I have been interested in designing and programming an omni directional robot for a few years now and so I decided to design an Octanum drive. I chose this over something like swerve or even an H-Drive due to it being simpler and given our capabilities. Hopefully I’ll be able to convince our Mentor to give us the funds to fabricate this now.
It runs on 4 Cims, 7fps traction wheels, 21fps mecanum wheels and weighs in at 50lbs. The pistons used to raise and lower the wheels are 1 1/16th" bore with 1.5" stroke.
I will most likely be uploading the CAD files in a few days if anyone is interested. Any and all critiques/suggestions are greatly appreciated.
It looks like the traction wheels are fairly close together, the drive may be tippy when in traction mode. You may be able to package the wheels better with the 4" mecanum wheels from Vex, but I don’t have any experience with their performance.
21 fps on mecanum mode may be geared a bit fast for 4 cims, the acceleration may be sluggish.
Also about the tippiness in traction mode, I can’t tell how much the wheels lift up, but is there any chance of rocking back onto the mecanums in a pushing match?
try a more realistic speed for the mecanum, like 14 fps.
I would recommend pivoting the traction wheel instead of the mecanum wheel. The reasons being that the mecanum wheel always applies sideways force and you will spend more time on the mecanum wheels.
Double check your physics for the piston to make sure it can lift a 150ish lbs robot.
I’m looking at gearing the mecanums rather high for this chassis just to have some fun with it. For a competition bot I would certainly lower this. That being said I will probably still reduce this a few fps for improved performance.
In terms of the tipping, I hadn’t really considered that. Now that I think of that though you are most certainly correct and I will be looking into the best solution for this.
The mecanum wheels are raised 1/8" above the ground when the traction wheels are down. So I would say that there is a good possibility of the mecanums hitting the ground should the bot start tipping around.
I would agree about your point on pivoting the traction wheels instead. The reason I have put them this way is due to size constraints due to the larger 6in wheels. However if I were to switch over to the 4in wheels like recommended above, than the size constraints would be no more.
I have just re-run my calculations and unfortunately it would appear I have made an error in my initial calculations however should I change my design to incorporate 4in mecanums, the change in dimensions would allow this piston size to work.
I would try to avoid this if at all possible. Not only do you loose traction and forward force, but mecanum wheels can do some unpredictable things when they don’t contact the ground evenly. This is one of the reasons that it’s more common to have the traction wheels at the corners rather than the inside of octocanum/butterfly/nonadrives.
Is there a particular reason that your chassis paneling needs to be so tall, extending above the tops of the 6" wheels? Seems you could save quite a bit of weight by chopping the top few inches off of the frame.
I would consider trying to integrate your initial gear reduction, or at least the final stage of it, into the module. This would allow you to use a dead axle for the module to pivot about, giving your sideplates and module much better rigidity exactly where they need it. This will be easier if you make the module pivot about the Mecanum wheel. Many similar designs are able to get away with a single stage reduction this way, by gearing to the faster wheel first, then incorporating further reduction into the belt run to the traction wheel.
I’d be concerned with the gap in the middle of the frame in conjunction with the thin, unflanged parallel plate construction. Your frame is four thin parallel plates in the middle, which can bend with each other easily. Either fill in this gap by making the front and rear chassis members one piece, or thicken or put some flanges on the chassis side rails to stiffen them up.
Agreed, those pistons look a little small to be lifting a robot on. I like what you’re doing with the single piston across the top though. Just make sure your hard stops are well defined and integrated, to ensure that it behaves as expected.
Are those 6" wheels? That 21 FPS number seems familiar to me. (Specifically 1102’s mechanum used 6" wheels this past season and those are the numbers they gave me. It worked out pretty well for them too.
Yes after all of the comments I have seen above (including yours) I will almost certainly be converting the design for a 4in mecanum. More pros than cons (cant see any cons at the moment). Also by only using a single reduction I can probably save a good amount of weight aswell apposed to using the toughbox mini’s.
The inner plates actually have two flanges on the bottom allowing the front middle support to be fastened aswell as the sides of the back support. We ran with this almost identical setup this year with a west coast drive and it was extremely durable. Now that I am switching over to the 4in wheels I may lower the top of the plates by an inch or two. The main concern was having the top of the mecanum showing out the top. By having it below it allows for an easier time trying to mount other subsystems to it.
there are issues with 4 in machanum, as far as i know vexpro is the only place to sell FRC ones, and they had serious issues with wear and functionality if you are going to try 4 inch wheels you should make the wheels vary easily replaceable (between matches you should be able to easily replace one, and you will probably want to make sure you can replace two in five minutes).
You have to remember, vex knows this is an issue. I am going to assume the people at vex are going to find a way to fix this for next season, so it might not be a problem.
Ok, so I have made some revisions to the initial design and am happy about most things but the pistons. In order to prevent having to re calculate the positions of the hard stops, I kept the stroke of the piston the same (1.5") and extended the rod length. My concern though is whether or not the pistons rods will flex under the load of having to lift the bot in the air. Here is a photo of the current setup:
Also, if you’d agree that this warrants concern, what might you suggest as a solution? I’d like to keep the single piston per side to help with keeping weight down and I can’t think of a workaround at the moment (perhaps I need more sleep)
Just as an update to the revisions made:
Spread the wheel modules apart by 4.5".
Swapped the toughbox mini’s over to a custom stackerbox.
Final speeds of 8.5fps on traction wheels and 21fps on mecanums.
Swapped to 4in mecanums, mecanums on the pivot with traction on the outside.
Ensured that the pistons now have enough force to lift bot.
Taken 1.25" off the top of the side plates to make them 5" (the height of bumpers).
All of the following modifications leads to a loss of 10lbs, totaling in at 40lbs. The only downside I see at the moment to these changes is that by going with the custom spinboxes over the toughboxes, is that I can no longer add 4 775’s for more torque if deemed necessary.
Also, on a bit of a side note, does anyone have an idea of when Vex will be re-releasing their 4in mecanums?
Adding 775’s to the drivetrain may not be such a good idea anyway. The RS-775 is cooled by a fan on its shaft, so when it stalls or is under high load it stops being cooled and will burn out much faster than a CIM or Mini-CIM, which use their high thermal masses to keep from burning out.
Your drivetrain is power limited in high gear so stall and near-stall conditions should be expected at some times. If you want better acceleration in high gear then you should consider an option that would allow you to add Mini-CIMs.
This Is something that could still be done with the toughbox mini. They have dual cim mounts. The way I would have incorporated 775’s is by using a cim-ile gearbox from vexpro. However yes, I had forgotten about the posts I had read about this topic and the 775’s might not be the best idea for continuous use.
This year, my team used a modified Octanum Drive that we called the “Dragonfly.” It also used 6" AndyMark HD Mecanums mounted on the outside of the modules with 4" x2" Colsons on the inside of each module. We had no issues with being passed onto our medallion wheels, and with a Vex Pro Ball Shifter with a CIM and a Mini-CIM driving each module,we had more than enough power to spare.
The thing that worries me about your concept is the placement of the pistons that actuate the modules. You could easily bend the shaft or just have one side of your robot on mecanum and the other using traction just from another not hitting you hard and pushing. We had one piston actuating each module with both the pistons on one side mounted to a machined aluminum block.
Our not was an expensive, power hungry monster, but it did its job, and it did it WELL. Here is a link to the CAD files.
And here is a picture of our competition bot! Good luck!
