Calculating bending strain and direction from multiple rosette strain gauges

[finnalli]

Dear all,

I have three strain gauge rosettes distributed evenly around the base of the circumference of steel pipe at 120, 240 and 360 degrees. The gauges inside the rosettes themselves are located at 0, 45 and 90 degrees, and thus align with the axial and circumferential directions of the pipe. I would like to use these 9 measurements in total to determine the maximum bending strain as well as the direction of maximum bending.

The model I've come up with assumes perfect bending and that bending strain varies sinusoidally around the pipe’s circumference. Therefore I can fit my three longitudinal strain measurements against a sine curves to determine a translational and scaling factor (the bending direction and maximum, respectively).

Does anyone have any other suggestions of how to use these 9 measurements to obtain a more comprehensive estimate of the bending strain and bending direction? I don't think Mohr's circle and the transformation equations are relevant here since the principal strain directions are known and the strain rosettes align with these axes, but I could be wrong.

Many thanks for your suggestions,
Finn

 

[SWComposites]

if you are referring to overall bending of the pipe section (and not local bending of the pipe wall), then all that is relevant are the 3 axial (0 degree) strain gage results.

Overall bending strain is linear about the neutral axis of the section. For bending about a given axis, Strain = Mc/EI

 

[rb1957]

there are references online for calculating stresses (normal directional, shear, and principal) from rosettes; you shouldn't assume a stress result ... let the gauges tell you what they are feeling and then you can make sense (or not) for these results.