Eucentric height and stage-disabling moves around search mag

[Willem Noteborn]

Hey Jonathan,

After a couple months more use we have a bit more feedback from our side. It seems that there are two issues that keep arising on our systems. They might be more related to our setups but maybe still good for you to hear.

1) It seems that the current method of determining eucentric height on individual gridsquares happens by using the search mag and probably not using the routines of serialem itself (which typically uses the view mag). We find that the current method keeps coming up with values that are around 10-20 um above the actual eucentric height compared to what we find manually and with serialem's integrated methods (even by using exactly the same low mag gridsquare FOV parameters as we would have in our typical search preset). This might not sound like a big problem at first as it can easily be compensated by some additional defocus but here comes the catch: on our system we have falcon camera's and they have dose protectors. These also act on large overfocus values (+20um overfocus and higher) and thus we find that after we hit a grid where the eucentric height is not found properly, the overfocus gets to large and from that point we only collect black blanked images. For now we have a workaround for this by tricking the dose protector a bit. If you have any suggestions on if this could be due to something we do in our setup or if you guys see this behavior as well on your systems would be very helpful.

2) Somehow, also during the eucentric height determination, it seems our stage likes to go in error/gets disabled. We don't really know why/how, but we notice it roughly every 10 squares it collects. We never see this behavior during long tilt series acquisitions with serialem nor with EPU sessions. For this issue, I have no clue where this might be related to. Just an observation from our side.

The upside is that both issues can be overcome/restarted!

Thanks for any help/suggestions in advance!

Willem

[Jonathan Bouvette]

Hi Willem,

Thanks again for the worthy feedback. For the falcon detectors and the eccentricity issue, you may use the SPA-precise or SPA-Ptolemy-precise protocol instead of their regular counterparts. This will use the View mag internal SerialEM eccentricity procedures. The search mag one is mainly to allow faster screening with non-falcon detector but we use the *-precise protocols for data collection.

If you do not have the precise protocols, you may need to update smartscope to at least v0.9.

Also, you may want to change the default protocols to the *-precise version when using the auto option during session setup. To do so, go to shared/smartscope/default_protocol.yaml and swap the SPA protocols by their -precise counterparts as shown below:
```

##Use this file to design rules for how to automatically set default protocol for a grid.
- conditions:
    - [holeType.name, NegativeStain]
    - [holeType.name, Lacey]
    - [holeType.name, MultiA]
  mode: any
  protocol: NegativeStain
- conditions:
    - [params_id.bis_max_distance, 0]
    - [params_id.tilt_angle, '!__0']
  mode: any
  protocol: SPA-precise
- conditions:
    - [holeType.hole_size, '!__None']
    - [params_id.bis_max_distance, '!__0']
    - [params_id.tilt_angle, 0]
  mode: all
  protocol: SPA-Ptolemy-precise
```

As for the other bug, maybe using the precise protocol will solve it. We have noticed some hanging during square re-centering on our end but not during eucentric height.

Best,

Jonathan

--
You received this message because you are subscribed to the Google Groups "SmartScope" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smartscope+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/smartscope/1d61d7bd-486b-448c-9f79-b3e15a1ea02fn%40googlegroups.com.

[Jonathan Bouvette]

Somehow my computer corrected eucentricity to eccentricity in my previous email. sorry about that.

[Willem Noteborn]

Ah yes, looking back at the most recent log file the stage issue is indeed happening during re-centering. Also, at the search mag. Intuitively it felt close to the eucentric height determination routine. But this indeed is correct.

Thanks for the suggestions and I will give it a try!

Op maandag 12 juni 2023 om 22:01:05 UTC+2 schreef jonathan...@gmail.com:

[Jonathan Bouvette]

It is good to know that you're also experiencing this problem with the square re-alignment. 

I'll try to look further into the re-centering hanging. I think it's related to directly grabbing the square image from the serialEM buffers without saving it to disk. It's the only place in that this method is used in SmartScope. I'll try to bring it up to the serialem team to see if they know anything about it.

I think we only ever saw this problem on our K2 system and not on the K3. What is your hardware where this happens?

[Willem Noteborn]

Hey Jonathan,

It seems I've missed this reply, sorry. We have it on our Krios G1 with K2+bioquantum filter / Falcon3EC system. The K2 is a bit grumpy on this system so we don't use it often. I haven't tried to set it up with smartscope yet. So for now everything is done on the Falcon3EC camera.

Best,

Willem

Op dinsdag 13 juni 2023 om 02:43:41 UTC+2 schreef jonathan...@gmail.com: