, without compromising the stability and performance of the original resin.
Organo-phosphorus compounds
The fire resistance of epoxy based composites can substantially be increased by the use of various additives. There are additives which can impart flame retardancy to the system but unreactive solid additives adversely affect the mechanical properties.
We have used a reactive organo-phosphorus compound to produce an inherently flame retarded epoxy system with no adverse effect on the mechanical properties. The plaques made by the resin pass UL-94 V0 standards. Although the fire test on carbon fibre composite pipes made with the resin composition in the lab was very positive, a test on a full pipe assembly was not undertaken during the project (in favour of pursuing a thermoset pipe solution) but could be scheduled for future technology development beyond the project.
2.3 Thermoset Pipes Manufacturing Process
For thermoset matrices a flexible tooling system composed of an inner silicone cord and an outer silicone tube is used as an expendable mould. The flexible tooling is supported by a rigid low cost external structure that defines the complete pipe geometry.
The main advantage of this system is the easy set-up of the tool and the greater control of pipe spring back. Pipe spring back is defined in this application as the level in which a bend in the pipe straightens after the mould or supporting feature is removed and can be difficult to predict with different sizes, bend angles and manufacturing processes of the pipes.
Thermoset resin impregnation was another method of pipe forming investigated by TWI. Braided carbon material was impregnated with a thermoset resin to produce a composite pipe. Oven and Infra-red curing techniques were developed for this method approach which were then utilised with other pipe preform methods later in the project.
It was found that, after a prototype batch of 5 pipes, this method would not be applicable for the overall application due to the inconsistencies in the wall thickness of the final article. Uneven areas of resin coverage which would ultimately lead to a part inconsistent in material properties and be very complex to qualify such a part to aerospace standards.
Overall, the project concluded that using thermoset resin systems to manufacture aero engine pipes would not be a satisfactory solution. The relatively low operating temperatures and fire resistance of most resins would limit the number of applications on aero engines and the manufacturing processes, in particular the need for individual mould tools for each pipe, would not be cost effective.
3. Thermoplastic Pipe
The thermoplastic pipe consists of a braided fibre reinforced polymer utilising the thermoplastic polymer PEEK as the matrix polymer. The carbon provides the basis of the strength of the pipe with the matrix polymer acting as a barrier to porosity. The material analysis process investigated various properties looking at commercial availability, temperature performance, qualification level and material strength performance.
Post braiding, the pipes are cured using proprietary technology and processes developed specifically by Sigma. Straight pipes are formed using modified CNC equipment to form complex 3D geometries. This method replicates metallic pipe forming procedures and the end fittings are applied after pipes are formed. During this reporting period, Sigma have developed a method of curing composite pipes and tooling for forming bent pipes.
The project investigated alternative technologies to braiding, most notable was filament winding which is a very similar technology to braiding. However the filament winding process did not yield prototype parts with sufficient hoop strength to meet the system pressure levels documented in the call.
Pre-prototype parts have been manufactured and cross checked for conformance with the existing Rolls-Royce Design rules listed in DRA54. Furthermore, testing has shown that no further design rule changes within DRA54 are necessary for the introduction of composite pipes in replacing existing stainless pipes. For fit and form, thermoplastic pipes can replicate the existing metallic technology.
Sigma has calculated values for the predicted strength of the pipe to ensure that the design requirements will be met by the prototype parts produced for testing. Classical laminate theory, material data within MIL-HDBK-17-2F and NASA reported values were used for the basis for this study. The design of the thermoplastic pipes have been proven to have a four times safety factor in the ability to withstand high pressure systems. The project call for proposal stated maximum pressures of 800psig and through requirements analysis with Rolls-Royce an additional target of 1020psig was suggested which has also been met with a safety margin. The maximum pressure achieved during pressure testing was 4000psig with the test halted as the end fitting bond leaked at this point.
3.1 Manufacture Methods
The project has concluded that lightweight engine dressing pipes can be manufactured from a thermoplastic matrix fibre reinforced polymer (FRP) and these pipes will replicate the current metallic pipe geometry and connections. Stock lengths of straight composite pipe will be manufactured and this straight material will then be used to produce bent pipe assemblies though application of post-forming processes.
The Patent Pending production process for the carbon fibre reinforced PEEK polymer pipe, arranged in a high level order from start to finish, is:
1. Preparation of materials and base mandrel.
2. Braid straight pipe preforms from tows/yarns made up of a mixture of thermoplastic polymer and a reinforcing material such as carbon fibres. Yarns can be either braided directly onto hard tooling (e.g. a metallic mandrel), or braids are manufactured ‘loose’ and later placed on/in a tool.
3. Wrap or coat the braided composite with a consolidation material and then ‘cure’ at an elevated temperature. The cure process melts the thermoplastic polymer allowing it to flow and wet-out the reinforcing fibres, the consolidating force on the composite is required to ensure good flow of the liquefied matrix and a certain consolidation pressure ensures a non-porous pipe. When the part is cooled the polymer re-solidifies thus producing a carbon fibre reinforced polymer (CFRP) material.
4. Reform the straight composite pipe into a complex bent shape equivalent to current metallic assemblies, this process is carried out on a modified CNC bending machine with bespoke tooling and control systems for forming thermoplastic composite. Bending is performed at elevated temperature so that the thermoplastic properties of the composite matrix can be utilised to aid forming and spring back. A complex computer model using the obtained properties of the thermoplastic material has been developed through the project to model the spring back level for each pipe design. This computer simulation model can be utilised in the NPI process of product launch to ensure that the post bend spring back is factored into the set-up of the bending programme to ensure the geometry requirements of the pipe are met.
5. Attach end fittings (manufactured in a separate process using traditional techniques) to the pipe. End fittings can be either directly comparable to current welded pipe fittings, or a new lightweight design. End fitting connections are made using a mechanical interface akin to a swaged connection. The end of the pipe is flared to allow the fitting to be inserted into the end of the pipe and a metallic crimp is then compressed down over the flair and fitting to provide a strong mechanical interface. The project team investigated the use of adhesives in end fitting attachment however due to the bonding nature of PEEK polymer, the strength of the bond was insufficient to meet the requirements of the call.
3.2 Precursor Material and Braiding
The most challenging aspect of material selection is the desired maximum continuous working temperature defined by Rolls-Royce; up to 350°C. Additionally, the pipe and fitting assemblies must demonstrate fire resistance according to RRES 90023 - Rolls-Royce Fire Test Specification.
During material identification and development activities, Sigma identified multiple types and sub-types of polymer available, e.g. PEEK, PPS, PEI, PEEKEK, PA6 etc. PEEK was ultimately selected due to having the highest processing temperature of all common polymers, therefore methods developed with PEEK can be more easily ‘read down’ to other materials. PEEK is also a very well established material in aerospace and readily available. PEEK has excellent mechanical and high temperature performance according to the information provided by the manufacturer.
For the fibres, three common fibres were investigated; Carbon, Glass and Aramids, all available in many types. For the project, uni-directional carbon was selected due to it being very well established in aerospace and was readily available in forms suitable for the selected process (i.e. braiding). The carbon fibres also demonstrate mechanical capability beyond that required in the product specifications.
The material for thermoplastic pipes utilises aerospace grade carbon fibres with superior consolidation and composite wet-out during processing. Unfortunately this material was difficult to braid and handle and Sigma had to develop processes and tooling to overcome many issues during manufacture. The process has been developed with automation as a key requirement to ensure that product can be produced for minimal cost.
Sigma set about improving the working process with the material to ensure that from a production perspective, the material could be utilised to produce volumes of up to 1500m of braided preform every day. This was identified by Sigma as a key enabling parameter for which to assess processes against to ensure that a production process could be commercially viable. Highly automated braiding reduces the labour costs of manufacture which enables the production of lower cost composite products, increasing the market applications for composite products.
Substantial trialling of different braiding and filament winding techniques have taken place to optimise the ratio of pipe strength to material usage. The goal is to utilise as little material as possible, making the pipe lighter and cheaper, but ensuring that the key criteria of strength and porosity are achieved. To optimise these parameters, investigations into different braid designs and pipe masses have been carried out resulting in a set of toolbox parameters.
The main barrier related to use of braided structures is related to the misalignment of fibres, unequal distribution of materials between the inside and the outside of the bend, which can result in:
- Porosity
- Wrinkling
- Change in mechanical properties.
Due to the fact that the product, an aero-engine, must be failsafe, standard manufacturing capabilities produce low quality structures; it was necessary to develop a toolbox to control the manufacturing limitations and the relationship between geometry and fibre distribution in the pipe.
Sigma have developed a toolbox for calculating the machine controls for a given braid, which is manually inputted into the braiding machine controller. A sample of different braid patterns was produced and their strength assessed in order to ascertain the optimal braiding parameters for the COMPipe product. In future a programmed recipe will control the braiding variables for each pipe design.
3.3 Consolidation Process
The next stage developed during the project is to take the braided preform and cure this to form a pipe which is consolidated under temperature. The output at this stage is straight composite pipe once removed from the mandrel.
Sigma developed a complex consolidation process for thermoplastic pipes which is in the process of being protected by a product Patent. The process allowed the consolidation of over 100metres of non-porous composite thermoplastic pipe as part of the project. The longest pipe produced was a total of 2200mm which surpassed the call requirement of 1800mm.
In support of the consolidation method, the project also investigated the means in which heat is applied to the composite to activate the process. A convection oven, induction heater and infra-red lamps were three of the technologies trialled to ensure a repeatable and production optimised process. A specific process to cure the composite pipe was developed as a result of the project.
The melting point of PEEK is 343°C so must be held at a temperature above this to allow the carbon fibre and PEEK to consolidate under the pressure of a tensioned tape. A high temperature furnace was used to prove the concept but the heating and cooling cycles were too long to make this a viable production process.
Both Sigma and TWI project teams investigated induction heating the composite as this is a very efficient heating process. A part is placed within an induction field which induces a current into the part which in turn generates heat in the part. Carbon is a very good conductor of current and therefore well suitable for induction heating however in the trials carried out for heat application for curing, the results were often inconsistent.
The most successful process was to use infra-red lamps to heat the pipe externally. Consistent results were produced using this method.
3.4 Forming 3D Geometries
The final process flow from finished straight pipes to completed pipe assemblies. In these final stages, the finished straight pipe is formed into complex 3D geometries and end fittings are applied to the pipe in a final stage.
Metal pipes are bent using a CNC bending machine. The machine consists of 4 electronically/hydraulically operated dies: a pressure die, a wiper die, a bend die and a clamp die. In order to make it possible to use with a thermoplastic composite pipe either extra tooling for heating the pipe directly must be added, or the tooling must be adapted to be heated.
Methods of conveying heat into the pipe were trialled but the heated dies method was chosen as the simpler and better controlled solution. Dies were modified for faster heating with heat flow simulations done to achieve a consistent temperature all around the dies. Dies were heated using a range of cartridge heaters.
In order to understand the key variables, their interactions and how they affect the process, a factorial design of experiments was proposed to investigate this. A list of variables was drawn proven through statistical analysis; examples are: Bend Speed, Bend Die Temp, and Pressure Die Temp.
The sample batch of pipes used for the experiments were all manufactured with the same process, and were bent using the same equipment.
Over 34 bends, all at 90°, the average spring back was by 3.31°, or 3.68%, with a standard deviation of 2.05° or 2.28%. Understanding the spring back characteristics when bending the pipe go towards achieving the correct geometry. The composite pipe can be bent to replicate current metallic geometries and has been checked against existing inspection fixtures for Rolls-Royce engine pipes. The spring back is also modelled in Abaqus software to prove the set-up of the machine and actual geometry of the bent pipe.
3.5 End-fitting joining
The final forming process is the joining of metallic end fittings to the pipe. Since adhesives are much less effective when used with thermoplastics than with thermosets, and have limited resistance to temperature and fire, an alternative joining method had to be developed.
Provisional trials have taken place as a proof of concept for the design which was proposed at the CDR. The fitting is mechanically interfaced with the pipe and the interference fitting takes place at elevated temperature in a swage process. The interlocked shape provides the ability to withstand the axial loads applied to the connection and the swage ring prevents any polymer creep through service life.
The process requires no welding or chemical adhesion which facilitates more simplified NDT processes than existing stainless steel end fitting joining.
Mechanical testing of this joint was performed (vibration, pressure & temperature tests, axial load tests) which proved successful at ambient temperatures. Unfortunately the joint leaked at elevated temperatures therefore an alternative design will have to be developed beyond the timescale of the project.
3.6 End-Fitting Design
Pipe end-fittings provide a method of joining a pipe assembly to other pipe assemblies or other pieces of equipment on an engine. End-fittings can take various forms (e.g. threaded connections, flanges, clamps) and are usually welded, brazed or swaged to metallic pipes. Owing to the higher temperature applications on aero engines the sealing mechanism is frequently metal to metal, requiring high compression loads. Separate seals (e.g. O-rings) have limitations in temperature and are susceptible to damage and therefore not commonly used in pipe systems. Current metallic end-fittings have historically been designed for ease of manufacture and therefore tend to be heavy.
The project examined a number of alternative end fitting concepts in different materials and with alternative sealing and connection mechanisms. Owing to engine operating temperatures, the conclusion was that metal to metal seals, and therefore predominately metallic, end-fittings were the most suitable.
Sigma has developed and owns the design rights for three bespoke end fittings for the composite pipe :
a. Lightweight End-Fitting – based on a current design already in service but optimised for minimum weight (about 30% lighter). They were also designed to be interchangeable with the current fittings on metallic pipe systems.
b. Hybrid End-Fitting – Similar to a. above but reducing weight still further by replacing certain metallic features with over-moulded thermoplastic. Sample components were made but were deemed to be too costly and less durable than the all-metal alternatives.
c. Band Clamp – an adaptation of a V-band clamp type of joint commonly used in aerospace. Although very light and compact, the manufacturing costs were prohibitive.
Final testing was performed on pipe assemblies using Lightweight End-Fittings.
3.5 Future Optimisation
The CfP states that the project requirement is to develop the product to TRL6 and to identify manufacturing process outlines for large scale production. Owing to the end-fitting joints leaking at higher temperatures and instability with the CNC Bending process, the project only achieved TRL5, as assessed by the Topic Manager.
Sigma have identified the following optimisation actions to improve the current designs and manufacturing processes to attain TRL6 and to become cost competitive with metallic pipe systems.
a. Braiding – minimise raw material usage and convert braiding from a batch process into a more continuous one.
b. Curing – The developed curing system is optimised for the pipe sizes required under the project however a significant optimisation can again be achieved by closely coupling it to the braiding process.
c. Pipe forming – The consistency, yield and speed of the CNC bending process must be improved using bespoke tooling and integrating it with the CNC program.
3.6 Process Stability
Key Process Parameters (KPP) are key set point levels or capabilities that must be met in order for the process or system to meet its operational goal. Through the development of the manufacturing capability of the project, Sigma has developed a critical control plan to understand the tolerance window for each KPP to ensure that pipes produced are within the specifications detailed on the drawings. The critical control plan documents the full manufacturing procedure and allowable process windows as part of the product quality control procedure. The development of the control plan and critical process window for each of the key processes in production has been a large element of the project. Without such a control on the complex manufacturing process, it would not have been possible to consistently produce pipes to the project TRL.
Topic Manager, Rolls-Royce, invested time and resource in supporting Sigma to develop controls and processes to mature the manufacturing capability of the product. Throughout the project the potential of the product was clear and the support from Rolls-Royce demonstrated their level of confidence in the technology and the abilities of the project team to produce a final product and process that meets both the call for proposal requirements and the Rolls-Royce detailed technical requirements.
4 Non Destructive Testing
The NDT technologies selected for the quality control of the products has to be related to the most likely defects in manufacturing. A careful assessment during manufacture was completed to assess the potential defects of both matrix approaches, thermoset and thermoplastic. The common major defects categorised by their potential to cause significant operational damage to the pipe system are:
- Porosity
- Matrix cracking
- Delamination
- Voids
Based upon these likely defects, a review and ranking of available NDT processes has been carried out. The results are documented in the project deliverable report, D3.1 Quality Control Procedures.
Computerised tomography is being used to analyse the prototype bent pipes. It has been found that this system provides an accurate method for the assessment of the inner deformation of the pipe, ensuring that the quality of the new products fulfil the standards defined for metallic pipes.
This process provides the necessary information for the qualification of the prototypes, but it cannot be considered for series production for cost reasons. It is envisaged that future developments in the use of ultrasonic Lamb-waves or laser shearography will be necessary for the inspection of thin walled composites in series production.
5. Conclusion
Based on the Topic Description of the Clean Sky Project “Non-metallic Pipes for Aero engine Dressings”, a new concept for pipes and end fittings has been developed. This concept can be categorised as a paradigm as it will replace the existing technology that has driven the design of aero engine dressings for more than half a century.
It was concluded at the project closure meeting attended by the project team and Rolls-Royce delegates that the project has come a long way and significant knowledge has been gained by all on the design and manufacturing of composite pipes.
The conclusion provided by Paul Broughton, COMPipe Design Review Chairperson within Rolls-Royce is that it is possible to make composite pipes that could be used for significant portions of the on engine piping system, with some of the major obstacles being overcome through the project, especially fire and pressure capability. The project has demonstrated that individually on a test rig the pipes can meet the objectives of this program, although the fire was not conducted with vibration and the remaining tests only fully conducted on straight pipes.
Due to the current manufacturing capability and control, a full suite of pipes is not currently available to offer to an engine development program. The control of the bending process in particular has been highlighted by the team as the area that needs more investment and work, along attachment of end fittings that leaked following thermal shock or cycling. The current analysis also shows that this technology will offer a similar weight to titanium which therefore makes the cost analysis a critical point for the exploitation of this technology. It is clear that the process developed can be scaled up to a production process and therefore there is a clear method of how this can be taken to production rate levels.
The project was awarded TRL5 status by the Topic Manager, on the basis where it can be shown that there is a manufacturing process that can produce each feature, the yield is currently low and there is still some work to do to get to a reliable production process. Moreover the key technical requirements have been demonstrated in rig tests to date to a maturity of TRL5.
It was concluded that there was good collaboration between all parties examining the various possible manufacturing techniques and also good investigation into the potential inspection methods. During all of the meetings the discussion has been open, which has resulted in a proactive approach to this project by all.
Ultimately the final COMPipe sample presented at the meeting demonstrated a weight reduction of 52% over an identical stainless steel pipe assembly. Whilst the cost to manufacture the product is more difficult to quantify directly due to the low manufacturing maturity it was proven to be comparable to the costs of Titanium parts with large reductions possible through increased capability growth and increased production volumes.
For future exploitation of the product and to reach greater TRLs the following elements need to be addressed to support continued exploitation.
1. Fire test under vibration.
2. Fire test at a bend, again with vibration
3. Pressure and Thermal testing of bent pipes.
4. Thermal shock and cycling on improved end fitting.
This activities have been recognised by the project and recorded as actions for exploitation beyond the Clean Sky COMPipe project.
Potential Impact:
During the period of the COMPipe project the market circumstances have changed significantly, with a number of key product introduction windows shifting to the right as a consequence of the post financial crisis market place. It is evident that the next major fleet technology insertion opportunities now need a step-change in performance capability more than before. This makes this Clean Sky COMPipe project all the more relevant and important to the continued and relentless pursuit of better performance and industry competitiveness.
On average, 12% of aeronautic sector revenues, representing almost €7 billion per year for civil aeronautics alone, are reinvested in Research and Development (R&D) and support around 20% of aerospace jobs. The European aeronautics sector is a vital contributor to society and economy. The industry accounts for approximately 3% of EU workforce, generates roughly €220 billion of the European GDP per year, and contributes positively to the EU’s trade balance with over 60% of its products exported. Every Euro invested in aeronautics R&D creates an equivalent additional value in the economy every year thereafter.
-The aeronautics sector is a key contributor to the European economy and the EU’s balance of trade
-Continued growth in demand for air travel raises new environmental and socio-economic challenges
- Research and innovation is core to EU competitiveness and drives sustainable value creation
Aviation is a vital enabler of our economy and society. The aviation market continues to grow significantly and the associated societal challenges persist. Air travel will remain essential to ensure EU’s economic growth, mobility, integration and cohesion. Air traffic is forecast to grow by 4% to 5% per year in the next decade leading to a near doubling of traffic by 2020. This poses environmental, societal and economic challenges that can only be tackled through an intense and sustained cooperation between public authorities, industry, research organisations and academia.
The socio-economic impact from the COMPipe is significant. For the COMPipe product, the implications of the technology are in line with governmental and airline policies for the reduction of carbon emissions from air travel. The technology delivers this in two ways: by reducing the overall weight of an aero engine, this in turn reduces the weight of the supporting structure, thereby reducing the weight of fuel required; secondly by using a thermoplastic composite as an alternative to thermoset composite, traditional used and found all over a modern aircraft, the opportunity for end of life recycling is increased.
Further to this, the manufacturing technology has expanded to include other applications including torque shaft, torsion bars, rotor containment sleeves, space frames and other bespoke applications, often replacing the slow manufacturing option of hand laid composites, with a more automated, repeatable, fast option of thermoplastic braiding. The viability of the technology as an alternative is huge, with the capability for one machine to produce around 1 mile of tubing per day. This would take composites into a far higher volume than previously possible. Hand laid composites can produce perhaps thousands of parts a year, automated layup methods such as filament winding increase this to tens of thousands per year. Braiding has the opportunity to increase this to hundreds of thousands of parts a year, opening up a part of the market that previously has been unavailable, in the mass automotive manufacturing business. This is an area in which the UK is a major manufacturer with several large manufacturers based here.
Sigma and TWI have been disseminating the technology and its advantages through a variety of shows and presentations, alongside press releases that have created a significant amount of interest in COMPipe and its technology derivatives.
List of Websites:
www.compipe.eu
Project Coordinator
Michael Andreae
Sigma Precision Components
+44 (0)1455 445500
The COMPipe project is developing non-metallic dressings for use in civil aerospace gas turbine engines. The project is being undertaken by a consortium of three members; Rolls-Royce Plc., Sigma Precision Components Ltd. (“Sigma”) and TWI Ltd. This project is part of the Clean Sky programme. High level project targets and development activities are given in two documents: The original project Call for Proposal (CfP) and the project Description of Work (DoW) document.
Primary development activities are being performed by Sigma and TWI. Rolls-Royce is the topic manager and has design authority for the engines being considered.
The object of this project is to update the concept of aero-engine dressing to the most advanced concepts already used in commercial aircrafts. The existing design concepts and components utilised within the engine dressings have remained largely static for a long period of time; there is little difference in the components and materials used in the engine dressings on the latest Trent engines to the early RB211’s from the 1970’s. The purpose of the topic call was to challenge the existing technology, by utilising non-metallic alternatives to save weight with minimal impact upon cost and functionality.
COMPipe allows to design and manufacture an agreed set of non-metallic pipes and support systems to replace traditional metallic variants in engine dressings. It has identified the manufacturing processes that could be used to scale up the production requirements to meet future delivery needs.
The scientific and technical achievements reached from the project are:
i. Design and development of a new concept of non-metallic pipes and fittings suited for aero-engines, specifically those for drains, scavenge, sensor, vent lines and oil.
ii. Selected materials meeting the required environmental constrains, both in the range of temperatures and in the chemical resistance to different products present in the aero-engine, such as Skydrol hydraulic fluid, oil and aviation fuel.
iii. Developed manufacturing practices to improve existing methodologies driving to lower cost production. Notably these have been development of a thermoplastic and thermoset matrix composite pipe.
iv: Additionally the project has managed to develop a solution which incorporates the technology required to meet the requirements of fuel carrying systems.
v. The project has produced a thermoplastic composite product standard that has been shown capable to reach zero wall porosity and high pressure retaining capability.
vi. The project has proven the ability to replicate the 3D geometries compatible with existing metallic technology.
vii. In the case of thermoplastic pipes, the present manufacturing capability and future full automation of the process has been assessed and future developments identified.
viii. To meet the requirements of fire performance, the project has developed new standards from the existing standards of Rolls-Royce in order to achieve measurable performance in the composite product.
ix. The project has provided technical results that can be used to upgrade existing elements of metallic pipe systems.
x. The project has demonstrated the applicability of existing technologies for fire resistant composite materials.
xi. Through a series of rig tests, the project demonstrated that composite pipe assemblies are capable of meeting, or even exceeding the performance and environmental requirements for an aero engine
Project Context and Objectives:
The Clean Sky Joint Technology Initiative (JTI) is the European Union's largest ever aeronautics research programme, with a total value of €1.6bn supporting demonstration of a wide range of aeronautics technologies for reduced emissions of CO2, NOx and noise.
The objective of Clean Sky is to demonstrate technologies in six programmes, all contributing to the ACARE (Advisory Council for Aeronautic Research in Europe) goals of 50 per cent reduction in CO2, 80 per cent reduction in NOx and 50 per cent reduction in perceived noise emissions by 2020. A Technology Evaluator will model each of the ITD (Integrated Technology Demonstrator) technologies and assess the contribution of the technologies towards delivery of the ACARE goals.
The Sustainable and Green Engine (SAGE) ITD of Clean Sky demonstrates five engine technologies contributing towards the environmental targets. There are six engine projects contained in the programme. Each one targets specific technologies and market sectors, led by a member of the European engines industry.
Clean Sky SAGE3 project aims to develop and demonstrate a large 3-shaft bypass engine Demonstrator. The topic manager is the major engine manufacturer, Rolls-Royce. The COMPipe project will develop non-metallic aero engine pipes and support systems to replace traditional metallic variants for engine externals. The objective of the project is to develop the technology and demonstrate to Technology Readiness Level (TRL) 6.
Over the last 30 years there has been little change to the existing metallic technology for pipes and supporting systems. Common pipe base materials are Titanium, Stainless Steel and Nickel Alloys (for example Inconel). Pipe fittings and details utilise the same materials and are commonly welded or brazed. The design of the end fittings have also largely remained static during this period. Applying the advances in computational analysis like CFD (computational fluid dynamics) and FEA (finite element analysis) into the design process for new end fittings can demonstrate significant potential for weight optimisation. Component Designers 30 years ago would have focussed on design for ease of manufacture and economical costings using tooling and machining available at the time. The objective of this project is to develop new end fitting designs utilising modern technology to dramatically optimise the weight of each component. In high temperature and high pressure applications, OEMs often elect to use Titanium materials, however long term stability and cost of Titanium has a high level of associated risk due to fluctuations in supply and sustainability of Titanium ore sources to support a retrofit of stainless steel technology.
The purpose of the topic call was to challenge the existing technology, by utilising non-metallic alternatives to save weight with minimal impact upon cost and functionality.
Specific objective criteria extracted from the call for proposal are:
• Operating range capability of -85°C to 165°C, but materials capable of operating in excess of this range are preferred.
• Materials are to be Fire-resistant to JES314-1 (which conforms to ISO2685) as a minimum and the system capable of continued operation in an engine environment.
• Pipe diameters to range from 0.25” to 1.5”
• Pipe bend radii of 1.5 x OD (later amended by Rolls-Royce to 2.5 x OD)
• Pipe length up to 1.8metres
• Pressure capability of -10psig to 450psig working and 800psig maximum.
The systems capability requirements are stated as follows:
• Non-metallic solution to be used for all Aero engine Air and Oil pipework. Specifically drains, scavenge, sensor and vent lines.
• Demonstrated resistance to Skydrol hydraulic fluid, oil and aviation fuel.
• Material porosity should be as close as possible to zero.
• Demonstrate how the non-metallic solution will combat handling damage in service.
• Core material and design must enable complex 3D shapes to be created.
• Demonstrate how potential inter-weave leaks at pipe bend radii are combated.
• Demonstrate alternative pipe support systems for practical integration and assembly on an engine.
• Demonstrate new and novel end fittings.
Manufacturing automation was regarded throughout the project as an additional primary objective. Typical large turbo fan engines have in excess of 200-300 different pipe designs and therefore any technology proposed by the project must be capable of meeting the technical specification whilst also having a flexible manufacturing technique. Throughout the project this objective developed further from focussing on designing manufacturing processes for ease of manufacture and with high levels of automation (to reduce inherent labour costs) to producing manufacturing methods with high levels of process stability to demonstrate a repeatable output from manufacture. This objective was deemed necessary to ensure suitable maturity of parts for testing to demonstrate that both the technology objectives could be met and that it could be met in large scale manufacture with a level of confidence.
Although the rate of adoption of composite materials in the gas turbine engine has not met the expectations of a decade ago, their superior specific strength and stiffness still promise major performance improvements. Indeed, there is a significant competitive advantage to be gained if the European aero engine industry can successfully develop new technologies and commercialise them rapidly. Complete airframes can now be produced in PMCs, and they are essential to modern helicopters. Temperature requirements limit their use in aero-engines, but most of the nacelle of a modern aero gas turbine is PMC, a component that accounts for around 25% of the weight and 20% of the cost of the power plant. Other PMC parts in current engines include fan blades, outlet guide vanes, bypass ducts, nose cone spinners, core engine fairings, annulus fillers and variable guide vane rings.
Until recently, the two major, inter-related, drivers for the application of composites in engines have been weight reduction and performance improvement. For example, MMC (Metal Matrix Composites) compressor drums have the potential for 80% weight saving over a conventional disc and blade assembly and PMC components typically provide 20-30% weight saving. If COMPipe can reduce the weight of 1 kg pipes by 25%, this equates to 3325 kg of CO2 during the life of an engine. Given predicted engine deliveries of 137,000 engines over the next 20 years, this is a lifetime saving of over 450,000 tons CO2.
Project Results:
1. Development of Two Approaches: Thermoplastic and Thermoset Fibre Reinforced Composite
The purpose of the call was to develop new technologies or to adapt existing ones to produce lightweight pipes and fittings for aero-engine dressings. Initially a short technical review of both the use of composites in aero-engines and the specific requirements related to the dressing was completed. The study and empirical data provided by the topic manager, Rolls-Royce formed the basis for the technical requirements for which the pipes must meet. These included factors such as strength, fatigue life, electrical conductivity, resistance to damage, fire performance, porosity, pressure retention, maintenance and installation and operating temperatures.
Currently aero-engines dressing are made out of metals such as titanium, nickel alloys for instance Inconel and stainless steel. This technology has remained steady for the last 30 years. A requirements capture phase demonstrated that non-metallic replacement technology must provide equivalent performances as the metallic components. Focussing on the specific strength or force per unit area at failure (σ/ρ) of stainless steel is typically 0.254 unreinforced polymer (Polypropylene) is 0.037 and a carbon/epoxy (61% resin) is 1.08. The project identified that the majority of ceramic materials have very poor fatigue and impact performance therefore the project concluded that the non-metallic solution must be a composite material in order to meet the strength requirement.
To put this into context, a composite material is a material made from two or more constituent materials with significantly different physical or chemical properties that, when combined, produce a material with characteristics different from the individual components. Examples include carbon fibre, fibreglass and concrete. In this aerospace application each material would not satisfy the requirements individually. It is the combined performance of the two materials that form the composite that realises a pipe with sufficient strength and performance characteristics.
Pipes have been the main focus of the call as these were considered throughout to be the most challenging element. Pipe systems have been trialled with composites in many different ways. The most common small scale production method found in a study of existing composite pipe technology identified many companies producing hand lay-up thermoset pipes formed in rigid moulds. This is a labour intensive process and requires complex tooling for each individual pipe geometry. The project identified that the most appropriate, for high production rates and cost effective manufacture, were found to utilise the following technologies or processes:
Thermoplastic Pipe
- Braided carbon fibre reinforced thermoplastic polymer
- Straight moulding tools
- CNC post formed into complex 3D geometries
- Mechanical fitting connections
- Metallic and hybrid polymer fittings
- External fire protection
Thermoset Pipe
- Braided carbon fibre reinforced epoxy polymer
- Chemically modified fire resistant polymer
- Moulded into final shape
- Chemically bonded fitting connections
- Metallic and hybrid polymer fittings
During the period of trialling, the project has looked at alternatives to the above solutions including materials, different preforming methods, 3D forming processes, end fitting designs and methods of joining them to the pipe.
The trials creating pre-prototype parts have demonstrated that the two pipe approaches detailed above were the preferred solutions as only they demonstrated capability to meet the complete technical requirements of the project. The project team reviewed a complete set of technologies and merits at the Preliminary Design Review, chaired by Rolls-Royce, prior to embarking upon the two thermoset and thermoplastic fibre reinforced composite options.
2 Thermoset Pipe
One method for the manufacture of a composite reinforced polymer pipe is to utilise a thermoset resin for the polymer material. Such a material can only be moulded into the final shape of the pipe, thus requiring individual rigid tooling to create the 200-300 pipe geometries for an engine. Thermoset resin pipes were produced as part of the project activities and TWI developed a modified thermoset resin to provide the pipe with enhanced fire performance capability. This is a completely novel application, combining the TWI technologies for producing thermoset pipes and fire resistant resins into a singular product. TWI successfully developed a tooling and production process for producing thermoset fibre reinforced pipes to the example 3D geometries supplied by Rolls-Royce. These pipes were tested for pressure retaining capabilities but due to the difficulties in achieving pipes to meet the large number of geometric shapes without the need for many complex moulding tools, the thermoset fibre reinforced pipes were not developed to the same TRL status as thermoplastic pipes.
2.1 Thermoset Material Selection
The most challenging aspect of material selection is the desired maximum continuous working temperature (165°C defined in the call but preferably up to 350°C per Rolls-Royce requirements). Additionally, the pipe and fitting assemblies must demonstrate fire resistance according to RRES 90023 - Rolls-Royce Fire Test Specification. A ranking exercise of available off-the-shelf materials was performed; scoring is based primarily on temperature performance.
For the thermoset resins the available materials included
TRC prepregs (A&P Braided Sleeve)
Resin System (in prepreg)
UF3357 Tg : 182°C
UF3350 Tg : 191°C
Reinforcement
Braided Carbon / Biaxial Sleevings (GAMMASOX™)
Biaxial Carbon Fabric (BIMAX™)
Triaxial Fabric (TRIMAX™)
P2SI prepregs
Material Description Process Information (Tg TCure)
P2SI® 635LM High temperature structural thermosetting polyimide prepreg with low melt viscosity.
500°F – 550°F service temperature applications. Out-of-autoclave cure capable.
Autoclave curing @ 85 psi. 335°C 343°C
P2SI® 4600 High temperature aerospace-grade epoxy. Out-of-autoclave cure. Post-cure at 200°C.
260°C 177°C
P2SI® T410 High temperature epoxy with low temperature initial cure. Composite tooling applications. Out-of-autoclave cure. Post-cure at 200°C.
Lower initial cure temperatures possible.
210°C 80°C
o Liquid Moulding Systems - Taylor-made Reinforcement by Sigma
▪ Hexcel
• RTM6 Tg : 200°C (Epoxy/Infusion)
• RTM651 Tg : 285°C (BMI/RTM)
▪ UBE
• PETI-330 Tg : 330°C (BMI/RTM)
▪ Cytec
• PRISM™ EP2400 Tg : 200°C (Epoxy/Infusion)
▪ Huntsman
• Tactic 742 Tg : 310°C (Epoxy - solid)
This exercise considered a range of thermoset matrices but, since the fire test is an extreme requirement, it was not included in the initial ranking of materials.
2.2 Thermoset Resin Fire Resistance
Combining the capabilities of TWI and the nature of the thermoset resins, two different approaches for producing composite pipes with fire resistance capabilities inherently engineered into the material were adopted.
The first is applicable to both thermoset and thermoplastic matrix materials. In this approach it is assumed that the matrix itself cannot be modified and it is necessary to isolate the polymer from the flame. Intumescent paint as well as other inorganic coatings, both metallic and ceramic, have been tested during the project with limited results.
The second approach is applicable to the thermoset resin used in the resin transfer moulding (RTM) process (Vitolane®). In this case the matrix has been modified using two different methods, both showing very promising results. Elements of this technology may also be developed to provide a fire resistant coating or additional protective layer for use in thermoplastic matrix composites.
Vitolane®
To improve the fire resistance of the epoxy resin RTM6 (HEXCEL®), which is used in the manufacture of the composite pipes, TWI has developed a novel and innovative technology based in the functionalization of silica - Vitolane® technology. Within this new technology it is possible to obtain high organic-inorganic hybrid materials. The addition of a proportion of this inorganic compound in high levels (50%wt in loading, approximately), can provide an improvement in the fire resistance of the organic resin. If successful, the developed RTM6 resin will provide to the core structure of the composite fire protection, without compromising the stability and performance of the original resin.
Organo-phosphorus compounds
The fire resistance of epoxy based composites can substantially be increased by the use of various additives. There are additives which can impart flame retardancy to the system but unreactive solid additives adversely affect the mechanical properties.
We have used a reactive organo-phosphorus compound to produce an inherently flame retarded epoxy system with no adverse effect on the mechanical properties. The plaques made by the resin pass UL-94 V0 standards. Although the fire test on carbon fibre composite pipes made with the resin composition in the lab was very positive, a test on a full pipe assembly was not undertaken during the project (in favour of pursuing a thermoset pipe solution) but could be scheduled for future technology development beyond the project.
2.3 Thermoset Pipes Manufacturing Process
For thermoset matrices a flexible tooling system composed of an inner silicone cord and an outer silicone tube is used as an expendable mould. The flexible tooling is supported by a rigid low cost external structure that defines the complete pipe geometry.
The main advantage of this system is the easy set-up of the tool and the greater control of pipe spring back. Pipe spring back is defined in this application as the level in which a bend in the pipe straightens after the mould or supporting feature is removed and can be difficult to predict with different sizes, bend angles and manufacturing processes of the pipes.
Thermoset resin impregnation was another method of pipe forming investigated by TWI. Braided carbon material was impregnated with a thermoset resin to produce a composite pipe. Oven and Infra-red curing techniques were developed for this method approach which were then utilised with other pipe preform methods later in the project.
It was found that, after a prototype batch of 5 pipes, this method would not be applicable for the overall application due to the inconsistencies in the wall thickness of the final article. Uneven areas of resin coverage which would ultimately lead to a part inconsistent in material properties and be very complex to qualify such a part to aerospace standards.
Overall, the project concluded that using thermoset resin systems to manufacture aero engine pipes would not be a satisfactory solution. The relatively low operating temperatures and fire resistance of most resins would limit the number of applications on aero engines and the manufacturing processes, in particular the need for individual mould tools for each pipe, would not be cost effective.
3. Thermoplastic Pipe
The thermoplastic pipe consists of a braided fibre reinforced polymer utilising the thermoplastic polymer PEEK as the matrix polymer. The carbon provides the basis of the strength of the pipe with the matrix polymer acting as a barrier to porosity. The material analysis process investigated various properties looking at commercial availability, temperature performance, qualification level and material strength performance.
Post braiding, the pipes are cured using proprietary technology and processes developed specifically by Sigma. Straight pipes are formed using modified CNC equipment to form complex 3D geometries. This method replicates metallic pipe forming procedures and the end fittings are applied after pipes are formed. During this reporting period, Sigma have developed a method of curing composite pipes and tooling for forming bent pipes.
The project investigated alternative technologies to braiding, most notable was filament winding which is a very similar technology to braiding. However the filament winding process did not yield prototype parts with sufficient hoop strength to meet the system pressure levels documented in the call.
Pre-prototype parts have been manufactured and cross checked for conformance with the existing Rolls-Royce Design rules listed in DRA54. Furthermore, testing has shown that no further design rule changes within DRA54 are necessary for the introduction of composite pipes in replacing existing stainless pipes. For fit and form, thermoplastic pipes can replicate the existing metallic technology.
Sigma has calculated values for the predicted strength of the pipe to ensure that the design requirements will be met by the prototype parts produced for testing. Classical laminate theory, material data within MIL-HDBK-17-2F and NASA reported values were used for the basis for this study. The design of the thermoplastic pipes have been proven to have a four times safety factor in the ability to withstand high pressure systems. The project call for proposal stated maximum pressures of 800psig and through requirements analysis with Rolls-Royce an additional target of 1020psig was suggested which has also been met with a safety margin. The maximum pressure achieved during pressure testing was 4000psig with the test halted as the end fitting bond leaked at this point.
3.1 Manufacture Methods
The project has concluded that lightweight engine dressing pipes can be manufactured from a thermoplastic matrix fibre reinforced polymer (FRP) and these pipes will replicate the current metallic pipe geometry and connections. Stock lengths of straight composite pipe will be manufactured and this straight material will then be used to produce bent pipe assemblies though application of post-forming processes.
The Patent Pending production process for the carbon fibre reinforced PEEK polymer pipe, arranged in a high level order from start to finish, is:
1. Preparation of materials and base mandrel.
2. Braid straight pipe preforms from tows/yarns made up of a mixture of thermoplastic polymer and a reinforcing material such as carbon fibres. Yarns can be either braided directly onto hard tooling (e.g. a metallic mandrel), or braids are manufactured ‘loose’ and later placed on/in a tool.
3. Wrap or coat the braided composite with a consolidation material and then ‘cure’ at an elevated temperature. The cure process melts the thermoplastic polymer allowing it to flow and wet-out the reinforcing fibres, the consolidating force on the composite is required to ensure good flow of the liquefied matrix and a certain consolidation pressure ensures a non-porous pipe. When the part is cooled the polymer re-solidifies thus producing a carbon fibre reinforced polymer (CFRP) material.
4. Reform the straight composite pipe into a complex bent shape equivalent to current metallic assemblies, this process is carried out on a modified CNC bending machine with bespoke tooling and control systems for forming thermoplastic composite. Bending is performed at elevated temperature so that the thermoplastic properties of the composite matrix can be utilised to aid forming and spring back. A complex computer model using the obtained properties of the thermoplastic material has been developed through the project to model the spring back level for each pipe design. This computer simulation model can be utilised in the NPI process of product launch to ensure that the post bend spring back is factored into the set-up of the bending programme to ensure the geometry requirements of the pipe are met.
5. Attach end fittings (manufactured in a separate process using traditional techniques) to the pipe. End fittings can be either directly comparable to current welded pipe fittings, or a new lightweight design. End fitting connections are made using a mechanical interface akin to a swaged connection. The end of the pipe is flared to allow the fitting to be inserted into the end of the pipe and a metallic crimp is then compressed down over the flair and fitting to provide a strong mechanical interface. The project team investigated the use of adhesives in end fitting attachment however due to the bonding nature of PEEK polymer, the strength of the bond was insufficient to meet the requirements of the call.
3.2 Precursor Material and Braiding
The most challenging aspect of material selection is the desired maximum continuous working temperature defined by Rolls-Royce; up to 350°C. Additionally, the pipe and fitting assemblies must demonstrate fire resistance according to RRES 90023 - Rolls-Royce Fire Test Specification.
During material identification and development activities, Sigma identified multiple types and sub-types of polymer available, e.g. PEEK, PPS, PEI, PEEKEK, PA6 etc. PEEK was ultimately selected due to having the highest processing temperature of all common polymers, therefore methods developed with PEEK can be more easily ‘read down’ to other materials. PEEK is also a very well established material in aerospace and readily available. PEEK has excellent mechanical and high temperature performance according to the information provided by the manufacturer.
For the fibres, three common fibres were investigated; Carbon, Glass and Aramids, all available in many types. For the project, uni-directional carbon was selected due to it being very well established in aerospace and was readily available in forms suitable for the selected process (i.e. braiding). The carbon fibres also demonstrate mechanical capability beyond that required in the product specifications.
The material for thermoplastic pipes utilises aerospace grade carbon fibres with superior consolidation and composite wet-out during processing. Unfortunately this material was difficult to braid and handle and Sigma had to develop processes and tooling to overcome many issues during manufacture. The process has been developed with automation as a key requirement to ensure that product can be produced for minimal cost.
Sigma set about improving the working process with the material to ensure that from a production perspective, the material could be utilised to produce volumes of up to 1500m of braided preform every day. This was identified by Sigma as a key enabling parameter for which to assess processes against to ensure that a production process could be commercially viable. Highly automated braiding reduces the labour costs of manufacture which enables the production of lower cost composite products, increasing the market applications for composite products.
Substantial trialling of different braiding and filament winding techniques have taken place to optimise the ratio of pipe strength to material usage. The goal is to utilise as little material as possible, making the pipe lighter and cheaper, but ensuring that the key criteria of strength and porosity are achieved. To optimise these parameters, investigations into different braid designs and pipe masses have been carried out resulting in a set of toolbox parameters.
The main barrier related to use of braided structures is related to the misalignment of fibres, unequal distribution of materials between the inside and the outside of the bend, which can result in:
- Porosity
- Wrinkling
- Change in mechanical properties.
Due to the fact that the product, an aero-engine, must be failsafe, standard manufacturing capabilities produce low quality structures; it was necessary to develop a toolbox to control the manufacturing limitations and the relationship between geometry and fibre distribution in the pipe.
Sigma have developed a toolbox for calculating the machine controls for a given braid, which is manually inputted into the braiding machine controller. A sample of different braid patterns was produced and their strength assessed in order to ascertain the optimal braiding parameters for the COMPipe product. In future a programmed recipe will control the braiding variables for each pipe design.
3.3 Consolidation Process
The next stage developed during the project is to take the braided preform and cure this to form a pipe which is consolidated under temperature. The output at this stage is straight composite pipe once removed from the mandrel.
Sigma developed a complex consolidation process for thermoplastic pipes which is in the process of being protected by a product Patent. The process allowed the consolidation of over 100metres of non-porous composite thermoplastic pipe as part of the project. The longest pipe produced was a total of 2200mm which surpassed the call requirement of 1800mm.
In support of the consolidation method, the project also investigated the means in which heat is applied to the composite to activate the process. A convection oven, induction heater and infra-red lamps were three of the technologies trialled to ensure a repeatable and production optimised process. A specific process to cure the composite pipe was developed as a result of the project.
The melting point of PEEK is 343°C so must be held at a temperature above this to allow the carbon fibre and PEEK to consolidate under the pressure of a tensioned tape. A high temperature furnace was used to prove the concept but the heating and cooling cycles were too long to make this a viable production process.
Both Sigma and TWI project teams investigated induction heating the composite as this is a very efficient heating process. A part is placed within an induction field which induces a current into the part which in turn generates heat in the part. Carbon is a very good conductor of current and therefore well suitable for induction heating however in the trials carried out for heat application for curing, the results were often inconsistent.
The most successful process was to use infra-red lamps to heat the pipe externally. Consistent results were produced using this method.
3.4 Forming 3D Geometries
The final process flow from finished straight pipes to completed pipe assemblies. In these final stages, the finished straight pipe is formed into complex 3D geometries and end fittings are applied to the pipe in a final stage.
Metal pipes are bent using a CNC bending machine. The machine consists of 4 electronically/hydraulically operated dies: a pressure die, a wiper die, a bend die and a clamp die. In order to make it possible to use with a thermoplastic composite pipe either extra tooling for heating the pipe directly must be added, or the tooling must be adapted to be heated.
Methods of conveying heat into the pipe were trialled but the heated dies method was chosen as the simpler and better controlled solution. Dies were modified for faster heating with heat flow simulations done to achieve a consistent temperature all around the dies. Dies were heated using a range of cartridge heaters.
In order to understand the key variables, their interactions and how they affect the process, a factorial design of experiments was proposed to investigate this. A list of variables was drawn proven through statistical analysis; examples are: Bend Speed, Bend Die Temp, and Pressure Die Temp.
The sample batch of pipes used for the experiments were all manufactured with the same process, and were bent using the same equipment.
Over 34 bends, all at 90°, the average spring back was by 3.31°, or 3.68%, with a standard deviation of 2.05° or 2.28%. Understanding the spring back characteristics when bending the pipe go towards achieving the correct geometry. The composite pipe can be bent to replicate current metallic geometries and has been checked against existing inspection fixtures for Rolls-Royce engine pipes. The spring back is also modelled in Abaqus software to prove the set-up of the machine and actual geometry of the bent pipe.
3.5 End-fitting joining
The final forming process is the joining of metallic end fittings to the pipe. Since adhesives are much less effective when used with thermoplastics than with thermosets, and have limited resistance to temperature and fire, an alternative joining method had to be developed.
Provisional trials have taken place as a proof of concept for the design which was proposed at the CDR. The fitting is mechanically interfaced with the pipe and the interference fitting takes place at elevated temperature in a swage process. The interlocked shape provides the ability to withstand the axial loads applied to the connection and the swage ring prevents any polymer creep through service life.
The process requires no welding or chemical adhesion which facilitates more simplified NDT processes than existing stainless steel end fitting joining.
Mechanical testing of this joint was performed (vibration, pressure & temperature tests, axial load tests) which proved successful at ambient temperatures. Unfortunately the joint leaked at elevated temperatures therefore an alternative design will have to be developed beyond the timescale of the project.
3.6 End-Fitting Design
Pipe end-fittings provide a method of joining a pipe assembly to other pipe assemblies or other pieces of equipment on an engine. End-fittings can take various forms (e.g. threaded connections, flanges, clamps) and are usually welded, brazed or swaged to metallic pipes. Owing to the higher temperature applications on aero engines the sealing mechanism is frequently metal to metal, requiring high compression loads. Separate seals (e.g. O-rings) have limitations in temperature and are susceptible to damage and therefore not commonly used in pipe systems. Current metallic end-fittings have historically been designed for ease of manufacture and therefore tend to be heavy.
The project examined a number of alternative end fitting concepts in different materials and with alternative sealing and connection mechanisms. Owing to engine operating temperatures, the conclusion was that metal to metal seals, and therefore predominately metallic, end-fittings were the most suitable.
Sigma has developed and owns the design rights for three bespoke end fittings for the composite pipe :
a. Lightweight End-Fitting – based on a current design already in service but optimised for minimum weight (about 30% lighter). They were also designed to be interchangeable with the current fittings on metallic pipe systems.
b. Hybrid End-Fitting – Similar to a. above but reducing weight still further by replacing certain metallic features with over-moulded thermoplastic. Sample components were made but were deemed to be too costly and less durable than the all-metal alternatives.
c. Band Clamp – an adaptation of a V-band clamp type of joint commonly used in aerospace. Although very light and compact, the manufacturing costs were prohibitive.
Final testing was performed on pipe assemblies using Lightweight End-Fittings.
3.5 Future Optimisation
The CfP states that the project requirement is to develop the product to TRL6 and to identify manufacturing process outlines for large scale production. Owing to the end-fitting joints leaking at higher temperatures and instability with the CNC Bending process, the project only achieved TRL5, as assessed by the Topic Manager.
Sigma have identified the following optimisation actions to improve the current designs and manufacturing processes to attain TRL6 and to become cost competitive with metallic pipe systems.
a. Braiding – minimise raw material usage and convert braiding from a batch process into a more continuous one.
b. Curing – The developed curing system is optimised for the pipe sizes required under the project however a significant optimisation can again be achieved by closely coupling it to the braiding process.
c. Pipe forming – The consistency, yield and speed of the CNC bending process must be improved using bespoke tooling and integrating it with the CNC program.
3.6 Process Stability
Key Process Parameters (KPP) are key set point levels or capabilities that must be met in order for the process or system to meet its operational goal. Through the development of the manufacturing capability of the project, Sigma has developed a critical control plan to understand the tolerance window for each KPP to ensure that pipes produced are within the specifications detailed on the drawings. The critical control plan documents the full manufacturing procedure and allowable process windows as part of the product quality control procedure. The development of the control plan and critical process window for each of the key processes in production has been a large element of the project. Without such a control on the complex manufacturing process, it would not have been possible to consistently produce pipes to the project TRL.
Topic Manager, Rolls-Royce, invested time and resource in supporting Sigma to develop controls and processes to mature the manufacturing capability of the product. Throughout the project the potential of the product was clear and the support from Rolls-Royce demonstrated their level of confidence in the technology and the abilities of the project team to produce a final product and process that meets both the call for proposal requirements and the Rolls-Royce detailed technical requirements.
4 Non Destructive Testing
The NDT technologies selected for the quality control of the products has to be related to the most likely defects in manufacturing. A careful assessment during manufacture was completed to assess the potential defects of both matrix approaches, thermoset and thermoplastic. The common major defects categorised by their potential to cause significant operational damage to the pipe system are:
- Porosity
- Matrix cracking
- Delamination
- Voids
Based upon these likely defects, a review and ranking of available NDT processes has been carried out. The results are documented in the project deliverable report, D3.1 Quality Control Procedures.
Computerised tomography is being used to analyse the prototype bent pipes. It has been found that this system provides an accurate method for the assessment of the inner deformation of the pipe, ensuring that the quality of the new products fulfil the standards defined for metallic pipes.
This process provides the necessary information for the qualification of the prototypes, but it cannot be considered for series production for cost reasons. It is envisaged that future developments in the use of ultrasonic Lamb-waves or laser shearography will be necessary for the inspection of thin walled composites in series production.
5. Conclusion
Based on the Topic Description of the Clean Sky Project “Non-metallic Pipes for Aero engine Dressings”, a new concept for pipes and end fittings has been developed. This concept can be categorised as a paradigm as it will replace the existing technology that has driven the design of aero engine dressings for more than half a century.
It was concluded at the project closure meeting attended by the project team and Rolls-Royce delegates that the project has come a long way and significant knowledge has been gained by all on the design and manufacturing of composite pipes.
The conclusion provided by Paul Broughton, COMPipe Design Review Chairperson within Rolls-Royce is that it is possible to make composite pipes that could be used for significant portions of the on engine piping system, with some of the major obstacles being overcome through the project, especially fire and pressure capability. The project has demonstrated that individually on a test rig the pipes can meet the objectives of this program, although the fire was not conducted with vibration and the remaining tests only fully conducted on straight pipes.
Due to the current manufacturing capability and control, a full suite of pipes is not currently available to offer to an engine development program. The control of the bending process in particular has been highlighted by the team as the area that needs more investment and work, along attachment of end fittings that leaked following thermal shock or cycling. The current analysis also shows that this technology will offer a similar weight to titanium which therefore makes the cost analysis a critical point for the exploitation of this technology. It is clear that the process developed can be scaled up to a production process and therefore there is a clear method of how this can be taken to production rate levels.
The project was awarded TRL5 status by the Topic Manager, on the basis where it can be shown that there is a manufacturing process that can produce each feature, the yield is currently low and there is still some work to do to get to a reliable production process. Moreover the key technical requirements have been demonstrated in rig tests to date to a maturity of TRL5.
It was concluded that there was good collaboration between all parties examining the various possible manufacturing techniques and also good investigation into the potential inspection methods. During all of the meetings the discussion has been open, which has resulted in a proactive approach to this project by all.
Ultimately the final COMPipe sample presented at the meeting demonstrated a weight reduction of 52% over an identical stainless steel pipe assembly. Whilst the cost to manufacture the product is more difficult to quantify directly due to the low manufacturing maturity it was proven to be comparable to the costs of Titanium parts with large reductions possible through increased capability growth and increased production volumes.
For future exploitation of the product and to reach greater TRLs the following elements need to be addressed to support continued exploitation.
1. Fire test under vibration.
2. Fire test at a bend, again with vibration
3. Pressure and Thermal testing of bent pipes.
4. Thermal shock and cycling on improved end fitting.
This activities have been recognised by the project and recorded as actions for exploitation beyond the Clean Sky COMPipe project.
Potential Impact:
During the period of the COMPipe project the market circumstances have changed significantly, with a number of key product introduction windows shifting to the right as a consequence of the post financial crisis market place. It is evident that the next major fleet technology insertion opportunities now need a step-change in performance capability more than before. This makes this Clean Sky COMPipe project all the more relevant and important to the continued and relentless pursuit of better performance and industry competitiveness.
On average, 12% of aeronautic sector revenues, representing almost €7 billion per year for civil aeronautics alone, are reinvested in Research and Development (R&D) and support around 20% of aerospace jobs. The European aeronautics sector is a vital contributor to society and economy. The industry accounts for approximately 3% of EU workforce, generates roughly €220 billion of the European GDP per year, and contributes positively to the EU’s trade balance with over 60% of its products exported. Every Euro invested in aeronautics R&D creates an equivalent additional value in the economy every year thereafter.
-The aeronautics sector is a key contributor to the European economy and the EU’s balance of trade
-Continued growth in demand for air travel raises new environmental and socio-economic challenges
- Research and innovation is core to EU competitiveness and drives sustainable value creation
Aviation is a vital enabler of our economy and society. The aviation market continues to grow significantly and the associated societal challenges persist. Air travel will remain essential to ensure EU’s economic growth, mobility, integration and cohesion. Air traffic is forecast to grow by 4% to 5% per year in the next decade leading to a near doubling of traffic by 2020. This poses environmental, societal and economic challenges that can only be tackled through an intense and sustained cooperation between public authorities, industry, research organisations and academia.
The socio-economic impact from the COMPipe is significant. For the COMPipe product, the implications of the technology are in line with governmental and airline policies for the reduction of carbon emissions from air travel. The technology delivers this in two ways: by reducing the overall weight of an aero engine, this in turn reduces the weight of the supporting structure, thereby reducing the weight of fuel required; secondly by using a thermoplastic composite as an alternative to thermoset composite, traditional used and found all over a modern aircraft, the opportunity for end of life recycling is increased.
Further to this, the manufacturing technology has expanded to include other applications including torque shaft, torsion bars, rotor containment sleeves, space frames and other bespoke applications, often replacing the slow manufacturing option of hand laid composites, with a more automated, repeatable, fast option of thermoplastic braiding. The viability of the technology as an alternative is huge, with the capability for one machine to produce around 1 mile of tubing per day. This would take composites into a far higher volume than previously possible. Hand laid composites can produce perhaps thousands of parts a year, automated layup methods such as filament winding increase this to tens of thousands per year. Braiding has the opportunity to increase this to hundreds of thousands of parts a year, opening up a part of the market that previously has been unavailable, in the mass automotive manufacturing business. This is an area in which the UK is a major manufacturer with several large manufacturers based here.
Sigma and TWI have been disseminating the technology and its advantages through a variety of shows and presentations, alongside press releases that have created a significant amount of interest in COMPipe and its technology derivatives.
List of Websites:
www.compipe.eu
Project Coordinator
Michael Andreae
Sigma Precision Components
+44 (0)1455 445500
