Ludl MAC5000 Z focus walking off during time lapse

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Ludl MAC5000 Z focus walking off during time lapse

nranthony
Hi all, I hope you can help.

I'm experiencing a focal drift from a Ludl MAC5000 setup whereby each time point appears to request the stage position, and this causes the z focus to move in increasing amounts.  On a time lapse, if I rest my finger on the stepper motor controlled dial, I feel nothing for the first few frames, and then before each time point it shudders a little, and after many time points the step gets larger and larger.  If I then abort the acquisition the stage jumps back to where it had started.

I'm running Win 7 32bit and MM ver 2.0.0-beta3 20170215.  I've attached a log of the last set of experiments.
CoreLog20170303T120529_pid2892.txt

I'm downloading the latest MM2 to give that a go, but any pointers would be much appreciated.  Please let me know if you need me to do any tests or have any questions.

Thanks
Neil

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Ludl MAC5000 Z focus walking off during time lapse

nranthony
Just to follow up, I installed nightly MM2 beta 20170305 32bit, and the issue appears to be reduced, but still present.  During a timelapse I see the z-position fluctuate now (instead of walking off in one direction), and on ending the timelapse I still feel a little jump back to the original position from the stepper motor.

Any ideas?  I'm going to chop away at the config to see if I can determine a combination that provides any clues...

Thanks
Neil
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Ludl MAC5000 Z focus walking off during time lapse

nranthony
It turns out that I got some different results depending on a few physical things I tried.

The Ludl z focus control has a heidenhain position sensor on the side (a plunger that references between the body and the position of the objective turret offset), and I think the system is trying to keep things stable while this reference is drifting a little.

Also, there's a red button on the end of the focus control stepper motor that appears to lock the position.  It's kinda intermittent, but I think it's a toggle for focus controlled vs manual focus.  It appears to 'let go' of the dial to some degree.  With this turn on, the wheel locks to one of the steps and this appears to help reduce the amount that things walk.  The system will still step back just a little at the end of the acquisition, so it's still keeping track of things as it moves and tries to move back.

Another thing that hopefully made a big improvement (I say hopefully, as some of this might be coincidental) is the fine/medium/coarse switch on the back of the xyz joystick/focus unit.  In the fine position it appears to walk less.

Also, to throw more things in the mix.  I think that I'm getting better results by always focusing up to the position I need instead of down.  I assume this is a standard gear backlash thing.

Thanks
Neil
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Ludl MAC5000 Z focus walking off during time lapse

m.losen
Dear Neal,

A few ideas:
If you're working with a fixed z-position during acquisition: Have you tried disabling the z-position form the position list? Or disabling the z-axis from the configuration?

Theoretically, the system has two options to control the z-position: Either just give directions to the stepper motor (blindly) or use the measurement of the Heidenhain linear encoder for a feedback control to operate the stepper motor.
I'm not sure in which mode MM is operating the stage.

A few thoughts & info of the z-stage below in line, I have the same stage here.
 
"The Ludl z focus control has a heidenhain position sensor on the side (a plunger that references between the body and the position of the objective turret offset), and I think the system is trying to keep things stable while this reference is drifting a little."

        Yes, that's correct. The Heidenhain sensor is connected directly to the stage controller (not the PC). I'm not sure if the MM device adapter makes use of this functionality. This should be easy to test by manually lifting the plunger a bit and see if the z-stepper motor tries to correct in the other direction when using MM."

"Also, there's a red button on the end of the focus control stepper motor that appears to lock the position.  It's kinda intermittent, but I think it's a toggle for focus controlled vs manual focus.  It appears to 'let go'
of the dial to some degree.  With this turn on, the wheel locks to one of the steps and this appears to help reduce the amount that things walk.  The system will still step back just a little at the end of the acquisition, so it's still keeping track of things as it moves and tries to move back."
       
        Yes, this button controls the torque of the motor (by keeping high current even if the motor is not turning). In high torque mode any slip of the z-axis should be much lower, but more heat is produced.

"Another thing that hopefully made a big improvement (I say hopefully, as some of this might be coincidental) is the fine/medium/coarse switch on the back of the xyz joystick/focus unit.  In the fine position it appears to walk less."
       
        During startup of the stage controller it reads the "0" position from the joystick, and presumably also the focus wheel. If for example the joystick springs are worn-out and the joystick is not centered well during startup, the x/y position will continually drift afterwards. I imagine something similar could happen to the z-axis.  

"Also, to throw more things in the mix.  I think that I'm getting better results by always focusing up to the position I need instead of down.  I assume this is a standard gear backlash thing."
        Maybe the Haidenhain sensor is not moving smoothly?


Hope this helps,
Mario


------------------------------------------------------------------------------
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Ludl MAC5000 Z focus walking off during time lapse

nranthony
Hi Mario, thanks for the details.

I tried moving the plunger during a timelapse, and then let it back to the original position, still during the timelapse, and once it had finished it moved the focus all the way to what appears to the maximum distance I moved it.  I'm guessing it compensates for any movement it experiences, but I don't understand why it would move all that way even after I returned the plunger during a single timelapse...  Kinda weird, but good to know that plunger is the main metric.

Good to know that the red button is the torque option.  I'll keep an eye out for any warming up etc.

Good point about the z dial being similar to the xy getting old.  I'll see if I can tape it or clamp it to see how much movement in the dial is affecting things.  Good tip ;)

I wonder if a little tiny touch of WD40 might help out?



Thanks
Neil


-----Original Message-----
From: Losen M (NP) [mailto:[hidden email]]
Sent: Wednesday, March 08, 2017 4:19 AM
To: 'Micro-Manager General' <[hidden email]>
Subject: Re: [micro-manager-general] Ludl MAC5000 Z focus walking off during time lapse

Dear Neal,

A few ideas:
If you're working with a fixed z-position during acquisition: Have you tried disabling the z-position form the position list? Or disabling the z-axis from the configuration?

Theoretically, the system has two options to control the z-position: Either just give directions to the stepper motor (blindly) or use the measurement of the Heidenhain linear encoder for a feedback control to operate the stepper motor.
I'm not sure in which mode MM is operating the stage.

A few thoughts & info of the z-stage below in line, I have the same stage here.

"The Ludl z focus control has a heidenhain position sensor on the side (a plunger that references between the body and the position of the objective turret offset), and I think the system is trying to keep things stable while this reference is drifting a little."

Yes, that's correct. The Heidenhain sensor is connected directly to the stage controller (not the PC). I'm not sure if the MM device adapter makes use of this functionality. This should be easy to test by manually lifting the plunger a bit and see if the z-stepper motor tries to correct in the other direction when using MM."

"Also, there's a red button on the end of the focus control stepper motor that appears to lock the position.  It's kinda intermittent, but I think it's a toggle for focus controlled vs manual focus.  It appears to 'let go'
of the dial to some degree.  With this turn on, the wheel locks to one of the steps and this appears to help reduce the amount that things walk.  The system will still step back just a little at the end of the acquisition, so it's still keeping track of things as it moves and tries to move back."

Yes, this button controls the torque of the motor (by keeping high current even if the motor is not turning). In high torque mode any slip of the z-axis should be much lower, but more heat is produced.

"Another thing that hopefully made a big improvement (I say hopefully, as some of this might be coincidental) is the fine/medium/coarse switch on the back of the xyz joystick/focus unit.  In the fine position it appears to walk less."

During startup of the stage controller it reads the "0" position from the joystick, and presumably also the focus wheel. If for example the joystick springs are worn-out and the joystick is not centered well during startup, the x/y position will continually drift afterwards. I imagine something similar could happen to the z-axis.

"Also, to throw more things in the mix.  I think that I'm getting better results by always focusing up to the position I need instead of down.  I assume this is a standard gear backlash thing."
Maybe the Haidenhain sensor is not moving smoothly?


Hope this helps,
Mario


------------------------------------------------------------------------------
Announcing the Oxford Dictionaries API! The API offers world-renowned dictionary content that is easy and intuitive to access. Sign up for an account today to start using our lexical data to power your apps and projects. Get started today and enter our developer competition.
http://sdm.link/oxford
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general

________________________________

This e-mail message (including any attachments) is for the sole use of
the intended recipient(s) and may contain confidential and privileged
information. If the reader of this message is not the intended
recipient, you are hereby notified that any dissemination, distribution
or copying of this message (including any attachments) is strictly
prohibited.

If you have received this message in error, please contact
the sender by reply e-mail message and destroy all copies of the
original message (including attachments).

------------------------------------------------------------------------------
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Ludl MAC5000 Z focus walking off during time lapse

m.losen
Dear Neil,

I tried if I could reproduce your findings with our system:
I used the "stage position list" with the functions "mark" and "go to" (not a timelapse) using the Ludl stage.
When the z-position is included in the position list (tick box at the bottom of the 'stage position list' is activated for the z-stage), the position of the Heidenhain sensor will affect what the stage does when "go to" is activated.
So the position reading is used by MM to control stage movement.
If the button "go to" is used multiple times without interfering with the Heidenhain sensor, the z-axis motor will still move miniscule amounts. (I would not expect this would affect the focus really).
I then disabled the checkbox for the z-drive in the 'stage position position list'and marked the position again.
This time hitting the button "move to" had no effect on the z-motor at all. I suspect this is what you might want to achieve.

This is also something you might want to check:
Apart from the z-motor is there is possibly a manual and independent course-focus on your microscope. If this is the case, there is typically a brake on it to prevent the stage sinking down by its own weight (by turning the coarse focus wheel; we had this problem once).
If this is the case, this could be the most simple explanation for the focus walking off, the Heidenhain sensor would actually correct for this problem by activating the focus drive connected to the fine focus.
It just needs adjustment of the brake.

Greetings,
Mario
 

------------------------------------------------------------------------------
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

RE: Ludl MAC5000 Z focus walking off during time lapse

nranthony

Hi Mario, thanks for your time taken to look at these things.

 

Thanks for those details on the GoTo and z-stage details.  I’ll check that things match here for completeness.

 

Also, thanks for the reminder on the course focus break.  I’ll definitely double check that.

 

Thanks

Neil

 

From: m.losen [via Micro-Manager] [mailto:ml-node+[hidden email]]
Sent: Friday, March 10, 2017 3:45 PM
To: Anthony, Neil <[hidden email]>
Subject: Re: Ludl MAC5000 Z focus walking off during time lapse

 

Dear Neil,

I tried if I could reproduce your findings with our system:
I used the "stage position list" with the functions "mark" and "go to" (not a timelapse) using the Ludl stage.
When the z-position is included in the position list (tick box at the bottom of the 'stage position list' is activated for the z-stage), the position of the Heidenhain sensor will affect what the stage does when "go to" is activated.
So the position reading is used by MM to control stage movement.
If the button "go to" is used multiple times without interfering with the Heidenhain sensor, the z-axis motor will still move miniscule amounts. (I would not expect this would affect the focus really).
I then disabled the checkbox for the z-drive in the 'stage position position list'and marked the position again.
This time hitting the button "move to" had no effect on the z-motor at all. I suspect this is what you might want to achieve.

This is also something you might want to check:
Apart from the z-motor is there is possibly a manual and independent course-focus on your microscope. If this is the case, there is typically a brake on it to prevent the stage sinking down by its own weight (by turning the coarse focus wheel; we had this problem once).
If this is the case, this could be the most simple explanation for the focus walking off, the Heidenhain sensor would actually correct for this problem by activating the focus drive connected to the fine focus.
It just needs adjustment of the brake.

Greetings,
Mario
 

------------------------------------------------------------------------------
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general


If you reply to this email, your message will be added to the discussion below:

http://micro-manager.3463995.n2.nabble.com/Ludl-MAC5000-Z-focus-walking-off-during-time-lapse-tp7588190p7588220.html

To unsubscribe from Ludl MAC5000 Z focus walking off during time lapse, click here.
NAML




This e-mail message (including any attachments) is for the sole use of
the intended recipient(s) and may contain confidential and privileged
information. If the reader of this message is not the intended
recipient, you are hereby notified that any dissemination, distribution
or copying of this message (including any attachments) is strictly
prohibited.

If you have received this message in error, please contact
the sender by reply e-mail message and destroy all copies of the
original message (including attachments).
Loading...