Beam shift in Leginon Tomogram data collection
Added by Long Gui about 10 years ago
Dear all,
I am also working with Leginon 3.0 installed in a Tecnai Spirit T12 TEM equipped with Gatan CCD.
Thanks to Anchi's help, now the Leginon client could recognize the TEM and the CCD very nicely and I have finished the calibration part of Leginon.
However, during my data collection of running "tomography" application on T12 with a magnification of 42,000X, the beam start drifting at ~ 45 degrees and my final image shows half dark half bright, so I have to stop my data collection.
It is interesting that I have tried to tilt the holder at FEI Tecnai User interface and the beam did not shift and keeps stable in the center of the screen.
And I have also tried to redo the "beam shift matrix calibration" in Leginon "calibrations" application, but it does not help.
Any suggestions I could apply to fix this problem? I have also attached a image to illustrate it too.
Thanks,
Long
| T12_beam_shift.pdf (137 KB) T12_beam_shift.pdf |
Replies (3)
RE: Beam shift in Leginon Tomogram data collection - Added by Anchi Cheng about 10 years ago
Beam moves off the camera at high tilts of tomography data collection if not related to "beam shift matrix calibration" but with HM_Beam_Shift_Image_Shift_Calibration
Often times, if the tilt model and eucentric height are not good, a large image shift needs to be applied in order to track the feature you are imaging as the the feature moves away during the tilt.
FEI scope's "image shift" is decoupled with "beam shift" once HM_Beam_Shift_Image_Shift_Calibration is properly calibrated.
Learn how to Reading_the_tomography_graphs will help you to see how to keep image shift minimal during tilt series acquisition.
Therefore, my suggestion is
- Perform HM_Beam_Shift_Image_Shift_Calibration
- Find out how much image shift is applied, and see if you can make it smaller.
If you need help understanding the graphs, please post a set here.
RE: Beam shift in Leginon Tomogram data collection - Added by Long Gui about 10 years ago
Hello Anchi,
Thank you for your reply for this question.
I have followed your instructions and re-perform the HM_Beam_Shift_Image_Shift_Calibration but the problem still exists.
So I started over to check this issue and I suddenly found that this problem is due to objective aperture. After I removed the objective aperture, the beam was not blocked anymore and the tomography could finish very nicely.
My question is : how could this objective aperture block the beam at high tilting angles? Do you think it might be a problem from the TEM itself? Should I contact the TEM engineer to check this FEI Tecnai spirit T12?
Thanks a lot,
Long
RE: Beam shift in Leginon Tomogram data collection - Added by Anchi Cheng about 10 years ago
There are two things contribute to objective aperture blocking the beam at high tilts:
1. The objective aperture is not centered. I am sure you have checked that.
2. Image shift required for tracking the target accumulated through the tilt series is too large. Image shift induces beam tilt, which translates to a position change in the diffraction space. When the beam tilt is large, you can observe how the diffraction pattern moves relative to the objective aperture if you are in diffraction mode.
The amount of beam tilt induced by the image shift is more or less constant. Therefore, the best thing you can do is to make sure that you don't need to apply too much image shift during the tilt. That translates to making sure that the grid is at Eucentric height as defined by the stage alpha tilt.
There are three things you may do if your Leginon default "Tomo Focus" and "Z Focus" are not giving you the eucentric height good for higher tilts:
It is possible that you find that using 1 degree stage tilt to adjust z height give you a very different z position from tilting it at 5 degrees or 15 degree. What you can do in that case is either
1. Change your "Tomo Focus" node focus sequence to make it tilt more.
2. Use beam-tilt-based autofocus to determine the z height if your rotation center is aligned while the focus is at the higher tilt Eucentric height.
If it is still rather hard to do this, and your tomography graphs look funny, then your best action would be to ask the service engineer to work on the goniometer. If you post a set of the graph from a good tilt series without the aperture, I should be able to tell you if it is "funny"