http://confocal-microscopy-list.275.s1.nabble.com/timeseries-large-images-and-focus-tp5747682p5751471.html
Central Imaging, IMCB, A*Star, 61 Biopolis Dr, 6-19B, Singapore 138673, sent from 9271.5622
> *****
> To join, leave or search the confocal microscopy listserv, go to:
>
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy> *****
>
> Hi Sarah,
>
> If PFS is engaged throughout your movie, your microfluidic chamber almost certainly has irregularities in the 'coverslip' width. If so then you can semi-solve the problem with a wide open pinhole and low NA objective. Alternatively, you could set up the 20 fields of view as two or three independent image series with their own set PFS position, and then stitch them into one long time series/image after using an ImageJ plugin like Stack Combiner. This could take a bit more work on your part (I do a lot of that with our lab's A1s) but it uses less imaging time and memory than a z stack.
>
> Moving between regions could be more problematic. As I recall PFS depends on the glass/water interface to work. In between fluid channels you most likely have a material whose index of refraction does not allow the infrared PFS beam to return useful information to its sensor. The other possibility is less dire: we have found that PFS can lose its 'lock' when you have the stage move too far too fast, most likely due to slight tilt/curvature of the coverslip. We solved this problem a couple of ways. If we have to make a long jump between XY positions, then we register a 'dummy' position midway along the jump that lets the PFS catch up. It's a slight waste of time and memory but it's better than losing a movie. We also found that the slowing down the automatic stage movement speed can help, sometimes quite a bit. I believe that most systems are set on max speed, and slowing ours a bit (you may need your service rep to do that) has not slowed down our time series imaging noticeably.
>
> Good luck and all the best,
>
>
> Tim
>
> __________________
> Timothy Feinstein, PhD
> Postdoctoral Associate, Vilardaga lab
> University of Pittsburgh Dept. of Pharmacology and Chemical Biology
> BST W1301, 200 Lothrop St.
> Pgh, PA 15261
>
>
> On Nov 17, 2010, at 8:29 AM, Sarah Chacko wrote:
>
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>>
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy>> *****
>>
>> Hello,
>>
>> I am using a Nikon A1 confocal and I am trying to take a time series over several hours, scanning across channels in a microfluidic chamber - my channels are 20 fields of view across, there are three channels.
>>
>> I am finding that even with one long (20 fields of view) image across one channel the focus is lost half way across, and moving to another region nearly always loses focus.
>>
>> I could increase the pinhole or take a small z stack so that being out of focus would matter less - I'm looking at bacteria ~1 micron in size, but don't need to resolve anything smaller, just distinguish between them using fluorescent labels.
>>
>> I wondered whether in principle this should work, using long image and PFS with several starting points, and if anyone has something similar working well, or whether I am expecting too much.
>>
>> Sarah
>> (student)
Note: This message may contain confidential information. If this Email/Fax has been sent to you by mistake, please notify the sender and delete it immediately. Thank you.