I just noticed something wild on my small e-outboard: when I surf down a wake or steep chop, the display flips to negative watts for a second or two. It looks like the prop is back-driving the motor just enough to push energy back into the DC bus before the controller cuts it off. That got me thinking-are we sleeping on “micro-regen” from wave/swell assists, and could we capture those short bursts without messing up the drivetrain or battery?
Idea: add a small supercapacitor bank on the high-voltage bus to soak up these fast spikes, then trickle them into the main LFP pack via a bidirectional DC-DC. The caps handle the instantaneous power and clamp bus voltage so the motor can actually stay in a controlled regen window. If the caps get full, a dump load (water heater element?) can burn off the surplus instead of kicking the controller out of regen.
Questions for the brain trust:
- Has anyone measured real-world regen potential from “gravity assists” on planing or semi-displacement hulls? I’m seeing momentary −200 to −600 W on a 48 V system at 12-16 kn, for 1-3 seconds. Any data on peak current/energy per event and how often these happen in steady swell?
- Controller behavior: which marine-capable controllers let you tune regen voltage/current limits and response so it doesn’t torque-thump when it transitions? Any success keeping regen active without triggering overvoltage fault when the bus rises fast?
- Bus architecture: better to tie supercaps directly to the traction bus with proper precharge and a balancing/monitor board, or put them behind a dedicated bidirectional DC-DC for tighter control? How are you handling isolation monitoring and ELCI when adding a second energy store on boats?
- Overvoltage protection: is a brake chopper/dump resistor a sane safety backstop on a boat? Anyone plumb one into a calorifier as a useful heat sink?
- Net gain vs drag: does allowing light regen when surfing actually increase overall efficiency on a passage, or does it just add drag and slow you in a way that costs more energy later? Any A/B testing with Wh/nm over identical sea states?
- Mechanical risk: any reports of torque reversals or cavitation spikes causing handling quirks when regen kicks in on fast hulls? Tips for ramp rates or prop sizing to keep it smooth?
- Data logging: I’m planning to log bus V/I, motor RPM, GPS SOG/COG, IMU pitch/heave, and sea state notes to correlate when regen happens and how big it is. What am I missing? Would anyone be up for sharing datasets to compare hulls and props?
Bonus angle: if this works even modestly, a small cap bank could also buffer sudden throttle chops or wave slams that otherwise hammer the pack and contactors. Feels like a neat reliability win even if the energy recovered is tiny.
If you’ve tried supercap + LFP hybrids on boats, or have a controller that plays nicely with light regen on a prop, I’d love to hear specifics: models, settings, schematics, gotchas. Also, any ABYC or other standards caveats when introducing additional energy storage on the traction bus?