Leginon

Sorry, apparently I got logged out automatically while composing the message so it got posted as Guest/ Anonymous,

William

"Anonymous" wrote: I've started trying to set up the MSI Application in Leginon, going through the steps in "MSI Quick-start" in the manual. I have a couple of questions about this. The first step is to align the microscope. The required alignments are described in chapter 4 of the manual. I'm not sure about the step for "HM Beam Alignment". It says to "check U-center and reset to Eucentric focus again for best result". I haven't done much microscopy in the past so I'm not sure what that means. Does that mean to move the specimen to eucentric height (e.g. using the z-height control and alpha wobbler on the Tecnai F20) and then reset the focus? (Presumably that should get eucentric focus. I'm not sure about actually using the eucentric focus button as I think that may depend on some other previous alignments.) Also, a few steps in MSI quick-start seem to be redundant as they seem to be done in the initial alignment (in step 1) e.g. step 10 is to correct beam pivot points and rotation center for LM mode, done after step 9 which is going to the sq preset (and finding true focus at eucentric height and set defocus to zero). Isn't this done in the LM Beam Alignment as part of the alignments carried out for step 1? Also, if I do this type of alignment after going to the sq preset; then if I do the alignment from the Alignment Control Panel in the Tecnai User Interface then the Alignment CP changes a lot of microscope state, including the magnification (and possibly intensity, beam shift and image shift as well). So do I need to do the alignment some other way at this stage?
William

1. Yes, check U-center and reset to Eucentric focus does mean "move the specimen to eucentric height (e.g. using the z-height control and alpha wobbler on the Tecnai F20) and then reset the focus". Just make sure that you actually focus the image before reset defocus to zero. Eucentric Focus button takes you to the factory preset focus which may be a bit off, so I don't normally use it unless I lose the beam completely from bad alignment.

2. Yes, some of the quick start protocol are redundant. If you have done the alignment at earlier step, it is normally not necessary again, especially ones in LM mode which does not need to be very good. With some experience, you would know if your microscope behaves well after alignment step 1 and know if you need to repeat even the HM ones in the protocol at a later time.

3. Alignment procedures in Alignment Control Panel tend to reset user defined values when they are saved. Therefore, the user level image or beam shift which leginon saves may have different meaning after the alignment. I use Direct Alignments as much as possible. It is certainly easier for beginners to use the Alignment Procedure for the same function. It also contains functions that are not accessible from direct alignments. Therefore, I think the best solution is to use alignment procedures at first before you set up the presets and then if you need to make adjustment later, use direct alignment. However, try to avoid BEAM SHIFT in direct alignment. It resets the user defined beam shift and reseting it means you will need to realign your presets. I also want to say that it is not necessary to do all alignments everytime you use the scope. In general, the calibrations done in Calibration Application are

stable after new alignment. The autofocus calibration done in Beam Tilt node works as long as your rotation center is set at eucentric focus. The state changes after alignment may affect preset parameters, either because it resets the image or beam shift etc. or simply because of hysteresis. You may cycle the presets a few times to see if it comes back before realign beam and image shift of the presets.

Anchi

Thanks. I have another question about alignment - unfortunately I've run into difficulties with the LM Image/ Beam Calibration alignment which is recommended in the Leginon manual. I attempted this alignment by going into the "Image LM/ Image/ Beam Calibration" alignment procedure from the alignment control panel. In step 2 of the alignment, the multifunction knob is set to image shift x and the instructions are to adjust it until the shift is greater than 50%, the coils are greater than 80% or the beam goes to the edge of the screen. However, I found that the beam disappears before any of those conditions are satisfied (some time before reaching the edge of the screen). This happens even when the specimen stage is taken out partly into parked position, to avoid obstructing the beam. We tried treating the situation as if the beam had got to the edge of the screen; but in this case step 3 instructs us to go back to step 2 where the beam disappears again. FEI sent out a service engineer to look at the problem but he had the same difficulties. He thinks that a fixed aperture is in the way. We expect to hear from FEI again but I'd be interested to know if you've ever had a similar problem? Also, can I possibly do without the LM Image/ Beam Calibration alignment? Is there another way to do that alignment? Do you know why this particular alignment is required for Leginon? I don't think the required controls are accessible to a direct alignment so that may not be a way to perform an LM Image/ Beam Calibration type of alignment. (By the way, the HM Image/ Beam Calibration was carried out without any difficulties),

William

Hi, William,

If you are able to align image shift of your LM presets (gr and sq)

to those in HM without beam cliping the image upon image shift

adjustment, you can do without LM Image Shift/Beam Shift Calibration.

HM one is the more critical one.

Is it possible to align the C1 aperture on the Tecnai F20? An FEI engineer I talked to seemed to think this required significant work involving getting into the column and taking out the C2 aperture (which cannot normally be removed) before aligning the C1 aperture; so it's generally best to stay away from adjusting the C1 aperture. However, I found a description of aligning the C1 aperture in a CM FEG manual, which may be different of course. Also, the description in the manual is aimed at applications (eg EDX work) that require a small C1 aperture. It says this means centering the smallest C1 aperture inside the largest C2 aperture and that centering the C1 aperture is done in the same way as the C2 aperture by focussing and defocussing the beam and ensuring that the beam expands and contracts concentrically. I'm not sure though if the smallest C1 aperture is centered by this method; then the largest C1 aperture would be (and the largest C1 aperture is what we normally use).

By the way, FEI still haven't been able to help us to carry out the LM Image/ Beam Calibration despite the FEI engineer trying a couple of times. It looks like this may not be so important fortunately. The clipping behaviour/ disappearing beam I described that occurs during the attempted LM Image/ Beam Calibration alignment; does not occur with the magnitude of image shifts used to align the presets,

William

Hi, William,

I asked a friend, Gary Ren about C1 aperture alignment. He has done it for

diffraction work. The method you described is the same as his. He

only did it for the smallest C1, though. The large C1 aperture we normally

use probably never be off by that much that you can see the effect through

even the largest C2.

Anchi

I looked at the manual for an automatic tomography package (from FEI I think - it may be Gatan) which uses a similar approach for autofocussing and the manual recommends some of the same alignments as the Leginon manual (including beam tilt pivot point and rotation center) but also some additional microscope alignments which were beam tilt calibration and diffraction shift, I think. I'm interested in trying these in case they are useful for Leginon. The F20 documentation on the beam tilt calibration mentions basing the calibration on the Bragg angle of gold diffraction rings in diffraction mode. I'm not sure what type of calibration grid I need for this? The EM supplies catalog I have lists some types of calibration grids with gold but I think these have (small) single crystals and surely I would need a (fairly fine) powder to get gold rings. The beam tilt calibration may be useful to ensure that our F20's beam tilt is comparable with other groups' microscopes and the values for beam tilt angle used in the Leginon manual (for controlling the behavior of the Beam Tilt, Focus and Z-Focus nodes). (It is possible though that the factory or service alignments are good enough.)

William

A recent release of Leginon added rotation center storage and retrieval to the Beam Tilt node. I'm not really sure how to use this in practice. (I suppose if I ignore it then things will not work as before?) If I have a recent alignment then presumably I can just go to the desired magnification (and high tension) and then just retrieve the rotation center without going through the a rotation center alignment again? I usually carry out my alignments in the alignment procedures in the Tecnai user interface (but often with slight modifications to the described procedure so some alignments are done with other than the default magnification or spot size) rather than using "direct alignment". Or maybe I could retrieve the rotation center by going into the alignment procedure, carrying out the rotation center alignment at the chosen magnification and then retrieving the rotation center in Leginon (even though the image on the screen will be jumping about due to some type of wobbler being set by the alignment procedure in the Tecnai user interface)?

William

Stored rotation center is used during autofocusing of a tilted specimen.

See leginon manual Chapter "Leginon "MSI" Application and Section "Special Operation setup. The rotation center value is required for a distortion correction that shapens the correlation peak.

You can used the store/retrieve for future alignment, but I have no idea how stable this number is over a long time.

When do you normally do the microscope alignment before using Leginon - immediately before using Leginon, in the same session on the microscope, or the previous day as some of the notes from the course on Leginon suggest? I've usually aligned the microscope immediately before using Leginon (just after putting the relevant cold grid in the microscope) because that fits in better with the microscope booking system in Leeds. However, I suspect that part of my method for aligning the microscope could actually be introducing contamination into the microscope. I usually carry the gun alignments with the cold holder in parked position (after inserting it with the cold grid and allowing it to pump down and also before opening the frost shield) so it is easier to see the beam for those alignments. I've looked into carrying out the gun alignment without any holder in the microscope but for X-ray safety reasons the Tecnai UI does not allow the gun alignments to be carried out at the smaller spot sizes - I'm not sure if this is problematic for the gun shift and gun tilt alignments but it does make it impossible to do the spot-size dependent gun shift alignment completely,

William

The general alignment can be done at any of the time you mentioned as long as the scope is not used for other conditions that affects the alignment between the alignment and the actual Leginon session. By that I mean a change in high tension,gun lens strength etc. What you described does seem practical for your situation.

On the microscope that is almost exclusively used by Leginon, we do not do the full alignment of the gun every time. We check the beam for its intensity and coherence on the day as the new user. If Leginon is used in the same way over a few days, we do not normally find it necessary to realign the scope except to adjust the exposure time if the beam intensity is not stable.

For alignment with the holder in the scope, we do similar thing as you do for the same reason. However, we do not put the holder all the way to the park position which is a few inches out. We just pull the stage out enough to insert a pen that sits across the cover of the compustage and so creates a space between the dewar and the goniometer face plate where the former is normally in contact with. It is not recommended by FEI but is most practical. If your alignment takes too long as in hours, you can align the scope with a room temperature sample first before inserting the cryo-sample. We have not found contamination an issue here as the time that requires the stage to be out is at most a few minutes. Another way to perform the alignment is to do it at an empty grid square where there is neither ice nor support. This works if the adjustment is small, as in most cases. The requirement of having no specimen only means that the beam should not be near any scattering material. Therefore, at high enough magnification, an empty grid square is good enough.

Anchi