First Drive Train, ADVICE NEEDED PLS!!

Technical CAD
1

Hello everybody! This is my first drive train design and i need some advice on how to improve it. My team has the pleasure of using sheet metal in making a robot, so I’m learning solid works. I’ve had experience in creo before, but I’ve never designed a drive train. This drive train is not a drop centre on purpose. One thing i see it that the gearbox on this drive train is too low to the ground. Please let me know if there are any other flaws!

Pictures:
https://gyazo.com/b20742974482074db8a7560d0be5246a
https://gyazo.com/7d573a25e41766961490683fba9e4df6
https://gyazo.com/ebcbec27977e3acaf2c01e70e3afaefb

2

Ok, so a few things stand out to me. Firstly, I will say, for the most part, this appears to be a pretty solid design, maybe even overkill in some places…

That said, I’m a bit concerned about the curved oval pockets around the large holes in the center of the frame ends and in the spacers on the drive modules. If this isn’t being used for some sort of attachment and is purely aesthetic, I would ditch them in favor of regular holes. These sort of pockets could be a huge strength issue with the frame, and they leave the holes inside them fairly unsupported. It’s also far easier to fabricate without them.

The center support looks good, but the top connection points could be a bit longer, they seem like the most likely point of failure for that part.

As for the drive base, the CIMple Box is certainly an inexpensive option, but personally I would opt for something that offers more gearing options and easier maintenance, like an AM Toughbox or one of VEXPro’s gearboxes. This is more personal preference, but still something I’d recommend you consider.
Generally, I find it’s better to keep the drive channels as narrow as possible, though in this case your belt selection may make this difficult. If you could find a way to narrow it though, it would make it easier to control the rigidity of the drive, since you don’t have to bridge as large of a gap with supports (it would also leave you more room inside the chassis for components)

The belt setup looks straight-forward enough, I assume you’re running the gearbox to the back wheel and then running the other wheels off of that. Assuming this is a drop-center wheel base, you may want to consider driving the center AND back wheels directly from the gearbox (you’d have to offset the back wheel slightly but that shouldn’t be an issue), as this will provide better redundancy in the event of a belt failure or some other issue.

Overall, like I said, it looks pretty good, not bad at all for a beginner. :wink:

EDIT: On the note of ease of fabrication, one thing we ran into with our chassis this year (first time sheet metal frame here too), was that the screw/rivet holes on the bent tabs of our frame were too close to the bends to hold their shape, which resulted in us having to modify the design to allow the holes to be positioned farther away from the bends. You might want to talk with whoever will be fabricating this to see what kind of limitations their equipment has in this regard. On a related note, if you’re having a sponsor cut/punch the holes, see what standard sizes they use and try to use as few different sizes as possible, this will reduce the number of tool changes and the time/effort required to make the parts.

EDIT2: On the note of frame rigidity, one thing I’ve seen a few teams do is design their belly pans in such a way that they not only connect to the inner frame, but the outer drive base frame as well, which results in a FAR more rigid chassis overall. Not sure if you would run into issues with the belts getting in the way, but if it wouldn’t take much to accommodate, you should consider doing that.

3

This is a really good first go at a sheet metal chassis.

I can’t see how your axles are mounted, but keep in mind the dead axle is a structural member of your chassis and keeps the spacing of your inside and outside panel. I strongly recommend using axle blocks to avoid local deformation of the panels from your axles. You can see what I mean in 1114’s Solidworks series.

Running belts with the braces you have in between the inside and outside panels is going to make maintenance a nightmare. You can’t assemble a belt like you can chain. We usually tie the inside and outside panels together above or below the wheels to allow maintenance, Otherwise this becomes hard to change wheels and if you break a belt you’re not pulling rivets out.

The lightening holes you have at the corners seem really close to the edge. Might want smaller holes. You really don’t save much weight from making those holes larger, but the big thing they give you is finger / tool access.

I don’t know if you’re water jetting this or punching it, but cutting circles on a waterjet is much easier / faster than cutting other shapes because circles always fall through themselves and hexagons don’t always. It’s common for water to push the cutout up at weird angles and if it’s not circular it can catch on the work piece / slats that hold the work piece. This means you have to stop the run and remove the piece instead of the circle just falling through, otherwise your cutting head can crash and that’s not cheap. It’s not something you have to do, but whoever is cutting your parts may appreciate it.

As Chris said you’ll likely want to run a belt from your gearbox to your middle and rear wheels rather than belting from the rear to the front. This puts less load on that belt and will give you some redundancy.

If you’re interested send me a PM and a I can share some of our sheet metal chassis CAD from last year.

4

My biggest question is going to be what is your motivation to make a custom chassis rather than using the KoP chassis? What you’ve designed here is style-wise and parts-wise very similar to the KoP chassis. (bent sheet metal design, flanges with regular hole spacing for mounting superstructure, TB mini gearbox, dead-axle, AM HiGrip wheels, belts connecting the wheels) The main difference that I see is that the gearbox is belted to the wheels instead of direct driving the middle one. You should definitely consider whether the time, effort, and cost of making a custom chassis in-season is worth it to your team for those marginal gains.

If you do decide to design your own, chassis, this is a pretty good starting point. A few small things:
• Do you really want all of the power for each drivetrain side to go through a single belt? If that belt should break/skip/unseat you’ve lost a whole drivetrain side. That’s one of the main advantages of West Coast Drives, which direct drive one of wheels.
• The lightening needs a bit of work. Triangles and small circles are strong. Large circles, slots, and hexagons aren’t. Also consider whether you can use thinner material instead of spending time designing and cutting the lightening pattern.
• Make sure you have a large enough ground clearance. With 4" wheels and that tall of a side plate, your ground clearance is probably pretty low. I’d say you want at least 1/2" even with a flat field to account for unevenness in the floor.

5

Thank you for the advice! I was trying to design this drive train with the rules from this year’s game. I didn’t want to make it a drop centre to get maximum stability when the elevator would go on it. Other than that I get your thinking about the drop centre and using different/better gearboxes.

6

If you’re not using a drop-center, you may want to swap out the front and/or rear wheels for omni-wheels. Having a flat drive with all traction wheels will create wheel scrub and make it difficult to turn in place (much like flat tank treads). Back in the day when we used flat drive, our entire frame would sort of bounce around on the floor when turning as the drive broke friction with the carpet, it worked, but was not the most efficient (killed battery life), and would defeat any stability gains you would get from the flat drive.

7

I totally forgot about the belts and braces issue. We are using a laser cutter to produce our robot. I will have to double check with our sponsor if they can do hexagonal cuts. You’re right about using circles instead, it does make it easier and faster to cut.

If you don’t mind sharing your some of your cad files , it would be much appreciated :).

8

I’m using omnis in the front. We’ve used the same idea on our robot this year and it worked. The pivot point of the robot (when it was turning in place) was more towards the back of the robot tho. I assume by using omnis in the front and back would solve that issue and have the robot pivot in place at the centre wheels.

9

Ok, didn’t notice that it already had omnis when I first looked at the picture.

As far as center of rotation goes, there’s a few things that come into play here: Traction and Center of Mass. So balancing traction by adding omnis to the back would help, but where your weight is also affects this, so if all your weight is on the back of the robot, it’s still going to have a tendency to turn around the back half of the robot.

Now of course, the trade-off here is that if you go to omnis on the front and the back, you may have a harder time controlling turns at high speeds and may be more susceptible to defense, so you’ll need to balance the need for control vs stability and maneuverability.

10

The main reason I want to learn designing a custom drive train is because it provides unlimited amounts of possibilities and creative freedom in design. I’ve made this design simple and similar to a KOP drive train to learn the basics and to get advice from more experienced people.

I agree with you on the single drive belt for each side and how it can be a problem.

If i move the gearboxes above the belly pan, the ground clearance in 0.92".

Thanks for the advice!

11

In that case, good job! I’d recommend for your next drivetrain project to look into a West Coast Drive (WCD). In my opinion, they’re more customizable, easier to make, easier to maintain, and more practical for mounting mechanisms.

12

Nice job!

If your using a laser cutter to fabricate the frame then I’d consider going with a lightening style that takes advantage of the laser better than just circles. The reason 1114 used circles in their sheet metal frame was because it was fabricated on a laser+punch combo machine so circles were the fastest/cheapest way to remove material. If your sponsor is just using a laser a more traditional triangulated cutouts will remove more weight while retaining more strength.