Issue with Centering/Calibration for X-ray WAXS Data in Raw

[Yu Hu]

Hi all,

      I encountered an issue while using Raw to process X-ray wide-angle scattering (WAXS) data. Specifically, the center calibration does not yield correct results for my dataset. I have tried fixing the wavelength, sample-to-detector distance, and detector tilt angles, but unfortunately, none of these adjustments resolved the problem.

Could you please advise whether this might be due to my misuse? I would appreciate any guidance or suggestions you might have.

Thank you very much for your support.

The data and screenshot of the GUI can be found in the attachment.

The sample-to-detector distance is 178.66 mm, the detector tilt angle is 72.41 degrees, the wavelength is 1.45, the detector is mar_165, and the sample is ZnO.

Best regards,

Yu Hu

[Jesse Hopkins]

Hi,

So there’s two possibilities here. The first and most simple one is that the calibration ring display in RAW isn’t set up for what you’re doing. It doesn’t draw actual ellipses, but circles. This is fine for standard geometries where you have small or no title angles, but with a detector tilt angle of 72 degrees would lead to a significant distortion in the display. This is a known (by me) limitation of that display, and hasn’t been fixed because, honestly, it’s a bit hard and no one’s ever wanted it (at least not done so and told me). The most likely option is that the fit is good but the display is bad. You could test this by trying to integrate your image to a 1D profile, if you get clean distinct rings at the expected q values for the calibrant then it’s probably good to go.

The other possibility is that in addition to the display not working right the fit is also bad. To test this you could use the pyFAI calibration tool (either GUI or command line) and see if it returns the same values as RAW (which is using that under the hood):

https://pyfai.readthedocs.io/en/stable/man/pyFAI-calib2.html

Note that in RAW we use what pyFAI calls the Fit2D geometry definitions, for historical reasons, so you’d want to make sure the output is in that format.

All the best.

- Jesse

----

Jesse Hopkins, PhD

Deputy Director

BioCAT, Sector 18

Advanced Photon Source

--
You received this message because you are subscribed to the Google Groups "BioXTAS RAW" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bioxtas_raw...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/bioxtas_raw/9aa3a7be-2110-40a5-8ade-a726acf44aabn%40googlegroups.com.
<waxs centering_done.png>

[Yu Hu]

Dear  Jesse,

       Thank you for your reply and suggestions. I did some further tests:

1. I tried integrating the image to a 1D profile, but I couldn’t get clean, distinct rings at the expected q values for the calibrant.

2. I also used the pyFAI-calib tool, but it didn’t return the same values as RAW.  

It seems that there are some differences between RAW and pyFAI at a lower level, even though both use pyFAI.geometryRefinement.GeometryRefinement for calibration.

Some details are included in the attached file.

Best regards,

Yu Hu

[Jesse Hopkins]

I appreciate the followup! I'll definitely investigate further, but it may be a bit before I have a chance to look into this in detail.

- Jesse

----

Jesse Hopkins, PhD (he/him)

Director, BioCAT

Sector 18, Advanced Photon Source

Research Associate Professor, Illinois Tech

To view this discussion visit https://groups.google.com/d/msgid/bioxtas_raw/5fc1f905-c62b-4104-951f-f7a493c2e117n%40googlegroups.com.