Engineering Failure Analysis

[Mark McLeod]

I have seen that servos can occasionally twitch with large random movements when the RC changes states or when it is reset. I've observed the random "twitches" to get worse or better depending on the master code that happens to be loaded and which PWM output is used. Victors controlling motors haven't suffered from a similar twitch since IFI fixed that part of the problem with Master Code v11.

Having the value already set to 255 during initialization then resetting it to the identical value later to solve your problem certainly is puzzling though. It'd be interesting to check the value just before the reset to see if it is occasionally not what it should be. (Makes me want to go play...)

[Alan Anderson]

Originally Posted by Mr. Freeman

We made sure to initilize it to 255 in the User_Initialization routine.

Did you also make sure not to "reinitialize" it to 127 at the beginning of User_Autonomous()?

[Gabe]

Activity: Removing pinion from two 3" Minibike motor shaft using heat from a propane torch, then striking with a punch and hammer.

Problem: Both motors were very squeaky afterwards and the shafts had more play pulling in and out.

Analysis: The height of the shafts were different from motor to motor. The motor smelled of plastic. Opening up the motors, the armature was hot.

Conclusion: Too much heat softened the adhesive fastening the motor shaft to the armature. The shock from the hammering then moved the shaft backward, where the adhesive rehardened.

Solution: Oddly enough, the solution came right after this incident. Out of frustration the pinion was hammered on an anvil, causing it to fracture and break away from the shaft cleanly. Test on the other pinion was also successful.

[CJV648]

We've had a lot of drivetrain analysis here, but not much on one of the interesting things I observed in this competition. Many teams specifically had spare gripper components ready for quick replacement. It looked like a lot of thought went into what to bring in order to prepare for gripper failure. Were these teams designing their grippers to fail at an easily repaired point or just going from hard experience or some combination of the two? How did they decide whether to upgrade design or simply repair the existing design when it was damaged?

[Mark Holschuh]

We made spare parts for our manipulator arm and claw for our regional competition. However, we came to realize that it would likely take about 90 minutes to replace the arm if it became necessary, as it would entail the disassembly of the support structure, and to reset a potentiometer. Our students redesigned the arm between the regional and the Championship so that they could replace it in 5 minutes if necessary. Our students learned a valuable lesson in "Design for Assembly."