I'm baaack

Wow it has been a long time.... who knew getting an engineering degree would be so much work [Sarcasm]. If anyone is wondering what i was doing these past couple years check out (www.sbusolar.com) . I just posted an Instructable HERE, so if you would all like to see these servos completed in the near future just go vote!!

Updates and size comparisons

It's been one hell of a first week back at school, and it does not appear that the workload will be easing anytime soon so I decided to get a bit of work done on these servos. I'm running through the assembly/disassembly and manufacturing side of the project to make sure these will work properly. Below you will find a comparison of the original servo block with standard sized servos and a block utilizing my servos; the size difference is considerable. I am sure some of you are wondering "Why the hell do you need 20 servos?!?"... Actually i need 24 but the other 4 need to have much more torque so I will be designing them later =). Now that I am back at the university I have access to the sheet metal shop so i should be able to whip up a prototype within the month (Hopefully).

Updates

You'll notice by the picture that a few things have been added since the last post. C-clips now hold the gears onto the shafts, i'm thinking of just press fitting the gears on with a .0010 - .0015 press fit; possibly heat the gear up and increase it to .002 press fit. This way I can stay away from keys and whatnot. The main gear is being held on with a two part shaft that also the piece that you will attach things to the servo. This servo is going to be a little tricky to machine without at least a basic mill or lathe; i'll go over the designs once i get a working prototype and simplify it all.

Gear train layout

Here is a short video showing all the gears in this design. It is missing the supports and pins, but it's the majority of the final servo. The gears appear to be moving a little funky but that is because the motor is spinning at 4000rpm and Inventor is having a hard time rendering it.





For size comparison here is my servo on top of a standard HITEC servo.

Gears!

I am finally back from the Solar Splash Competition in Arkansas, I got in at 3am and after some sleep I started right back up with the servos.

I have been narrowing my search for gears, and came across some nice gear stock from WM BERG(link below). These bars are 10" long and cost about $10 for a length. Meaning that if you were to use .38" of stock on each gear it would cost you 38 cents per gear. There are 2 of these gears (one 10 tooth and one 20 tooth) so that's 76 cents for this portion of the gear train, plus ~$1 for each motor; so with the motor and 2/3 of the gear train we are still under 2 dollars for each servo. The following still needs to be finished:

1. The main carrier plate needs to be finished so i can calculate cost.
2. The main gear/output shaft needs to be designed
3. The control circuitry

I should be able to get the carrier plate finished by the end of the week.

http://www.wmberg.com/catalog/catpage.aspx?url=pdf/B05B116.pdf

Finals > Servos

Progress has been slow the past couple of days, between studying for finals and working on the boat I haven't had much time to do anything. One of the biggest issues I can foresee with this project is making the main mounting plate (pretty much the most important part) easy to manufacture. So to remedy this I am going to go through the process of MY servo first, and then once i get a working model i'll produce a second design that can be made with simple hand tools (drill,hacksaw,hammer,etc...).

Finding gears the right size and ratio are proving difficult, I'm trying to design these servos so that they can be produced for under $10 each; and the gears seem to be adding up quickly. There are a few options that I am considering:

1. Machine a set myself, and cast a rubber mold to produce ABS/resin gears
a. This would allow you to make sets as you need them
b. However you can only make a small amount at a time
c. The other drawback is that the gears will be weaker then machining metal versions.
2. Design the gears to be machined, and run a large batch
a. Gears can be made quickly and in large batches
b. You can't just make one set, you'd have to produce 10-20 sets to be cost effective.
3. Design the servo around available gears
a. you can buy sets as you need them
b. Expensive

The third option is at the end of the list for a reason, doing this would almost double the size of the servo; which frankly will not work with my prosthesis. I'm going to consult the machinists handbook and design some gears myself when i get a chance.

Don't expect any other updates until early June, I leave for the "Solar Splash" competition the 25th and will not get back until June.

Gear Layout

Here is my current design for the gear layout, it's missing alot of supports and bushings at the moment but you'll get the idea. The motor connects to the blue gear, which meshes with the green gear equating to a 2:1 ratio. The green gear then spins the red worm gear which is meshing with the yellow gear with a ratio of 50:1. The yellow gear is also the output gear (shaft is underneath in this view), which will give you a 100:1 ratio * .44 oz/in = 44 oz/in of stall torque.

Just to give you some sense of scale, that blue gear is .25" in diameter.

Next step is to add the rest of the bushings/supports.