This is a notice for all teams who are using the wooden barrier for their prototypes. We made a steel barrier to specifications and the differences are significant. We were easily able to go over the wooden barrier but the robot gets stuck halfway across the metal barrier. We highly suggest that teams attempt to procure an accurate sample to simulate the barrier conditions at competitions.
If any teams who are close enough want to test their robot but are not able to get a barrier, PM me and we can set up a time to test your robot at KING TeC’s build site.
It is the radius corners of the metal along with it’s smooth surface compared to the square corner and tactile surface of the wood which is the difference and it is big.
The bump is made out of sheet steel bent on a break, so the corners are a radius on the edges. giving that some robots may not be able to grip smooth corners
If you’re looking for a cheap solution, though I’m not sure how it compares to the real steel, what our team did was grabbed a whole bunch of pieces of angled aluminum and just put them on one vertex of our wooden barrier. Seems to work for us. Just gotta drive the robot around each time instead of going over and back.
I just went back through the field drawings to look at how the steel barrier is made, and a 4x6 steel tube might be an accurate and relatively easy to find substitute for some teams. I believe we have some 4"x6" or at least 4"x4" tube sitting on the Material Rack at my job, I will try to get some dimensions of the corner radii to compare.
That is true that team test there prototypes onto he wooden barriers and are successful but I can bet that some of those team will have a problem at their district/regional competitions. That is why I am telling my team don’t rely on the success of the wooden barrier.
We covered our barrier with aluminum flashing to simulate the coefficient of friction and rounded edge in an affordable manner – it made no difference to our drive train…
But yeah, a lesson I had to learn a half-dozen times in FIRST and am now paranoid about is this: never engineer anything so that it barely works. All values are nominal, and assume that if they’re off, they’ll be off in a manner disadvantageous to your robot.