Overview & comparison of current focus stacking systems

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

Moderators: rjlittlefield, ChrisR, Chris S., Pau

jeremylavine
Posts: 16
Joined: Sat May 18, 2019 7:13 am
Location: Washington, DC

Overview & comparison of current focus stacking systems

Post by jeremylavine »

Hello all. I've just ordered the Laowa 25mm 2.5-5x to use with my Sony a6500 and I'm looking for focus stacking solutions that will work for this setup and cover me if I want to go further in the future, e.g. with a 10x microscope objective.

It seems that mounting the camera or lens tripod mount to an automated motorized rail is the most common approach for ultra-macro focus stacking in a studio. StackShot, WeMacro, and MJKZZ seem to be the popular prepared solutions for this. I'm a bit confused by the current offerings I see on the MJKZZ store, and what level of rail I'd need for different magnification levels. I've seen the SR-90P kit described in discussions on this forum as a good entry-level option comparable to the WeMacro and StackShot, and the Industrial Rail Conversion Kit for use with THK rails available used off eBay as a more solid option better for higher magnifications. But does anyone know how the new "Qool Rail 250" compares with the SR-90P kit, and whether the Ultra Rail Kit is the same as the Industrial Rail Conversion Kit plus a rail? Is there any consensus regarding the magnification level (e.g. 10x) at which the wobble from the rails in the SR-90P, WeMacro, and StackShot kits becomes a problem, and where the THK/HIWIN rails provide noticeably better results?

I see that WeMacro and MJKZZ offer battery packs for their motorized rails. Does anyone have any experience using these in the field? Would this be too unwieldy to carry around on a tripod and use in the park, and if so would any manual rails be easier to handle while still offering enough precision for 2-5x magnification?

I've also seen a microscope focus mount referred to as an alternative to a focus rail, e.g. on the Zerene website here: http://zerenesystems.com/cms/stacker/do ... versusrail, but I haven't found any details on this beyond this page on butchering a microscope to get the mount, which seems like a bit much: http://extreme-macro.co.uk/microscope-stage/. Are there off-the-shelf mounts? Do the WeMacro specimen stages (e.g. https://www.wemacro.com/?product=xyr-st ... men-holder) fit this purpose (allowing you to focus stack with the camera fixed)?

What prepared options are there for planar panoramas? Would the WeMacro rail mounted to the WeMacro vertical stand (https://www.wemacro.com/?product=vertical-stand) do the trick?

Sorry if this is overly broad or if it has been covered elsewhere already. If the mods want to split it up or move it the beginners forum, I'd understand. And thanks in advance for any and all responses.

Lou Jost
Posts: 5987
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

After working with rails for several years, I now rarely use them. Instead I use the WeMacro MicroMate to turn a microscope fine focus knob. This is immensely more stable than a rail, and easier to mount unusual or big lenses. Just lay them face down the microscope focusing stage. Microscope focusing blocks come up often on ebay. So do cheap whole microscopes you can salvage parts from . And it is easy to modify a microscope x-y stage (also cheap and common) for limited manual x-y stitching:

www.photomacrography.net/forum/viewtopi ... highlight=

Edit: Your particular 2.5x-5x lens is one of the few lenses which do not mount well on this system, because of its tapered rather than flat front. But still, I have used it successfully on this rig. I think I found wide tapered tubes that the lens+camera could be inserted into.

JohnDownie
Posts: 115
Joined: Tue Oct 18, 2011 6:57 am

Post by JohnDownie »

Lou,

What is the microscope in the set up you linked to?
Lou Jost wrote:After working with rails for several years, I now rarely use them. Instead I use the WeMacro MicroMate to turn a microscope fine focus knob. This is immensely more stable than a rail, and easier to mount unusual or big lenses. Just lay them face down the microscope focusing stage. Microscope focusing blocks come up often on ebay. So do cheap whole microscopes you can salvage parts from . And it is easy to modify a microscope x-y stage (also cheap and common) for limited manual x-y stitching:

www.photomacrography.net/forum/viewtopi ... highlight=

Edit: Your particular 2.5x-5x lens is one of the few lenses which do not mount well on this system, because of its tapered rather than flat front. But still, I have used it successfully on this rig. I think I found wide tapered tubes that the lens+camera could be inserted into.

Lou Jost
Posts: 5987
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

It is a leftover column and head of a Nikon UM11, and a standard Nikon focus block from some other Nikon microscope.

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

Jeremy,

I've used rails from Stackshot, Wemacro, MJKZZ and THK. I also use the Wemacro Vertical Stand with these rails, and with a Stack and Stitch System I'm developing. My subjects are chips that can range from 100's of microns to semiconductor wafers which are 100's of mm.

The Wemacro Vertical Stand is an excellent portable stand for V and H use (with option), as well as a good base for a S&S System (see posts on this subject). All the rails mentioned are good, with the Stackshot being the most versatile but more expensive, the Wemacro is the most popular and a very good value, while the MJKZZ uses a single guide design whereas the Stackshot, Weamcro and THK use dual guides, so the MJKZZ tends to have more rock & wobble than the dual axis designs. The THK KR20 are precision rails that can be adapted to our use with the MJKZZ kit or DIY for use with standard 400 step NEMA 17 stepper motors. With these rails you'll want to consider a more advanced controller like the ones I'm developing (see posts on this) to take advantage of the rails precision, although they do operate well with the standard controllers from Stackshot, Wemacro and MJKZZ.

With the advanced controllers I'm developing the THK KR20 on a Wemacro Vertical Stand produces some amazing results as shown here at 800X!!

https://drive.google.com/file/d/19wE-B- ... sp=sharing

Here's a few videos showing the setups with these components, first set is a single axis,

https://drive.google.com/file/d/1elblFj ... sp=sharing

https://drive.google.com/file/d/1j8hIgv ... sp=sharing

Then the developmental S&S System in the next videos.

https://drive.google.com/file/d/1XHR1tI ... sp=sharing

https://drive.google.com/file/d/18GxU2U ... sp=sharing

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

ray_parkhurst
Posts: 3436
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

mawyatt wrote:the rails mentioned are good, with the Stackshot being the most versatile but more expensive, the Wemacro is the most popular and a very good value, while the MJKZZ uses a single guide design whereas the Stackshot, Weamcro and THK use dual guides, so the MJKZZ tends to have more rock & wobble than the dual axis designs. The THK KR20 are precision rails that can be adapted to our use with the MJKZZ kit or DIY for use with standard 400 step NEMA 17 stepper motors.
In general I have not seen best performance from these "dual rail" machines. I assume it's because the bearings are of the "pillow" type. These circular bearings have no pre-load, so have a bit of wobble and play. The THK rails have 4 orthogonal bearing surfaces, all with pre-load, and are thus the most stable of the types. The mjkzz rail is between the two, in that it has a dual-bearing/single-rail mount, so although those bearings are pre-loaded, they can't withstand much torque without movement. The best rails use two widely-spaced single-rails, with two carriages per rail, to spread the torques such that virtually zero play is allowed. The Parkers and a few custom rails I've seen use this design, and can achieve sub-100nm wobble performance.

All that said, IMO any of the 3 commercial options will work fine at 2.5-5x and even at 10x, so I'd look for other aspects such as cost, software support, etc to decide between them.

jeremylavine
Posts: 16
Joined: Sat May 18, 2019 7:13 am
Location: Washington, DC

Post by jeremylavine »

Lou,

Do you have any more detailed description and other photos of your setup, or do you know of any existing detailed descriptions of such a setup? I'm not really familiar with microscope mechanics, components, and variations, and it's not clear to me how this works. Do you have the stage mounted where the nosepiece would be? Is the camera looking through a panel of glass on the stage to the subject on the desk below?

You say that this is far more stable than a rail, but have you compared this to THK or HIWIN rails?

I'm not finding anything searching Google or eBay for Nikon UM11 or UM-11. I do see results for Nikon MM-11 (e.g. http://www.photomacrography.net/forum/v ... hp?t=17656 and http://www.photomacrography.net/forum/v ... hp?t=27667) and Nikon UM2.

Lou Jost
Posts: 5987
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Sorry, that's right, it was a UM2, close relative of the MM11.

I have only tried the purpose-built macro rails, the WeMacro rail and Stackshot. I am sure that there are industrial rails that are as stable as one could want.

Using a Micromate plus focus block increases the precision of a motor, because each step of the motor produces only a very small rotation of the fine-focus knob. For my set-up the step size of the MicroMate is about 1/10 the step size of a similar motor used to drive a rail.

The big flat black and aluminum thing is the x-y stage; I've build an aluminum lens-mover where normally there would be an aluminum slide-mover. That moves in the y direction while the black rectangle moves in the x direction. Both directions are controlled by two coaxial knobs under the stage.

I have sawn a large hole in the middle of the x-y stage. The x-y stage sits on a metal ring about 60mm diameter which is attached to the focusing block (they are often sold together). So the tube lens rests on the x-y stage and the objective protrudes under it, through that ring.

jeremylavine
Posts: 16
Joined: Sat May 18, 2019 7:13 am
Location: Washington, DC

Post by jeremylavine »

Mike,

I saw your interesting comments comparing the Stackshot, WeMacro, and MJKZZ systems and THK rails in these earlier discussions ...
"Advice with motorized rail" (May 2017), "Wemacro vs. Stackshot" (Aug. 2017), "Wemacro or Mkjzz rail?" (Oct. 7, 2017)
... and am happy to see you respond here.

You commented that the Stackshot Controller with Modified THK KR20 was by far the best overall rail system. Do you still consider this to be the case?

Is the option to control it from Zerene Stacker the main benefit of the Stackshot controller over the WeMacro and MJKZZ controllers, or are there other significant benefits? You commented : "If you are on a budget the WeMacro system is a bargain. However if you are a serious user doing lots of stacking sessions, soon you will get frustrated with the limited capability, especially during setup. I believe William at WeMacro is considering addressing these limitations, so we may see an updated controller/software in the future." Were there any updates along these lines that you're aware of? I don't see any revision history on the WeMacro site.

Regarding rails: I see comments from you and from Niels Geode saying that the wobble from the Stackshot, WeMacro, and MJKZZ rails can be seen even at 5x magnification, with you making the point that this is particularly relevant when not using Zerene's auto-alignment when stitching. This makes me hesitate to buy any of these rails even to start with for lower magnifications, and makes me think it would be wiser just to start with a THK rail.

I see comments from you about using rubber bands for off-axis loading to reduce wobble. To what extent did you find that this fixed the problem? I assume you don't find the need to do this with the THK KR20 rail; is that correct?

I see a few results on eBay for THK KR20 for $100-200: 250mm, 200mm, 162mm. Should all of these work, or are there some specific criteria I need to look for?

Regarding your stack and stitch system, I assume you mean these posts of yours: Single Axis Stacking System Video and Automated Stack & Stitch System Videos. This is a very interesting system you've assembled! It almost makes me want to try rolling my own, too. I'm comfortable with soldering and writing C code (or Python), and your system looks a lot neater than the mess of boards dangling off wires I imagined for a DIY project. I'm also bought into the Arca-Swiss mounting system and Godox/Adorama Flashpoint flash & strobe system. I'm curious what advantages and disadvantages you see with this compared to the Stackshot, WeMacro, and MJKZZ controllers.

Lou Jost
Posts: 5987
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Jeremy, what kinds of subjects do you want to work with?

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

ray_parkhurst wrote:
mawyatt wrote:the rails mentioned are good, with the Stackshot being the most versatile but more expensive, the Wemacro is the most popular and a very good value, while the MJKZZ uses a single guide design whereas the Stackshot, Weamcro and THK use dual guides, so the MJKZZ tends to have more rock & wobble than the dual axis designs. The THK KR20 are precision rails that can be adapted to our use with the MJKZZ kit or DIY for use with standard 400 step NEMA 17 stepper motors.
In general I have not seen best performance from these "dual rail" machines. I assume it's because the bearings are of the "pillow" type. These circular bearings have no pre-load, so have a bit of wobble and play. The THK rails have 4 orthogonal bearing surfaces, all with pre-load, and are thus the most stable of the types. The mjkzz rail is between the two, in that it has a dual-bearing/single-rail mount, so although those bearings are pre-loaded, they can't withstand much torque without movement. The best rails use two widely-spaced single-rails, with two carriages per rail, to spread the torques such that virtually zero play is allowed. The Parkers and a few custom rails I've seen use this design, and can achieve sub-100nm wobble performance.

All that said, IMO any of the 3 commercial options will work fine at 2.5-5x and even at 10x, so I'd look for other aspects such as cost, software support, etc to decide between them.
I speak directly from experience as I have a Stackshot, 2 Wemacro and 2 MJKZZ SR90. The Stackshot rail is the best of the three from a backlash and wobble standpoint. The original Wemacro and SR90 have more wobble and backlash than the Stackshot, later Wemacro improved the bearings and is now better with wobble but still has backlash, slightly more backlash but less wobble than the SR90 (none of these are in the THK KR class though).

Agree that any of these work fine for general usage, especially with backlash compensation and/or off-axis loading. When I use any of these rails I often use off axis loading since it improves backlash and wobble but doesn't cost anything (few rubber bands and bolts).

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

ray_parkhurst
Posts: 3436
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

I've tested multiple examples all of them but the StackShot, and unfortunately each of the rails seems to be a bit different, even within the same model. I had one WeMacro that wobbled unacceptably, and others that were quite good. I attribute this to the variability of the pillow block bearings.

To the OP's question about microscope focus block quality, for sure this is going to be better than almost any other solution from backlash and wobble perspective.

Lou Jost
Posts: 5987
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Focus block backlash seems completely undetectable at 20x or less in good Nikon focus blocks (with fine focus knob--that is really important, don't bother with a block that doesn't have a fine focus knob along with the coarse focus knob).

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

jeremylavine wrote:Mike,

I saw your interesting comments comparing the Stackshot, WeMacro, and MJKZZ systems and THK rails in these earlier discussions ...
"Advice with motorized rail" (May 2017), "Wemacro vs. Stackshot" (Aug. 2017), "Wemacro or Mkjzz rail?" (Oct. 7, 2017)
... and am happy to see you respond here.

You commented that the Stackshot Controller with Modified THK KR20 was by far the best overall rail system. Do you still consider this to be the case?

Is the option to control it from Zerene Stacker the main benefit of the Stackshot controller over the WeMacro and MJKZZ controllers, or are there other significant benefits? You commented : "If you are on a budget the WeMacro system is a bargain. However if you are a serious user doing lots of stacking sessions, soon you will get frustrated with the limited capability, especially during setup. I believe William at WeMacro is considering addressing these limitations, so we may see an updated controller/software in the future." Were there any updates along these lines that you're aware of? I don't see any revision history on the WeMacro site.

Regarding rails: I see comments from you and from Niels Geode saying that the wobble from the Stackshot, WeMacro, and MJKZZ rails can be seen even at 5x magnification, with you making the point that this is particularly relevant when not using Zerene's auto-alignment when stitching. This makes me hesitate to buy any of these rails even to start with for lower magnifications, and makes me think it would be wiser just to start with a THK rail.

I see comments from you about using rubber bands for off-axis loading to reduce wobble. To what extent did you find that this fixed the problem? I assume you don't find the need to do this with the THK KR20 rail; is that correct?

I see a few results on eBay for THK KR20 for $100-200: 250mm, 200mm, 162mm. Should all of these work, or are there some specific criteria I need to look for?

Regarding your stack and stitch system, I assume you mean these posts of yours: Single Axis Stacking System Video and Automated Stack & Stitch System Videos. This is a very interesting system you've assembled! It almost makes me want to try rolling my own, too. I'm comfortable with soldering and writing C code (or Python), and your system looks a lot neater than the mess of boards dangling off wires I imagined for a DIY project. I'm also bought into the Arca-Swiss mounting system and Godox/Adorama Flashpoint flash & strobe system. I'm curious what advantages and disadvantages you see with this compared to the Stackshot, WeMacro, and MJKZZ controllers.
Jeremy,

Wow, lots of questions. I'll do my best answer them, ping me if I miss something.

The Off-Axis loading really helped the Stackshot, Wemacro and SR90 rails. I didn't need to do any backlash nor wobble compensation with the superb THK KR types, they are real good as you can see in the video. In the 800X you can see that I'm leaving the spot in one direction and returning in the other direction without any backlash nor wobble compensation.

Any of these controllers mention will operate the THK KR rails with a proper motor adapter cables. I've run them all, you can even run the one brands motor with another controller, they all work together with mentioned proper cable adapters. The Stackshot controller is the most versatile of these, and working with Zerene even better.

Originally set out to do something different with my controller endeavors though, that being to attempt to control motor/rail movement induced vibration with electronics, but after doing a good job mechanically. I wanted to also achieve 1~2 second with high precision between images and waiting for things to settle down was in the way. The Pololu Tic-500 are nice controllers and work very well, but still have some limitations with regard to control algorithms and parameters, so I decide to attempt to tackle the Trinamic devices. I wanted high precision, but also high speed from the same controller and rail, these seem to be mutually exclusive. All this proved to be a difficult task, I felt like a 170lb linebacker trying to tackle a speedy 240lb running back :shock:

I have limited software skills but finally succeeded in talking to the Trinamic devices via SPI and now have these Trinamic devices working. After working with the various parameters for motor current control, an added benefit was the large reduction in noise as well as induced vibration. Built up a few prototype hand wired boards for evaluation, and eventually got things working very well and decided to offer this to the community here and asked for some help with a PCB layout. No one accepted the challenge so I tacked it myself and designed a few boards and got them fabricated.

Still have a couple of the PCBs for the TMC5130 left (one shown in video), these support single axis and S&S efforts with 2 additional Tic-500 which operate via USB. This produces a superb fully automated S&S System and with the THK KR20 on the Z axis is deadly accurate and quite, but also works extremely well with the Weamcro, Stackshot and SR90 rails. These boards have optically isolated Camera and Strobe trigger and support a Honing sensor for Z axis zero homing.

I've delivered a few of these to some folks here, have another going out tomorrow.

https://www.photomacrography.net/forum/ ... hp?t=39473

https://www.photomacrography.net/forum/ ... hp?t=39396

https://www.photomacrography.net/forum/ ... hp?t=38511

https://www.photomacrography.net/forum/ ... hp?t=38604

BTW the Adorama/Godox strobes are an excellent choice.

Now I'm working on high speed, current and voltage controller/drivers. Also just developed a Hall Effect Motor current sensor for reading and displaying accurate motor Sine and Cosine coil currents. I need this to help with test evaluations and parameter optimization of various motor/rail combos, "seeing" the motor currents is a big help. Also allows me to verify the different modes of operation and algorithms. I'll probably design a PCB and offer this is folks are interested.

https://www.photomacrography.net/forum/ ... 630#249630

PM me if you are interested in any of this.

Hope I've answered all your questions, ping me if I didn't.

Best,
Last edited by mawyatt on Wed May 29, 2019 1:42 pm, edited 1 time in total.
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

ray_parkhurst wrote:I've tested multiple examples all of them but the StackShot, and unfortunately each of the rails seems to be a bit different, even within the same model. I had one WeMacro that wobbled unacceptably, and others that were quite good. I attribute this to the variability of the pillow block bearings.

To the OP's question about microscope focus block quality, for sure this is going to be better than almost any other solution from backlash and wobble perspective.
Wonder if that was an original version of the Wemacro, the old one I have has similar characteristics, but the new version is much better?

Agree, the microscope block is not going to have much backlash and wobble. I chose the rail route because the subjects I generally work with can vary considerably in size from 100's microns to 100mms, but want to do high resolution and high pixel count on the larger subjects, thus S&S.

Best,
Research is like a treasure hunt, you don't know where to look or what you'll find!
~Mike

Post Reply Previous topicNext topic