This paper presents an overview of NDT at TWI and European-sponsored NDT research projects being undertaken at TWI. These include NDT projects recently completed, currently underway or about to commence. The paper also presents the total value of these projects and routes to exploitation of the results.

1. Introduction to TWI’s NDT Facilities

TWI is one of Europe’s largest independent research and technology organisations. Based at Abington near Cambridge since 1946, TWI provides those in industry with technical support in welding, NDT and associated technologies. TWI has over 430 skilled staff and a turnover of £27 million. During the last decade, TWI has participated in over 40 European projects including Framework V, CSG and CRAFT projects. TWI employs a total of 35 qualifed NDT specialists and inspection engineers with considerable academic, research and industrial experiences.

NDT at TWI is spread across four business units as follows:


• The NDT Technology Section. Mainly concerned with research, development and consultancy.
• Plant Integrity Ltd. A wholly owned subsidiary of TWI mainly concerned with sales and service provision using low- frequency ultrasonic systems (Trademark Teletest) for the pipeline industry.
• NDT Training and Examination (includes TWI South East Asia, Malaysia). Provides training and certifcation of NDT personnel worldwide.
• The Test House Ltd. Provides routine NDT services, for example flm-based radiography, manual ultrasonics, penetrants, magnetic particle inspection etc.

The combined turnover of these business units has been steadily increasing at an average of 30% per annum since 1996 and was £3.4 million in the year 2001.

TWI has the following NDT research equipment and facilities:


• Eight laboratories dedicated to NDT research.
• Latest NDT equipment and systems including phased array systems, P-scan, low-frequency ultrasonic (Teletest) systems, thermography systems, ACPD and ACFM systems, eddy current systems and real-time and conventional X-ray equipment amongst others.
• The largest dedicated underwater NDT test and training facility in Europe (TWI, Middlesborough).

The total investment in NDT research equipment and facilities is estimated to be greater than £5 million. In addition, TWI will move into its new £22 million facility at Granta Park in South Cambridge.

2. European-sponsored NDT Research at TWI

Table1: European sponsored NDT research projects at TWI

Table 1 shows a summary of the European-sponsored NDT projects recently completed, currently underway or about to commence. The value of these European NDT research projects is more than £15 million, and £2.4 million of project work is being carried out at TWI.

The European NDT projects shown in Table 1 involve a total of 77 European companies and research providers including:


• 31 SMEs (Small-to-Medium Enterprises) in the feld of NDT, automation and robotics,
• 25 large enterprises,
• 11 research institutes, mostly in NDT-related felds and
• 10 universities in NDT or related felds.

A significant number of the above organisations are involved in more than one NDT research project with TWI. The great potential for technology transfer and dissemination is obvious. Further collaborations are also facilitated by the relationships developed during the current projects.

3. Details of TWI’s European NDT research projects

A description of each project is given below.

3.1. The Development and Validation of Non-Destructive Testing Techniques for Butt Fusion Joints in Polyethylene (PE) Pipes.
Acronym: WINDEPP. Budget: £623,000

Background
This project has recently been completed and has developed a butt fusion welding machine with an integral ultrasonic non- destructive examination (NDE) module, which has the potential for providing complete confidence in the long-term quality of each weld produced.

Even with good site practice, it is impossible to totally eliminate all possible flaws in butt fusion welds in polyethylene (PE) pipes. Such flaws could include airborne dust and sand, water, grease, air pockets and cold welds produced by deviations from set welding parameters. Consequently, there is a need to determine the existence of any flaws in the weld through NDE and to establish whether such flaws are likely to reduce the service life of the pipe system. Such information would eliminate the need for destructive testing of the welds which, in turn, would reduce costs and allow the quality of the actual installed PE pipe system to be determined.

Objectives
The objectives of this project were:


• To optimise three ultrasonic NDE techniques (time-of-fight diffraction, tandem and creeping wave) for butt fusion welds in PE pipes.
• To determine the limits of detection of an NDE system that combined the above three techniques.
• To determine critical faw sizes and levels of particulate contamination in PE pipe butt fusion welds, ie sizes/levels above which the long-term integrity of the joint is reduced.
• To design and manufacture a prototype welding/NDE machine.

Fig 1: The prototype butt fusion welding machine manufactured as part of the project.

Fig 2: The prototype NDT module developed as part of this project showing a combined TOFD, tandem and creeping wave system.

Fig 3: The combined welding and NDT system.

Project results
Results have shown that a combination of three different ultrasonic NDE techniques is capable of detecting planar faws down to 1 mm signifcant dimension and sand contamination at levels down to at least 3% by area, in butt fusion welds in PE pipes of diameters up to 315 mm outside diameter (OD). Neither of these types of faw can be detected reliably by visual examination or manual testing of the external weld bead.

The work has also shown that ultrasonic NDE cannot detect either fine particulate contamination or cold welds produced by non-standard welding conditions. However, fortunately, both of these types of flaw can be either detected or eliminated by other techniques. Fine particulate contamination can be detected using a manual bend-back test on the removed external weld bead, and conditions that produce cold welds should not be possible using an automatic butt fusion welding machine, with process control and monitoring.

The project has also determined the minimum size of planar flaw and minimum levels of fine and coarse particulate contamination that cause premature failure of butt fusion welds in PE pipes of 125 mm and 315 mm OD, using a combination of specimen and whole pipe tensile creep rupture tests on welds containing known sizes/levels of different flaws. This information, together with the combined automatic butt fusion welding machine/ultrasonic NDE system developed in this project, followed by examination of the removed external bead, should allow all possible flaws that are likely to occur in butt fusion welds in PE pipes in the field to be either detected or eliminated.

Although the work in this project used one particular grade of PE, welded using one particular welding procedure, the advantage of a combined TOFD, tandem and creeping wave system is that it should be applicable to any PE pipe material made using any welding procedure since, unlike some other ultrasonic systems, it does not rely on a specific shape or size of weld bead. Figure 1 shows the prototype butt fusion welding machine manufactured as part of the project. Figure 2 shows the prototype NDT module developed as part of this project. The final NDT module was manufactured by sponsor SME, Ultrasonic Sciences Ltd. Figure 3 shows the integrated welding and NDT system. This formed the main deliverable of the project.

Project Consortium


Coordinator:

TWI Ltd (UK)

Industrial partners:

Simplast Spa, Pelerma (Italy), ATEV SAS di A Guiffrida, Sicily (Italy), Joseph Sauron SA, Bondoufe (France), SGD Engineering Services Ltd., Stoke-on-Trent (United Kingdom), Ultrasonic Sciences Ltd., Hampshire (United Kingdom), Advantica Technologies Ltd, Loughborough (United Kingdom), Gaz de France, St. Denis Plaine (France), George Fischer Plastics Ltd., Huntingdon (United Kingdom), Health & Safety Executive, Bootle (United Kingdom), Sade C.G.T.H. Secteur Sud, Le Plessis-Robinson (United Kingdom), Solvay Polyolefns EEEurope – Belgium SA, Brussels (Belgium) and Conzorzio Catania Richerche, Catania (Italy).

3.2. Development of a Robotic System for the Inspection of Aircraft Wings and Fuselage.
Acronym: ROBAIR. Budget: £1.23 million

Background
Regular and periodic non-destructive testing (NDT) is mandatory for civil airlines throughout the world. Most inspection is currently carried out manually, such that operator fatigue can lead to mistakes. Increasingly, airlines require a hard copy of inspection results to eliminate operator subjectivity. Furthermore, the requirement for 100% inspection of vital structural features is becoming more commonplace and the slow rate of manual inspection is therefore prohibitively expensive. The cost of NDT is increased further where X-ray inspection is used, since components to be inspected must be removed from the aircraft. The objective of this programme is to develop a robotic inspection system, which will walk over large areas of an aircraft structure, carrying out automatic data collection and interpretation to identify all structural faws, without the need to dismantle components.

As well as using conventional NDT sensors, the project involves development in the following technologies; acoustic camera, phased arrays, thermography, dry contact ultrasound and eddy currents. It is anticipated that the prototype system to be developed will be commercialised soon after project completion.

Objectives
The technical objective is to develop a robotic NDT inspection system capable of rapid automatic scanning of large and complex structures. This will be achieved by developing novel NDT techniques for aircraft inspection and deploying these techniques using a robotic system.

Fig 4a: view on aircraft fuselage	Fig 4b: Close up schematic of robotic system on wing
Fig 4: Schematic of Robotic Inspection System for Aircraft Inspection

Description of work
The multi-tasking robotic NDT system for aircraft inspection (see Figure 4) is being developed by the consortium in six work packages (WPs):


1. System specifcation and provision of defect samples of aircraft wings and fuselage:
   Samples and associated data have been supplied by British Aerospace, British Airways and RAF. The robot inspection system and faws to be detected have been technically specifed.
2. Development of NDT techniques:
   The developments in NDT techniques using phased arrays, eddy currents, thermography and dry contact ultrasonic are nearing completion. The project is investigating the following:
   • Acoustic camera: The use of an acoustic camera for large area inspection on aircraft is being investigated (see Figure 5). The acoustic camera generates real time high resolution images (C-scan) over an area. Images are typically presented at 30 frames/second. A current C-scan that currently takes several hours can now be done in minutes, producing high-resolution images. The basis for this technology is a novel two-dimensional imaging array that creates immediate, high-resolution images of subsurface faults. The output is a standard video image that can be video taped or fed directly into a PC for further processing. This camera will be further developed to enhance its pulse echo operation.
   • Eddy currents: Figure 6 shows eddy current images of aircraft fasteners used to locate and inspect fasteners automatically. Kontrol Technik (partner SME) is supplying the eddy current system to the project. The equipment will aim to use a multi- frequency technique to simultaneously inspect for surface and sub-surface faws. Some results are shown in Figure 6. Optimisation of this technique is being carried out.
   • Phased array inspection: Figure 7 shows manual phased array inspection of fasteners. A phased array probe capable of carrying out a scan to detect fatigue crack at the root of the fastener hole has been designed. This carries out an angle probe scan covering the necessary circumferential portion of the fastener hole from a single location. The development of wedges made from hydrophilic material developed by Sonatest is being attempted by TWI to enable dry coupling of the phased array probe. Technical University of Sofa is carrying out the ultrasonic modelling work for this task.
   • Thermography: The use of thermography is being investigated by TWI with SME Horton Levi to detect water ingress in composite aircraft structures. The technique is expected to be applied onboard the robot.
   • Dry contact ultrasonic inspection: Sonatest are designing a new dry contact wheel probe to fit on the scanner.
3. Development of NDT sensors and systems for deployment by the robotic system:
   These are being developed specifcally for deployment by the scanner module (schematic shown in Figure 4). The instrumentation for the improved NDT techniques (for example phased array probes, eddy current instrumentation, thermographic camera, ultrasonic dry contact wheel probes and acoustic camera) will be further developed to miniaturise NDT sensors, eliminate/reduce wiring and reduce sensor weight. The NDT sensors will be further refned for specifc interface and deployment by a NDT scanner module.
4. Development of a NDT scanner module (schematic in Figure 4):
   The scanner module is at its design stage. This will be a mechanical scanner carrying NDT sensors and will be attached to the mobile climbing vehicle shown in Figure 4(b). The scanner will be able to perform inspection on a limited number of structural geometries with a variety of scanning routines. It will use sensor-based control to maintain the NDT sensor at desired surface contact forces or stand-offs and allow real-time path modifcation to follow unknown surface contours. Appropriate control algorithms will be developed for the scanner. The scanner will be developed by SME Zenon and will locate the fasteners using information from the eddy current scans.
5. Development of a mobile climbing vehicle:
   The mobile climbing vehicle (to be developed by South Bank University) is at its design stage. This vehicle will be designed to provide mobility to the NDT scanner module. The vehicle will be able to move on the wings and fuselage of aircrafts and will deploy the specially developed NDT scanner module on to the inspection surface with speed and accuracy. The vehicle will be controlled by an operator through visual feedback using an on board video facility. A new feature of the vehicle will be a suction wheel, which will enable it to move much faster over the surface than some of the ‘walking’ designs envisaged at the outset. The vehicle is expected to have a payload of 18 kg.
6. System integration and testing:
   This work package will begin at the end of year 2002. The NDT sensors will be integrated with the scanner module. The scanner module will be integrated with the vehicle. Laboratory testing will be carried out followed by feld tests at TWI and RAF.

Fig 5: Acoustic camera to be utilised for inspection of composites on aircraft. Courtesy of NDT Consultants Ltd and Imperium Inc.

Fig 7: Fastener inspection using phased arrays at TWI. Wing and fuselage samples provided by RAF.

Fig 6: Eddy current C-scan images of fasteners on an aircraft wing used to identify and then inspect rivets. Courtesy of Kontrol Technik.

Deliverables
This project will deliver a prototype NDT robotic system for aircraft inspection that will be able to inspect large areas of aircraft wing and fuselage rapidly and without the subjectivity of an operator.

Project Consortium


Coordinator:	Sonatest plc (UK)
Industrial partners:	NDT Consultants Ltd (UK), Zenon Ltd (Greece), Kontrol Technik (Germany) and Horton Levi (UK).
Research partners:	TWI Ltd (UK), Frontier Systems (Greece), South Bank University (UK) and Technical University of Sofa (Bulgaria).
End User Panel:	RAF (UK), British Airways (UK) and British Aerospace (UK).

3.3. Development of Novel Non-Destructive Testing Techniques and Integrated In-line Process Monitoring for Robotic and Flexible Friction Stir Welding Systems.
Acronym: QUALISTIR. Budget: £1.24 million

Background
A new ‘environmentally friendly’ welding technology invented by TWI and called Friction Stir Welding (FSW) has recently emerged as a very important process for welding of high-strength aluminium alloys (previously unweldable and important to the aerospace and automotive industry). FSW is now also beginning to make an impact in other materials.

Common weld flaws in fusion or other welds, such as porosity, lack of penetration or fusion can be detected by conventional NDT methods. As for all welding processes, the FSW process produces specific flaw types of its own. For example, at the root of an improperly welded friction stir weld, a so-called ‘kissing bond’ can be created. That is, at the weld root a very short length of the weld interface, as small as 30 to 50 micrometers, may be in intimate contact but without true metallurgical bonding. Even this small flaw can drastically reduce mechanical properties. Currently no commercial NDT instrumentation exists that has been proven to detect kissing bonds. Current methods of inspection thus include:

1. stressing the component to ‘open up’ the kissing bond and then detecting it with common NDT methods such as standard ultrasonics, eddy currents or radiographic techniques; and
2. batch destructive testing of components. Both these inspection processes are destructive and cannot be integrated into a factory production line.

The following technical limitations are therefore hampering an even more widespread use of FSW:


• Lack of proven non-destructive testing (NDT) and in-process monitoring techniques, and
• Lack of manufacturing fexibility. Currently only fat panels can be welded in industrial environments. This is because of the high forces required for FSW, which are diffcult to apply using currently available industrial welding robots.

Objectives
The technical objectives of the above project are to develop:

• NDT techniques based on phased array ultrasonics for the detection and characterisation of faws in FSWs; the most important of these faws being ‘kissing’ bonds.
• New in-process monitoring systems to improve joint quality by reducing manufacturing faws.
• Flexible robotic systems for the fabrication of complex shaped welds.

Description of work
A novel robotic and fexible FSW system, integrated with NDT and in-process monitoring, suitable for welding complex 3D geometries will be developed (by the consortium) in six work package (WPs):


• System and Design specifcation with participation from End Users:
  The consortium has produced a full specifcation of the phased array sensor and system used to detect faws in friction stir welds, the in-process monitoring system, the customised robot and the customised FSW machine. This task was performed with considerable participation from End Users such as Airbus and research organisations such as TWI and GKSS.
• Development of Phased Array NDT techniques and System:
  Initial work has commenced on the design of novel phased array sensors and systems for the detection of ‘kissing bonds’ in friction stir welds. This task will deliver NDT techniques and a NDT system based on novel phased array design. This system is likely to be able to detect all faws associated with friction stir welding and will be designed to be easily interfaced with the robot and the FSW machine. All common welding methods have design codes with associated NDT requirements. As FSW is a new method these codes do not exist presently. This project aims to improve the NDT so that structurally signifcant faws can be detected and qualifed.
• Development of In-process monitoring techniques and system:
  Initial work has commenced on the design of in-process monitoring sensors and systems. This task will deliver in-process monitoring techniques and a system based on sensors, which measure key weld parameters. The system is likely to be able to monitor the friction stir welding process by monitoring key weld parameters and will be designed to be easily interfaced with the robot and the FSW machine.
• Customisation of Robot to carry Phased Array NDT and In-process Monitoring Systems (Modules):
  This task plans to deliver a customised standard Tricept 805 robot (manufactured by Neos, a sponsor SME in this project) for friction stir welding of 3D industrial components and structural elements (Figures 8, 9 and 10) and will provide the Robot with appropriate interfaces for NDT and in-process monitoring.
• Customisation of a FSW machine to carry in-process monitoring and Phased Array NDT modules:
  This task plans to deliver a customised standard FSW machine for friction stir welding of 2D industrial components and structural elements and provide the machine with appropriate interfaces for NDT and in-process monitoring. The machine will be situated at TWI.
• Factory Trials with End User Participation:
  This task plans to deliver:
  1. a complete robot for FSW with in-process monitoring and NDT.
  2. a complete FSW machine with NDT.
  The customised robot and FSW machine will be tested for the fabrication of aerospace and automotive friction stir welded components.

Fig 8: A general assembly drawing of the multi-tasking FSW robot system (showing the NDT and in-process monitoring module) Key: Tricept 805 FSW robot (manufactured by Neos Robotics). Aerospace component being friction stir welded. In-process monitoring module. This will contain sensors, which convert process data (e.g. temperature, vibration, rotation speed, forces, tool heel plunge depth, pin dimensions, workpiece thickness and dimensions) into electrical outputs. Schematic of scanning mechanism for the NDT sensors (phased array probes). Phased array probes are shown being scanned. The couplant feed and retrieval mechanisms and various attachments are not shown. Electronic systems are not shown. The NDT sensors will be part of the NDT module. Friction stir welding tool being applied by the robot through the in-process monitoring module.

Fig 9: Tricept 805 Friction Stir Welding robot to be utilised for welding and NDT. Courtesy of GKSS

Fig 10: The robot ‘end effector ’ to be integrated with the phased array sensors and system.

Research sub-contractors, who are leaders in NDT, in-process control and monitoring and friction stir welding will execute the technical research tasks. The SMEs consisting of NDT companies, probe/sensor manufacturers, process automation and control, and manufacturers of friction stir machines and robots will develop NDT and monitoring systems and integrate them into the friction stir robots and machines.

Fig 11: Phased array testing of friction stir welds at TWI.

Fig 12: Lack of root penetration flaw in friction stir weld. Courtesy of R/D Tech.

Deliverables
The project plans to deliver a customised robot and a customised FSW machine, each will incorporate Phased Array NDT and in-process monitoring modules for quality control.

Figure 8 shows a schematic of the final system. Figure 11 shows phased array testing of friction stir welds (FSWs) at TWI. Figure 12 shows NDT of friction stir welds by RD-Tech.

Project Consortium


Lead and coordinator:	R/D Tech (France)
Industrial partners:	Vermon (France), ISOTEST (Italy) and Neos Robotics (Sweden).
Research performers:	TWI (UK), GKSS (Germany) and Technical University of Sofa (Bulgaria).
End User Panel:	Airbus and other aerospace companies. These companies represent the frst potential customers of the SMEs in the project.

3.4 Train-Mounted Sensors and Systems for the Inspection of Rails.
Acronym: RAIL-INSPECT. Budget: £1 million

Background
The last three decades have seen continuous increases in train traffic, train speeds and tonnage carried on European rail networks. These have put an increasing amount of strain on the rail tracks. In the last few years an increase in accidents due to broken or cracked rails has occurred in Europe. This has resulted in loss of life, severe delays in services and loss of revenue for the train operators. The public confdence in this economical and environmentally friendly mode of transport has also reduced as a result of the above.

Objectives
The major technical objectives of this project are to:


• Further develop current automated NDT techniques and eliminate subjective manual NDT methods.
• Improve ‘probability of defect detection’.
• Provide a mechanism to inspect the entire volume of rail head and web.
• Carry out on-line NDT from on-board a test train at signifcant speeds.

Description of work
The project commenced on the 1st of January 2002. The multi- NDT system (to be carried by a test train) for rail inspection is being developed in seven work packages (WPs):

• System specifcation and provision of defect samples of rails:
  The consortium has recently developed specifcation of the NDT system and defects. This includes specifcation of:
  1. likely faws occurring;
  2. identifcation of candidate NDT techniques;
  3. NDT sensors;
  4. design of mechanical probe holder including size, dexterity and load carrying capability;
  5. design of location system to locate the faw within the rail cross- section; and
  6. defect visualisation and presentation.
  Several EU railway companies have agreed to provide the project with lengths of rail containing some faws. Some of these samples are already at TWI. See Figures 13 and 14 for faw types in rail.

  Fig 13: Flaw types in rail head that can cause broken rails Key: (A) Head surface flaws. These usually prevent ultrasonic techniques in inspecting areas under these flaws. (B) Squat flaws running below and parallel to the rail surface. (C) Roughness of rail head due to missing material (caused by breaking/slipping wheels) prevents ultrasonic inspection. (D) Vertical longitudinal split flaws. (E) Star cracks at bolt holes. (F) Diagonal crack in web of rail. (G) Horizontal flaws. (H) Flaws in thermit welds eg. lack of fusion and other weld cracks. (I) The gauge corner crack. This is difficult to detect and can become a cause of rail breakage that result in derailments. (J) Bolt holes (although not flaws) can provide initiation points for star cracks (see e)..

  (a) Shows a star crack starting from a bolt hole	(b) A broken rail caused by a gauge corner crack (see I in Figure 13)	(c) A major rail head crack.
  Fig 14: Pictures of rail faws Key:

  Fig 15: A test train. The intended position of the scanner module is shown in green.

• Development of NDT techniques:
  Improved NDT techniques and instrumentation (phased arrays, dry contact ultrasonics, EMATs, real time X-rays and low frequency eddy currents) are being investigated and developed for rail inspection. Rail inspection is highly complex for a number of technical and logistical problems. One example of these technical problems, which this project will address, is rail damage and wear. Wear changes the profle of the rail head diverting the ultrasonic beams from the designed directions. Phased array ultrasonic beams can be realigned to compensate for these profle changes. By this process the designed inspection quality can be maintained
• Combination of NDT techniques and development of data fusion methods:
  Sophisticated data fusion techniques will be developed based on results from each of the NDT techniques. This is likely to result in:
  1. coverage of the majority of cross-section of the rail;
  2. an increase in the probability of detection of a particular faw by using a combination of several NDT techniques;
  3. a decrease in the number of false positive indications; and
  4. removing the need for manual follow-up, which is currently needed to locate and sentence the faws detected.
• Development of man-machine interface and defect visualisation:
  This will consist of a high level man-machine interface for real-time recording and display of defect severity and location providing verifable records. This will constitute a high level faw detector.
• Development of NDT sensors and systems:
  The NDT sensors and systems will be developed specifcally for deployment by the test vehicle/train. The instrumentation for the improved NDT techniques will be further developed to ruggedise NDT sensors, reduce wiring and develop instrumentation to work in harsh environments. The NDT sensors will be further refned for easy interface and deployment by the test vehicle/train.
• Modifcation/Customisation of a test vehicle/train for attachment of NDT sensors and systems:
  A test vehicle/train will be modifed to provide mobility to the NDT sensors. The test vehicle/train will move at signifcant speeds carrying out NDT inspection of the rail.
• System Integration and testing:
  The NDT sensors will be integrated on a laboratory test vehicle and laboratory trials will be carried out. The NDT sensors will then be integrated with the test vehicle/train. Field trials will be carried out on rails and will include travelling over deliberately induced realistic faws. The NDT system will provide on-line and recorded faw status. These trial results will be fed back into the design of the probes and equipment to provide a working prototype system, which is applicable for site rugged conditions.

Deliverables
The fnal result will be a feld prototype one-stop inspection system operating on-board test trains of National rail companies in Europe and throughout the world. This will establish automation of inspection tasks that have been performed previously either with limited automation or in most cases entirely manually. The prototype system will be designed to be easily integrated onto any European test train. Figure 15 shows a schematic the placement of the final system.

Project Consortium


Lead and coordinator:	Sonatest plc (UK)
Industrial partners:	Kontrol Technik (Germany), Computerised Information Technology (UK), Imasonic (France), Zenon (Greece) and Technitest (Spain).
Research performers:	TWI (UK) and Technical University of Sofa (Bulgaria).
End User Panel:	European rail companies have been approached to join the end user panel in due course.

3.5 Manufacturing and Modelling of Fabricated Structural Components.
Acronym: MMFSC. Budget: £5.84 million

Background
The European aerospace industry is one of the Community’s leading industrial strengths, competing successfully in world markets and ensuring the employment of some hundred thousand personnel across Member States. The proposed programme seeks to enable a step-change in the process of design and manufacture of aero- engine structures, leading to signifcant reductions in lead time, materials use, and cost.

Deliverable of Project
The programme will deliver:


• A robust methodology for optimal design of structural fabrications;
• Development of automated control and NDT;
• New, pertinent physical analysis methods;
• A data management protocol;
• Increased confdence in fabrication as a method of manufacture;
• Increased confdence in weld strengths;
• Greater material properties awareness; and
• European structural fabrication technology to compete world wide against single-piece castings.

TWI’s contribution
NDT is a small part of this project and the objectives of the work relevant to TWI’s NDT activities, is to improve weld quality and productivity through NDT. This will be achieved by the development of a real time weld monitoring system (partly based on an ultrasonic phased array system) that will enable welding parameters to be monitored and controlled automatically and continuously. The deliverables will be procedures and techniques for obtaining welded joints that are free from signifcant faws and geometrically accurate. This will enable aero-engine components to be manufactured more cost-effectively.

Project Consortium
The consortium partners are listed below:
Rolls-Royce plc, Ferroday Limited, Brunel University, Fundacion ROBOTIKER, The Queen’s University of Belfast, The University of Nottingham, University of Southampton, Werkzeugmaschinenlabor der RWTH – Aachen, Volvo Aero Corporation AB, Luleå University of Technology, Fundacion Tekniker, Motoren- und Turbinen-Union GmbH, Heriot-Watt University, Universidad de Cantabria, The Welding Institute (TWI Ltd), CLFA Groupment d’Etude et de Recherche pour les Applications Industriels des Laser de Puissance (GERAILP), Industria de Turbo Propulsores, S.A, Société Nationale d’Étude et de Construction de Moteurs d’Aviation and Defence Evaluation and Research Agency.

3.6. Production Line Integrated Sensor System for Porosity Quality Control of Magnesium Castings.
Acronym: MAGCAST. Budget: £1.24 million

Background
Magnesium (Mg) castings have tremendous potential in a number of industries, for example, aerospace, automotive and telecommunications, owing to their combination of strength and light weight. However, their use is being hindered by the diffculty of casting Mg components of consistent acceptable quality with respect to internal porosity.

Traditionally, hot chamber high-pressure casting has been used for producing Mg parts of complex shapes to very high tolerances. This process has always suffered from the problem of internal porosity in the casting. Two forms of porosity are common, shrinkage porosity and gas porosity. Porosity is a problem as it can lead to structural failure in safety critical parts (for example steering column housings in cars and aerospace components for example airframe structures, engine parts, gear boxes, and fuel hydraulic components etc).

The NDT methods currently used for Mg applications are visual and dye-penetrant inspection. These methods are only effective for the detection of surface flaws. Therefore, X-ray film radiography has been used for the detection of internal porosity in aluminium castings. However, this method has the following limitations:


• Labour- and time-intensive process thus extremely expensive and susceptible to subjective and differing evaluations among inspectors.
• Interpretation errors from operator fatigue resulting from viewing thousands of radiographs.
• Provides limited information in critical areas of complex shape, because of diffculties in correctly placing the flm and/or orientating the X-ray beam.
• Speed of inspection is too slow for high-volume continuous casting lines.
• High costs due to the extensive use of flm radiography. X-ray flms have to be developed, washed, fxed, dried and then viewed on an illuminated screen and manually interpreted involving high labour and material costs. The written procedure for the X-ray NDT inspection may be 200 pages long, and 1000 or more radiographic exposures may be required. In the aerospace industry, this level of inspection can add more than 20,000 Euros to the cost of a single cast component part.

Objectives
The technical objectives of the projects are the development of:


• A real-time and digital X-ray sensor system for the accurate identifcation and quantifcation of porosity;
• Software for operating the inspection system and computing QC data; and
• Real-time feedback from the inspection system back into the casting process to stop the casting process, thus stopping the production of defective castings.

Description of work
The real-time X-ray in-line inspection system for magnesium castings will be developed in fve work packages commencing March 2002:


• Design and manufacture of specimens:
  This work package will produce test specimens containing various known faws (concentrating on porosity), which are representative of the types of faws encountered when casting magnesium.
• Development of radiographic techniques:
  This work package will develop radiographic techniques and establish and validate the limits of detection of these techniques for the various types of castings and the various types of faws considered.
• Development of real time digital X-ray hardware:
  This will include the following:
  1. development of a stable X-ray generation hardware;
  2. development of ‘direct X-ray conversion’ digital detector;
  3. integration of X-ray generation and detection hardware to form the real time X-ray digital hardware; and
  4. customisation of a manipulator to handle the casting during exposure.
  This work package will provide the hardware to be used to take radiographic images of casts.
• Development of high-speed X-ray image reconstruction and automated faw recognition algorithms:
  This will include the following:
  1. a high-speed X-ray image reconstruction algorithm will be developed. This will combine the X-ray images and the positioning information from the manipulator controller to reconstruct a 3D representation of the casting part inspected; and
  2. an automated faw recognition algorithm will be developed. This will automatically process the reconstructed image to evaluate the extent and position of porosity in the casting, and sentence it as acceptable or rejectable based on manufacturers’ previously established criteria. This work package will integrate all the hardware and software components of the system into an integrated prototype. This prototype will be used to carry out laboratory and feld trials.
• Laboratory and feld trials:
  The elements of the system will be integrated and tested in the laboratory followed by feld trials at the casting installation of UKRC (SME manufacturer of magnesium castings).

The main innovations of the project will be:


• New and novel X-ray techniques for the detection of porosity and other faws in magnesium castings. This project will contribute towards a European standard for the inspection of magnesium castings.
• Development of novel X-ray digital detectors consisting of direct X-ray imaging tiles. These tiles (invented by partner SME, Simage in Finland) convert X-ray photons into electron pairs in the substrate layer and the ionising radiation is collected for image formation resulting in images of higher quality requiring shorter radiation exposure times. Present-day digital detectors convert X-rays to visible light photons and then detect the visible light (using a Silicon Intensifed Tube camera or a CCD device). This two-stage process results in lower resolution images.
• Stable mini-focus X-ray generators unaffected by continuous use or high duty cycles.
• Defect recognition software, which includes fast X-ray image reconstruction algorithms that can reconstruct 3D information about the component structure and identify porosity/faws automatically (automatic faw recognition algorithms), without operator assistance.

Deliverables
The project will deliver a fully tested real-time and digital X-ray prototype system for the in-line inspection of magnesium castings. It will provide an input for the real-time control of the casting process.

The Project Consortium


Lead and coordinator:	Computerised Information Technology Ltd (UK)
Industrial partners:	Balteau NDT (Belgium), UK Racing Castings Ltd (UK), Simage (Finland) and Spree Engineering Ltd (UK)
Research performers:	TWI (UK), CEA LETI (France) and Technical University of Sofa (Bulgaria).

3.7. Development of a Robotic System for the Inspection of Large Steel and Aluminium Plates in Industrial Plants.
Acronym: ROBOT INSPECTOR. Budget: 1.24 million

Background
Steel and aluminium plates are inspected extensively during manufacture, fabrication into welded structures and in-service when subject to damage from corrosion or fatigue. There is a need for automated inspection systems that can be applied throughout the product life cycle and which are versatile, portable and able to perform in hazardous environments with little human intervention.

Objective
The major technical objective of this project is to develop a robotic NDT inspection system for large plates, capable of detecting:

1. corrosion on both sides of plates;
2. faws in welded joints; and
3. faws under thick coatings.

Description of work
The development of the autonomous robot inspection tool for plates and welded plate joints will be conducted in seven work packages to commence in March 2002:


• System specifcation and provision of defect samples of plates and welded plate joints.
• Development of NDT techniques:
  Dry coupled ultrasonic techniques with a wheeled probe and compression wave and Lamb wave techniques will be developed. Phased array NDT techniques for ultrasonic testing of both steel and aluminium plates and welded joints will be investigated. Advanced sensors for eddy current testing of plates through coatings will be developed.
• Development of NDT sensors and systems:
  The NDT techniques will be developed further for deployment from a scanner module on the robot.
• Development of NDT scanner module:
  A scanner module that is able to mimic the manual scanning motions of NDT sensors will be developed for operation from the mobile robot vehicle. It will include instrumentation for collecting NDT data and displaying results.
• Development of robot vehicle:
  To operate autonomously via an umbilical.
• Development of robot positioning and guidance system based on laser guidance:
  The system will be able to track the robot across fat plate surfaces and along welded plate joints and around the annular rings of stage tanks.
• System integration and testing:
  All the components of the system will be integrated and subjected to laboratory and then feld trails.

Deliverable of project
A prototype robot scanner and vehicle will be developed for the remote inspection of large metallic plates as in:

1. tank foor plates and welds;
2. oil tanker ships; and
3. steel/aluminium plates at the manufacturing plants.
Figure 16 shows a schematic of the proposed robotic NDT system.

Project Consortium


Lead and coordinator:	Zenon Ltd (Greece)
Industrial partners:	Sonatest plc (UK), Isotest (Italy), Tecnitest (Spain), Spree Engineering and Testing (UK) and Estampaciones Mayo S.A.E. (Spain).
Research performers:	TWI (UK), Noemon Ltd (Greece) and Technical University of Sofa (Bulgaria).

3.8. Smart Structural Diagnostics using Piezo- Generated Elastics Waves.
Acronym: PIEZODIAGNOSTICS. Budget: £1.8 million

Background
The proposed three-year R&D project aims to develop a new structural diagnostics environment based on piezo-generated wave propagation with the following key features:


• A long-distance wave generation technique, based on advanced, low-frequency piezo-transducers,
• An advanced signal processing system able to identify location and intensity of multiple faws affecting elastic wave propagation.

The vision for the future is to fully integrate localised monitoring with remote assessment via tele-communications technology. One can envisage real-time decision-making on infrastructure integrity over the Internet. The expected exploitation time is medium term (five years).

Objectives
The objectives of the proposed project include:


• To develop a smart piezo-electric diagnostics technology with the potential of long-distance operation (wave propagation for distance of hundreds of meters) to overcome requirement for close proximity between wave generator, sensors and potential faws,
• To develop an associated data processing methodology enabling real-time monitoring, multi-damage detection and its precise identifcation (dispersed locations and intensities),
• To develop a prototype installation for remote monitoring,
• To evaluate the developed monitoring system, in laboratory and industrial conditions,

Fig 16: A schematic of the proposed robotic inspection system carrying out inspection inside an oil storage tank. The annular plate and the foor plate welds will be inspected by the robot

Fig 17: TWI patent income versus expenditure.

Fig 18: Teletest instrument.

Description of work
The project workplan is structured around six technical work packages and began in February 2002:


• Requirements for smart structural diagnostics to provide formulation of detailed concept and specifcation for Smart Monitoring System, defnitions of working environment, operational requirements and monitoring objectives.
• Piezo-sensors to develop new type of composite sensing transducers and its mounting techniques
• Piezo-wave-generators and piezodiagnostics (PD) technology for specialised piezo-electric actuators, and a prototype hardware of the PD monitoring system consisting of composite sensing transducers, amplifed piezo-wave-generators, network connections and local control unit.
• Data processing system and damage identifcation methodology for signal processing tailored to the PD technology (software of the PD system). Damage detection, location and identifcation based on a highly effcient solution strategy will be developed.
• Experimental testing of the developed sensing technology in laboratory conditions
• Field application to evaluate the performance in industrial conditions.

Deliverables


• New technology for continuous, long distance interrogation of structural condition. Working prototypes of individual and networked piezo-devices.
• New, associated methodology for faw identifcation.

The Project Consortium


Coordinator:	CEGELEC NDT (France)
Industrial partners:	TWI Ltd (UK), WS Atkins Consultants Ltd (UK), CEDRAT RECHERCHE SA (France), Centrum Diagnostyki Rurociagow i Aparatury SP Z O.O (Poland), International Center for Numerical Methods in Engineering (Spain), Intitute of Fundamental Technological Research, Polish Academy of Sciences (Poland), Ecole Centrale de Lyon and Alstom CERG (France).

3.9. Effective Application of TOFD Method for Weld Inspection at the Manufacturing Stage of Pressure Vessels.
Acronym: TOFDPROOF. Budget: 1.08 million

Background
TOFDPROOF project aims at producing a coherent package of EU agreed documents (procedures for applying TOFD with related acceptance criteria and recommendations for training and certifcation) based on a Round Robin test performed on welded specimens and validated through site trials.

The project is focused on all aspects allowing the effective application of the TOFD as a stand-alone method for the weld inspection during manufacture of pressure equipment. Technological, regulatory, human factors are considered.

The performance of TOFD will be compared with conventional NDT as defined by European standards for testing pressure vessels at the manufacturing stage. This evaluation will be carried out by means of a Round Robin test on welded specimens. All the results generated by this Round Robin exercise will be stored in a database set up on a website. Results will be in a suitable form for implementation in CEN standards and dissemination among NDT specialists. Specific tools will be developed in order to enable a quick and reliable comparison of the TOFD results with those obtained by conventional NDT. This comparison will normally be performed using Probability of Detection (POD) curves and statistical analysis tools. At the same time, optimised TOFD procedures and specific related acceptance criteria will be developed.

Fig 20: 24" (610 mm) diameter tool mounted on an insulated road crossing. The flaw detector unit is also visible. (Photo courtesy ARCO Alaska Inc).

Guidelines for training and certification will be written and distributed to the NDT society, the relevant standardisation CEN technical committees and the EU companies dealing with weld inspection. TOFD with the corresponding acceptance criteria will then be applied on site on welded components in order to demonstrate the technical efficiency and cost competitiveness compared with conventional NDT. The website will allow general feedback from any EU citizens on the proposals resulting from the project.

Fig 19: Transducer mounting tool for 4’ nominal bore (114 mm diameter) pipe.

Objectives
TOFDPROOF will allow the manufacturer of pressure vessel to use TOFD as a stand-alone NDT technique for weld inspection, through the following objectives:

1. To compare TOFD performance with conventional NDT as applied according to the European standards defned by CEN TC 121 ‘Welding’;
2. To defne the feld of application of TOFD, highlighting weaknesses and strengths, if applicable, the need to use TOFD in combination with other NDT techniques will be explained;
3. To optimise the methodology of application in order to ensure reproducible inspections with different pieces of equipment and inspectors;
4. To verify how TOFD allows for detection of transverse faws;
5. To develop acceptance criteria; and
6. To defne a framework for operators qualifcation and certifcation.

Recommendations for the training and certification of the operator will be provided and the assessment of the influence of the objectiveness of the inspector interpreting the results will be determined.

The project is expected to commence in March 2002.

Project Consortium


Coordinator:	Institut de Soudure
Industrial partners:	TWI Ltd, IS Service, Sonovation, Mitsui Babcock Technology Centre, Staatliche Materialprefungsanstalt (MPA) Stuttgart), Tecnatom SA, VTT, Instituto de Soldadura e Qualidade and TUV Suddeutschland Bau und Betrieb GmbH.

4. Exploitation of Technologies Developed through European-sponsored Research

4.1. Exploitation of TWI’s intellectual property
TWI exploits its intellectual property daily through the selling of know-how and expertise to member companies world-wide and participation in various funded research projects. TWI has a long history of developing new ideas and inventions, for example the CO2 laser, gas assisted laser cutting and CTOD (Crack Tip Opening Displacement Concept) were all invented at TWI and subsequently transferred to industry.

Prior to 1990, TWI did not actively exploit its generated and owned intellectual property, in fact TWI held patents that were costing more to maintain than they generated through licence income, refer to Figure 17. As a consequence, TWI began to develop an exploitation strategy by first learning how to protect, defend and exploit ideas through talking to similar organisations in the market, patent attorneys, licensing executives, and attending conference and work-shops on IP related issues. This is an on-going activity in TWI to ensure that the policies and strategies put forward for adoption align to our mission statement, to the market and to best practice within industry.

Due to TWI’s size and resources available, only commercially exploitable intellectual property is protected through patents and/or trademarks and a stage / gate procedure has been formalised to manage inventions. As a consequence of this, licence income has exceeded expenditure since 1996. TWI is currently progressing 28 inventions through the various stages of the stage / gate process.

4.2 Exploitation of intellectual property generated in European-sponsored NDT research
The exploitation and dissemination of intellectual property generated in European-sponsored research is critical to the success of the projects run. TWI has experience of exploiting directly such intellectual property and of assisting organisations in the development of exploitation strategies for generated intellectual property under their control (as contractual conditions differ from programme to programme relating to the ownership and exploitation rights in place for generated intellectual property). This has taken many forms from a presentation of key considerations to the detailed plan for exploitation (including their impact on dissemination activities if patent protection is progressed) and assistance in determining licensing strategies.

Some examples of our activities in the exploitation and dissemination of European-sponsored projects are now presented.

Example 1. WINDEPP – The Development and Validation of Non-Destructive Testing Techniques for Butt Fusion Joints in Polyethylene (PE) Pipes
TWI co-ordinated the development of the partners’ exploitation plan and publication strategy adopted as the generated intellectual property was protected. Customers of the WINDEPP technology were involved in the project from the outset and provided valuable technical and commercial input. A prototype machine incorporating the NDT module was developed and trialed within the project and a close-to-market machine is currently being produced. The technology has been widely disseminated including its presentation at a variety of conferences.

Example 2. Development of Low Frequency Ultrasonic Guided Wave Technology for the Rapid Survey of Pipes
This technology was initially developed through a THERMIE European project led and managed by TWI. The three-year project was completed in 1997. Initial exploitation included commercial prototyping and extensive feld trials. This resulted in the Teletest® low-frequency guided wave technique and system currently marketed by Plant Integrity Ltd (a wholly-owned subsidiary of TWI). The principal advantage of Teletest® (Figures 18, 19 and 20) is that pipelines of signifcant lengths, (30 m or more in each direction), may be examined from a single test point. The benefts are:

• The ability to examine very long lengths of pipe very quickly from a limited number of test points.
• Reduction in the costs of gaining access to the pipes for inspection by avoidance of removal and reinstatement of insulation or coatings (where present), except for the area on which the transducers are mounted.
• The ability to inspect inaccessible areas, such as under clamps and sleeved or buried pipes.
• The whole pipe wall is tested, thereby achieving a 100% examination.

Since the launch of this product in 1997; Plant Integrity Ltd has dramatically increased its turnover per year between the years 1997-2001. Plant Integrity’s success has led other European companies into this market. Plant Integrity and other European suppliers of this technology are making significant inroads into the USA and Japanese markets as well as in other world markets, making this a significant European success story.

Example 3. LINFRIC™ Low cost linear friction welding machine
The LINFRIC machine was developed in a CRAFT project, funded by the European Commission under the Framework IV programme, which was successfully completed in September 2001. The project aimed to develop a prototype hydraulic linear friction welding machine at a lower cost than those currently available on the market in order to make the technology more accessible to potential users.

TWI assisted the project partnership in the development of an exploitation strategy, which included obtaining the name ‘LINFRIC’ as a trademark which will be licensed by TWI to the project partnership. The prototype will reside for a period of time at TWI as the partnership agreed that it was the ideal location to promote the machine due to the large number of companies that visit TWI where they are introduced to new and novel developments in many fields.

The LINFRIC machine itself is being offered to the market by Blacks Equipment Ltd, Doncaster (United Kingdom) and Klaus Raiser GmbH, Eberdingen (Germany). Both were project partners in the above CRAFT project.

5. Conclusion

The above-mentioned European funded NDT research projects at TWI have the following common themes:


• Further development of novel NDT techniques.
• Automation of the application of these techniques.
• Elimination of labour intensive and monotonous inspection tasks.
• Elimination of the need for an operator to work in hazardous environments.
• Elimination of operator stress and error caused by the need for great attention to detail and NDT process variability.
• Elimination of subjective data interpretation.

The achievement of the goals of these European projects comprising of some 77 European companies will in a small way contribute to Europe’s competitiveness in NDT against USA and Japan. The successful collaborations in the above projects have facilitated the application of new proposals to the European Commission. The NDT Section at TWI is currently applying for the following 3 CRAFT projects:


• Development of NDT techniques for inspection under coatings (budget 2 million Euros);
• Development of NDT techniques for inspection of composites in aircraft structures (budget 2 million Euros); and
• Development of NDT techniques for the inspection of printed circuit boards (budget 2 million Euros).

It is important to note that the EC CRAFT projects are targeted towards SMEs and any Intellectual Property developed through these projects remains the property of the SMEs.

Contacts

The following persons can be contacted for further information on the above projects or for launching further EC projects:

For technical enquiries contact: Dr Aamir Khalid, Manager, NDT Technology Section, Structural Integrity, TWI, Granta Park, Great Abington, Cambridge, CB1 6AL UK. Tel: 00 44 (0) 1223 891162; Fax: 00 44 (0) 1223 890689; E-mail: aamir.khalid@twi.co.uk

For enquiries regarding commercialisation and exploitation, contact:

Ms Laura Barrett, Intellectual property, TWI, Granta Park, Great Abington, Cambridge, CB1 6AL UK. Tel: 00 44 (0) 1223 891162; Fax: 00 44 (0) 1223 890689; Email: laura.barrett@twi.co.uk

Dr A Khalid is the manager of the NDT Technology Section at TWI Ltd.He has a BSc(Hons)from University College London,MSc in NDT from Brunel University,PhD in NDT from Cranfield University and an MBA from South Bank University.

© NDT.net - info@ndt.net

|Top|