DIY solution to spinning disc banding artifact

[Gary Laevsky]

***** To join, leave or search the confocal microscopy listserv, go to: http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy Post images on http://www.imgur.com and include the link in your posting. *****

Hi All,

Well known artifact result of mismatched exposure and disc field
acquisitions.

Perking Elmer used to have a "solution" using the ProSynch box (or
something like that).

We're using a CSU21 that has the synch BNC cable output.

We have triggerable cameras (sCMOS (both Flash and Zyla)).

We're using Elements, so we can't exactly fine tune the exposures per acqusition (can any software?).
We're using a macro to spin up the disc to 5K.  We can change the speed of
the disc, but momentum happens, so we can't do it instantaneously.

We need a way to synch the camera with the disc.  Acquisitions are sub 30
ms.

I imaging we will take a signal off the disc and use this to trigger the camera.

Can someone please point me to a solution?

Thanks so much.

--

[George McNamara]

***** To join, leave or search the confocal microscopy listserv, go to: http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy Post images on http://www.imgur.com and include the link in your posting. *****

Hi Gary,

To go faster, you can also acquire a smaller region of interest ... orientation of ROI matters for sCMOS. If I recall correctly, FLASH ROI can be full width (X=2048), make Y less to go faster.

If NIS Elements does not let you acquire fast, try MetaMorph.

George

On 12/20/2016 5:00 AM, Gary Laevsky wrote:

***** To join, leave or search the confocal microscopy listserv, go to: http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy Post images on http://www.imgur.com and include the link in your posting. *****

Hi All,

Well known artifact result of mismatched exposure and disc field
acquisitions.

Perking Elmer used to have a "solution" using the ProSynch box (or
something like that).

We're using a CSU21 that has the synch BNC cable output.

We have triggerable cameras (sCMOS (both Flash and Zyla)).

We're using Elements, so we can't exactly fine tune the exposures per acqusition (can any software?).
We're using a macro to spin up the disc to 5K.  We can change the speed of
the disc, but momentum happens, so we can't do it instantaneously.

We need a way to synch the camera with the disc.  Acquisitions are sub 30
ms.

I imaging we will take a signal off the disc and use this to trigger the camera.

Can someone please point me to a solution?

Thanks so much.

--

[Nico Stuurman-3]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Garyand others,


Not sure about the BNC connector on the CSU21, but the one on the CSU10
was designed to adjustthe disk speed to a video synchronization signal.
The CSU10 manual suggests that the speed of the disk can be changed
slightly by providing synchronization pulses at around 30 Hz (the CSU10
spins at 1800 rpm= 30 rps). I have never used that input.

However, it does help to set the exposure time to exactly n times the
time it takes for the disk to scan the field of view .  The CSU10 covers
the field of view in 30 degrees, or 12 times per full rotation.  So, if
the disk spins at 30 rps, it will cover the field of view 360 times and
it takes 1/360 = 2.78 ms to cover the field of view once.  Since the
disk speed may be slightly different, it is best to play with the exact
exposure time (either using a slide that is fluorescent throughout, or
using trans-illumination) to find out exactly how long to expose the
camera to cover one field of view.  Once you have that number, simply
multiply by an integer to go to longer exposures.  I do not know about
Elements, but exposure time on the Flash 4 and Zyla can certainly be set
with sub-millisecond accuracy.  I wrote a script for our
Micro-Manager-based system to round the exposure time to the nearest
good value just by pressing a key on the keyboard.

With the Flash4.0 you then have the additional issue of dealing with the
rolling shutter.  You can work around that by setting up emulated
globalshutter (or whatever the term is for that approach these days):
use the ttl output that fires when the complete chip is exposing and
build an AND gate to your laser illumination.  That will of course add
10.0 ms to the exposure time you have to set.


Best,


Nico



On 12/20/16 3:00 AM, Gary Laevsky wrote:

> ***** To join, leave or search the confocal microscopy listserv, go
> to: http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy Post images
> on http://www.imgur.com and include the link in your posting. *****
> Hi All,
>
> Well known artifact result of mismatched exposure and disc field
> acquisitions.
>
> Perking Elmer used to have a "solution" using the ProSynch box (or
> something like that).
>
> We're using a CSU21 that has the synch BNC cable output.
>
> We have triggerable cameras (sCMOS (both Flash and Zyla)).
>
> We're using Elements, so we can't exactly fine tune the exposures per
> acqusition (can any software?).
> We're using a macro to spin up the disc to 5K.  We can change the speed of
> the disc, but momentum happens, so we can't do it instantaneously.
>
> We need a way to synch the camera with the disc.  Acquisitions are sub 30
> ms.
>
> I imaging we will take a signal off the disc and use this to trigger
> the camera.
>
> Can someone please point me to a solution?
>
> Thanks so much.
>
> --
> Best,
>
> Gary Laevsky, Ph.D.
> Director, Confocal Imaging Facility
> Nikon Center of Excellence
> Dept. of Molecular Biology
> Washington Rd.
> Princeton University
> Princeton, New Jersey, 08544-1014
> (O) 609 258 5432
> (C) 508 507 1310

[Guy Hagen]

***** To join, leave or search the confocal microscopy listserv, go to: http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy Post images on http://www.imgur.com and include the link in your posting. *****

Hello:

The trick is that the laser has to be off during camera readout. With the right logic, one can use the camera signals to control the laser AOTF. Then you should be fine as long as the camera exposure time is longer than 1 complete disk rotation.

best,

Guy Hagen

On Tue, Dec 20, 2016 at 11:35 AM, Nico Stuurman <[hidden email]> wrote:

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Garyand others,


Not sure about the BNC connector on the CSU21, but the one on the CSU10 was designed to adjustthe disk speed to a video synchronization signal. The CSU10 manual suggests that the speed of the disk can be changed slightly by providing synchronization pulses at around 30 Hz (the CSU10 spins at 1800 rpm= 30 rps). I have never used that input.

However, it does help to set the exposure time to exactly n times the time it takes for the disk to scan the field of view .  The CSU10 covers the field of view in 30 degrees, or 12 times per full rotation.  So, if the disk spins at 30 rps, it will cover the field of view 360 times and it takes 1/360 = 2.78 ms to cover the field of view once.  Since the disk speed may be slightly different, it is best to play with the exact exposure time (either using a slide that is fluorescent throughout, or using trans-illumination) to find out exactly how long to expose the camera to cover one field of view.  Once you have that number, simply multiply by an integer to go to longer exposures.  I do not know about Elements, but exposure time on the Flash 4 and Zyla can certainly be set with sub-millisecond accuracy.  I wrote a script for our Micro-Manager-based system to round the exposure time to the nearest good value just by pressing a key on the keyboard.

With the Flash4.0 you then have the additional issue of dealing with the rolling shutter.  You can work around that by setting up emulated globalshutter (or whatever the term is for that approach these days): use the ttl output that fires when the complete chip is exposing and build an AND gate to your laser illumination.  That will of course add 10.0 ms to the exposure time you have to set.


Best,


Nico

On 12/20/16 3:00 AM, Gary Laevsky wrote:

***** To join, leave or search the confocal microscopy listserv, go to: http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy Post images on http://www.imgur.com and include the link in your posting. *****
Hi All,

Well known artifact result of mismatched exposure and disc field
acquisitions.

Perking Elmer used to have a "solution" using the ProSynch box (or
something like that).

We're using a CSU21 that has the synch BNC cable output.

We have triggerable cameras (sCMOS (both Flash and Zyla)).

We're using Elements, so we can't exactly fine tune the exposures per acqusition (can any software?).
We're using a macro to spin up the disc to 5K.  We can change the speed of
the disc, but momentum happens, so we can't do it instantaneously.

We need a way to synch the camera with the disc.  Acquisitions are sub 30
ms.

I imaging we will take a signal off the disc and use this to trigger the camera.

Can someone please point me to a solution?

Thanks so much.

--
Best,

Gary Laevsky, Ph.D.
Director, Confocal Imaging Facility
Nikon Center of Excellence
Dept. of Molecular Biology
Washington Rd.
Princeton University
Princeton, New Jersey, 08544-1014
(O) <a href="tel:609%20258%205432" value="[phone]" target="_blank">609 258 5432
(C) <a href="tel:508%20507%201310" value="[phone]" target="_blank">508 507 1310

[Gary Laevsky]

***** To join, leave or search the confocal microscopy listserv, go to: http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy Post images on http://www.imgur.com and include the link in your posting. *****

Thanks for the advicel.

The triggered acquisition does blank the laser during readout.

My impression was the video synch IN to the disc was not very effective, as you can't change the speed of the disc instantaneously, which is what would probably be needed to synch with a camera running at 100 Hz.

I was hoping to find a way to change the camera exposures, dynamically, to match the disc.

You can change the exposure to sub-millisecond exposure in Elements, but as you say, the disc isn't always spinning at the same speed.  And those sub-millisecond tweaks do get rid of the artifact, for a time.

We're looking for a device that could read the disc speed, and trigger the camera.  I THINK the camera acts as master (through the dac) to aotf  blanking when triggering, so that may maintain the good gate during exposure.

If the BNC on the disc is a synch in, i need another way to read the disc.

On Tue, Dec 20, 2016 at 3:56 PM, Guy Hagen <[hidden email]> wrote:

***** To join, leave or search the confocal microscopy listserv, go to: http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy Post images on http://www.imgur.com and include the link in your posting. *****

Hello:

The trick is that the laser has to be off during camera readout. With the right logic, one can use the camera signals to control the laser AOTF. Then you should be fine as long as the camera exposure time is longer than 1 complete disk rotation.

best,

Guy Hagen

On Tue, Dec 20, 2016 at 11:35 AM, Nico Stuurman <[hidden email]> wrote:

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Garyand others,


Not sure about the BNC connector on the CSU21, but the one on the CSU10 was designed to adjustthe disk speed to a video synchronization signal. The CSU10 manual suggests that the speed of the disk can be changed slightly by providing synchronization pulses at around 30 Hz (the CSU10 spins at 1800 rpm= 30 rps). I have never used that input.

However, it does help to set the exposure time to exactly n times the time it takes for the disk to scan the field of view .  The CSU10 covers the field of view in 30 degrees, or 12 times per full rotation.  So, if the disk spins at 30 rps, it will cover the field of view 360 times and it takes 1/360 = 2.78 ms to cover the field of view once.  Since the disk speed may be slightly different, it is best to play with the exact exposure time (either using a slide that is fluorescent throughout, or using trans-illumination) to find out exactly how long to expose the camera to cover one field of view.  Once you have that number, simply multiply by an integer to go to longer exposures.  I do not know about Elements, but exposure time on the Flash 4 and Zyla can certainly be set with sub-millisecond accuracy.  I wrote a script for our Micro-Manager-based system to round the exposure time to the nearest good value just by pressing a key on the keyboard.

With the Flash4.0 you then have the additional issue of dealing with the rolling shutter.  You can work around that by setting up emulated globalshutter (or whatever the term is for that approach these days): use the ttl output that fires when the complete chip is exposing and build an AND gate to your laser illumination.  That will of course add 10.0 ms to the exposure time you have to set.


Best,


Nico

On 12/20/16 3:00 AM, Gary Laevsky wrote:

***** To join, leave or search the confocal microscopy listserv, go to: http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy Post images on http://www.imgur.com and include the link in your posting. *****
Hi All,

Well known artifact result of mismatched exposure and disc field
acquisitions.

Perking Elmer used to have a "solution" using the ProSynch box (or
something like that).

We're using a CSU21 that has the synch BNC cable output.

We have triggerable cameras (sCMOS (both Flash and Zyla)).

We're using Elements, so we can't exactly fine tune the exposures per acqusition (can any software?).
We're using a macro to spin up the disc to 5K.  We can change the speed of
the disc, but momentum happens, so we can't do it instantaneously.

We need a way to synch the camera with the disc.  Acquisitions are sub 30
ms.

I imaging we will take a signal off the disc and use this to trigger the camera.

Can someone please point me to a solution?

Thanks so much.

--
Best,

Gary Laevsky, Ph.D.
Director, Confocal Imaging Facility
Nikon Center of Excellence
Dept. of Molecular Biology
Washington Rd.
Princeton University
Princeton, New Jersey, 08544-1014
(O) <a href="tel:609%20258%205432" value="[phone]" target="_blank">609 258 5432
(C) <a href="tel:508%20507%201310" value="[phone]" target="_blank">508 507 1310

--