Differential Concept

Differential Hood Concept

This is a CAD model I spent a few days working on. I was unable to finish it due to the limitations of my computer. Either way, I came up with this idea a few days after the game reveal when I worked with @Xenon27 on his CI2D. This in tandem with discussions of some strategies with the SCA for what an ideal robot should have for this year and the idea of a trap door came to mind. The CAD itself was made over the course of the early season in bits in pieces because my computer doesn’t know how to cooperate.

A trap door is essentially a subsystem on a flywheel that passively (or actively) cycles out unwanted balls from within the robot. This enables for an overall more efficient collection and scoring system because the robot does not need to move from scoring position to cycle out unwanted balls. For my particular design, I had the flywheel hood act as a “trap door” so to speak. The balls will be intaked upwards and plopped out the back of the robot when the hood shifts forward.


I was initially inspired by the discussions being had on discord and the forums. My solution was based on a slightly different mechanism, but they were functionally the same. I’m referring to this.

This video showcases late-season 169Y & A as well as 9605a. These teams had angling hoods in turning point and I thought it was an efficient solution for this game given the early season meta.

I also saw the use of differentials and saw how similar the motion of an angling flywheel hood is to something like a chassis to tilter differential or mobile-goal lift to chassis differential. I also knew I wanted a flywheel to be powered by two motors. You can see the overlap here clearly. A hood to flywheel differential seems like a good application for this.

Why a differential?

First is the trap door which was explained above.

Though this mechanism isn’t really fast, it definitely can be with proper tuning. The benefit of using a differential for this is you now have an active hood. Meaning you can shoot balls and change the angle the balls are shot at at the same time.

Additionally, staying within the size limits will be really easy. This wasn’t the main goal behind this design, but it is definitely a plus to consider. You can very easily change the state of the hood.

Design Philosphy

So, when I designed this robot, I wanted speed to be prioritized. I know this game was designed for fast cycling and strategic match play. This is why I chose a hood bot. There is a lot of conversation on the benefits of a hood bot. As can be seen:

That being said, the drive is geared 3:7 on 600 rpm motors. This totals out to 257 rpm on 4" wheels. This seems to be an ideal ratio for this years game, but I wouldn’t be opposed to something faster.

The chassis also features an elevated gearbox that is as space-efficient as possible. This was inspired by the Pilons ITZ robot and 9605a TP robot. This allows the middle of the chassis to be entirely open to the game elements for possibly a greater capacity. There are also tracking wheels. I was just interested to see if I’d be able to fit them on a robot like this and it turns out they fit just fine. This was mainly because I am a programmer, and I also want to encourage others to explore that daunting field.

The flywheel itself is on a differential. Both motors (on the hood and flywheel) power the flywheel at once, or the hood at once. The setup could also move both the flywheel and hood simultaneously but with significant power drops. I believe it will take a good programmer to make this work well.

The flywheel is more of an overall indexer. There are no intakes on this robot, but mainly because that was never the point of this robot. I just wanted this to be a proof of concept for a possible motor sharing technique. The motors powering the flywheel would be 200 rpm making the fkywheel spin at 1400 rpm at top speed, but it might not need to go that fast. Mind you, it is on 2 motors. With good build quality, there are minimal friction losses.

If I were to make intakes, I’d probably get somewhere between 600 rpm - 300 rpm for an ideal ratio. It’s too soon to tell what will be best. I also think omni wheeled intakes tensioned inwards would work well. This is because the intake would adapt to the curvature of the ball as the ball is being intaked. The intakes would be compliant. This wouldn’t work well for fielding, but it’d work really well for descoring and cycling. Just remember W = f*d, do more work with the same amount of force on the ball by covering more distance. I think a secondary intaking system like a scraper would be necessary for these kind of intakes. More on that in another thread though.

Anyways, more on differentials here.

More renders

Screen Shot 2020-06-06 at 11.59.53 PM
Screen Shot 2020-06-07 at 12.00.11 AM
Screen Shot 2020-06-07 at 12.00.28 AM
Screen Shot 2020-06-07 at 12.58.34 AM
Screen Shot 2020-06-07 at 12.58.54 AM
Screen Shot 2020-06-07 at 12.57.32 AM

Screen Shot 2020-06-06 at 11.59.36 PM
Screen Shot 2020-06-07 at 12.57.52 AM
there is one thing to note: *I intentionally made the dimensions of the CAD off in size in a few places that will make this robot impossible to hole count So don’t try it lol


I like the concept! Nice job! Can’t wait to see it in action!

1 Like


Overall, that’s a really nice bot you’ve got over there! Great Job!


On the note of your computer having issues, don’t put in motors until the end as they are a lot of parts. The same goes for Omni wheels.


I’d add if you really need the motors and wheels, you can hide them pretty easily. This should boost your performance as the motors and wheels are made up of a lot of components.

Really nice design! I think this will be a popular choice!


The CAD looks great, it’s definitely unique from all other designs so far.


You absolute legend


this is probably better than a trapdoor, because it will have less friction than a ratchet (or three ratchets) if built properly. It also have the potential to create more opportunities using the adjustable hood, perhaps you could control the trajectory of your shots so you can have like a foot or two of distance from the goal where you can still land shots.


If u use the library from the CAD discord, motors are only one part

1 Like

I feel like you went too far here Taran, but I completely agree with everything else you said. I have to admit, I have come a long ways from my first year in VEX (TP). That is because of people like Taran who are willing to just put their working designs (while not always perfect) out there for people to question, criticize, study, and learn from. Nothing is EVER going to be perfect in it’s first design, and most of what is put here in the VEX Forum is in its first design. It’s so people can improve it, or even make their own designs which were INSPIRED by the first-stage designs. If you are going to put something out there, don’t make it a lie. Put it out there in its true form.

Honestly, VEX has become WAY more competitive than when I started (only 2 years ago!). The original purpose was for kids interested in programming, building, and learning about robotics (or STEM in general) could come to learn and collaborate. We need to come back to that stage where we are not afraid to help others in their passion.


It really hasn’t.

Just because you weren’t aware of that more competitive scene doesn’t mean it didn’t exist.


bruh taran none of this is to get recognition.

I never hide anything on my designs, in fact most of my robot is completely public. And this topic was just made to demonstrate a concept, without giving a tutorial on how to build it.

this has nothing to do with matt, so don’t even try to bring him into this.

This topic does help the community, as do all the SCA topics. If you have a problem with how @mvas presents his ideas, take it up with him, don’t attempt to immaturely derail his helpful and insightful topic because he didn’t want people to holecount.


Just because they’re avoiding holecounting doesn’t mean they’ve “fallen.” It serves as a proof of concept, while not allowing people to directly copy. Directly copying is not learning. I mean yeah you see how it works, but there has to be some degree of originality. Plagiarism is illegal, because copying someone else’s work and calling it your own is wrong, but for educational purposes it means you ARE NOT LEARNING ANYTHING. If anything @mvas is promoting learning opportunities by forcing people to see ideas, and recreate them originally, as opposed to copying and not learning anything. Releasing all or some info is completely up to the discretion of the designer, and no one else…


I’m going to be real here, I don’t see any aspect of whatever @mvas has posted here that isn’t “sharing ideas” or “helping others”. If anything, I’d argue that whatever he’s doing is trying to mitigate the impact that holecounting can have on regions and instead guide other teams towards basing their designs off another robot.


Let’s not bring this to a personal level here.

1 Like

BY THE WAY , I’m not even competing in Vex EDR next year for anyone wondering. This was just a fun little side project. I will be planning on joining a Vex U team whose robot will probably be mainly 3D printed anyways. So for anyone wondering if this had some personal deeper agenda to it, it really doesn’t. Quarantine is just really boring.


We should stop arguing and let it be,
@mvas This looks like a good idea I am excited to see how this turns out


I’m defending personal attacks on SCA made by taran, he’s the one that made the attacks.

SCA exists to help each other, and to better the vex forum community. This topic is a great example of a helpful post, that outlines a well thought out concept. I personally don’t take any measures against holecounting, because I don’t care if someone tries to holecount my stuff. And just because @mvas did take measures, he’s doing it because he doesn’t want people to copy and not learn anything. This is a concept, not a tutorial. It isn’t meant to give teams instructions on how to build. It’s meant to give inspiration and ideas to team looking to up their complexity.

and tbh, if you need to holecount to have a robot, you shouldn’t be attempting a differential anyways.

Now can we please get this back on topic?


I wasn’t saying that to anyone personal. That was for all around. I didn’t realize that I hit reply to you. Geesh.

Thank You (20 chars)


Just a friendly neighborhood reminder to keep it on topic and civil. Love the idea @mvas! Could be pretty useful!