Project

General

Profile

atlas acquisition taking a really long time

Added by Anonymous almost 19 years ago

Atlas acquisition has started taking a really long time. I go to the Grid Targeting node, press Calculate Atlas and then Acquire. It is still running after almost one and a half hours - with a busy wheel spinning for the Square Targeting node. The last message for that node is "displaying targets...". Previously, atlas acquisition has taken about twenty to thirty minutes at the most. This started happening after I changed all the presets to use an intensity level that spreads the beam to cover the screen and also a lower spot size (to make up for the lower brightness from spreading the beam more). I don't know if this is connected but it might make a difference to exposure levels and suitable exposure times. (Also, I have now done beam tilt node defocus and stigmator calibrations for the focus and hole presets),

William


Replies (4)

- Added by Anonymous almost 19 years ago

This problem appears to have stopped happening, after changing the presets so they have the old larger spot sizes and the intensity adjusted so the beam spreads enough to cover a bit more of the screen than the larger circle (rather than the entire screen). Spreading the beam more was in retrospect really a work around for not doing the beam tilt pivot point alignment properly. I re-did the alignment and went back to the old preset settings and things seem to work okay now. I am still curious about the possibility of spreading the beam more. Also, the beam still seems to shift a bit in the autofocus presumably due to beam tilt - even with a recent beam tilt pivot point alignment.

beam shift during beam tilt - Added by Anchi Cheng almost 19 years ago

I do not know a way that will completely remove small beam shift during the beam tilt. It happens to us, too. leaving beam some what larger does help, although a more important reason to do this is that you will see some significant beam shift when you image shift a big distance even if you do your best in beam shift/image shift calibration.

atlas creation - Added by Anchi Cheng almost 19 years ago

William,

Could you tell us how many images were involved in creating the atlas that took forever. What were the dimension of each gr image? What was the size of the atlas you made leginon to create into mosaic. Did it get stuck at the beginning after the first gr image was taken or was it after a few? Changing spot size and intenisty of the beam only affects the acquisition. Your problem is at the level where it scales the images. Therefore, the parameters I request here are more important for the program. Not to mention that the problem went away after you reset the presets. If you give me more info, I will try to repeat it here and see if I can identify the problem.

Re: atlas creation - Added by Anonymous almost 19 years ago

Each gr image is 512x512 with 8x8 binning. I'm not sure how I can find out the size of the atlas. In mosaic settings in the Square Targeting node the option is checked to scale the image to 512 pixels in the largest dimension and the radius for mosaic in the settings in the Grid Targeting node is 0.0009 m. (I've used these settings in previous, and more recent, runs of Leginon without the described difficulties - these are the settings recommended in the manual, I think.) I'm not sure but I think it did get stuck at the beginning after the first gr image. I may need to try to reproduce the problem by running again with the earlier change in the presets (spot size and intensity) some time,

William

"anchi" wrote: William,
Could you tell us how many images were involved in creating the atlas that took forever. What were the dimension of each gr image? What was the size of the atlas you made leginon to create into mosaic. Did it get stuck at the beginning after the first gr image was taken or was it after a few? Changing spot size and intenisty of the beam only affects the acquisition. Your problem is at the level where it scales the images. Therefore, the parameters I request here are more important for the program. Not to mention that the problem went away after you reset the presets. If you give me more info, I will try to repeat it here and see if I can identify the problem.

    (1-4/4)