Butterfly Drive Train with 2 Pneumatics?

Hello,

I was looking around at butterfly drive trains, and everyone that I see uses 4 pneumatics. Is it at all possible to do the same thing with 2 slightly larger pneumatics? I was thinking to set them vertical in between the modules to translate force downward onto a bar which turns both modules at the same time on each side. Thanks!

I have seen it done with two before. One way you could do it is if you had a piston with a 10" stroke, you could build a rack n pinion type of actuator and extend the pneumatic long enough to actuate both of the pods on one side simultaneously. Another option would be to 3D Print or CNC Laser Cut an actuator that would combine the force a single pneumatic and thrust out the 2 pods per side. It does get tricky if you have more than 2 motions though. If you do it in a way that you would have it engage 4 mecanum wheels on one actuation and then 4 standard performance wheels on another actuation that would be fine. If you wanted to ever engage all 8 wheels at once though you would have to make that the starting position and then actuate the command after you start the robot up. Then you can convert to an 8 wheel drive. Let me know if you have any more questions.

Certainly viable, but I see several reasons why teams dedicate one cylinder per module. For one thing, the mechanical linkage you described would not be as simple as you think. To have an even, consistent force applied to two different modules at the same time is not as easy a task as it first appears. Also, teams often utilise the space between wheel modules on each side. Having a mechanism in there to connect the modules takes up some of that precious space. Also, the suspension provided by individual cylinders can be very beneficial for weight distribution and for consistent movement.

But if using connected wheel modules fits your design requirements better, go for it!

Like you could, but why? There are some very simple modules out there (versadrop plz) if you did want to go that route, and the bigger actuators would likely be more weight consuming and design time intensive, as well as potentially less functional as Corthon mentioned. With a bit of smarts you can also wire all 4 actuators to the same solenoid so the actuations are perfectly in sync.

Check out 1086 basically any year they used octocanum (2017, 2014, 2013). It’s hard to see from the top, but from underneath you can see there is only one cylinder on each side.

https://scontent.ftlv1-1.fna.fbcdn.net/v/t31.0-8/fr/cp0/e15/q65/17504464_10155139194134282_4419264118352063821_o.jpg?efg=eyJpIjoidCJ9&oh=79c9950d460f236b29e492018f6d08cf&oe=5A24ED76

The trick with using one cylinder per side is you want to limit the size of the stroke as much as possible and just get bimba to add an extension for any additional length you need. The way 1086’s modules are structured, you actually end up with a smaller, more compact system with one cylinder running between the modules than two individual cylinders.

I feel like the bigger issue here isn’t the mechanics of getting it to work but the available pressure of the pneumatics. Remember that the pneumatic cylinders are effectively supporting the weight of the entire robot when deployed, so you would need a cylinder with a larger bore if you go to just 2 cylinders. You may find such pneumatic cylinders may be larger and more expensive than you expect.

You don’t need Bimba to make extensions for you at all. There’s nothing magic about what attaches to a cylinder! You can easily make your own structure to do this.

One example would be 488’s geometry in 2011, if you dig in this thread here you can see the example. You don’t want to have the piston push straight down as proposed in this thread - you ideally want a geometry such that the weight of the robot doesn’t as easily cause the piston to deflect