I have Kernel average misorientation data to analyze the distribution of retained strain. However, data has been achieved using different working distances for different samples. How can I normalized my data?
The kernel average map classifies somehow the local orientation gradient. If you change the working distance (distance of the diffracting volume from the pole piece) there is no change in orientation. Therefore I would also think there is no affect of WD to the kernel, except the indexing is still comparable. I strongly believe that you don't have a tiltable detector of Bruker so that any (remarkable) change in WD will shift your pattern. Thus, different bands are perhaps used for the indexing which might affect the reliability of indexing. But this is quite theoretical and has never been proven until now. I am sure it has some impact but then for each orientation in a different way.
What might be interesting with respect to WD is a change in magnification. If you are talking about this, then you are absolutely right. In fact it will only point out a remarkable change in step width. However, if the step width is exactly the same I would recommend to spend time for something else. It shouldn't have any effect.
Working distance is sometimes also used for the sample-EBSD screen distance. Then it has some influence since the angular resolution of each pixel is related to the accuracy of each orientation measurement. This will certainly have some impact but also here I don't assume a remarkable influence. On the other hand it has been never exactly investigated (as far as I know).
In fact you have to ask yourself what your KAM will show you exactly. I would recommend to improve the acquisition, i.e. frame averaging and using the full dynamic range of the signal in order to get the best result. And then make sure that you still have no surface near deformation coming from the preparation.
Calibration is certainly important. I wouldn't say essential since I do not know the resolution of the images and the Hough transform. Of course it shouldn't be totally wrong :-). Do you use a new automatically calibrating system? Why you are using different WDs? If you have the same material I would always select a very similar WD. Not because of the impact on KAM but because of the comparable backscattered signal. It is more important to get the same signal (background position (WD dependent) and EBSD signal (preparation and deformation dependent)).
I would like to emphasize a point made by Gert Nolze above, it is spacing between points of the scan that should affect your answer, and working distance should have a trivial effect lost in the noise. Keep your raster spacing equal between scans and you should be fine.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Material Characterization