Technique de mesures 3D industrielles

GOM
 
Systèmes de Mesure | ESPI | Applications | Essais Matériaux | Simulation Numérique
Simulation Numérique 
  
Measurement of deformation during the welding process 
ARAMISESPI
 
Systèmes de mesure: ARAMISESPI
 
Keywords: welding seam
 
In order to determine welding techniques which minimize distortions, numerical simulation of the welding process are being developed. Experimental investigations which determine the strains and displacements in some experiments are then necessary for the verification of this newly developed Finite Element Method (FEM) -analyses. Due to their high local resolution and accuracy, optical deformation methods such as speckle-interferometry (ESPI) and the grating method (ARAMIS), are well suited for this application. Due to its high sensitivity, ESPI is used to determine the development of the deformations in short intervals, during the welding process. The determination of the complete deformation and the deformation during the cooling-down period is measured with the grating method.

In the following, the experimental set up and the results taken during the welding of a seam using inert arc welding with non-consumable electrode are discussed:

 
3D-ESPI sensor head at the welding tablePrinciple of the measurement set-up
3D-ESPI sensor head at the welding tablePrinciple of the measurement set-up
 
Deformation measurement using ESPI during the welding

In the displayed results (strain perpendicular to the welding direction) the deformation of the part in 3 seconds was observed at different states of the welding process. The black spot marks the current position of the welding spot in the individual images, the welding direction runs horizontal from right to left through the image. The images display almost the entire sheet metal.

In fig. 3 (65.-68. sec.) areas of distinct compression can be detected directly above and below the position of the welding spot. Another area of compression is created in the welding direction in front of the welding spot. In the diagonal between this spot and the lateral areas of compression, areas of small tension can be found.

In fig. 4 (99.-102. sec.) all of these compression and tension areas have moved with the welding spot. In addition an area of strong tension is visible direct behind the welding spot.

In fig. 5 the welding spot has almost reached the end of the specimen. Here too the lateral areas of compression are still beside the current position of the welding spot and keep its original size. The area of tensile strain following the welding spot spans the entire welded area of the seam and has increased in size in its maximum, which lies just behind the welding point.

In the displayed results the data could not be determined directly in the welding spot, as the thermal radiation causes decorrelation due to the high local rise in temperature caused by the welding.

The displayed results can be used to verify the greatly varying (in both local and temporal aspects) developments of the deformations over the entire area of measurement. In addition the aptitude of the ESPI measurement technique is demonstrated for welding applications.

 
results
Fig. 3: 65. - 68. Sec
Fig. 4: 99. - 102. Sec
Fig. 5: 152. - 155. Sec.
 
Deformation measurements using ARAMIS grating method during welding

The grating method ARAMIS is used to measure the absolute deformation introduced by the welding. Shown is here the underside of the specimen. The welding seam lies on the x-axis of the images and the welding was carried out in the direction of the positive x-axis (from left to right in the result images).

The displacement plot displays the deformation and its direction and is dominated primarily by rigid-body displacements. Local compressive strains can be detected in the immediate area of the welding seam. The lateral shrinkage caused by the welding process can be seen clearly in the distribution of displacements in y-direction.

The upper half of the specimen moves down and the lower half moves up. This effect increases in the direction of the welding from the starting point of the weld to the middle of the specimen, and then decreases only slightly. The deformations are symmetrical to the welding seam, but clearly asymmetrical from the starting point (left) to the end (right) of the welding. The maxima of the lateral movements lie in the third quarter of the welding seam, to the left and right of the seam. The image of the distribution of strains in y-direction shows clearly that maximum displacement gradients are not to be found directly below the welding seam, but slightly above and below (left and right when viewed in welding direction) the welding seam.

 
Welded specimen with structured surface
Fig. 6: Welded specimen with structured surface (300x200x10mm³)
 
X DisplacementY DisplacementZ DisplacementStrain distribution
X DisplacementY DisplacementZ DisplacementStrain distribution
 
In the image of the x-displacements the immediate seam area has shifted, in relation to the specimen as a whole, against the direction of the welding. This effect significantly increases in the first half of the specimen and then remains approximately constant. The image of the z-displacements mainly displays the upwards movement of the two halves of the specimen that occurs left and right of the seam. Since this effect slightly increases in the direction of the welding, the distribution of the z-displacements is slightly saddle-shaped.

The ARAMIS grating method measurement of the specimen's underside has been used to detect that, in spite of the time-dependent extensive development of strains detected by ESPI, the final strains resulting from the welding process are concentrated in the areas close to the welding seam. This result can also be obtained from the upper surface of the specimen where, outside a very slim area of the seam and the seam's edge, no strains can be detected.

Summary

ESPI and ARAMIS can be used even under extreme thermal and experimental conditions like the welding process to carry out successful analyses of deformations. The results of the measurement is displayed in quantitative as well as in graphical display, to make the interpretation and analysis an easy job.

By courtesy of the Institut für Messtechnik und Experimentelle Mechanik TU-Braunschweig

  
 

GOM mbH
Mittelweg 7-8
38106 Braunschweig
Germany
Tel.:+49 531 390 29 0
Fax.:+49 531 390 29 15
info@gom.com
www.gom.com
 
ContactsImpression