Features for a motorized stage (focus stacking)

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: Pau, rjlittlefield, ChrisR, Chris S.

jac B
Posts: 31
Joined: Tue Jul 14, 2020 2:01 am

Features for a motorized stage (focus stacking)

Post by jac B »

Hi,

I read that features for a motorized stage are e.q. travel range, velocity (max, min). acceleration (max, min, at a load), resolution, bidirectional repeatability, flatness and straightness over full travel range, horizontal and vertical load capacity (max), min achievable incremental movement, home location accuracy, absolute on-axis accuracy, backlash, etc.

It’s my intention to understand, not to discredit a product: Are the values for former features known for the Cognisys StackShot? Is it reasonable to ask for?

Regards,

elf
Posts: 1398
Joined: Sun Nov 18, 2007 12:10 pm

Re: Features for a motorized stage (focus stacking)

Post by elf »

It's reasonable, but you should first define your tolerances for each of those fields.

rjlittlefield
Site Admin
Posts: 21127
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Features for a motorized stage (focus stacking)

Post by rjlittlefield »

I am not aware of any studies for most of your parameters.

In that quite long list, the only value that I have ever cared much about is "min achievable incremental movement".

See https://www.photomacrography.net/forum/viewtopic.php?t=16323 for a study of that, using one of the earliest StackShot controllers.

More recent StackShot controllers of the same nominal model have notably better performance, due to using a newer motor control chip that gives much closer to pure sine/cosine drive currents. In particular, in High Precision mode, the largest microstep in each sequence is now much closer to the average step than it was on the old controller. I have never published the data, but checking my notes, "There's some wandering above/below a straight line, with the result that the biggest step is about 1.3X larger than the average. But it's only about 1.3X. With my old controller, that ratio was more like 3X, so it's really a big improvement in the limit of small steps." According to recent discussions with Cognisys, any units sold after June 15, 2015, should have the new chip with the better stepping.

The default value for backlash is set by Cognisys to 0.22 mm. I expect that will vary quite a bit from one unit to another. I don't worry about backlash because I always set limits and shoot the stack while moving in the same direction, so backlash is taken up by the time any important positions are reached.

--Rik

jac B
Posts: 31
Joined: Tue Jul 14, 2020 2:01 am

Re: Features for a motorized stage (focus stacking)

Post by jac B »

elf wrote:
Tue Aug 11, 2020 12:16 pm
It's reasonable, but you should first define your tolerances for each of those fields.
Hi elf,
Can you explain, why should I first define my tolerances for each of those fields for the StackShot, before that someone mention the values for those fields for the Stackshot?
Regards, jac

jac B
Posts: 31
Joined: Tue Jul 14, 2020 2:01 am

Re: Features for a motorized stage (focus stacking)

Post by jac B »

rjlittlefield wrote:
Tue Aug 11, 2020 12:51 pm
I am not aware of any studies for most of your parameters.

In that quite long list, the only value that I have ever cared much about is "min achievable incremental movement".

See https://www.photomacrography.net/forum/viewtopic.php?t=16323 for a study of that, using one of the earliest StackShot controllers.

More recent StackShot controllers of the same nominal model have notably better performance, due to using a newer motor control chip that gives much closer to pure sine/cosine drive currents. In particular, in High Precision mode, the largest microstep in each sequence is now much closer to the average step than it was on the old controller. I have never published the data, but checking my notes, "There's some wandering above/below a straight line, with the result that the biggest step is about 1.3X larger than the average. But it's only about 1.3X. With my old controller, that ratio was more like 3X, so it's really a big improvement in the limit of small steps." According to recent discussions with Cognisys, any units sold after June 15, 2015, should have the new chip with the better stepping.

The default value for backlash is set by Cognisys to 0.22 mm. I expect that will vary quite a bit from one unit to another. I don't worry about backlash because I always set limits and shoot the stack while moving in the same direction, so backlash is taken up by the time any important positions are reached.

--Rik
Hi Rik,

Thanks, I didn’t get a hit on that link, when searching the subject.

Maybe I misunderstand, but I have the impression that (most of) these parameters are used to compare motors and choose which motor(s) is/are under/best/over dimensioned for (a) certain focus stacking task(s)/application(s).

(known or uncertain) tolerances for backlash and (an) unknown value(s) for (an) other parameter(s) can be part of the comparison and choice for (a) motor(s).

What are your focus stacking task(s)/application(s), where only “min achievable incremental movement” is relevant?

Regards, jac

elf
Posts: 1398
Joined: Sun Nov 18, 2007 12:10 pm

Re: Features for a motorized stage (focus stacking)

Post by elf »

jac B wrote:
Tue Aug 11, 2020 1:17 pm
elf wrote:
Tue Aug 11, 2020 12:16 pm
It's reasonable, but you should first define your tolerances for each of those fields.
Hi elf,
Can you explain, why should I first define my tolerances for each of those fields for the StackShot, before that someone mention the values for those fields for the Stackshot?
Regards, jac
If you don't know what you want, how will you know what you need :shock: It's not a matter of defining them before someone else gives you the data, it's about how the data will relate to your requirements.

lothman
Posts: 494
Joined: Sat Feb 14, 2009 7:00 am
Location: Stuttgart/Germany

Re: Features for a motorized stage (focus stacking)

Post by lothman »

jac B wrote:
Tue Aug 11, 2020 7:48 am
Hi,

I read that features for a motorized stage are e.q. travel range, velocity (max, min). acceleration (max, min, at a load), resolution, bidirectional repeatability, flatness and straightness over full travel range, horizontal and vertical load capacity (max), min achievable incremental movement, home location accuracy, absolute on-axis accuracy, backlash, etc.

It’s my intention to understand, not to discredit a product: Are the values for former features known for the Cognisys StackShot? Is it reasonable to ask for?

Regards,
supposing you will work on magnification bigger than 1:1 I would say:
- most stacks are beyond 1 cm so travel issue is for comfort on your rig not for the result of your stacked image
- acceleration (and how smooth it is) is for vibration can be compensated to pause 2 seconds before exposure, then vibration will have faded out
- resolution is related to Riks recommendation for minimal stepsize when you want to use high magnification optic with high NA, then you might want very consistant tiny steps may be 0.5µm
- bidirectional repeatability is not an issue for 99,9% of us since we do not measure or position at given coordinates (what would be the case if we would machine something)
- flatness and straightness over full travel range is also not an issue since the stacking software like Zerene adjusts the single frames to each other
- horizontal and vertical load capacity (max), should be bigger than the weight of your camera+optics+fixation
- home location accuracy, absolute on-axis accuracy, this would be important if you have to reach a certain position repeatably (like on machining) but for stacking you start some above Start and end somewhere beyond End that's it so no absolute position is required. Of course the more precise a stage is specified the less the chanse it will vibrate and wobble.
- backlash if you do your stack only in one direction of movement this is also not an issue, but may be important to focus forth and back at high magnification because the backlash is annoying

rjlittlefield
Site Admin
Posts: 21127
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Features for a motorized stage (focus stacking)

Post by rjlittlefield »

jac B wrote:
Tue Aug 11, 2020 1:55 pm
Maybe I misunderstand, but I have the impression that (most of) these parameters are used to compare motors and choose which motor(s) is/are under/best/over dimensioned for (a) certain focus stacking task(s)/application(s).
I'm not sure what you mean by "motor". Most of the parameters on your list will be controlled by things other than the motor itself. Backlash is caused by mechanical slop in the drive train. The accuracy of microstepping is dominated by sine/cosine shaping of the drive currents. Distance per step is determined by screw pitch and/or gear ratios. Flatness and straightness over full travel range will be controlled by rigidity of the rail and quality of the bearings or bushings. Absolute on-axis accuracy is controlled by the drive screw (or gears, if using a microscope focus block -- see http://www.photomacrography.net/forum/viewtopic.php?t=27549 for more about that).
What are your focus stacking task(s)/application(s), where only “min achievable incremental movement” is relevant?
It's pretty hard to characterize my applications, because I play everywhere from landscapes down to 100X on sensor. I maintain a list of postings, at http://janrik.net/RiksLinks.html , if you have lots of time to spend browsing.

In any case, when focus stacking, my main concern is always to avoid visible artifacts while shooting something close to a minimum number of frames. At high mag, especially 20X and above, that means getting small and uniform step sizes, especially not ever getting an unusually large step that might cause a band of blur. As noted in the thread that I linked earlier, my early StackShot was usable but not ideal at 40X and 50X, because the variability in step size meant that I had to shoot many more frames than if step size had been uniform.

These days, I use a motor-driven microscope focus block for most studio work that requires 10 mm or less of stack depth.

Straightness of travel and absolute accuracy have been of little interest to me because they have not been limiting factors. Other people have noted that with their StackShot rails, nutation becomes a significant problem at high magnification. It seems reasonable to expect less nutation with better slides that use ball or roller bearings as opposed to sliding bushings, but I am not aware of any careful comparisons.

--Rik

jac B
Posts: 31
Joined: Tue Jul 14, 2020 2:01 am

Re: Features for a motorized stage (focus stacking)

Post by jac B »

elf,
I ask for values.
You try to impose your opinion, did not give values, don’t know values :shock:
Jac

lothman
Posts: 494
Joined: Sat Feb 14, 2009 7:00 am
Location: Stuttgart/Germany

Re: Features for a motorized stage (focus stacking)

Post by lothman »

jac B wrote:
Wed Aug 12, 2020 1:28 am
elf,
I ask for values.
You try to impose your opinion, did not give values, don’t know values :shock:
Jac
Elf gave you a very well considered advice. Your question for values without any background make no sense to answer.

So why blaming Elf? Why are you asking for values, do you have commercial interests and hope others will do the work for you?

jac B
Posts: 31
Joined: Tue Jul 14, 2020 2:01 am

Re: Features for a motorized stage (focus stacking)

Post by jac B »

lothman,
I agree that we disagree.
You did not give values, don’t know values for the StackShot.
The answer on your speculations are in my other topics.
Jac

jac B
Posts: 31
Joined: Tue Jul 14, 2020 2:01 am

Re: Features for a motorized stage (focus stacking)

Post by jac B »

Rik,
Thanks.
I meant motorized stage, instead of motor.

I want (as a hobbyist) to compare and choose (the values for parameters from) motorized stages which are under/best/over dimensioned for (a) certain focus stacking task(s)/application(s).
Motorized stages like e.g.:
- https://cognisys-inc.com/focus-stacking ... -pkgs.html
- https://www.thorlabs.com/navigation.cfm ... gJFnPD_BwE
- https://www.physikinstrumente.com/en/pr ... lectionbox, https://www.physikinstrumente.com/en/pr ... d-drivers/

I try to get an impression of the characteristics of focus stacking applications and setups:

Subject applications with e.g.:
-handling constraints (e.g. a pinned museum specimen), so positioning limitations;
-a surface in mm, cm or micron;
-the level of detail (5x, 10x, 20x, 50x magnification);
-one stack;
-stacks of one column or row; stitching (linear or rotation panned) stacks;
-stacks of multi columns and/or rows; stitching (linear or rotation panned) stacks; stitching (top-, front-, side-, rear-, bottom) views of a subject;

Equipment setups like e.g.:
-camera objective moving on the Z-axis or X-axis; subject moving on the xy-axis or yz-axis;
-a motorized (focus stacking) stage for the equipment or subject;
-manual and/or (fully) motorized stages for multi columns and/or rows and/or subject views;

Constraints between subject application and equipment setup: …

Start (with a) simple (application, setup), but with expandable equipment.

Jac

Macro_Cosmos
Posts: 993
Joined: Mon Jan 15, 2018 9:23 pm
Location: Sydney

Re: Features for a motorized stage (focus stacking)

Post by Macro_Cosmos »

Alright, you're a hobbyist after a motorised rail for focus stacking, and you want some very specific specifications of the rail itself, ie exact numbers.

You're asking for too much. Numbers can be typical, but sample-to-sample variation exists. The stackshot is a product with photography in mind, it's not used to position say slides for fully automated imaging, what you've listed mostly doesn't matter at all. It seems to me that you saw these categories on Thorlabs without understanding what you actually want, therefore I don't think you can characterise most of it. Correct me if I'm wrong though, but you did put down absolutely sound advice. Without actually answering elf's question, we can't give any advice that's not anecdotal, this includes the numbers you want.

In such a case, it will be far better to contact Cognisys' support, they might have the numbers. They gave me pin-out diagrams and 3D models so I can make my own bottom plate to replace that useless included one.

So here's your list?
-travel range: Their website has it. 100mm or 200mm.

-velocity (max, min): Why does this matter? I have the numbers, explain to me why you want them. By the way, the velocity will depend on vertical or horizontal, and the load. Without that provided, none of us can provide those numbers. This can also be calculated with relative ease.

-acceleration (max, min, at a load): What load? 4kg? 10g? Same as above.

-resolution: reliably 1-2um. Rik's notes states at most 1.3x, this is I assume a statistical outlier, but it will show up. You didn't explain what kind of magnification you want and/or the resolution requirements.

-bidirectional repeatability: Definitely more than 5um, but why? It's for focus stacking, not CNC.

-flatness and straightness over full travel range: Again, why? This is broken into pitch&yaw variation which is an angle, something even my workplace that deals with lasers and nano-adjustments of detectors doesn't care about. There's also axial accuracy which again is something that I find a hard time believing anyone would care about.

-horizontal and vertical load capacity (max): https://cognisys-inc.com/focus-stacking ... l-pkg.html

-min achievable incremental movement: Well I can split one pulse into 10 and slap 0.1um onto it, this is a meaningless figure unless the rail can reliably achieve such a step.

-home location accuracy: It has lots of backlash, and backlash differs from model to model. This will change, just because mine is 15um doesn't mean yours will be 15um. It also depends on load which ties in elf's question you brushed off.

-absolute on-axis accuracy: How does this relate to amateur focus stacking? This is a term that I've only seen Thorlabs use. It really seems like you simply lifted parameters from them. Oh, and doesn't this fall under "flatness"? How flat the travel is... ie how well does the rail stick to the central axis. :-k

-backlash: enough to make most measurements taken by a person only specific to that one unit they have, because eh... sample variation.

Unless you're extremely well-off in finances, there's no reason to buy Thorlabs' motorised rails. They require coding, they require an expensive K-cube controller ($800 I think) and the power supply is also $300, you need a controller/interface that goes beyond "move rail to POS A", you code your own. Want the controller to also take pictures? Code it yourself using a relay circuit. You're looking at a good $2000 to get something basic running when the Stackshot (and wemacro etc) has everything already done for you. PI is several steps above Thorlabs, so I'd times that $2000 by 5. It's gonna be a fun project and if you know Matlab, it won't be difficult... so there's that.

Oh, and in order to characterise numbers that small, one needs some very good metrology equipment. My dial 1um resolution indicator positioned using magic arms for example isn't one of them so my numbers won't even be that good. Temperature will also matter since we're talking about sub-micron measurements.

lothman
Posts: 494
Joined: Sat Feb 14, 2009 7:00 am
Location: Stuttgart/Germany

Re: Features for a motorized stage (focus stacking)

Post by lothman »

jac B wrote:
Wed Aug 12, 2020 3:53 am
lothman,
I agree that we disagree.
You did not give values, don’t know values for the StackShot.
The answer on your speculations are in my other topics.
Jac
What about your speculations?
Only why I do not share information with you does not mean I do not have those information!

so my speculation on you are:
You have nothing to "give" to the community only want to "take" and I consider you as rather rude, so no further input from me.

jac B
Posts: 31
Joined: Tue Jul 14, 2020 2:01 am

Re: Features for a motorized stage (focus stacking)

Post by jac B »

lothman,
you can’t bear criticism, acting rude by (keep) speculating, generalizing.
our communication tells about both of us.
jac

Post Reply Previous topicNext topic