SOUNDWELD: Quality inspection using acoustic emission monitoring

Last modified: 
12-11-2019

SOUNDWELD will investigate a new promising real-time non-destructive examination technique for welding processes, based on acoustic emission during the weld cycle. The acoustic emission technique is based on the detection and conversion of high-frequency elastic waves into electrical signals. Acoustic emission monitoring (AEM) is currently being used for the surveillance of industrial processes or structures. In this project, the application of this technology will be expanded towards welding processes.

 

Acoestic emission monitoring

Nowadays acoustic emission (AE) is widely used for many different applications, ranging from monitoring welding processes to controlling the integrity of bridges during their lifespan. By using AEM as an in-line quality control system, it is possible to listen to the sounds emitted by materials during the investigated process. The main goal of AEM is to surveil industrial processes or structures in a non-destructive way.

The AEM technique is based on the detection and conversion of high-frequency waves into electrical signals. When a metal is stressed, for example during plastic deformation, fracture or other local instabilities, low-level sounds are emitted. The energy for these sounds originates from the stored elastic energy in the object or from externally performed work. The waves will cause a displacement at the surface that can be measured with a sensor. In order to accurately distinguish the signals originating from the AEM source, external sounds should be excluded. This can be done for example by looking into the frequency domain, since the sound waves of the material have relatively high frequencies.

AEM is currently being used for the surveillance of industrial processes or structures.

Acoestic emission monitoring of welding processes

Welding industries are faced with the need to monitor the weld quality and system integrity more frequently, in order to guarantee the structural functionality of the products. Hence, weld quality is becoming increasingly important as customer expectations increase. A primary concern is to detect weld defects fast, reliable and cost-effectively. Current destructive and non-destructive techniques are time-consuming and expensive and are not always appropriate for assessing the weld quality. AEM as an in-line quality control system allows to overcome the current limitations of the conventional characterization techniques.

AEM could eliminate or considerably reduce the post-production selective inspection, reduce the number of destructive tests and increase the reliability of the assembly process.

Acoustic emission monitoring adds a new dimension to NDT of welds. The wide-ranging applications of acoustic emission monitoring are illustrated by examples of real-time data from submerged-arc, gas tungsten-arc and resistance spot welding.

 

Advantages of this technique

  • Cheaper and faster non-destructive control of welds
  • Lower cost for optimizing welding parameters or determining the parameter window
  • Real time system: Early and fast detection of welding defects possible
  • Less destructive testing is possible

More information

Project goals

SOUNDWELD will investigate a new promising real-time NDT method, based on acoustic emission during the welding process. AEM will be examined in a structured way for a variety of welding processes: 3 pressure welding processes will be in the focus of this project; one conventional and 2 innovative welding processes :

  • Arc welding (MIG/MAG)
  • Resistance spot welding
  • Magnetic pulse welding
  • Refill friction stir spot welding: The resulting interface in these welds differ significantly from conventional welds, resulting in a higher risk for overlooking defects when using conventional NDT methods.

The following sub-objectives are identified:

  • to investigate the reproducibility of the AEM signals.
  • to determine the appropriate AEM settings.
  • to recognize weld defects based on AEM measurements.
  • to develop a non-destructive weld quality monitoring system based on AEM.

Project description

Input requested

If you believe that this technique could be applied to materials or products in your company, share your vision and help us to explore the possibilities of this new development. Even if you want to follow up the further evolution of this technique (without an immediate application in your company), you can subscribe to this project.

Please contact Irene Kwee for further information.

Benefits for participating companies

The registered companies will get to know the technique in all its aspects, and can follow up the new developments that will increase the applicability of AEM. In addition, they can also steer the project, e.g. choice of materials, demonstration parts, testing, etc., and can also introduce their own case study into the project (i.e. their workpieces are welded and tested).

Every 6 months, an information meeting is organized where the results are presented. Only registered companies are entitled to the reports and results of this investigation (contractually defined by VLAIO).  

Results

User committee meetings

The next UC meeting will take place on Monday, 25th November, from 13h30 – 17h, at KU Leuven De Nayer.  

Address:

KU Leuven – Campus De Nayer
Jan de Nayerlaan 5
2860 Sint-Katelijne-Waver
Belgium

You can sign up at the reception desk and ask for Patrick Van Rymenant or Oleksandr Kurtov. Via this link, you can find attached a plan of the campus.
 

Program:

13u – 13u30 : Welcome & registration
13u30 : Introduction of the SoundWeld project
14u : Latest results of acoustic emission measurements applied onto:

  • Electromagnetic pulse welding (TFF & AGT)
  • Resistance spot welding (KU Leuven)
  • Semi-automated arc welding (BWI)

15u : Machine learning applied onto semi-automated arc welding (Oqton)
15u30 : Demonstration of acoustic emission meausrements applied onto resistance spot welding
16u : Discussion

 

Please confirm your presence before Tuesday 19th November by notifying Irene Kwee

Project results

  • You can download the preliminary results here