After developing MecaTrak for the 2016 game Stronghold, https://www.chiefdelphi.com/forums/showthread.php?t=146426&highlight=MecaTrak
we decide to improve upon this drive train for the 2017 season. We looked at this drive train and listed out the pros and cons to determine how we could improve MecaTrak to fit our needs for whatever may come for the 2017 game. In the end we decided MecaTrak was a very good drive train but very heavy.
After the game reveal in January, 3026 set out to make the lightest version of MecaTrak that we possibly could. After many iterations, we finalized our new drive train. We call it Posi-Canum.
This drive train has two components to it. The LOCKING MECANUM WHEEL (Patent Pending) and the drop down third wheel.
3D printed air cylinder mounts to attach around sim motors
Drop Down Third Wheel
2 High Traction wheels per side
1/2 inch stroke at 60 PSI to apply wheels
Connected by belt to back mecanum wheels, powered by 2 Cims
Given the versatility of this drive train, implementing a Nav-X, and our 4 camera vision tracking we are able to have many drive modes
Robot Centric Mecanum
Robot Centric Tank Drive
Field Centric Mecanum
Field Centric Tank Drive
The joystick gets vectorized and if the joystick turns less than 90 degrees in either direction the robot will rotate to that position, if it is greater than that the robot will switch what it thinks it front end is and adjust the rotation accordingly.
Boiler Centric (Rotates to Boiler)
Peg Centric (rotates and strafes to peg)
Peg and Boiler Centric (Rotates to Boiler, strafes to Peg)
Rope Centric
Gear Centric
The pushing power of tank drive and the mobility of mecanum has proven to be one of team 3026’s best creations.
Any questions about the drive train we will be happy to answer. Good luck to everybody this season.
Team 190 did locking mecanum wheels in 2010 using a similar concept. We had issues with the disk brake wearing away at the rollers and causing roller failures. How much drive time have you put on this?
An air cylinder pushes against the brake disc. That wedges against the rollers similar to a cone clutch. The rollers are then no longer able to rotate and the mecanum wheel essentially functions as a normal wheel. This brake disc and air cylinder application has been filed.
It’s been a while, but iirc the most common failure for our rollers was caused by voids in the rubber we were using. Unlike your implementation (3026), our module used custom made mecanum wheels. These included custom hubs, forks, and most notably, rollers. We machined molds that held a small titanium insert, and poured a liquid rubber (no idea on specifics, though I can go digging if anyone would like) that cured and formed a roller with integrated shaft.
While the molds were beautifully machined, they did not allow for proper evacuation of air, and as such it was incredibly difficult to get rollers with homogeneous composition. When these voids occurred at edge locations on the insert, or on the outer surface of the roller, tears would often form and propagate after minimal use. Replacing the rollers was simple, just two number eight screws on top of the fork retained them, but this didn’t change the fact that we burned through rollers during practice and competition.
Like Evan, I am interested to hear how your wheels are holding up. I would expect much better than ours did lol. I think that if we had had time to refine the molds, and test different rubber, we could have greatly improved the life of the rollers, but such is build season.
I’ve included some pictures of 190’s modules from 2010 via an imgur album link. I intended to embed them in my post, but I was having a tough time trying to resize them.
Can´t wait to see it in action… if your system can really fix the problem of meccanum robots being pushed, this will be a huge breakthrough and more teams will be using this.
I’m not so convinced. While it is a beautiful mecanism, it seems too complicated for the average team to justify building this in leu of a normal mecanum (or an 8-wheel mecanum if you want additional pushing force) or even a butterslide drive.
To answer some of the questions that many are having, we have 2 videos to show and some explanations to how the wheel locks.
This drive train has already been field tested at the Duluth, Minnesota Northern Lights Regional. The rollers on the mecanum wheels have had no damage to them after a full regional and practice with the robot in the shop during the build season. They have held up to the wear and tear of competition beyond our expectation.
To see the pushing power that this drive train gives our robot we have put together a video of it playing defense at that regional competition. This regional had a very good mix of tank drive robots and mecanum drive robots (along with one butterslide) and this video will show us playing defense against both mecanum and tank drive.
On the subject of how the wheel works. The brake locks the rollers on the mecanum wheel to turn it into a standard tank drive wheel, as stated by todd.bolzer@liv. We would love to post a video of a close up of the wheel locking, but our robot is in the bag and we do not have any documentation on it. However we can show the concept behind how the wheel works with the video below. Switch the screwdriver with pneumatic cylinders and you have the concept of how the wheel locks its rollers.
How do you manage peg centric driving? That would be immensely useful if we could do it to. Our field-centric drive is already set up, but how do we set it to “know” where the peg is?
Well - not with the butterslide engaged. The robot is still on a strict diet to try and make weight for 10K Lakes so we can add the actuators.
But for those that are questioning this posi-canum - I have seen this thing in person and it is incredibly well engineered and built. Kudos to 3026 for creating this - if you get a chance to talk to them you will notice how incredibly giddy they are (and should be) about this drive. Now if they could just get some time on the field to calibrate their vision…
10K will be a lot of fun - at least for 3026 (hopefully we are ready too).
