pic: The Beast Box

Well, here’s a render of my latest CAD project.

Most details are welcome to be shared.

Low speed of 3 fps, high of 9. The intention is to have the driver have high as a default, and switch down in order to politely escort robots across the field.

The side plates are .125 thick plates of 7075. The track system is the OutBack Track system.

Questions and comments welcome. (And unlike the other thread when I posted something like this, I really don’t want a rule discussion… I just want pointers on my design.)

Is there any reason why you’re only using (it seems) one motor per side? Or is there a second set of gearboxes on the other side of the chasis?

There’s two motors there, you just can’t see the other one, as it’s directly below the one on top.

Sweet gearbox!

You might want to consider using one pneumatic cylinder instead of 2. If you put one in the middle and hook both shift bars to it, you get the same effect as two, but use half the air. (unless of course you’re planning to use a compressor or utilizing the space between the two cylinders)

Just a thought; our team did that with our pneumatic shifter this past year and it worked great. It helped us with weight by eliminating the extra cylinder and potential air tanks. :wink:

I notice that you are using 0.25" face width gears in the transmission. Did you already do the calculations for this and it will work? I am currently thinking on a single speed transmission that is small and light weight, and using smaller face width definately would save weight.

I tried that on my 05 bot, and wasn’t too pleased with it. It takes a bit more tuning to get the drive system to shift evenly, which makes driving a bit of an issue. Plus, each cylinder is a half inch stroke, 3/4 inch bore, so the air useage is minimal.

Yeah, these are 24 dp gears. The pinion and the first gear are 20, as that is the gearset that ends up being hit with the most force and wear in a drive system of this type. They’ll work just like 20 dp, but they’re smaller.

Sorry, I noticed that you appear to have quarter inch face width gears in your transmission. I woud like to know if you did any gear strength calculations with this decision, or if you are just making an educated guess. Hope that was clearer.

not to hi-jack the thread, but could you post the drawings of your set-up. it sounds very interesting.

Why 7075 plates? And why only 1/8"?

Also, you may want to run some lewis bending numbers on those 24dp gears… you may have already, but at first glance it looks iffy. Depends entirely on the material, though.

Otherwise looks good!

Edit: Also, you said you used 20 DP on the initial reduction off the CIMs, and jump down to 24 on the other stages? That is counter intuitive to me, as you are gearing it down each stage you move away from the motor, putting more torque through the teeth of those gears.

I think because it’s strong enough. We had the tracks this year and 1/8" is plenty enough to hold everything together. No point making them thicker if it’s not needed.

Those two support beams running through the center, are those 3/4" L’s? or are they squares?

Instead of lining up the motors perpendicular to the transmission, why not put worm gears on them and line them up parallel to the transmission? This way, it’ll give you more usable space and the tracks would be self-locking. No more rolling down the ramps after the power is cut. A robot with self-locking tracks from Outback would not be moved by anything at anytime unless it wants to.:smiley:

Why dont you integrate the tracks into the total length of the frame? Make them removeable but you could extend them the entire length & have them detach as a single unit for easy working while the rest of the team works on whatever mechanism is on top of the bot. In the drawing the plates just get welded onto the 2x1 (guessing) at the ends but I think you might run into issues with that.

I personally have no experience with worm gears, and so venturing there on a system this expensive would not be a good use of budget for me.

Those are 1x1 box extrusion pieces.

I personally have no clue how to run numbers like that. I’m a junior in high school, and if I learned to do that, I’d be seriously happy (so if you wanna teach me, then yay!). But I’m going from experience here. In the past, when we take apart an old tranny, the pinion set is always the one with the most wear and tear on it, and so I’ve kept that one nice and buffed up.

They actually are removeable, by two bolts per end of the tread, similar to what 997 did this year. However, I’m too lazy to put those bolts in at this CAD stage.

There was an excel spreadsheet posted here a few years back [only last year actually] that took the basic lewis bending formula and had a few boxes where you input the material, torque input, and face width, etc and it gave you aprox safety factors.

For those unfamilar with strength of materials and the like, it was a very good resource.

Let me try to find it…

Edit: Here it is… http://www.chiefdelphi.com/forums/attachment.php?attachmentid=3873&d=1137918029

Thank Matt Adams for that one!

We (753) used the same kind of transmission last year, as well as outback tracks. I would reccomend using servos to shift instead of pistons. we ran into problems with them. Good luck!!!

I know all about the problems that the Outback system had, and did a bit of CAD work to fix these issues. What you see here is a previous iteration of a design. The more modern version is much better, involving modded SuperShifters, or some equivalent.

Have you thought of using NBD (Nothing But Dewalts). We’ve used them since writing the paper from 647. We’ve had very little trouble (some faulty transmissions from the manufacturer) other then that they worked great. And we’ve used them with a few of the other motors in the KOP. In Fact that’s all we used in 2005. No Pnuematics.:ahh: