I've been messing around getting my mindset for the season. I came up with this while messing around the last two nights
It moves at 8ft/s w/ a single speed transmission using an CIM and a motor coupled with the AndyMark AM Planetary Reducer(ratio is .367 overall).
Still playing with which motor I want to use for the AM Planetary. I am also messing around with the numbers to figure out the force it would apply.
This is really my first attempt at a gearbox using actual numbers so please feel free to comment…any tips are appreciated
First of all, good work so far. I do have a suggestion how ever. You may want to consider enlarging the outside plate (closer to the sprocket). This does 3 things:
you can support the motor output axles in 2 places
you can move the cross beams higher, above where the motors mount, to help strengthen the motor mount plate
makes machining easier, as you could make the same plate twice rather than making 2 different ones.
Again, good work so far, and congrats on posting your first gearbox on CD, that takes some courage to do.
In your design, I would make the small gears smaller and the big gears bigger. This will give your output a lower speed and higher torque and you will not have to have as much of a chain reduction aftwerward. You might even be able to couple it right to a wheel (depending on what size the wheel is).
Side Plate
One thought is on the plate where the sprocket is. Trying to remove that plate in the heat of competition from the robot would be especially difficult with the holes for the screws to mount to the frame where they are (imagine trying to get a screwdriver down there during finals when a gear shears). Flip the plate around and have the frame mounting screws on the outside.
Motor Position
Its a bit scary to see the motor so high on a thin piece of metal sticking up without any support. A solid whack from a random robot or team member and that could potentially bend just out of interfacing with the gear. Possibly consider moving the motors lower in the design and centralizing it between cross supports (I think this may have been touched on earlier).
I agree with Foley, I’d make that outerplate taller and support your motor shafts…and/or not put your gears so far out onto the motor’s shafts… set up a CNC mill/ or a prototrac and you could bust out enough of one plate…instead of many different plates…
I try and work and use a mirrored set of plates…it makes machining numerous times easier…
Pat,
As far as the gearing goes:
I’m positive you don’t need 2 stages for 8 fps from a 5500 RPM output motor.
Try dropping it to one stage, and utilizing a much larger initial reduction.
With a large single stage gear reduction, you should be able to get a nice sprocket reduction which scales to any wheel size (ie: 10:20 for a 4" Wheel, 10:30, for a 6" Wheel, 10:40, for an 8" Wheel, etc).
I’m in love with 10 tooth output sprockets for exactly this reason. It allows for a larger chain reduction (by sacrificing chain wrap, I know…)
Also, #35 - 10 tooths come with a TON of bore sizes.
This is really my first attempt at a gearbox using actual numbers so please feel free to comment…any tips are appreciated
A few comments on drive train / transmission design:
Design everything around maximizing power output. Essentially, this means that at the point where your wheels slip, your should be peaking current draw from your motors (40 amps each). You need to employ a factor of safety in there somewhere, perhaps 10%.
This implies that there will be different ratios for a design with the Fisher-Price or Drill motor. You need to pick one of these to design around.
People have mentioned having the shaft supported on both sides. This indeed generally a good rule of thumb. However, the real reason it’s good to have a shaft supported on both sides is to reduce shaft deflection. The moment of inertia of a shaft, which is a factor in it’s deflection is a function of diameter to the 4th power. Larger shafts mean you may not need to have it supported on both sides, especially early in the transmission where torques / forces are lower. Also, having a gear moving closer to the side of the transmission will reduce deflection.
A Karl Kreder-ism (college advisor on our team last year) is that the round posts like you have are difficult to work with, having ledges (ASCII sketch below showing top view) work better to encourage overall alignment and stiffness, which is critical for gears.
ASCII Sketch
* *
* *
* o o o o o o o o o o o o
* o o o o o o o o o o o o
*'s - walls with ledge notch cut in them.
o's - wall that fits on ledge.
Just some ideas. Ask for more details if you need 'em.
Those plates are going to be hell to manufacture. You’re going to need CNC machine tools to do so.
Consider making all rounded/angled surfaces 90* angles and you could do it manually. Also, the right side plate is going to be REAL hard to make, seeing as you’re going to have to start out with a solide aluminum block and mill it down so that it’s got the angle on the bottom, unless such a material is already sold that I’m not aware of.
As far as supporting the motor shafts… Place the gears as close to the plates as possible and no support is needed for the cim. The andymark reducer should not need support either based on how they are using it in their gearbox.
In fact, place all gears as close to the plates as possible to reduce deflection. As for your plates, it may be tough to find angle material that doesn’t require that most is cut off on the short leg. If you have a source, I am interested. I have stayed with 1/4" on my main plates and my latest design has the motors up high like yours.
The nice thing about Inventor is it lets you draw anything, the next step is to think like a machinist and decide if pretty or functional is more important. We perform weight loss on our plates with the bandsaw and drill press after the gearbox is proven functional.
The season is almost here, I hope everyone has built something by now. The last thing we all want is robots that do not move on the field!
Actually, the side plates would not be that tough to do on a manual mill. It might be a bit easier if everything were squared off, but the layout for this design is not too bad. The right side plate can just be made from a standard size piece of aluminum angle (a guesstimate based on the image is that a 3"x3" angle of 6061-T6 aluminum would do nicely, and they are readily available from On-Line Metals). Just cut down the flange on one side to make the horizontal mounting base. For the rounded top, align the position of the main axle on the plate with the rotational center of a rotary table mounted on the mill, offset by the radius of the curve, and cut the curve. Probably a maximum of a 45-minute job, including mounting and aligning the rotary table. Everything else is just X-Y coordinates to locate the holes to drill. Very easy if you have a DRO attached to the mill; a little more work but still very straightforward if you don’t have a DRO and have to count handle revolutions (i.e. the same technique that has worked just fine for machinists for the last 100 years). The left side plate would be done in a similar fashion, just making 45-degree cuts in two locations instead of the rounding out the top. Making those cuts is just a matter of locating the work properly on the mill table, setting the angle of the workpiece with a tri-square, clamping it in place and cutting the piece.
