Taking Sawppy to Maker Faire Bay Area 2019 was going to be three full days of activity, more intense than anything I've taken Sawppy to before. I didn't think it was realistic to expect a completely trouble free weekend and any brea7kdowns will be far from my workshop so I tried to anticipate possible failures and packed accordingly.

Despite my worries, the first two days were uneventful. There was a minor recurring problem with set screws on shafts coming loose despite Loctite that had been applied to the threads. I had packed the appropriate hex wrench but neglected to pack Loctite. So I could tighten set screws back down, but lacking Loctite I had to do it repeatedly. Other than that, Friday was completely trouble-free, and Saturday rain required deployment of Sawppy's raincoat. But Sawppy got tired by Sunday morning. Driving towards Dean Segovis' talk, I noticed Sawppy's right front corner steering angle was wrong. At first I thought it was just the set screw again but soon I realized the problem was actually that the servo would turn right but not left.

With the right-front wheel scraping along the floor at the wrong angle, I drove Sawppy to a clearing where I could begin diagnosis. (And sent call for help to Emily.) The first diagnostic step was pushing against the steering servo to see how it pushes back. During normal operation, it would fight any movement off of its commanded position. With the steering behavior I witnessed, I guessed it'll only fight in one direction but not another. It didn't fight in either direction, as if power was off. Turns out power was off: the fuse has blown.

I replaced the fuse, which immediately blew again. Indicating we have a short circuit in the system. At this point Emily arrived on scene and we started methodically isolating the source of the short. We unplugged all devices the drew power: router, Pi, and all servos. We inserted third fuse, powered on, and started testing.

Sawppy dead servo 29

We connected components one by one, saving the suspected right-front servo for last. Everything was fine until that suspected servo was connected, confirming that servo has failed short. Fortunately, a replacement servo is among the field repair items I had packed with me, so servo replacement commenced. When the servo was removed I noticed the steering coupler had cracked so that had to be replaced as well.

Using a spare BusLinker board and the Dell Inspiron 11 in my backpack, I assigned the serial bus ID of my replacement servo to 29 to match the failed front right steering servo. Then I pulled out a servo shaft coupler from the field repair kit and installed that on my replacement servo. We performed a simple power-on test to verify the servo worked, plugged everything else back in, and Sawppy was back up and running.

https://youtu.be/DBcx4Wwspks