818 Gearbox and Chassis Design

[NolanV]

This offseason our team has worked on designing a gearbox and chassis to use in the coming years. We have always used the kit chassis for a 6 wheel tank drive setup. Our goals for the gearbox were to have a common design for 4", 6", and 8" wheels at whatever speed gives us more pushing force than our wheels can hold at 40A, 12ft/s, and ~18ft/s. We achieved this with a 3 stage shifting gearbox with reverse mounted motors, this helped keep the gearbox as thin as possible. The third stage is what allows us to keep the same robot speed with different sized wheels. Here is a link to the google sheet where we did the math for the gearbox https://docs.google.com/spreadsheets/d/1AprbYyvjUoRbo5IgMOXDJVuAzUCjZJ_YA-TuTIEV9Ow/edit?usp=sharing

After finalizing the gearbox design, we wanted to integrate it into a parametric chassis. There is nothing particularly notable about the chassis other than it is fully parametric.

Test Chassis 20191220.STEP (71.0 MB) 818 Custom Gearbox 20191220.STEP (53.1 MB)

[nmatthes]

Have you thought about power consumption on running 6 falcons on just your drive train? Not including mechanisms, the compressor running to shift, and any other motors.

[ProPain37]

A. That looks like a heavy gearbox with all those gears. You could likely reduce weight by using 8T pinions rather than 12T pinions so you wouldn’t have to reduce the gear ratio as much. You might even be able to get rid of the 3rd stage.
B. I’m gonna say that 3 Falcons is likely overkill considering a lot of teams do just fine with 2 CIMs.
C. Rather than have one gearbox that can adapt to the change in game, make a new gearbox each year. This way the gearbox would be lighter and more game specific and more tuned to what wheels you are going to use.
D. I would pocket the gearbox plates to save further weight
E. Your bellypan looks like it could be lighter as well.

[NolanV]

Yes, we have considered it, we are going to implement power management code. We are going to limit the current of the drive motors in different situations, as well as using motion profiles that include jerk in our moves to reduce the amount of current draw when starting motion. Also, with the gear ratios we picked the motors shouldn’t have to work terribly hard to get the robot moving.

[typeusernamehere]

That thing is gonna kill some poor defense bot

[NolanV]

Thanks for the feedback!
A. The gearbox with motors and all except for the wheel is 7.21 lbs. I am not a fan of the 8t pinions as they are not true 20DP tooth profile, 12t is the smallest gear with the true profile. The 3rd stage is what allows us to change wheel size easily. For 4" wheels the ratio is 1:1, for 6" wheels the ratio is 1.5:1, and for 8" wheels the ratio is 2:1. The first large cluster gear is the size it is to be able to space the motors out far enough so that they don’t hit each other. Then we are somewhat limited on shifting stages. There was a lot of consideration behind what gears to use for each stage to try and hit our 3 target speeds. We are able to swap out our shifting stages to achieve top speeds of 6.5, 12, and 17 ft/s. This allows us to decide for each game what options we want and be able to change our minds mid way through the season and only have to swap out two gears. So we order a full set of gears to work for any game and we don’t have to worry about gears being in stock or shipping delays (this burned us really bad a couple years ago).

B. The style of our team is to have a strong drive so that we can play defense well and move other robots on the field, similar to 217 this year. We select the highest COF wheels we can every year (within reason) this usually ends up being AM plaction wheels with blue tread. We have been using 3 mini cims for the last few years, and have been very happy. We had the debate about 2 vs 3 when they came out, but the design was originally started with NEOs and then slightly modified for the Falcon so we left the option for 3 based on the fact that they are still way lighter than what we had been using in the past.

C. Our team doesn’t meet on weekends and are somewhat limited in our capacity as a team, so having one less thing to really worry about every year will be a huge help for us. We use 4, 6, or 8 inch plaction wheels with tread depending on desired ground clearance or 8" Pneumatic wheels if the game requires. We also determined that we will only want a combination of a pushing gear, a 12ft/s gear for mostly half field games, and 17 ft/s for full field games. This also allows us to have parts stocked for our gearbox before we have made any decisions and just have to make a new set of mounting plates and spacers and we are off and running very quickly.

D. Our intent was to build the chassis and gearbox during the off season and then make a final design with a little bit of light weighting, but we didn’t get a chance to build them. Also, all three of the plates combined are only 1.64lb so we won’t be able to save much weight.

E. The belly pan is currently 5.47lb, we could go a little thinner with our spacing between the diamonds, but this design was mostly about the gearbox. We are also going to be adding mounting features for our electronics on the bellypan, so it is also not final. Also, going back to our play style, the more rigid our base is the better.

Overall, the design does not meet the needs of every team, but based on our goals for the gearbox design and the needs of our team this design is what we need.
-It gives us flexibility for wheel size
-flexibility for robot speed
-gives us shifting in a more compact package than anything I could find from Vex or WCP
-it is relatively easy to assembly and service with all of the bolts being accessible and going the same direction

[NolanV]

That is the goal!
Or to be a really good defense bot.

[Van_DanderWide]

Looks like a great job of matching a relational design to a team strategy! And there is flexibility to integrate the Game Strategy into it too. We have a similar design, but not as parametric.
I see you place the Belly-Pan on the top of the frame rather than below. Can you share your thoughts on this?
Also, it looks like you are using the 18t sprockets, and refer to the ‘jerk’ motion. I am concerned about this being amplified when using 8" wheels, and the increased moment on the small diameter sprockets. Do you see any reason to use larger sprockets with the larger wheels?

[NolanV]

We went back and forth quite a bit on bellypan on top. Putting it on top allows for electronics to be mounted under the robot (if the field allows it) and it allows us to access the chains easier if we have any issues. We are using jerk as in the derivative of acceleration so we ramp up our acceleration. This reduces torque when starting from a stop and makes the robot more stable and less tippy. The sprockets is one thing that we did not put super careful consideration into. Is your concern the tooth can’t handle the load under heavy acceleration with an 8" wheel. If so my answer is we will see, but I don’t expect to have any issues. The smaller sprockets also package well under the bellypan.

[jackoleary]

Looks like it’s gonna be pretty heavy… Is there any reason not to pocket those plates?

Also, you can eliminate that small plate if you opt to run chain in tube and mount the middle plate directly to the tube

[NolanV]

The gearbox with motors and all except for the wheel is 7.21 lbs. The 3 plates only weigh 1.64 lbs so pockets won’t save us much weight. We had been using 3 minicims on our robots which is 6.5 lbs, so our whole assembly is much lighter.

If we were to go to chain in tube our motors and shifter would stick out over the wheels and maybe outside of the frame. Also the shifter is mounted to the small plate so we can not eliminate it.

[Burritoman]

Not a particular issue or anything, but are you planning to have your bed pan purposely over the frame instead of under?

[Connor_McBride]

The only thing that comes to my mind is that the electronics might be mounted upside-down.

[Matt_G]

Should be fine as long as you write your code upside-down as well.

[Connor_McBride]

Okay but if you live in Australia would it then be right side up?

[jackoleary]

Sticking out over the wheels is perfectly fine as long as you know not to hit stuff sideways. As far as pocketing goes, if you can save a pound or two you may as well but it’s up to you.

[AriMB]

Not a particular issue or anything, but are you planning to have your bed pan purposely over the frame instead of under?

Not to be pedantic here, but if your frame has a bedpan in it then you probably have more serious problems

[Ryguy7824]

Yeah, we’re purposely going over for easy access to the chain. If one pops off, we just need to flip the robot over and pop it back on. If the belly pan is under the frame, we have to work around the electronics and anything else that is there around the sprockets.

We also have the option to mount our electronics underneath, should the game field allow it and we need the extra space in the middle of the robot, just like @Connor_McBride said.