Bug #962
closedbeam tilt value from some version of tecnai or TEM scripting is not in radian
100%
use a larger beam tilt knowing that it is not in radian. The two scopes we saw this need a factor of 6.
Description
Some tecnai(TEM) scripting gives beamtilt value through Illumination.RotationCenter in radian. Some version gives it in some unknown unit. We do not know when the change occurs and if this is true for all versions after a certain time.
Here is one way to tell while using Leginon to calibrate defocus in Beam_Tilt node:
The default 0.01 radian beam tilt with -2 um defocus and -4 um defocus should separate by 20 nm in theory, assuming Cs=2.0 mm. This is equivalent to 100 pixel on a 2 Angstrom pixel-size CCD. For a 4kx4k camera, it means 1/40 of the length of image. If the correlation peak position during defocus calibration barely moved from the origin, your scope does not give radian value by this function call.