Keywords: residual stress, measurement, strain gauge, three-dimensional finite element analysis, ANSYS.

An extensive experimental programme has been developed to study the interaction of the fabrication technology, imperfections and experimental behaviour of welded plate girders with single-sided fillet welds between the web and flanges [1]. In this paper the focus is on the residual stress measurements using a modified hole-drilling method, the "one-gauge two-hole" method, in which a single uniaxial strain gauge and two through-holes on the two sides of the gauge is employed. This method is much easier to execute and less sensitive to inaccuracies than the more commonly used hole-drilling method with a strain gauge rosette and a single blind hole drilled at the centre of the rosette. The latter also requires specialist equipment due to its high sensitivity to inaccuracies in the drilling process.

The actual measurements, including all the inaccuracies, are numerically reproduced using the general purpose finite element software ANSYS [2] to ensure an accurate evaluation of the experimental results. High resolution photographs are taken of each measurement location and using an image analysis software more commonly used in medical imaging the exact position of the gauge and the two holes can be measured. With the help of these data and a fully parametrised three-dimensional finite element model the measurements can be reproduced with great accuracy. Comparing the initial stress applied to the finite element model and the stress obtained from the average relaxed strain over the area of the strain gauge after the simulation of the drilling of the two holes gives the ratio of the actual residual stress and the measured one. This ratio is then used to correct the measured residual stresses in the two specimens. The results of the measurements in the two types of girder show that using a single-sided fillet weld results in significantly lower levels of residual stresses when compared to a double-sided fillet weld. This indicates that the zone of high residual stresses is much narrower in girders fabricated with single-sided fillet welds.

Apart from evaluating the actual measurements, the measurement methodology itself is also thoroughly studied to fully understand how the inaccuracies of the execution influence the readings of the strain gauges and to develop an easy way to calibrate future measurements. After defining the key inaccuracy parameters a parametric study is performed to evaluate their effect on the measurements. The results of the study can be summarised in a calibration table that can be applied to any future measurements using strain gauges with the same active grid size and holes with the same diameter. The same methodology can be easily applied to develop similar tables for other types of strain gauge and different hole diameters.

1

L. Dunai, G. Jakab, J. Nézo, B.H.V. Topping, "Experiments on welded plate girders: Fabrication, imperfection and behaviour", First International Conference on Advances in Experimental Structural Engineering, Nagoya, Japan, 19-21 July 2005.

2

"ANSYS User's Manual", Release 10.0, 2005., ANSYS Inc., www.ansys.com.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £145 +P&P)