pic: FRC 1983 Skunk Works Robotics 2015 Preseason Drive Base

Meet our preseason drive base Slippery Biscuit. In the past we have designed a rectangular chassis and it has worked well for us. Because of the nature of last year’s open field, we ran into major issues with friction pinning. These issues became more prominent when we competed against robots with 3 CIM transmissions versus our 2 CIM transmissions. So, for this year’s preseason project, we choose to design and build a drive base that contained some good solutions to friction pinning to determine how well they work and how to build them during build season. This design has redundant friction pin solutions. The link the the CAD model is https://drive.google.com/open?id=0BycP6Cx4kCRpUDlwUVJ1NjA0VkE&authuser=0

The drive base features a sheet metal, 6 wheel tank drive with a dropped center wheel for improved maneuverability. It has an octagon shape with an outward slope of 25 degrees yielding a larger platform and a geometry that allows steering out of a friction pin situation. This is inspired by FRC Team 971, Spartan Robotics. All wheels are VEXPro 4 inch traction wheels using Blue Nitrile Rough Top tread. The front and back wheels are 1 inch wide riding on dead axles while the center wheels are 2 inches wide on a live axle. The center wheels are driven by a chain off the transmissions. The chain is tensioned using a cam that positions transmission. The outer wheels are driven off the center shaft using 9mm wide, 5mm pitch HTD belts on 42 tooth VersaPulleys. The belts are tensioned by adjusting the position of the dead axles.

The transmissions have custom water jetted and milled aluminum bearing plates and standard Andymark Supershifter gears/shafts with 3 in-line CIMs. A dog-shifter enables a high gear ratio of 4.82:1 (16 fps) and a low gear ratio of 12.35:1 (6 fps). Speeds are calculated at 81% of CIM free speed. High speed reduction is 12:40, 28:35, and 19:22. Low speed reduction is 12:40, 15:48, and 19:22. The only custom gear in the transmission is 3-D printed idler gear that drives a Grayhill encoder.

In between the outer and center wheels are drop omni-wheels that are pneumatically deployed. This enables the robot to pivot out of friction pinning situations. This feature is inspired by FRC Team 118, Robonauts. This design features a departure from our traditional two-color Velcro bumper system. Here we are studying a single piece bumper system that is easily removed, yet rugged under heavy defense. Over-centering toggle clamps retain the bumper by pinching a flange attached to the bumper against the top of the chassis.

Things we would like to investigate further:
Instead of using a chain to drive our center shaft, we would like to drive it using gears. We will not place the air tanks in the frame. We thought it would be a good use of space since we wanted a box frame in the front and back, but the tanks are difficult to service if a leak occurs. As for belts, they are nice and light but they are difficult to remove/replace.

Looking good, can’t wait to see it in action! Great job on the model.

It was looking pretty far along last time I visited, how soon until the Slippery Biscuit hits the carpet?

Looks great.

My only suggestion would be to replace that center chain with gears for sure. We have had alot of trouble in the past with a single chain driving an entire side of the robot. That chain fails, and you are dead in the water the rest of the match. Plus moving an entire transmission to tension a chain is a pain.

Out of curiosity, how much does this weigh, sans battery and electronics?

Besides a few missing parts for the belt tensioners, the robot mechanically is done. Since we decided to use roboRIO for it after we had made the design, we are lacking few electrical components for it and won’t be able to install those until the beginning of build season. I think we should be looking at operating it by the end of the first week of build season.

Of course, this will not be used for competition. It will be a test bench for the programming team.

The drive base with battery, electronics, and pneumatics weights approximately 54 pounds. So this doesn’t include the bumpers. The drive base frame by itself weighs about 15 lbs.

What type of 3d printer are you printing your gear off of?

Also I see you included handles in your design. Out of curiosity what material are you planning on using there and where could I buy it.

In all looks like a really well made design. I bet no matter what the game you won’t have to worry about using and abusing it through a whole season. Solid drive train.

We have found handles to be an important feature of our robots. they go into the model at the beginning of the design to be sure they are not an afterthought.

The 3-D printer we use I believe is a Stratasys Dimension 1200es or a similar model. As for the handles, we got them from McMaster-Carr.

Here is the link http://www.mcmaster.com/#standard-pull-handles/=v61cau

I would look into Team 148s off season hex drive train. They did sheet metal as well and it seems as they found that any more angle than 8 degrees didn’t help much with friction pins. I’m not sure in that though so don’t quote me, but do check it out.

I thought you could not use your off season stuff for the upcoming build season?

The rules may have changed, but I remember there being something about being able to use pre-season work so long as it’s publically available before kickoff (feel free to correct me if I’m wrong). We’ve never completely reused a design, though, since there’s usually much more integration of the drivebase and with the rest of the design, and we get curveball rules that require some dimension changes. Even still, we make a new model during build season (though it ends up being pretty similar to anything we previously did)

They won’t be competing with this drivebase, so it doesn’t matter.

Interesting concept.

I really like that teams are focusing on mobility. I know that a lot of the FRC world likes defense but I’m a big fan of offense and that means getting out off pins.

Overclocked built Rover with mobility in mind with many of the features you discussed, specifically continuous bumpers which are a big advantage imho.

I suggest you consider sailcloth as an added feature. The Rover project has convinced me that sailcloth is the way to go (watch the video for a demo),

Great preseason project.

Dr. Joe J.

Great work skunks! I can’t wait to see what you guys come up with for this season.

On the topic of sail cloth, I remember that last year you guys played a balanced mix of defense and offense. If you want to continue with that in 2015, I would recommend using sail cloth on the sides of your robot, and a higher friction bumper material on the front and back.

I’m fairly sure any design or code you do in the pre-season may be re-used so long as you publish it.

Physical components may not be re-used unless they are unmodified commercial off-the-shelf components.

Sailcloth is definitely on our radar. We have purchased some for testing, but we were a little late in the preseason to actually perform any tests.

We have enjoyed the convenience of permanent bumpers with convertible colors. They can look a little tattered and can violate floor clearance rules if the Velcro isn’t placed well. A particular difficulty for us has been recruiting a new adult with sewing skills and equipment to make bumper covers. I say “new” because the person who did it “last” year will inevitably not want to do it “this” year. It’s a lot of work and difficult sewing. We want a bumper system that can be completely assembled by students.

As far as bumper retention goes, we have looked at latch solutions from 254 and 971. Many teams use fasteners of one type or another. The over-toggling clamps shown in the model seem robust and are easy to operate. We like the type with the clamp-lock for security. They are pricey, however. Look for removable bumpers (red and blue) from us in the future.

Of course, this assumes we will have bumpers this year. Who knows, maybe that’s the “change” that’s coming.:yikes:

It’s also legal to use a prebuilt drivebase for your prototype robot if you build one. I don’t know if that’s what 1983 is planning on, but it’s a good way to have something to test code on early in build season if you have the capability.

Also - this thing looks sweet! Definitely something I’ll be keeping an eye out for at PNWs.