Project Otto Session 7

Huge progress.

My apologies if anyone got this sent more than once. I ended up hitting publish instead of save draft but the website cranked it all out. Sorry about any confusion.

I just finished a 10 day “staycation” in which I managed to get a LOT of work done on the motion rig.  I managed to get quite a bit done, its very exciting to see this come together and to see that it actually works!  I’m not exactly surprised it works, I put a lot of time and thought into this, but there is always that fear that there are things I miscalculated or did not measure correctly,  and sometimes these errors can be very expensive.

Please excuse the messy house, this happens when I get busy on projects. You should have seen it when we were running the kickstarter for the Lens Apparatus, I had a mountain of cardboard boxes.  😉

Below is the X axis Gantry. This part sits on the bearings on top of the main support rails and provides a slide for the X axis. This entire thing will slide along the Y axis.  Pretty big, but it was not too much work. Luckily I measured every thing correctly the first time and it went together fairly well.

Ball screw and motor went on without much effort by using some basic 4 inch angle brackets. More about those brackets later.


Below: The gantry with the outer casing of the elevator, ball screw, and brackets.


Below, a good view of where the winch will attach on the elevator.



The winch is turning out to be a real pain in the ass. It is a Nema 23 stepper, 425 Oz Inch 4.2 amp motor. So pretty beefy to say the least.  The box on the end is a 30:1 worm gear drive.  Worm gears kick ass, they have virtually zero backlash when done correctly, and the 30:1 means the strength of the stepper increases by a factor of 30, while slowing down by a factor of 30. Slower is good. I don’t want to raise the elevator so fast it turns inside out. It may be too slow. I figure with 425oz inch, that is about 25lbs of lifting if the weight is set 1 inch from the output shaft.  With the 30:1 and lets be conservative with a 50% loss of efficiency,  that still ends up being 375lbs at once inch, and the reel will only be .75 inch radius not a full inch.

I have no confidence in my math or understanding about gears, but that is about what I came to by reading online.  I am hoping i can get speeds that lift the Pan/Tilt at 2 inches per second.  Depending on how well it does with load testing, I may opt to swap the gear box out with a 15:1 or even 10:1 ratio.


However, it would be good if I don’t cut my hand off trying to figure this out.

I have a friend who has a machine shop and i asked if i could pay him to help me out,  but my understanding is he usually does things like bifurcating large things like train engines that have seized, this is not really his specialty. He ended up referring me to another person but I never heard back from that guy. The origional intent was to get a piece of 2 inch aluminum round stock and machine it into a spool that I could attach to the gearbox.

I decided I could probably just cut a couple 1/2 inch segments off that stock, and then another 1.5 inches from another 1.5in diameter aluminum stock i have, drill and tap, then bolt together and make a spool that way. However believe it or not i really have no idea what I am doing and was trying to handhold that piece of aluminum while slicing through it with my miter saw and it got about half way done and the teeth grabbed in and got stuck in the aluminum, yanking it out of my hand and slamming it into the miter saw.


Nice bent blade.

I then decided that was a bad plan.

About 10 minutes of digging around online and for under 30 bucks shipped i found a shop that would cut the pieces for me. Once i get them in all i have to do is drill and tap them and i should have a spool. 🙂

After completing the gantry I installed it on the Y axis bearings and slid everything around, and it works like a charm. One gentle finger push can slide it from end to end with ease 🙂

The size of the gear box for the Z axis and the low height of the ceiling pose some challenges.  In order for this to work the ballscrew has to be ABOVE the winch gearbox, so I really had to sit the Y axis steppers to about an inch from the ceiling! You can see the temp bracket I mocked up for the stepper motor to test fitment in the image below.

(The room looks purple due to the grow lights for my carnivorous plants)20161010_161611
It is really up there, and looking at the image below you can see why. That screw on top is the Y axis. Just barely over the Z axis gearbox.


And another picture of the Y axis motor.



The couplings were in bad shape. They came with the ballscrew and were made for 5mm shaft and these motors have an 8mm shaft. I drilled them out by hand knowing they would wobble, but they would work for fitment and testing while I waited for the new couplings to arrive in a couple days.

Next I mounted the Pan tilt, ran the wires, and gave it a test and everything works as expected. Nice fluid motion, very stable, very rugged.

WHen moving the X and Y, that was a different story. Ok, so X worked pretty good, but the Y had a MAJOR shuddering problem.  The coupling was hand drilled and causing a wobble, and the weak flimsy brackets for the motor was exasperating the situation.  At this point I was pretty tired and took a day off. The next day the proper couplings arrived as did some more aluminum stock.

So I got back to it and built some proper brackets.


You can see what a ghetto rig I had at first.


These brackets were much much sturdier and easy to space everything out perfectly to align the ballscrew and motor.



Looking good!


The studder/shuddering problem was resolved! Now back to the Pan/Tilt.

It was too goddamn big. I underestimated the size of this behemoth, which is not a big deal but the main problem i have on this is the order of axis from far to close in relation to the camera SHOULD be

Pan —> Tilt —> Roll —> Camera


The problem with the way it is shown is that the TILT is affected by the roll. If the camera is angled up and you roll the camera, it will arc across the room, not spin the camera.  I figured i would fix both of these issues with a single solution by modifying it so the ROLL is now a Tilt, and change the TILT to a roll.

Here was the idea.  You can see the camera is sitting sideways from where it was, but now it rolls on the old pan assembly and the tilt is what used to be the roll. Now no matter where the camera has panned and tilted, the roll is ALWAYS inline with the sensor.


A few more cuts and modifications and I managed to move the camera much closer to the axis of the pan.  I may shorten things a bit more, but right now it is far more manageable.


And that is where I am currently at. Z axis is next. Cable management parts are en route so I can keep things nice and tidy,  i still have to wire up a few other motors, then when everything is built and working 100%, I get to rip it all apart and paint all the silver pieces black.  (well, except the bearings and drawer slides, but the brackets, aluminum square tubes, etc all get painted)
With any luck, This rig will be done by the end of october.

knock on wood….


2 thoughts on “Project Otto Session 7”

    1. Yeah, this is what happens when somebody who has a propensity for building things spends most of their time single. 😉

Leave a Reply