NewBoatDrone Mk3 and Mk4: Water Time to be Alive!
Learning from Past Mistakes
In the previous post we made some progress towards a working airboat, using the BeeBrainBL FC to drive two 1202 motors in a differential thrust configuration, but the short flat balsa hull was prone to flipping right on it’s face with any sudden stops. One of the biggest annoyances with FPV drones is having to put down your goggles and do the walk of shame when you can’t fly out of a crash, I can guarantee you, the swim of shame is worse!
The main reason I chose an airboat in the first place was that it would be the least likely to get tangled and stuck, but capsizing is equally problematic and needs to be fixed.
Back to the drawing board then, I designed a longer hull with a raised prow, to prevent it nose-diving. You can see similar shapes on some full-scale airboats. The boat is also designed with a flat rear, which is convenient for printing, and I made sure that none of the angles resulted in more than 45 degrees of overhang when printing
This hull accepted the same “thrust pod” from the previous boat, so that I didn’t have to move any electronics around. Making sure that your components are modular and re-usable is a huge benefit when experimenting and iterating towards a goal.
Watertight Hollow 3D Prints
The images above show how the slicer is used to make the hull hollow, by printing without any support or infill, and only two perimeter walls, we get a relatively lightweight and robust result. Ensuring that the hull is watertight can be tricky with thin-wall prints, but there are a few things you can do to help
Print hotter (to ensure good bonding between layers)
Print slower (for layer bonding too)
Over-extrude slightly (I used about 5%)
In the end my PETG print had two pinprick holes which I fixed with superglue, but an alternative solution would be to paint/varnish the whole thing.
Testing - Mk3
This iteration was the first to be able to reliably putter around a pond without getting stuck or capsizing, which is definitely a minor win in boat-building… but it never was able to get up to and significant speed since it has way too much drag, and the weight was too far forward. I believe the thrust of the motors, being so high above the water-line relative to the short hull, was creating a pitching moment forwards, preventing the nose from coming up on to a plane.
Testing Mk4
I reasoned that I would be able to get the boat up on to a plane if the motors thrust was pitched down slightly, effectively “pushing” the stern downwards and therefore pivoting the bow upwards, so I redesigned the thrust pod to have a 10 degree tilt (it turns out this was way too much - real airboats use trim angles in the single degrees or fractions of degrees!)
I also moved the thrust pod as far backwards as it would go to take weight off the bow.
These two changes made for a boat that could finally reach something close to a planing speed under ideal conditions (flat water, fresh battery, etc), but had a negative effect too: there is a massive difference in nose-up pitch between low and high throttle, making it very difficult to pilot via FPV.
Giving up yet?
Heck no! The solution is obvious; build Mk5!