Here is a link to the CAD files: https://drive.google.com/file/d/0BwkVdOFz3p39Q3BrSlprdUxyOTQ/view?usp=sharing
We have really appreciated all of the resources that other teams have shared regarding their designs. We especially want to thank Team 1640 Sab-BOT-age and Team 2067 Apple Pi for your excellent resources.
We welcome any feedback.
I don’t have access to the CAD models at work, but great job on tackling swerve in the off-season! I’m sure it was a fun exercise for your students to go through.
Did you guys get a chance to manufacture and assemble it yet? Looking forward to seeing the real deal.
We just put it through its first drive test yesterday and it seemed to handle great. Here is a quick video clip of its first run:
We were able to machine all of the parts in our shop. We definitely learned a lot.
I have attached a couple of close ups of the actual modules on the chassis.
http://i59.photobucket.com/albums/g314/mwright60/IMG_2024.jpg](http://s59.photobucket.com/user/mwright60/media/IMG_2024.jpg.html)
http://i59.photobucket.com/albums/g314/mwright60/IMG_2025.jpg](http://s59.photobucket.com/user/mwright60/media/IMG_2025.jpg.html)
http://i59.photobucket.com/albums/g314/mwright60/IMG_2026.jpg](http://s59.photobucket.com/user/mwright60/media/IMG_2026.jpg.html)
Those look great Michael! We found that the longest thing to manufacture are the module tubes that connect the wheel module with the stationary plate. We have used ball bearings, needle bearings, and sleeve bearings in the past. We can complete all the swerve module parts in just a few days, but the lathe gets back logged with those tubes.
How is your bearing system set up in the tube?
Do you find it takes a lot of time to manufacture those?
The module tubes definitely took the longest to make and we had to get creative with them. They were also the most fun. We have a HAAS TL-1 Lathe and it was a life saver.
We bored out the ID with a boring bar and press fit bronze thrust bushings into them with an arbor press. We then parted the excess bushings off on the lathe. The thrust axle houses the bearing.
We used our mill to drill out the mounting holes. There are several operations involved but we think we can do it efficiently now if we decide to use swerve for this season.
Looks like you can get rid of the spur reduction stage and do all your reduction w/ the timing belt.
Adam,
I see that now. We could use a 24t and 60t pulley or an 18t and 48t to achieve the same reduction. We may be able to rearrange the setup which may also reduce our footprint with the mounting plate.
How would you suggest mounting the hex axle directly to the CIM motor? We don’t have a keyway broach yet but that may be an option. Thoughts?
Do you have a lathe? If so getting the reamer and broach set required for interfacing w/ CIMs is not crazy expensive.
Also, you can purchase pinions from BBman.com, sdp-si or motion industries to get a smaller pinion that what vex sells. You’d have to ream and broach afterwards.
Yeah. We have lathes. I was just looking on McMaster to find the broach and bushings. We have never broached key ways before but it doesn’t look too difficult.
Any concerns with mounting the .500 in Hex directly to the CIM shaft?
Here is what I think we would need to buy in order to broach correctly:
8mm reamer- http://www.mcmaster.com/#8851a23/=10a37ir
2mm keyway broach- http://www.mcmaster.com/#8805a11/=10a3730
8mm collared bushing- http://www.mcmaster.com/#8804a71/=10a36rr
Looks correct.
We’ve pressed 3/8 and 1/2" hex onto CIM shafts as well if you do want to stick with the hex interface.
Great. Thanks! We will try this out.
Vex also sells this part http://www.vexrobotics.com/vexpro/hardware/217-3255.html
Thanks for the heads up. We are actually using these on the current module. They are great little pieces of hardware. We may have to make some longer versions of them though depending on how everything lines up.
We did the same in our first iteration. We had no trouble pressing them on. We pressed with and without taking the motor shaft out. We even found a way to reverse the polarity of a CIM motor in the process. Lets just say the potentiometer was destroyed when the code ran!!! We had a box full of destroyed pots. When we switched to encoders on the second iteration, we could survive mistakes like that. We also run our code with only feedback loops on the turn, we don’t get feedback from speed, we just go on percentage of the available voltage. We haven’t noticed any issues with doing it this way.
Here is a first person video of the upgrades in code and move to encoders from Fall of 2013. This is the second iteration and unfortunately still had chains. You can hear those throughout video. We have since upgraded to belts and direct drive.
The last 90 seconds is my favorite!
hmm. if you got rid of the spur gear, the belts wouldnt fit around the versaplanetary.
BUT! what if you add a SECOND CIM. If you place them side by side the reduction could be done with the belts AND it would fit around the steering motor.
The reason i had this idea is because ive seen a nearly identical vex EDR swerve
But if you were to change the shape of the module and put the CIM in a different spot (like making it more of a corner module) and arrange the CIM, steering, and wheel shafts in a L configuration, you might be able to make it work. Also it might package better and smaller. Just a thought. This thing is awesome!
That’s exactly what we are looking to do for the next iteration. We already made a sketch and it will fit in a smaller footprint. We are also going to try the new VexPro encoders for the planetary gearboxes.
I will tell you all that we are really maxed out on power with our current gearing. It is geared pretty aggressively and we did have some brown outs during testing on carpet. We are in the process of adding current monitoring, ramping controls, etc. But we are going to increase the reduction from the thrust motors from 4:1 to around 5:1.
Our Vex Versaplanetaries are running at 100:1 with the bag motors and that is probably pushing it too far as well. We love the 1:1 with the steering gears so hopefully with our other adjustments it will still work.
That’s what the off season is for though!
