m_cam (1) optomechanics

[michael_r]

Infected by the macro virus i decided to build a rig for high mag photomacrography and after some steps on the learning curve i arrived at the m_cam described below.
m_cam is an acronym that officially stands in abbreviation of „modular camera system for photomacrography“, my personal interpretation is „michis cam“ of course. ':D'

I learned a lot from the descriptions given in this forum - in particular the article on the bratcam by Chris S was extremely helpful to me. However i did not want to rebuild this excellent system as i had some other design goals: I wanted

1. a versatile system that can be used with different cameras - the system should in particular be compatible with my Sinar P2 camera. As it is not possible to mount a sinar P2 on a stackshot (or something comparable) this requires that the camera is fixed and the object moves. I found this appealing as this approach allows for a tight mechanical coupling of the camera and lens combo to the rig and therefore should allow to circumvent vibration effects caused by mirror and shutter.

2. a setup that can be used in horizontal as well as vertical position (e.g. with a wall adapter to be developped)

3. a system that does not suffer too much from vibrations (there are no „good vibrations“ in this business)

4. a flexible system that allows for the illumination of opaque objects as well as transparent objects with a condensor system (to be developed)

5. a system that can be automated with the stackshot controller (and zerene) as i did not want to develop software for automated stacking with e.g. a newport controller (if this were not the case i would have chosen one of those motorized newport stages with submicron step size).

6. a system that allows to easily use components in other applications (e.g. on a tripod out of doors)

As it is a modular system i will describe first the base unit and then some of the units on the camera and the object side that can be attached to it. I will keep it short as the present state is determined to a substantial extent by „ebay coincidences“: Price matters and therefore i am somewhat reluctant to buy the pricey optomechanical components in new form - and on ebay with limited budget of time and money you have to a certain extent to „want what you get“. Another issue are „historical coincidences“ - from various camera systems that i happened to own during the last decades there were interesting components that i wanted to employ (as soon as you do not want to focus to infinity you can mount almost everything that produces a real image to a camera body).

The Base Unit

For the realization of the base unit i choose a 9x9 alu profile as an optical bench; such profiles are common in industry automation and not too expensive. They can be easily combined with standard components. I found a length of 90 cm to be sufficient for my purposes.




For greater flexibility i wanted to be able to perform a coarse positioning of the camera in the y-direction. For that purpose i added a X95-profile from LINOS to the base unit which allows to vertically translate the camera. On the sled (LINOS) an angle and an alu adapter plate is mounted which holds a 100 mm clamp from Heijnar photo to which the camera side modules can be attached


A smaller profile was also added in order to mount flash, diffusers, etc. Camera side is left, object side is right. On both sides i have 100 mm arca swiss style clamps to mount various modules.




I took two measures to circumvent problems associated with vibrations in order to get rid of internal (vibrations induced by mirror and shutter of the camera) as well as external sources of vibrations.

Internal vibrations. I opted for a rigid coupling between camera and a large mass in order to get rid of internal vibrations. The large mass is the base unit described above.

External vibrations. In order to get rid of external vibrations i introduced a two-stage decoupling. In the first stage thin air-inflated seat cussions are used under a plate of black granite. On this plate the base unit is placed separated by sorbothane bumpers which form the second stage in external vibrations decoupling.

That i put so much effort in this is partly explained by the fact, that the first objective i got was a 50x MTTY (one of the ebay coincidences) - and that in my first setup everything was dancing on the monitor in 100 % life view operation.

Object side modules are attached to the optical bench with the help of an adapter plate equipped with a 100 mm clamp from Heijnar photo.

Camera side modules.

The camera modules were defined to be applicable on a tripod as well. A shift of the camera unit in z-direction for coarse grain positioning is possible.



small bellows unit. A minolta bellows is mounted on a short rail for coarse positioning in z-direction (in the direction of the optical axis).





large bellows unit. For longer extensions up to 30 cm a pentax 6x7 bellows (a reminiscence of former 6x7 days) can be attached . Coarse positioning in z-direction is enabled by the bellows.




Infinity objectives unit. I use my old pentax 6x7 lens as tube lense for various infinity objectives from Mitutoyo and Nikon and think that the results are not too bad. I have some alternative tube lens candidates, but an investigation on which to choose best is still pending. With all lens combinations i want a rigid coupling of the combo at two points to the underlying rail: camera with an arca swiss style clamp and objective with a DIY aluminium ring equipped with nylon srews. No chance for vibrations!


Sinar camera unit. Not shown here as this unit devoted to „stack and stitch“ as it will not be used before i have a camera with electronic shutter.


Camera moving. The rig can be also applied to objects that can‘t be moved. In this case it is possible to adapt either the stackshot z-stage shown below or the z-stage that i use for manual positioning of the camera with macro lens or bellows on a tripod on the camera side.


Object side modules

The object side modules allow for fine translations in xyz-direction as well as rotations around two or three axis. The object side modules are a combination of a stage that performs fine z-positioning (the z stage) and a stage for the other movements (the object positioning stage)

I have three z stages: one for manual positioning, one with a stackshot and one for motorized stacking with high resolution

Manuel z stage. A newport stage allows fine positioning in z_direction and is coupled to one of the object positioning stages with arca swiss style clamp. I do not show it here.




stackshot z stage. The stackshot can be directly mounted with the 100 mm clamp on the object side which also allows for coarse positioning. I adapted a 60 mm rrs clamp to it in order to mount one of the object positioning stages




Newport stage. I had similar ideas as seta666 has published last year and had already considered the kardanic coupling demonstrated in his post. I use a newport M436 which is a reallly sturdy linear rail. It came with a micrometer which i replaced with a100 tpi Newport screw that is driven by step motor from cognisys that is compatible with the stackshot controller. The drive reduces the number of turns of the screw in comparison with the rotations of the stepper by 1:7: Bye bye microsteps (i do not particularly dislike the microsteps - its more the noise that comes with them)! It works really fine - but if you want to travel over a longer distance you need some patience. Fortunately this can be avoided in high mag work.




sinar object orientation module. A modified stage from a sinar P2 camera allows for object positioning in three translational axes as well as rotation around two axes. I think it works fine, but it is not really what you want for positioning in micrometer steps. So i would recommend it if magnification stays below a maximum value of five. It fits e.g. to the stackshot z-stage via an arca swiss style adapter plate.




Goniometer stage. I wanted to add a module with goniometers identical to the top layer of the bratcam to be mounted on one of the z-stages. I already had a M481A for almost one year but never ever appeared a GNL20 on ebay. I began to accept that i had to buy one new, when - a couple of weeks ago - Chris S (thanks Chris) pointed me towards a seller in Korea who had Kohzu stages (a brand that i didn‘t know until then). Well i ordered a pair, waited patiently, added new handles and combined the goniometers with some other ebay „prey“ to the precision positioning module shown below. The assembly is „brand new“ and i have not yet decided if this is the most practical configuration. The idea however is as follows: M481A and goniometer stage are aligned with the optical axis of the camera such that the optical axis is 6cm above the DIY aluminium plate (thickness 10 mm) as the Kohzu have a fixed point there. The horizontal sled in the top layer is used to bring the object into this point. The other two sleds allow then to scan the surface.


And finally the total setup (add some tissue diffusers in your mind) for work with microscope objectives.




I admit that during this project the focus shifted somewhat from photography to optomechanical tinkering. If i think of Riks „cork drive“ or his setup for the feather photo (which produced excellent results) i also have to admit that my setup is possibly somewhat „overengineered“ - but that doesn‘t bother me as i had a lot of fun in assembling this beast. Drilling holes, cutting threads (my first ones - in alu its easy) and arriving at something that is fun to operate was an experience of its own to me and now the focus can shift back to photography. If you found something useful in this report: you‘re welcome.

Michael

[ChrisR]

A well thought-out rig, very well explained. Thankyou for that Michael!

[rjlittlefield]

Agreed!

I have added this thread to the FAQ list at http://www.photomacrography.net/forum/v ... 5311#55311.

I am particularly interested to see the sleds mounted above the goniometer, to allow bringing subject exactly to center of rotation. That's a very nice touch. Also the stabilizing ring around the end of the tube lens. (To pick only two of many nice design points...)

--Rik

[seta666]

Very nice indeed!! My set up now does look really simple

[Chris S.]

Bravo, Michael!

I've spent quite a bit of time studying your post. There are lots of really nice design and construction touches beyond the ones you pointed out. The engineering and execution are wonderful.

One thing I like is your two-stage vibration isolation approach. You could probably tune each of these stages for isolation from a different range of vibration frequencies, right? With the seat cushions, this would be especially easy, as the inflation is probably adjustable. (I don't use any form of vibration isolation, as vibration isn't a problem in the environment where I shoot, so long as I turn off the nearest furnace. But I've experimented with vibration isolation. The ability to adjust for current conditions seems useful; the ability to cancel two different vibrational frequency ranges at once, doubly so.)

I also admire your approach to the timing belt/timing pulley drive (though I confess I don't see the Cardan joint to which you refer). With your 7:1 drive ratio, assuming you have a 0.9 degree stepping motor, my back-of-the-envelope calculation suggests you are getting 0.18 microns per full step. But I did this quickly--is the number right? Like you, I always work in full steps.

Your goniometer stage--with the three-axis linear translation on top of the goniometers, is, as far as I know, the first example we have seen at PMN of fully kinematic six-axis subject movement. (Portions of that phrase made a good search term a couple of years ago, but this seems to have changed, and I don't know a more currently useful one.) The Bratcam's subject stage is semi, not fully, kinematic. I've pondered adding axes as you have done, and will be interested in your experience with them. (You expressed some question about this in your post.) You're probably gaining positioning control, at some loss of flexibility in lighting and background control. You'll have to tell us, as you use the M-cam, how you view this tradeoff. : D

Am looking forward to seeing how you implement your vertical attachments, and the condenser for transmitted light.

Hats off to you!

--Chris

[dolmadis]

Congratulations and thank you for posting.

I am currently considering my first macro rig and it is most helpful to see your design and learn about the background to some of your design decisions.

I have also been looking at the collected experience in the FAQ and marvel at the diversity.

I do not have the budget or the skills to copy or emulate the engineering of many of these designs but I would be most interested in learning what you all consider to be the critical points of any design. In other words what are the "must be delivered" and pitfalls to be avoided.

I have a splendid focus block on its way which I have chosen with a great deal of welcome help from ChrisS.

ChrisS has also put the value of Arca S mounts to me.

On a specific point on this design could I ask please what the length of the horizontal rail is in photo 2 above? Is it the 90cm that you refer to as Base Unit?

I am asking this because I am unsure of the scale of a useful rig.

How vital are the goniometers please?

Thanks

John

[Craig Gerard]

Nicely done!


Craig

[g4lab]

I never got tired of looking at these. I love this forum!

[michael_r]

Thanks for the appreciation.

Chris: i agree with the 0.18 micron but the motor is the one distributed by cognisys last year and it has 200 steps per turn (1.8 deg per step) gives me 1400 steps per turn of the screw which corresponds to 0.254 mm per turn or 0.18 micron per step.

Dolmadis: I think with the focus block you already have the most important component. I also think that it is a good idea to get some arca swiss style components: You can extend your arsenal with time. If you consider to design a rig with an alu rail as in my base unit you have the advantage that this is easy to modify and scalable: If you want to change a distance - just open four screws by one turn then you are able to shift the plate to the new position and fix it there again by four turns. Or you may remove the plate and replace it by another one. The costliest portion of the base unit was the LINOS portion that i bought new. But this was added later. In my first setup the camera was mounted on a block and the distance of the optical axis to the alu rail was kept fixed. This should be enough if you want for example only to work with microscope objectives. Concerning the length of the rail i think you should look at the last image of my post to see that with 90 cm your on the safe side - but i would not go below 60 cm. Such rails can by bought with user specified length in various online shops.

I'm not a craftsman and i had to chose an approach that is compatible with my very limited skills and machinery. What i did is really simple: The alu plates were bought with specified dimensions from an ebay seller in Germany and were not expensive. I only had to drill a couple of holes - that was not really demanding as in most cases i added 0.5 mm to the diameter of the screw.

Don't try to include everything from scrach: let your setup grow. Start with a simple configuration and add components as you need/get them. Look at the 100% live view on the monitor of your computer - if you walk through the room and think that what you see on the screen is not acceptable: add some sorbothane bumpers. I now use a set from Edmunds which is not expensive and can be recommended. And concerning goniometers: they are not a must have but are really nice to have. Have good luck with your project.

Kind regards from Germany

Michael

[dolmadis]

Hi Michael

Thank you for your thoughts and advice.

I had already started looking at aluminium rails and had thought about buying a cine camera slider off the shelf for minimal modification. But the prices were prohibitive.

Then your posting came along and my interest in aluminium rails was rekindled.

Unfortunately I do not seem to be able to find supplies of rail and plates in my Ebay searches.

Could you share your link with me to explore please? If you would prefer please PM or Email me.

Thanks again.


John

[michael_r]

Hi John

i bought my stuff from www.motedis.com (located in germany) which is somewhat cheaper than Bosch-Rexroth. They have a 100 page pdf catalog that shows a lot of cross sections etc.

I do not know where you live - therefore a german seller may not be of interest to you considering transport fees. But i think that you should have a seller of Bosch Rexroth material in your country. Try to google it.

Michael

[dolmadis]

Hi Michael

I am in the UK. This is very helpful and puts me on the right track (no pun intended !!)

John

[ricardosp]

Hello Michael,

I have been studying your design, and really I like it. I have taken a lot of ideas for my future set-up.

I would appreciate if you could indicate me the model of the Newport adjustment screw, as I was unable to locate it.

Also I would like to know the model of the piece that fix the screw to the pulley.

Is there any update to your set-up or any feedback about your experience with it?

Thanks

[michael_r]

Hi Ricardo,

i didn't keep the bill, but i'm pretty sure that the newport screw was from the AJS100 series that i found in my 2012 catalog (didn't check whether these are still available), the aluminium cylinder is DIY.

News? Not really - to me the "m-cam" is a good approach towards horizontal stacking. I first wanted to hang the whole system to the wall to allow for vertical stacking, but the wall in my hobby corner is covered with thermal insulation that does not allow to hold such heavy equipment. My present project therefore is a separate vertical stand that also uses an oly chc focus block, which i think is very handy for quick manual positioning (i use my camera also for viewing purposes as a digital microscope) - concerning automated driving i intend to benefit from Rik's ingenious vinyl connector published a couple of years ago in order to be able to quickly place/remove a stackshot-controlled stepper motor. For that purpose i want to mount the stepper on a short arca swiss rail in series to the knob to be able to quickly place or remove it. The parts are somewhere in the mail - if i think that the results are worth mentioning i will post one or more pictures from the setup in Riks post on automating the chc block.

Kind regards

Michael