Community Research and Development Information Service - CORDIS

Final Report Summary - BONDTEST (Development of an ultrasonic testing technique to characterise diffusion bonds in layered structures)

Executive Summary:
As component design evolves in complexity, the associated manufacturing technologies such as joining and welding have to respond, given that many new leading-edge manufacturing solutions require joining of dissimilar materials and thin coated metal layers. The most suitable method to address these needs is the so called diffusion bonding, because unlike the more usual welding processes, it involves atomic diffusion at the joint interface, without any significant deformation of the components, not requiring melting of the materials being joined. As a consequence, a clear industrial need has arisen for a device that would provide on-line inspection of diffusion bonded components to ensure that the desired level of bond integrity is obtained.
Since mechanically testing selective components doesn’t ensure the whole production lot from being bonded to a satisfactory integrity level, a successful development of a Non-Destructive Testing (NDT) technique can provide significant time and cost reduction. This is the main goal of the BondTest project: to deliver a high sensitivity Phased Array Ultrasonic Testing (PAUT) system capable of operating with high near-surface resolution with Full Matrix Capture (FMC) techniques and Automatic Defect Recognition (ADR) software. The scientific objective was to model and compute the interaction of the ultrasonic beam with the joint interface in diffusion bonded joints of dissimilar metals. The project has formulated the technical and functional specifications - materials, configuration and component thickness - for the development of the PAUT technique required for optimal defect detection in the designated diffusion bonded test pieces. The main benefit of the PAUT technique is the electronic focusing abilities, which allow for more sound energy at the investigation point, resulting in better signal-to-noise ratio and better detection capabilities, by lowering the noise and enhancing the signal features.
In order to achieve it, the BondTest project has developed a novel ultrasonic NDT system and technique that offers speed, safety and innovation for the testing of diffusion bonds and that can be commercialised and made available to the industry. The BondTest system, and its component parts and software, including the above-mentioned ‘automated defect recognition’ (ADR) algorithm package, will be available to product manufacturers who use diffusion bonding welding technology, such as inspection/NDT companies, end-user plant operators and their mechanical maintenance companies.
The BondTest system is capable of full volume automated ultrasonic scanning of the critical diffusion bonding region in various plate like structures. It uses a curved physical focusing in the passive direction, which enables focusing in both the active and passive directions, maximising the spatial resolution.
The prototype has been assessed through laboratory validation of the complete system, software and techniques on several representative diffusion bonded samples. These trials focused on the validation of the equipment performance - in terms of defect detection and assessment/evaluation capability, diagnostic area coverage and the quality of the component and defect displayed image resulting from the PAUT system – while comparing it with the existing systems and techniques.
Beam steering can be used for mapping components at appropriate angles or depths, which greatly simplifies the inspection of components with complex geometries. Another benefit is the ability to sweep the beam without moving the probe. One caveat though, is that a fast inspection technique is required since the mounting, scanning and assessment needs to be performed during a short period of time to be cost effective.
The BondTest prototype brings the following benefits:
• Covers a large range of materials and geometries;
• Has sufficient memory to store inspection data;
• Allows the measurement results to be downloaded to a computer;
• Drives complex ultrasonic phased array probes;
• Has a rapid and integrated data analysis presenting a pass/fail answer.
The validation trials were highly successful, which means that the consortium has successfully designed, developed and manufactured a complete manual encoded/automated ultrasonic inspection prototype system for the inspection of diffusion bonded planar or slightly curved plates. This system also demonstrated improvements in terms of minimised setup time, higher inspection speed and coverage.

Project Context and Objectives:
Project context

Diffusion Bonding is an advanced solid state welding process by which two metals (which may be dissimilar) can be bonded together. Diffusion involves the migration of atoms across the joint, due to chemical composition concentration gradients. The two materials are pressed together at an elevated temperature (usually between 50% and 70% of the melting point) so that the process can be applied to finished machine parts. The pressure will relieve voids that may occur due to the different surface topographies and irregularities.

Diffusion Bonding is often used on sheet metal structures. Typical materials that are welded include titanium, beryllium, and zirconium. It is usually used on low volume work (batch production) pieces mainly for aerospace, nuclear fuel, and electronics industries. The process has been used most extensively in the aerospace industries for joining materials and shapes that otherwise could not be made — for example, multiple-finned channels and honeycomb construction.

Superplastic forming/diffusion bonding (SPF/DB) have been developed specifically within the aerospace industry. The process is used commercially for titanium and its alloys, this material being one that exhibits superplastic properties at elevated temperatures within defined strain rate conditions.

The conditions of temperature and pressure coincide with the conditions required for bonding, and therefore the two processes have been combined into one manufacturing operation either in sequence or together. This process (SPF/DB) is used to produce stiff sandwich structures for airframe parts, or the wide chord, hollow fan blades for aero-engines. Both of these processes involve skins with internally bonded structures as reinforcing elements. For aircraft, the increased application of diffusion bonding will help to allow for the conservation of expensive strategic materials and the reduction of manufacturing costs.

The diffusion bonding process is also used for the manufacture of advanced titanium alloy aero-engine structures and components for the production of both static fabrications and complex rotating parts. In order to utilize diffusion bonding processes in a production environment, the process parameters which contribute to consistent formation of joints of the required strength have been critically examined.

Process variables include temperature, pressure, time, surface roughness and interlayer composition, density and thickness. Mechanical testing (tensile, impact and fatigue) complemented by metallography has predominantly been used to identify the permitted variations in the processes for the realistic and economical production manufacture of high quality lightweight aero-engine fabrications. The development of a high integrity bond via optimized diffusion bonding processes has been fundamental to the development of Rolls-Royce's unique wide chord fan design concept for large jet aircraft engines.

Hot forming of dissimilar metals for the aircraft, aerospace, automotive, engine and medical business sectors will also form part of the project itinerary, as will magnesium, aluminium, titanium, Ti-Al, steel and nickel dissimilar metal DB combinations as they are all used for solid state bonded components for a range of applications.

TWI manufactured between 15 and 20 DB samples in steel, whilst the End-User Group provided a further 10 samples, including appropriate DB samples in similar/dissimilar metal/material combinations including titanium and steel combinations and SPF/DB components and assemblies used for sandwich fuselage panels, and hollow fan & compressor blades with reduced weight and optimisation of service performance. Sandwich structures are manufactured with SPF components that are joined by the DB process.

Approximately 10 of the total number of samples were of Ti 6AL 4V specification, this being the most prominent titanium material used in aircraft construction due to its high creep strength, lower density, improved weldability, higher formability and higher static strength.

The need for DB arises from the growing number of applications of solid state welding, particularly diffusion bonding, due to the process advantages of producing very clean welds with little or no distortion and residual stresses. Furthermore, high integrity applications, particularly those that are designed for a long service life, may be limited because of small defects with the potential to grow in-service (e.g. through fatigue) being undetected at the component manufacturing/fabrication stage.


The objectives at the start of the project were:

• To research and model the interaction of directional beam ultrasonic waves with the joint interface in diffusion bonded joints between agreed target combinations of dissimilar metals including the capability to fine focus the ultrasonic beam at close proximity

• To build a PAUT unit and pulser/receiver to examine diffusion bonded welds in a variety of materials including metals, ceramics and dissimilar metal joints to a high sensitivity (defect detectability for small defects of 0.1mm) with beam focusing at < 10mm from the surface.

• To develop an FMC technique where the transmission from a single element of the array is received on all of the elements, before repeating the process throughout the entire array. The entire test data set is processed to generate the defect image with an increased SNR and corresponding enhanced defect image and characterisation (shape, size and position) performance.

• To develop ADR algorithms that will automatically classify the specified detected defects and categorise their acceptability or otherwise. This will be applied to ‘clear-cut’ unambiguous defect types, e.g. linear lack of fusion/bonding on the joint line and will provide accurate and consistent sentencing criteria when applied over the component production run.

• Development of a pre-production prototype integrated PAUT system with data processing and display software and FMC technique and ADR software with operating procedures describing the application of the developed technique capable of testing diffusion bonded joints to the required industrial level.

• To evaluate the inspection performance of the BondTest ultrasonic inspection system by testing and evaluating the on representative dissimilar metal/material diffusion bonded samples (WP1) and if practical actual in-situ bonds that are operating in-service.

Project Results:
See attachment - Appendix 1

Potential Impact:
See attachment - Appendix 2

List of Websites:

Reported by

United Kingdom
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top