Captain’s Log 06/27 – 07/26

June 29, 2021

I’ve run enough testing on the boat to determine that the charging system seems to be supplying enough power to the battery. However, the battery management system which ensures equal charging to the 4 cells wired in series with each other seems to not be doing its job correctly. Only 3 of the 4 cell banks are charging, which only gives me a maximum voltage of 12.6v, which happens to equal the maximum voltage of 3 cells wired in series (4.2v x 3 = 12.6v).

Also, during the last two days of motor tests, whenever I turned on the motor the boat seemed to go the opposite direction of where it was supposed to go, so there could be some issues with the rudder now as well. Its been drifting for long enough where I’m sure its had some things happen to it. I actually noticed another motion event recently before I rebooted the boat for a new waypoint…

Either way, ill try to run the motor from time to time but overall ill probably just let the boat drift and see how long I get telemetry updates. Ill lower the update frequency down to 6 to 8 hours when I go full drift mode.


June 30, 2021

8:35 pm: LoCARBs backup thruster is still functional albeit running at a much lower speed (despite setting it at full throttle) since the max voltage it can run at is 12.6v (most of the time it runs around 12v). The problem of the boat navigating the wrong way has gone away after a full power cycle of the Pixhawk autopilot. In fact, today the boat kept its position despite fighting wind/current, but wasn’t able to fully overcome it (it went backward slightly). As soon as I turned off the motor, the boat drifted much more quickly towards the opposite direction it was trying to fight against.

One of the most interesting aspects of this project involved running into problems that I hadn’t even thought about. I never once thought I would have a situation where the boat would still operate, but at a lower voltage than the full 13.5v-16.8v. Even if I did however, I am not sure what I could do to completely mitigate against it. The only thing I could think of would be to add a secondary waterproof barrier/bag around the battery pack and essential charging components which is actually more difficult to do than in concept. That’s food for thought for the next water-based project though.

LoCARB is 615 miles from the launch site and about 520 miles from the coast of Baja California. It has however traveled a total of 750 miles if counting all directional movements.


July 6, 2021

Changed the update interval to 8 hours and letting LoCARB drift. I don’t have any hope for any meaningful propulsion with the voltage as it is. I wonder where it will end up!


July 10, 2021

10:44 pm: Setting update interval to 6 hours. Will turn on backup thruster in the morning since there doesn’t seem to be any wind.


July 11, 2021

The Px4 Pixhawk 1 autopilot definitely has trouble when powered on for any long amount of time. When not rebooted after being powered on continuously it seems that after a day or two, it will lose its mind and decide to go the opposite of where it should.

My secondary thruster start procedure:

  1. Power cycle the PX4 to reset its sensors
  2. Reboot Arduino after the PX4 resets to resync serial communications (one sat message can do #1 and #2)
  3. Disable motor and monitoring
  4. Set Motor to 2nd backup motor and disable auto throttle
  5. Change to full throttle

July 19, 2021

1:55 pm –  Autopilot has lost GPS lock for the past two updates, but somehow still responds and sends back heartbeats to the Arduino. These clone pixhawk 1 2.4.8 autopilots really arent reliable being powered on for long durations. I sent a command to power cycle the Pixhawk autopilot and Arduino with new waypoint sent as well. Motor Off and update back to 8hr interval. LoCARB has traveled over 800 miles from the launch site, and is still slowly heading on a South-Western course.

I am definitely impressed with the KST brushless servo with contactless position sensor used for the rudder, its been powered on and attempting to correct the boat for over 7 weeks, and as far as I can tell, still working well!


July 26, 2021

10:30 pm – The boat has been powered on for 1 week with no apparent issues, but I sent commands for tomorrow to power cycle autopilot, Arduino, set new waypoint, and resume 8hr updates. Since July 13th, the boat has traveled about 140 miles southwest due to the favorable winds but is now heading on a more southern course due to a shift in wind direction. I hope winds continue to work in the boats favor. I calculate that LoCARB will make Hawaii around December-January.