It’s Week Six (of six) of the 2010 FRC season.

As if the title of this blog isn’t enough of a cliché, how about this – the problems that you experience are always those that you don’t plan for. Sounds simple and obvious, right?

Week Six (of six), Day Two (Tuesday):

The basics of the robot are in place. The drive train looks great, with a very impressive Mecanum-based drive system. The ‘bot can drive like a tank, or strafe like an omni-wheel system. A simple button click on the analog joystick allows the driver to select between the two modes. Everything has been considered – small enough to crawl under the tunnel with ease, yet rugged enough to overcome the 12″ high “bumps” as if they’re nothing more than a Wisconsin winter-induced pot hole. Multiple test driving opportunities have proven that everything is status quo.. nothing to be concerned about.. time to focus on the soccer game—gathering the ball, kicking it across the field, blocking opponents’ actions—the strategy couldn’t be more complete.

As the team puts the finishing touches on the “ball handling” features, I begin planning the obstacle course intended to challenge would-be drivers so we might select the critical four team members to ultimately represent Team 930 at the Wisconsin Regional, and—dare I say—National competitions. The course is laid out, rules established, and it’s time to experience the final adrenaline rush of the FRC season.

The first candidate sits in front of the joystick. With skill, Mr. San Miguel delicately maneuvers the robot forward and back, strafes left and right, and attempts to rotate the ‘bot on its central axis. But something isn’t quite right.. the wheels aren’t behaving as expected on the newly laid out carpet. But that’s not possible – the optical encoders used to sense the wheels’ motion, and the closed-loop control methods employed by the controls team, should guarantee that the wheels behave like a well-trained dog. What could be happening?

Perhaps we have a weak or dead battery…. “Let’s put in a fresh battery,” I suggest. Eagerly, a team member replaces the battery, commenting “Wow, the wires are really hot!!” With a newly-charged 12V battery in place, the ‘bot exhibits the same behavior. Mr. Josh Holat (Firmware Team Lead) announces that the system is reporting 8V on our fully-charged 12V battery.

End of case history. The next 1-2 hours might represent some of the best “real world” lessons possible, and yet, none of this timeframe required a single volt or amp from a battery, rotation of a motor, or stutter of robot movement, to arrive at a conclusion.

The conclusion was simple – the wires were overheating, and the motors were overheating. This could only happen if the forces on the wheels are so strong, the motor cannot overcome them, even with a 12.75 to 1 gear ratio. The motors were stalling out! This causes not only a lot of heat, but a lot of current in the stall condition. Result: sudden drop in battery voltage, lack of response from the motors, and very hot parts! And yet, how could such a cohesive team, with many experienced engineering mentors, not recognize a design flaw of this magnitude until the final week?

Enter “Welcome to the Real World.” New product design teams, especially constrained by limited timelines, funding, and human resources (i.e, FRC season), always take risks. One must anticipate or predict the greatest risks, and focus on mitigating those risks. Speaking for myself, I would never have predicted that torque would come into question – and as a result, some signs of the problem (obvious in hindsight) were either ignored, or attributed to other root causes. Case in point:

• Wheels didn’t always behave as expected when trying to “spin” … but the battery voltage reading often indicated a weak battery. Solution: replace the battery!!
• Closed-loop wheel control system creates inherent stability challenges. If “requested” wheel movement doesn’t match “actual” wheel movement, one might assume that the control system is to blame.

Hiding behind these seemingly logical root causes might have cost Team 930 two weeks of productive design improvements.

Sounds so simple, right?
Hindsight is 20/20?

Welcome to the Real World.