Perfect Focus and MDA

classic Classic list List threaded Threaded
16 messages Options
Reply | Threaded
Open this post in threaded view
|

Perfect Focus and MDA

Kyle Karhohs
Hi,
Does anyone know if there are known issues or pitfalls with PFS and multi-dimensional acquisition? The last time I attempted to use MDA with PFS offset set for each position it did not seem to work. All the settings were configured correctly, as far as could tell, but the microscope went out of focus anyways after only an hour of imaging. I hear the PFS activate and deactivate at each position, but the focus is still off. Has anyone else had troubles like these?
Thanks,
Kyle

------------------------------------------------------------------------------

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Knecht, David
I have also had problems with PFS in MM.  What I normally do is:
1. Get the sample in focus
2. Turn on PFS and refocus with the PFS wheel then turn off PFS
3.  Record each position, with PFS off and using Z drive to set position of needed.  Only the XY and Nikon Z selected in the check boxes (not PFS unless you want a different relative Z position for each sample which you usually don’t want).
4.  Go back to position 1 and make sure PFS is off.   This is really important.  I find that if it is not set up at the beginning position with PFS off, the sequence gets screwed up and does not work.
5. Run Acquire

This usually works, but I still find I get focal drift over time sometimes.  Dave


On Mar 30, 2014, at 12:37 PM, Kyle Karhohs <[hidden email]> wrote:

Hi,
Does anyone know if there are known issues or pitfalls with PFS and multi-dimensional acquisition? The last time I attempted to use MDA with PFS offset set for each position it did not seem to work. All the settings were configured correctly, as far as could tell, but the microscope went out of focus anyways after only an hour of imaging. I hear the PFS activate and deactivate at each position, but the focus is still off. Has anyone else had troubles like these?
Thanks,
Kyle
------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general

David Knecht, Ph.D.    
Professor and Head of Core Microscopy Facility
Department of Molecular and Cell Biology
U-3125
91 N. Eagleville Rd.
University of Connecticut
Storrs, CT 06269
860-486-2200
860-486-4331 (fax)






------------------------------------------------------------------------------

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Kurt Thorn




On 3/31/2014 5:56 AM, Knecht, David wrote:
I have also had problems with PFS in MM.  What I normally do is:
1. Get the sample in focus
2. Turn on PFS and refocus with the PFS wheel then turn off PFS
3.  Record each position, with PFS off and using Z drive to set position of needed.  Only the XY and Nikon Z selected in the check boxes (not PFS unless you want a different relative Z position for each sample which you usually don’t want).
4.  Go back to position 1 and make sure PFS is off.   This is really important.  I find that if it is not set up at the beginning position with PFS off, the sequence gets screwed up and does not work.
5. Run Acquire


There are some issues with the PFS in the current Micro-Manager release - I just worked with the Micro-manager team to document them and hopefully they will be fixed soon. 

Currently, multidimensional acquisition with the PFS behaves very differently when the acquisition is started with the PFS locked or unlocked. If started with the PFS locked, the PFS is not properly synchronized with the acquisition, and the autofocus essentially does not work (I can provide more details if your interested). So the critical issue is to make sure the PFS is unlocked when you start the acquisition. 

Kurt

This usually works, but I still find I get focal drift over time sometimes.  Dave


On Mar 30, 2014, at 12:37 PM, Kyle Karhohs <[hidden email]> wrote:

Hi,
Does anyone know if there are known issues or pitfalls with PFS and multi-dimensional acquisition? The last time I attempted to use MDA with PFS offset set for each position it did not seem to work. All the settings were configured correctly, as far as could tell, but the microscope went out of focus anyways after only an hour of imaging. I hear the PFS activate and deactivate at each position, but the focus is still off. Has anyone else had troubles like these?
Thanks,
Kyle
------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general

David Knecht, Ph.D.    
Professor and Head of Core Microscopy Facility
Department of Molecular and Cell Biology
U-3125
91 N. Eagleville Rd.
University of Connecticut
Storrs, CT 06269
860-486-2200
860-486-4331 (fax)







------------------------------------------------------------------------------


_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general


-- 
Kurt Thorn
Director, Nikon Imaging Center
http://nic.ucsf.edu/blog/

------------------------------------------------------------------------------

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Sam Lord
In reply to this post by Kyle Karhohs
Kyle Karhohs wrote
I hear the PFS activate
and deactivate at each position, but the focus is still off.
I've found that unchecking the "Autofocus" altogether in the MDA window is typically the best option for multi-position imaging with PFS. Instead of turning off and turning back on between each position, PFS stays on and simply maintains the objective in the same position relative to the sample as the stage moves around. (For this to work, I think it's best to not include z in the positions list, so just uncheck that option at the bottom of the positions window.)

Of course, PFS can still fail, especially if you move the stage a very long way or run out of oil. But that's probably going to happen no matter what.

But it sounds like Kurt has also found some specific issues.
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Sam Lord
Sam Lord wrote
I've found that unchecking the "Autofocus" altogether in the MDA window is typically the best option for multi-position imaging with PFS.
One more detail: do not include the nosepiece in the group, or else PFS will not stay on.
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Knecht, David
In reply to this post by Sam Lord

On Mar 31, 2014, at 1:23 PM, Sam Lord <[hidden email]> wrote:

Kyle Karhohs wrote
I hear the PFS activate
and deactivate at each position, but the focus is still off.

I've found that unchecking the "Autofocus" altogether in the MDA window is
typically the best option for multi-position imaging with PFS. Instead of
turning off and turning back on between each position, PFS stays on and
simply maintains the objective in the same position relative to the sample
as the stage moves around. (For this to work, I think it's best to not
include z in the positions list, so just uncheck that option at the bottom
of the positions window.)

Of course, PFS can still fail, especially if you move the stage a very long
way or run out of oil. But that's probably going to happen no matter what.

PFS routinely fails with our slide chambers if we leave it on all the time, which is why we have it go off and on.

But it sounds like Kurt has also found some specific issues.



--
View this message in context: http://micro-manager.3463995.n2.nabble.com/Perfect-Focus-and-MDA-tp7582535p7582540.html
Sent from the Micro-Manager mailing list archive at Nabble.com.

------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general

David Knecht, Ph.D.    
Professor and Head of Core Microscopy Facility
Department of Molecular and Cell Biology
U-3125
91 N. Eagleville Rd.
University of Connecticut
Storrs, CT 06269
860-486-2200
860-486-4331 (fax)






------------------------------------------------------------------------------

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Knecht, David
In reply to this post by Kurt Thorn

On Mar 31, 2014, at 12:52 PM, Kurt Thorn <[hidden email]> wrote:





On 3/31/2014 5:56 AM, Knecht, David wrote:
I have also had problems with PFS in MM.  What I normally do is:
1. Get the sample in focus
2. Turn on PFS and refocus with the PFS wheel then turn off PFS
3.  Record each position, with PFS off and using Z drive to set position of needed.  Only the XY and Nikon Z selected in the check boxes (not PFS unless you want a different relative Z position for each sample which you usually don’t want).
4.  Go back to position 1 and make sure PFS is off.   This is really important.  I find that if it is not set up at the beginning position with PFS off, the sequence gets screwed up and does not work.
5. Run Acquire


There are some issues with the PFS in the current Micro-Manager release - I just worked with the Micro-manager team to document them and hopefully they will be fixed soon. 

Currently, multidimensional acquisition with the PFS behaves very differently when the acquisition is started with the PFS locked or unlocked. If started with the PFS locked, the PFS is not properly synchronized with the acquisition, and the autofocus essentially does not work (I can provide more details if your interested). So the critical issue is to make sure the PFS is unlocked when you start the acquisition. 

Kurt
Hi Kurt- Not sure what issues you sent them.  The start synchronization problem we seem to agree on.  Why we still get focus drift over time I don’t understand.  I would have expected that if the Z position of the sample changed due to temperature changes, PFS would correct for that by setting the z position relative to the current bottom of the sample.  Thus recording the Nikon z position should not be needed if PFS is doing its job.  I am still confused on whether the Nikon Z should even be recorded when setting up the positions.  As I understand PFS, it should not matter unless the Z position has moved so far that you are out of PFS range when PFS comes on.  

This usually works, but I still find I get focal drift over time sometimes.  Dave


On Mar 30, 2014, at 12:37 PM, Kyle Karhohs <[hidden email]> wrote:

Hi,
Does anyone know if there are known issues or pitfalls with PFS and multi-dimensional acquisition? The last time I attempted to use MDA with PFS offset set for each position it did not seem to work. All the settings were configured correctly, as far as could tell, but the microscope went out of focus anyways after only an hour of imaging. I hear the PFS activate and deactivate at each position, but the focus is still off. Has anyone else had troubles like these?
Thanks,
Kyle
------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general

David Knecht, Ph.D.    
Professor and Head of Core Microscopy Facility
Department of Molecular and Cell Biology
U-3125
91 N. Eagleville Rd.
University of Connecticut
Storrs, CT 06269
860-486-2200
860-486-4331 (fax)







------------------------------------------------------------------------------


_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general


-- 
Kurt Thorn
Director, Nikon Imaging Center
http://nic.ucsf.edu/blog/
------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general

David Knecht, Ph.D.    
Professor and Head of Core Microscopy Facility
Department of Molecular and Cell Biology
U-3125
91 N. Eagleville Rd.
University of Connecticut
Storrs, CT 06269
860-486-2200
860-486-4331 (fax)






------------------------------------------------------------------------------

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Sam Lord
In reply to this post by Knecht, David
Knecht, David wrote
PFS routinely fails with our slide chambers if we leave it on all the time, which is why we have it go off and on.
I see. Thanks for your solution. I'm going to write that down in case we start having issues with PFS in the future.
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Kurt Thorn
On 3/31/2014 11:19 AM, Sam Lord wrote:
> Knecht, David wrote
>> PFS routinely fails with our slide chambers if we leave it on all the
>> time, which is why we have it go off and on.
> I see. Thanks for your solution. I'm going to write that down in case we
> start having issues with PFS in the future.

My guess is that the PFS fails here because of the refractive index
change associated with the chamber walls.  Particularly if you are using
an oil or water-immersion lens, when you cross into the well of a
multiwell plate or chambered coverslip, the PFS will fail because the
reflection of the IR beam from the glass-water interface is lost.

Kurt

>
>
>
> --
> View this message in context: http://micro-manager.3463995.n2.nabble.com/Perfect-Focus-and-MDA-tp7582535p7582544.html
> Sent from the Micro-Manager mailing list archive at Nabble.com.
>
> ------------------------------------------------------------------------------
> _______________________________________________
> micro-manager-general mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/micro-manager-general
>
>


--
Kurt Thorn
Director, Nikon Imaging Center
http://nic.ucsf.edu/blog/



------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Kurt Thorn
In reply to this post by Knecht, David
On 3/31/2014 11:17 AM, Knecht, David wrote:

On Mar 31, 2014, at 12:52 PM, Kurt Thorn <[hidden email]> wrote:





On 3/31/2014 5:56 AM, Knecht, David wrote:
I have also had problems with PFS in MM.  What I normally do is:
1. Get the sample in focus
2. Turn on PFS and refocus with the PFS wheel then turn off PFS
3.  Record each position, with PFS off and using Z drive to set position of needed.  Only the XY and Nikon Z selected in the check boxes (not PFS unless you want a different relative Z position for each sample which you usually don’t want).
4.  Go back to position 1 and make sure PFS is off.   This is really important.  I find that if it is not set up at the beginning position with PFS off, the sequence gets screwed up and does not work.
5. Run Acquire


There are some issues with the PFS in the current Micro-Manager release - I just worked with the Micro-manager team to document them and hopefully they will be fixed soon. 

Currently, multidimensional acquisition with the PFS behaves very differently when the acquisition is started with the PFS locked or unlocked. If started with the PFS locked, the PFS is not properly synchronized with the acquisition, and the autofocus essentially does not work (I can provide more details if your interested). So the critical issue is to make sure the PFS is unlocked when you start the acquisition. 

Kurt
Hi Kurt- Not sure what issues you sent them.  The start synchronization problem we seem to agree on.  Why we still get focus drift over time I don’t understand.  I would have expected that if the Z position of the sample changed due to temperature changes, PFS would correct for that by setting the z position relative to the current bottom of the sample.  Thus recording the Nikon z position should not be needed if PFS is doing its job.  I am still confused on whether the Nikon Z should even be recorded when setting up the positions.  As I understand PFS, it should not matter unless the Z position has moved so far that you are out of PFS range when PFS comes on. 

I think you should always record the Nikon Z.  If your positions are widely separated, it's possible that the Z coordinate will change enough from one position to the next that it will fall outside of the locking range of the PFS.  Furthermore, the Z coordinate will be updated after each autofocus cycle so that at the next time point the microscope always goes to the previously best found Z.  Otherwise, over time the position can drift out of the PFS locking range. Updating the Z coordinate is critical for maintaining PFS lock in long time courses or in samples with lots of movement.

Here's what I reported to Micro-manager (Mark has reproduced this and has more details):

On a Ti with PFS and a piezo Z-stage, set the Core-Focus to the Ti Z drive. This allows us to move the stage using the piezo to simulate focus drift.  Set up a multi-point acquisition with two points using the PFS, and start acquisition of these two points in a timelapse with 10 sec delay.  At some point during the wait period in the acquisition, move the piezo Z stage.

The behavior differs depending on whether the experiment is started with the PFS locked or not.

 If the PFS is not locked at the experiment start, Micro-manager behaves as I expect: after the piezo stage move, when the PFS locks, the Ti Z drive moves to the new focused position for each point. At subsequent time points, the Ti Z drive moves to the previously found focused position, and all images are captured in focus.

If the PFS is locked at the experiment start, Micro-manager does not update the Ti Z drive position after the piezo stage move and always goes to the initially programmed Z position. Additionally, although the PFS locks, the images are captured out of focus.

Kurt



This usually works, but I still find I get focal drift over time sometimes.  Dave


On Mar 30, 2014, at 12:37 PM, Kyle Karhohs <[hidden email]> wrote:

Hi,
Does anyone know if there are known issues or pitfalls with PFS and multi-dimensional acquisition? The last time I attempted to use MDA with PFS offset set for each position it did not seem to work. All the settings were configured correctly, as far as could tell, but the microscope went out of focus anyways after only an hour of imaging. I hear the PFS activate and deactivate at each position, but the focus is still off. Has anyone else had troubles like these?
Thanks,
Kyle
------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general

David Knecht, Ph.D.    
Professor and Head of Core Microscopy Facility
Department of Molecular and Cell Biology
U-3125
91 N. Eagleville Rd.
University of Connecticut
Storrs, CT 06269
860-486-2200
860-486-4331 (fax)







------------------------------------------------------------------------------


_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general


-- 
Kurt Thorn
Director, Nikon Imaging Center
http://nic.ucsf.edu/blog/
------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general

David Knecht, Ph.D.    
Professor and Head of Core Microscopy Facility
Department of Molecular and Cell Biology
U-3125
91 N. Eagleville Rd.
University of Connecticut
Storrs, CT 06269
860-486-2200
860-486-4331 (fax)







------------------------------------------------------------------------------


_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general


-- 
Kurt Thorn
Director, Nikon Imaging Center
http://nic.ucsf.edu/blog/

------------------------------------------------------------------------------

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Mark Tsuchida-2
Hi All,

On Mon, Mar 31, 2014 at 11:35:54AM -0700, Kurt Thorn wrote:

> On 3/31/2014 11:17 AM, Knecht, David wrote:
> >On Mar 31, 2014, at 12:52 PM, Kurt Thorn <[hidden email]> wrote:
> >>On 3/31/2014 5:56 AM, Knecht, David wrote:
> >>>I have also had problems with PFS in MM.  What I normally do is:
> >>>1. Get the sample in focus
> >>>2. Turn on PFS and refocus with the PFS wheel then turn off PFS
> >>>3.  Record each position, with PFS off and using Z drive to set
> >>>position of needed.  Only the XY and Nikon Z selected in the check
> >>>boxes (not PFS unless you want a different relative Z position for
> >>>each sample which you usually don't want).
> >>>4.  Go back to position 1 and make sure PFS is off. This is really
> >>>important.  I find that if it is not set up at the beginning position
> >>>with PFS off, the sequence gets screwed up and does not work.
> >>>5. Run Acquire
> >>
> >>There are some issues with the PFS in the current Micro-Manager release
> >>- I just worked with the Micro-manager team to document them and
> >>hopefully they will be fixed soon.
> >>
> >>Currently, multidimensional acquisition with the PFS behaves very
> >>differently when the acquisition is started with the PFS locked or
> >>unlocked. If started with the PFS locked, the PFS is not properly
> >>synchronized with the acquisition, and the autofocus essentially does
> >>not work (I can provide more details if your interested). So the
> >>critical issue is to make sure the PFS is unlocked when you start the
> >>acquisition.

Just to clarify, MDA behaves very differently depending on whether the
PFS is on when starting _if_ you have Autofocus checked in the MDA
window (with the current autofocus being set to PFSStatus).

The behavior when starting an MDA-with-autofocus with the PFS already on
is buggy.

> >Hi Kurt- Not sure what issues you sent them.  The start synchronization
> >problem we seem to agree on.  Why we still get focus drift over time I
> >don't understand.  I would have expected that if the Z position of the
> >sample changed due to temperature changes, PFS would correct for that by
> >setting the z position relative to the current bottom of the sample.  Thus
> >recording the Nikon z position should not be needed if PFS is doing its
> >job.  I am still confused on whether the Nikon Z should even be recorded
> >when setting up the positions.  As I understand PFS, it should not matter
> >unless the Z position has moved so far that you are out of PFS range when
> >PFS comes on.
>
> I think you should always record the Nikon Z.  If your positions are widely
> separated, it's possible that the Z coordinate will change enough from one
> position to the next that it will fall outside of the locking range of the
> PFS.  Furthermore, the Z coordinate will be updated after each autofocus
> cycle so that at the next time point the microscope always goes to the
> previously best found Z. Otherwise, over time the position can drift out of
> the PFS locking range. Updating the Z coordinate is critical for maintaining
> PFS lock in long time courses or in samples with lots of movement.
>
> Here's what I reported to Micro-manager (Mark has reproduced this and has
> more details):
>
> On a Ti with PFS and a piezo Z-stage, set the Core-Focus to the Ti Z drive.
> This allows us to move the stage using the piezo to simulate focus drift.
> Set up a multi-point acquisition with two points using the PFS, and start
> acquisition of these two points in a timelapse with 10 sec delay.  At some
> point during the wait period in the acquisition, move the piezo Z stage.
>
> The behavior differs depending on whether the experiment is started with the
> PFS locked or not.
>
>  If the PFS is not locked at the experiment start, Micro-manager behaves as
> I expect: after the piezo stage move, when the PFS locks, the Ti Z drive
> moves to the new focused position for each point. At subsequent time points,
> the Ti Z drive moves to the previously found focused position, and all
> images are captured in focus.
>
> If the PFS is locked at the experiment start, Micro-manager does not update
> the Ti Z drive position after the piezo stage move and always goes to the
> initially programmed Z position. Additionally, although the PFS locks, the
> images are captured out of focus.

I have very little to add to this, other than very technical internal
details. In the case that the PFS is locked at the start, what actually
happens is that the Z position is overridden by the Z drive position in
the position list, _after_ the PFS had locked in the right position.
This is the bug, and the workaround is to simply make sure to turn off
the PFS when running an MDA using the MDA's Autofocus feature.

In summary, there are mainly two ways to use the PFS with Micro-Manager
MDA (in 1.4.16):

1. Continuous. Turn on PFS on the microscope stand and uncheck Autofocus
in the MDA. This works well for simple samples (e.g. a single coverslip)
when you are imaging a simple XY position, or multiple XY positions with
moderate Z change between positions. There must also be no
focus-disrupting barriers between positions (such as well edges).

2. Per-frame. Check Autofocus in the MDA windows, configure Autofocus to
use PFSStatus, and make sure to turn off PFS on the microscope stand
before starting. If using multiple XY positions, make sure to include
the Z drive in the position list (see Kurt's explanation above).
Including the PFSOffset in the position list is optional (include only
if you want different offsets per position).

Please let us know if you find an edge case that is not covered by these
two. In particular, I have not recently experimented with PFS in
combination with Z stacks.

As for fixing the bug, we are contemplating a quick fix that will
unconditionally turn off the PFS ("continuous autofocus" is
Micro-Manager's general term, applicable also to other autofocus
devices) at the beginning of the MDA if Autofocus is checked. Does this
seem reasonable to everybody?

A longer-term solution might be to add radio buttons (Continuous vs
Per-frame/position) to the MDA window's Autofocus section. The
continuous option would only be available for hardware autofocus devices
that support that mode of operation.

Best,
Mark


--
Mark Tsuchida
Micro-Manager Team (USCF Vale Lab)

------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Sam Lord
Mark Tsuchida-2 wrote
As for fixing the bug, we are contemplating a quick fix that will
unconditionally turn off the PFS ("continuous autofocus" is
Micro-Manager's general term, applicable also to other autofocus
devices) at the beginning of the MDA if Autofocus is checked. Does this
seem reasonable to everybody?
Yes, that seems reasonable. Basically, this would treat the first position as it does all subsequent positions, by turning off then back on the PFS, right?

A longer-term solution might be to add radio buttons (Continuous vs
Per-frame/position) to the MDA window's Autofocus section. The
continuous option would only be available for hardware autofocus devices
that support that mode of operation.
I like this solution. I think that the Autofocus option is confusing for many, because if you want continuous PFS, you must *uncheck* Autofocus in the MDA. That's counter-intuitive for many. Having that radio button options makes sense. But how would that radio button behave if you choose a software-based autofocus instead of PFS?
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Mark Tsuchida-2
Hi Sam,

On Tue, Apr 01, 2014 at 11:15:51AM -0700, Sam Lord wrote:

> Mark Tsuchida-2 wrote
> > As for fixing the bug, we are contemplating a quick fix that will
> > unconditionally turn off the PFS ("continuous autofocus" is
> > Micro-Manager's general term, applicable also to other autofocus
> > devices) at the beginning of the MDA if Autofocus is checked. Does this
> > seem reasonable to everybody?
>
> Yes, that seems reasonable. Basically, this would treat the first position
> as it does all subsequent positions, by turning off then back on the PFS,
> right?

Yes.

> > A longer-term solution might be to add radio buttons (Continuous vs
> > Per-frame/position) to the MDA window's Autofocus section. The
> > continuous option would only be available for hardware autofocus devices
> > that support that mode of operation.
>
> I like this solution. I think that the Autofocus option is confusing for
> many, because if you want continuous PFS, you must *uncheck* Autofocus in
> the MDA. That's counter-intuitive for many. Having that radio button options
> makes sense. But how would that radio button behave if you choose a
> software-based autofocus instead of PFS?

The idea would be to gray out the "Continuous" radio button when the
current autofocus method is a software autofocus plugin or is otherwise
incapable of continuous operation. The reason why this is a
"longer-term" solution is that there is currently no reliable way to
determine this.

Best,
Mark

--
Mark Tsuchida
Micro-Manager Team (USCF Vale Lab)

------------------------------------------------------------------------------
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

JB Lugagne
Hi,

I am using the MMCore API, and in particular the enableContinuousFocus()
function with the Nikon PFS system on a Ti2. Unfortunately, when I try to
set the Z position with the setPosition function, the PFS is disabled and I
have to reactivate it with enableContinuousFocus again. Is there a way to
avoid this? I am not sure if this is slowing down my acquisition or not, but
since I am trying to acquire hundreds of positions every 5 minutes every
millisecond counts for me.

Thank you,
Jean-Baptiste



--
Sent from: http://micro-manager.3463995.n2.nabble.com/


_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

Nico Stuurman-2
In reply to this post by Kyle Karhohs
Hi Jean-Baptiste,

> I am using the MMCore API, and in particular the enableContinuousFocus()
> function with the Nikon PFS system on a Ti2. Unfortunately, when I try to
> set the Z position with the setPosition function, the PFS is disabled
> and I
> have to reactivate it with enableContinuousFocus again. Is there a way to
> avoid this? I am not sure if this is slowing down my acquisition or
> not, but
> since I am trying to acquire hundreds of positions every 5 minutes every
> millisecond counts for me.

I have not worked with the Ti2, but with the Ti, this is baked into the
microscope firmware: any movement of the Z-position disables the PFS.  I
suspect that the same is true for the Ti2. Since you want the PFS
activated in any case, can you simply skip the Z position movement, and
rely on the PFS to establish the new position?

Best,

Nico




_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|

Re: Perfect Focus and MDA

JB Lugagne
Hi Nico,

I see, that's what I suspected.
If I understand how the Z-Drive and the PFS interact correctly, since I will
be covering a large surface there is a high chance that I will eventually
move out of the PFS range no?
In this case I can try a hybrid approach of adjusting the Z position every
few positions.

Thank you,
Jean-Baptiste



--
Sent from: http://micro-manager.3463995.n2.nabble.com/


_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general