Analysis of weld pool oscillation in P-GMAW by Means of Shadowgraphy Image Processing

This paper presents a methodology for studying the weld pool oscillation by means of measuring its oscillation frequencies using both the welding voltage signal and the analysis of weld pool profle images. The intense arc brightness is reduced by using the technique of shadowgraphy and images are ob...

Full description

Saved in:
Bibliographic Details
Published inSoldagem e Inspecao / Welding and Inspection Vol. 18; no. 1; pp. 39 - 49
Main Authors Ramos, Esdras Godinho, De Carvalho, Guilherme Caribe, Alfaro, Sadek Crisostomo Absi
Format Journal Article
LanguagePortuguese
Published 01.03.2013
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:This paper presents a methodology for studying the weld pool oscillation by means of measuring its oscillation frequencies using both the welding voltage signal and the analysis of weld pool profle images. The intense arc brightness is reduced by using the technique of shadowgraphy and images are obtained at 1000 frames per second by a high speed CMOS camera. The geometry of the pool is viewed in profle and frequency spectra are obtained by Fast Fourier Transform using MATLAB image processing. The methodology was applied to analyze the weld pool oscillatory behavior that results from a pulsed GMA welding process. Two different welding current pulse types were tested: a) the conventional trapezoidal pulse (with only one level of peak current) and b) the trapezoidal pulse with the so called "detaching current" (the welding current presents a modifed trapezoidal shape, in which, at the end of the peak current time, the welding current is reduced to an intermediary level and kept in that during some time, after which it reduces further to the background current level). The results show that it is possible to measure the puddle oscillation using this methodology and, also, that the measurement by images present itself as more reliable than the analysis by the voltage signal.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0104-9224
DOI:10.1590/S0104-92242013000100006