Support #2524
closed
exposure targeting suffers bad compustage
Added by Jian Shi about 11 years ago.
Updated almost 7 years ago.
Affected Version:
Appion/Leginon 2.1.0
Description
My compustage suffer intermittent large move error (>10um, more than half of field of view of sq). Stage reproducibility run gives bad results too.
The exposure targets are totally missed very often. I attached the settings in exposure targeting, exposure and focus for your reference. To my distress, I was trying to use navigator in focus, navigator seems never been activated. Use navigator in exposure sometimes helps, but seems cause drifting problem.
Your suggestion are very welcome. Thanks
Files
More elaboration: I think my compustage cripps from time to time. While moving, the motor might miss one stroke and the image jerks once then move on. The consequence is there is large systematic error out of it. When I recall the saved positions, it won't go back to the same place at all. I tested compustage with reproducibility run. When it works, it quite accurate. When it is bad, it can be as bad as 10um plus.
Lowest sq is 1700X 20um FOV, hl is 6500X 5um FOV. Manual navigation in sg and hl preset works perfect. I used all parent images for focus node, however Adjustment node only calculates the transform based on parent image and only move once without further correlation. Due to the large systematic error occurred before, it will end up complete different places for hl images. So my targets on lacey carbon in that hl image are completely missed (see attached example). But the leginon somehow goes on and take exposure on wrong places.
The Navigator option in focus seems only effect for taking final images. I wish it also works for initial hl target centering as well. It will useful to deal with errorneous compustage I have now. I will have FEI engineer to check on it, but I won't bet on them.
You got it right that target transformation does not use navigator iterative move. I am bothered by that, too. This will require some development effort although I do have some idea on how to do it.
It seems that stage reproducibility does not meet the spec for any scope from that you've described. In such situtation, we normally ask them to repair to meet their spec.
- Status changed from New to Closed
Also available in: Atom
PDF
