New VirtualArc® software increases productivity and reduces costs of robot welding

Industrial Robot

ISSN: 0143-991x

Article publication date: 1 October 2003

74

Keywords

Citation

(2003), "New VirtualArc® software increases productivity and reduces costs of robot welding", Industrial Robot, Vol. 30 No. 5. https://doi.org/10.1108/ir.2003.04930eaf.004

Publisher

:

Emerald Group Publishing Limited

Copyright © 2003, MCB UP Limited


New VirtualArc® software increases productivity and reduces costs of robot welding

New VirtualArc® software increases productivity and reduces costs of robot welding

Keywords: Welding, ABB

Improves process implementation, final weld quality and productivity

Full "off-line" welding parameter tuning of MIG/MAG process

ABB has launched VirtualArc®, a powerful software tool that gives robot welding engineers full "off-line" control of the MIG/ MAG process (Plate 4).

Featuring an easy-to-use software interface, VirtualArc® can be operated from a single PC or laptop and provides users with efficient "off-line" welding process tuning to predict a wide range of results including weld shape and penetration, weld quality and possible welding defects.

Predicting such results brings substantial benefits including shorter implementation time, better optimised welding productivity and quality, as well as better documented welding procedures. In turn these considerations contribute to lower product costs.

As well as saving time and money through better production, VirtualArc® is also an efficient training tool for robot operators, programmers and production/welding engineers, and an excellent platform for retrieving and storing arc welding process information for future developments. VirtualArc® is also useful for comparing different weld procedures in cost/metre weld.

The technology behind the VirtualArc® software is based upon the combination of arc physics, welding simulation, arc welding practical experience and experimental measurements.

Weld profile and quality predictions are obtained using Bayesian neural network tools. Predictions from arc simulations, and heat and mass transfer to the workpiece, are used as input to the neural network to predict weld quality, weld profile and weld defects. The accuracy of predictions and robustness of VirtualArc® software has been verified by applying the methodology to a large number of welds. It was found that the system consistently reproduced arc dependent properties as well as final weld penetration profile and quality.

Mark Oxlade of ABB, comments "ABB strongly believes that the need for production enhancing tools like VirtualArc® will become more and more important as the number of skilled welding engineers and welders are decreasing on the market. ABB's development of VirtualArc® software and technology is an important step forward in helping arc welding customers to automate their production in an efficient manner."

For further information, please contact: David Marshall, ABB, Auriga House, Precedent Drive, Rooksley, Milton Keynes MK13 8PQ. Tel: +44 (0) 1908 350300; Fax: +44 (0) 1908 350301; E-mail: david.marshall@gb.abb.com

Plate 4

Related articles