Quote:
Originally Posted by Dmentor
Team 148 has clearly spent a lot of time tailoring and streamlining their engineering process which enables them to spend more time in the early design steps. My occupational experience (and research in defect analysis) tells me that dedicating more time up front provides the biggest return on their labor investment. In order to generate this time, they have become extremely efficient in modeling, production, integration, and to some extent testing.
|
BINGO... ie DO NOT TRY THIS AT HOME!!! If any regular team spent 3 weeks prototyping, they would be in a world of hurt because their parts would never get finished in time to even build a robot. I know this because in 2006 1511 spent that 3 weeks prototyping, and finally abandoned all prototypes and went with a concept we figured would work, no design, we just built it... fortunately we survived, but we most certainly did NOT come out with an elegant machine, and did not have enough test or programming time. Dmentor hit it on the head, its that they have trained their students and team so well on design, and have great manufacturing support and capability, and THAT allows them to prototype for that long to come up with the best solutions. Until you get to that level of engineering process, in my mind you CANNOT spend 3 weeks prototyping or you will shoot yourself in the foot.
I would still suggesting picking 1 design right up front and racing to finish it as fast as you can so you have debug and programming time at the end, unless you have an insanely well trained and well oiled team (hint use preseason wisely). My bet is that once they pick their designs, teams like 148 & 217 can crank out full designs in 4-5 days, manufacture in 3-4 days, and then still have time to test. 1511's best robot design was probably 2008, and that took us 4 weeks of CAD design, because we are NOT proficient enough. It worked for us because we picked the design by day 2, designed for 4 weeks, had parts built along the way and were left with about a week to debug & program, but there is no way we had time for 3 weeks of prototyping.
Maybe John or Paul could post a sample schedule that they follow...
