Magnification changes in "hl" preset
Added by Mario Borgnia almost 11 years ago
Hi Anchi,
I continue to set up Leginon on our Krios equipped with a K2 camera at the end of the energy filter.
I have chosen the following presets:
gr 175X
sq 360X
hl 6500X
en,ef,fc,fa 105KX
They are all set up with a beam size that is supposed to give parallel illumination.
The problem that I am finding is that in "hl" preset the suggested defocus leads to a significant change in pixel size. The pixel size at Gaussian focus is 2.26E-9 m, whereas at a defocus of -1.5E-4 m, as suggested on the documentation, the pixel size is 2.52E-9 m. Is this what you experience in your instruments as well?
I can only complete the calibrations successfully when I set the preset with a defocus value and its corresponding px size. Is this what is expected?
Thanks
Mario
Replies (1)
RE: Magnification changes in "hl" preset - Added by Anchi Cheng almost 11 years ago
It is true that magnification changes when a large defocus is applied. We have the same experience.
Because pixel size values for the lower mag presets are only used as a rough estimate during calibrations to warn you about unreasonable results and to put a scale bar on the image in
the web viewer. It is not critical to be right on.
If your pixel value is off, and resulting a rejection of the matrix calibration Leginon is doing but you are confident that the correlation peak found from the image movement during the calibration is genuine, other than what you have done, you could just increase the tolerance of acceptable error in the node settings. It would probably save you more time this way. However, since you've already done the work, you probably have a better pixel size value for the true magnification at this defocus : )
The large underfocus we recommend is applied because the low contrast at this low mag makes it hard to see the holes on an ice grid. If you don't have this problem, feel free to use a value closer to zero.