Dane.Kouttron

[09.23.20]  Photography Turn-Table

This 


What?





Conclusion Image Directory

Some Project Background
Halfway through the pandemic found photogrammetry, which is the (art?) of gluing together a number of static photographs into a 3D model. This works both for inside-out and outside-in activities. Initially I spent some time working on a 3D model of a facility for training purposes (take a look at to where things are etc), but this process also works wonders for making a 3d model mesh out of imagery alone. Branching out further its also convenient to film from a static location and get a rotary video of an item. By interfacing to a remote shutter, I could rotate-trigger-rotate, or set the camera to record video, begin spooling and stop recording after a complete rotation.

This project is actually two parts, a rotary table and something to drive the rotary table. 






So lets put everything together

For the first time in a while I actually cleaned my workbench. <choir noises>
First step, some rollers:
For roller wheels, I opted for some quickly-available mcmaster-carr widgets, with a built in sleeve bearing, held up on 3d printed legs. This isn't a high speed contraption, and the largest things I'd probably put on this were in the 20kg range, plenty for a few 3dprinted parts under compression.
Assembling the rider-wheels was fairly quick, these wheels are a simple 1/4-20 bolt, two washers and a nut. An internal bushings in the wheels forms the bearing surface. Rinse and repeat twice more and the three roller wheels are up and running. For a base I used a home-depot 'round wood thing' that actually had been my home towel stand forever. I painted it white and sanded it a bit. Wood is convient, as attachments are quick and flexible in location. 
Time for a gear-motor and a motor-holster.
The gear motor for this project was a random-lab-cleanout widget, 24v ~ 8rpm. I opted for the motor itself to not be the support, just to transmit rotary force, the thought was to design the rollers to sit slightly higher than the center rotary shaft, then trim and adjust the shaft collar that transmits rotary torque to the surface. Either way these mounts are really simple and take advantage of the threaded posts on the gear motor.
With the roller wheels assembled, placement of the motor-holder assembly was up next. Note that the center of the motor-gearbox is just the output shaft, I took a few tape measure points to locate the center of the white wooden round. With the motor mount roughly centered, wood screws were used to attach it to the base-plate. The motor hangs from the 3d printed mounts. 
With the three wheel assemblies done, it was time to space them out equidistantly. To do this eaisly i cheated, i knew the minimum sized rotary disk i'd be using, and placed one wheel assembly perpendicular to the motor mount. The remaining two wheels were positioned equally apart, by taking a measurement between the three. For each to be 120 degrees apart their spatial distance would also have to be equal.
With spatially equal positioning, everything is attached with some stubby 3/4" wood screws. Notice the small plate on the top of the gear motor? Thats the small magnetic coupler for the motor side.
Magnetic Coupler motor side
Mating from the motor shaft to the rotary platform, I chose to use a shaft coupler press-fit into a 3dprinted part. Going with a purely printed part would probably end in headaches as the small 8mm shaft D-shape doesnt provide a lot of contact area. Shown in the animation is the 3d printed part, shaft collar and magnets not shown.
Magnetic Coupler rotary table mount
To allow easy changing of the rotary platform (lets say I want to use an different color / material) I opted for a magnet mount to mate to the platform of choice. Four magnets, each set for north facing one side, are pressed in, the remaining holes are extra contact area for epoxy to mate with the rotary table surface. You can choose whatever epoxy works for you, I like the somewhat compliant DP 270. 
Epoxy TimeThe carrier for the rotary table is magnet-coupled. This has a benefit of being removeable (can change the color of the rotary stage) and also works as a way to prevent jamming, if for whatever reason the rotary platform gets snagged it will just mechanically de-couple not destroy its gearbox. DP270 was used to adhere the printed part to the table. Small 1/4" magnets were used, each facing the same direction for the coupling.
Granite Circles
You do not need a waterjet to get round granite, there are many 'cheese cutting boards' out there [link]. I had the opportunity to cut one due to some great timing, and it worked out fantastically
Large Acrylic Round
While at the ever-excellent hobbyshop, I cut out a 23.5" circle from a sheet of 1/4" 24" square acrylic stock. This time I was smart and also made a small center hole for allignment of the magnetic coupler.



With the two options cut and glued we have the basic setup

I added a piece of masking tape on top of the magnet assembly, mostly so the surface was easier to clean metal particulate off of. Being able to swap out rotary platforms is really nice.
granite surface plate, rotary table and white plastic rotary table
close up view of the rotary table assembly
Top view of rotary platform mechanism
I was fairly surprised how well the centering worked out for the location of the motor shaft. The granite rotary surface is smaller than the base, by approximately 1" in diameter,
top view of rotary platform mechanism
top view of rotary table
The white rotary table surface is fairly gigantic, which works well for taking photos of larger items

One of the first things i took photos of was a 48v hybrid motor, which was rather large. Even without a time-lapse controller (running directly from a 12v battery) the shot was great.




















S





Lets build a simple controller, using available components. 
Simple Software State Machine
There's really only two major functions here, spin in place and film with a video camera, or control an external Panasonic camera via a 3mm remote connector. The latter is interesting as it opens up possibilities for doing photogrammetry, or taking lots of photos of something and re-combining into a physical model. The # of frames / revolution isn't really 'accurate' as the gear-motor does not have position feedback, as such it will most likely be time driven, possibly measuring v-bus to compensate if a battery is running low.


Hardware Overview
I didn't want to re-invent the wheel, so I opted for an arduino, an LCD + Button Shield, and a motor driver shield.  There are lower cost options, but this worked fairly well for getting a prototype up and running in an afternoon. Everything stacks up and can be modeled into a small printed case. I'm a big fan of a simple power switch, which completely disconnects the battery pack from the setup, and I had a curious push-on-push-off illuminated switch in the spare parts bin. To actuate the camera shutter, I'm opting for a small isolated relay board and a resistor, while its possible to use a bidirectional opto-coupler, I had the small relay driver board on hand, and for whatever reason enjoy 'click' sounds.

Enclosure Time
The goal all along was to print out a case, as it can happen in the background and can be whatever i want. I've been really enjoying thermal-insert nuts, and may have gone a bit bananas on this build. Every part is metric, aside from the rubber feet, which only seem to exist with an imperial thread.

Design thoughts for the enclosure
I wanted to be able to remove the electronics for testing,



(There's other photos in the photo gallery)
Concluding Remarks:

If you have questions or comments, ask below or send over an email.
Comments:
HTML Comment Box is loading comments...
Stay safe when working with electrons in aqueous / semi-aqueous environments. Also wear sunscreen, I'm not responsible for your newly aquired winter-farmers tan : ]

Dane.Kouttron
Rensselaer Polytechnic Institute 
Electrical & Electrical Power
631.978.1650