Bug #4898
openAlternate tilt scheme produces larger error than sequential and tracking failed between 30-40 degree on Krios but not Arctica
0%
Description
Reported by Gabe Lander
Hi Anchi, We're collecting some tomography on the Krios right now & noticing that the alternating tilt scheme yields much worse results than the sequential. The first attached image contains the plots from a tomogram collected using the sequential scheme, the 2nd is using the alternating scheme. As you can see, the shifts in the y axis are much more dramatic using the alternating scheme, and usually by 30-40 degrees the imaging has wandered off the hole in one of the directions. We didn't seem to have this problem on the Arctica when we used the alternating scheme, so I'm wondering if this is simply a goniometer issue, and that our Krios goniometer can't handle the back-and-forth tilt movements? Or are we missing something here that we can fix? Best,
Updated by Anchi Cheng over 7 years ago
Got a chance to try this and concludes that Krios/Arctica/Tecnai all behave a bit differently due to weight distribution and goniometer design.
On Krios, a permanent delta 3 degree of backlash correction works the best.
On Arctica/Tecnai/Talos, the swing of the gear alone can be used as the correction, although I suspect that Tecnai/Talos need a bit different scheme from Arctica because of the weight difference of side entry stage and Autoloader stage.
I will write code to separate the two cases.
Updated by Alex Noble over 7 years ago
Hi Gabe,
I also see a greater error in alternating collection compared to sequential on all of our Krioses. I didn't realize that this was a goniometer-based issue, as Anchi pointed out.
Anchi, I'll be able to test out the code when you're ready.
-Alex
Updated by Anchi Cheng over 7 years ago
I committed the first version of the change. Alex and I tested it out last week but it did not improve alternate mode accuracy. Will need to refine.
Updated by Alex Noble over 7 years ago
Hi Anchi,
The test was on Krios3, which has an apparent goniometer issue around 20 degrees. The next time I have a bit of time (like a few hours) on krios 1 or 2 I will do a more proper comparison.