AutoFocus Issues on Opera

13 views
Skip to first unread message

Santosh

unread,
Sep 16, 2010, 2:26:57 PM9/16/10
to Opera High Content Imaging
Hi,
Has anybody faced auto focus issues with the Opera QEHS system? We
face auto focus errors when we try to image 384 and 1536 plates. Are
there other users who have faced such issues in the past? Is there a
simple fix for this one?

Regards,
Santosh

John Moffat

unread,
Sep 17, 2010, 12:54:00 AM9/17/10
to Santosh, Opera High Content Imaging
Hi Santosh,
How pervasive is the autofocus problem? Complete failure or sporadic? Specific to one objective?
Incorrect plate definition can lead to complete autofocus failure. So can having the plate incorrectly seated in the table.
I've had more trouble with 20x air than any other objective, - though I only use that if someone comes to me with the wrong (i.e. non-CellCarrier) plate.
How's your autofocus laser power ? If that's dropped it can also make the pfoc less robust.

One interesting glitch I ran into recently was 10x water autofocus failure when plates were loaded with the plate::works robot or by hand with the table extended, but not when the lid was lifted and the plate placed directly on the table. after watching closely, I realized a loop of the immersion objective tubing was getting in the way of the objective cup re-seating properly after the table moved out and in. I trimmed the tubing a tad and looped it over instead of under the adjacent 20x objective tubing, and everything was fine.

John






--
You received this message because you are subscribed to the Google Groups "Opera High Content Imaging" group.
To post to this group, send email to Oper...@googlegroups.com.
To unsubscribe from this group, send email to OperaHCI+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/OperaHCI?hl=en.


Emmanuel Gustin

unread,
Sep 17, 2010, 1:59:47 PM9/17/10
to Opera High Content Imaging
Hi Santosh,

The Opera software installation should include a simple curve plotting
tool. If you open that and put it on top of the Opera screen, the
software will show the track generated by the autofocus sensor: You
should have two clear peaks, from the reflections at the two surfaces
(top and bottom) of the plate bottom.

Plate definition indeed matters, as it defines the search range of the
autofocus scan. If you are not using a predefined plate, you can
consult the manual for instructions on how to create a new one. With
some combinations of plates, plate definitions and objectives, it is
possible for the objective to collide with the plate bottom or its
skirt, lifting it out of the plate holder. It ruins the measurement
and of course it is bad for the focus drive, so it is best adjust the
settings in such a way that it is avoided, although that can be hard
because the working distance of some objectives is short. IIRC there
is also a setting that allows you to exclude the rows and columns at
the outside of the plate for some objectives, if the plate isn't low-
skirted.

We have had problems with the water immersion objectives because of
bubbles in the immersion fluid, which generate spurious scatter
signals. Because of the location of the immersion pump, part of the
circuit is at lower pressure. Make sure all connections are airtight,
sonicate your immersion water before use to de-gas it, and make sure
to flush long enough before you start measurements.

Best Regards,

Emmanuel

Santosh

unread,
Sep 24, 2010, 1:23:05 PM9/24/10
to Opera High Content Imaging
Thank you for your responses. Our auto focus issues is not restricted
to any Objective. We tried both 20 and 40 x air objective, with or
without plateworks, the result was the same. The problem is with the
issues is its randomness. We were not able to reproduce the same auto
focus issue on the same plate when we re imaged it with the same
settings.



Regards,
Santosh

We often faced this issue with the Aurora and Griener plates (Haven't
been able to test the Others).

On Sep 17, 10:59 am, Emmanuel Gustin <Emmanuel.Gus...@telenet.be>
wrote:
Reply all
Reply to author
Forward
0 new messages