Submitted by RCFisher on Mon, 05/17/2010 - 00:34
Posted in
I am building an OpenMoco timelapse rig and have gotten one tiny piece of the project done. My dolly. I decided to build a ladder dolly system mainly due to it's strength and light weight. I can also find a ladder just about anywhere if traveling on a project. I also decided early on to use 8020 aluminum profiles for the dolly. The goal was to be light weight and very adjustable. I could have made the dolly lighter with 1" sq tubing but the adjustability provided by the profiles is pretty important. Most of the parts are 20" or less so they fit in a pretty small case to ship. check out the images here . . .
The ladder is really filthy but even so the dolly rolls really smooth.
Now it's on to building the drive. The only problem now is the really long lead times for the stepper and gearbox from Anaheim Automation. Backup plan is to also purchase some Meade drive motors as a quickie solution if my project this is being built for comes all of a sudden. Another thought is to buy a 1rpm gear motor and use a Pololu Jrk motor driver as a continuous drive mechanism. I could use the same gears and shafts just a slightly different motor mount. Also this week I need to start futzing with my Arduino and get the electronics sorted out. Still mulling over ideas for the enclosure. I had thought about just putting everything in a Pelican case. I had decided in the past few days to put the motor drivers on or near the motors and just output S/D and camera shutter from the control box. That way motors can be tuned for their purpose, drive or pan/tilt.
Well many things to do and not enough time or money!
Cheers
Nice looking rig! Definitely
Nice looking rig!
Definitely keep those drivers closer to the motors - not only does it keep the heat away from your control unit, it also helps minimize power loss on longer cable runs.
!c
I have worked in film visual
I have worked in film visual effects for about 25 years and a lot of motion control rigs have the drivers close to the computer or in the same racks. But a lot of other rigs do it the other way, drivers by the motors. I think for my rig drivers by the motors is a better direction to head in. Also I could consolidate all the S/D channels in one cable to keep the amount of stuff to be connected, or forgotten about, down. That may not be very practical though for flexibilities sake.
I am also rethinking my enclosure for the arduino. I had wanted to build into a Pelican case but now maybe a smaller box that can go into a case with cables and stuff.
BTW: My part in the film biz is lighting for effects and live action.
Cheers
Also thanks Church!
Also thanks Church!
Hi Church Well with the
Hi Church
Well with the Anaheim Automation gearmotors hard to get in a timely manner I was thinking of making my own reduction with a worm drive. I know you have said that you liked 120:1 for the P/T but what would a good ratio be for the dolly drive. 20:1 or more? I figure I can whack together a worm and motor for around $120, gears can be pricey. Your thoughts.
Cheers
To figure out the right ratio
To figure out the right ratio for the dolly, you have to consider the drive distance before calculating the right gear ratio. I'm going to (for the purposes of illustration) assume that you're using a belt-drive with a pulley.
So, if my pulley is 1" in diameter, that means it's circumference is roughly 1*3.1416 or 3.1416". So, every time the pulley makes a full revolution, you'll move 3.1416 inches. Without any gearing, that means each step would result in 3.1416/1600 (200 steps * 8x microstepping) or roughly 0.002" of travel. Knowing that the minimum step distance one can plugin between shots is 1 step, that means your slowest speed will be 0.002" between each shot. That sounds like it'll work, right? But what if you're doing 1 shot per second? That means the slowest speed is 0.12" per minute. If that's a good enough speed for you, and the motor has enough torque, then you don't need any gear reduction =) Of course, you can always reduce the diameter of the pulley as well, as that results in smaller travel per step.
However, to talk about "how much torque is enough". The AA motors are listed in oz/in torque. So, a 1" diameter pulley results in a 0.5" lever (roughly - the lever length is approximately determined by the radius of the gear). At 100 oz/in, that means you're putting 50 oz's of torque onto the drive or about enough torque to move 3lbs from stand-still =) In that case, we want to move the torque up, even if we don't need a slower speed. In this case, we could increase the diameter of the pulley, and increase speed but increase the length of the lever and therefore the available torque.
If the camera weighs 5lbs, the cart weighs 8, and the motor has 100 oz/in of torque, and you decide a 1" pulley is "just right", then we can figure the gear reduction we need at a minimum: (5+8)*16 = x*(100*.5), where 'x' is the gear reduction ratio. So, we get 133=x*50, or x = 2.66:1 reduction ratio. That will be _just enough_ torque to move the camera on level ground, if you have a perfect no-friction bearing assembly. Those don't exist =) So, multiply by 2 to be safe. So about 6:1 would be the target gear ratio for that setup. That's way too minimal to justify worms, unless you're looking to get the holding friction provided by a worm gear setup (so you can cut power on vertical moves, for example). That could easily be done using a transition stage of a couple of pulleys. It's rare to find a gear ratio of greater than 4:1 using two pulleys (that would be a 4" pulley and a 1" pulley =) but you could transition from smaller to larger, and smaller to larger again in a 2:1 increments to get 4:1. Remember, at 4:1, your slowest speed is now four times slower, so now our min speed goes from .12"/min @ 1shot/sec to 0.03"/min at 1shot/sec.
I know the answer seems kind of long-winded, but I want to make sure you understand how I go about making these determinations, rather than just throwing a number at you and hoping it works. This way to have the ammo you need in case that perfect 1" pulley can't be sourced, or your stuff weighs twice as much as I think =)
(For the engineers in the house: yes, I know everything above is in-exact and approximate, it's better to be close and done today than perfect and done six months from now =)
!c
Thanks Church, that was a bit
Thanks Church, that was a bit more than I asked for but appreciated. In my passed real time moco work we always used a min 5-1 gearing for the lowest and sometimes a 40-1 for P/T. Anyway 20-1 sounded like a good place to start for the dolly, especially since I don't think this will ever be a real time rig. If it is a different drive would be in the works. I also liked the idea of cutting or sleeping the motors between moves with no creep, one of the better reasons for the worm. At least now I have a good place to start noodling it out.
Thanks again.
by any chance, has anyone
by any chance, has anyone set-up a spreadsheet, or google doc that you can plug in all these calculations? I'm not sure I'm following everything?
disney, check out this page:
disney, check out this page: http://bftgu.solarbotics.net/tutorials_mech_torque.html
It's a good tutorial on calculating torque.
!chris
very cool dolly. I built one
very cool dolly. I built one kinda like that but i love how you used the 80/20 so it can be adjustable. what size are the extrusions? the 10s? Or larger. I want to build one that can withstand about 40lbs. I shoot with a canon slr but i also want to be able to use a large Beta style HD camera.