Hello all, my first post here, although I've been looking in from time to time over the past few months. I've been playing with electric propulsion for a fair while now and am in the final stages of developing a low power propulsion system, suitable for an electric version of Winsome.
I took one of my prototype brushless units over to Swallow Boats last Autumn and Nick and I drove around the estuary for ten minutes or so with it driving the Storm 15 at hull speed, in calm water, on around 350 watts. Winsome needs much less than 100 watts for river cruising, so my plan is to have solar charging that will keep the battery topped up whenever (if ever....) the sun shines.
Tony started this thread with a "wish list", which is exactly how I set out to design a complete system about 10 months ago. My wish list was:
At least 6 hours endurance on battery power
Cruise at 4kts on rivers and other inland waterways
Ability to solar charge at as high a charge rate as practical without spoiling the look of the boat
Total system efficiency (from charge to thrust) greater than 70%
Propulsion system weight as low as possible
Battery cycle life at least 1000 cycles
Zero battery maintenance
Low battery hazard rating (no toxic materials or risk of fire)
So far, I have a sealed battery system that provides just over 1kWh of usable power (enough for around ten hours of cruising at the expected power level) in a sealed box that weighs 11.6kg. It uses Lithium Ferrous Phosphate batteries, that are very safe, unlike other some other lithium chemistries. Cycle life should exceed 2000 cycles and calendar life should be around eight to ten years.
Cruise speed power is determined by the hull resistance; as Winsome was designed to do around 4 kts with only pedal power (probably around 100 watts) it was a near-ideal hull to use.
Getting good system efficiency is a great deal of work though. Ordinary boat props are fairly poor, they waste around 30% or more of the power you put in to them. The answer to this problem was to make a prop, but it turns out that the thin, high aspect ratio, blades are superb weed catchers. The way around this problem was to make the prop blades fold, so that if a bit of weed gets caught the throttle can be eased, the blade folds back and the weed (hopefully) slips off. The prop design I have now (12" diameter, 10" pitch, 700 rpm at full power) should be around 84% efficient, just about as good as it's possible to get with a practical prop.
Next, I looked at motor efficiency. Electric motors are generally pretty good, but the best are, without a doubt, the brushless permanent magnet ones developed for high powered model aircraft. These are tiny, yet deliver upwards of 6kW (8hp) in a package around 3" in diameter and weighing about 1.5kg and cost around $100 or so (take a look at this one, for example:
http://www.hobbycity.com/hobbyking/store/uh_viewItem.asp?idProduct=5142&Product_Name=Turnigy_80-100-B_130Kv_Brushless_Outrunner_%28eq:_70-55%29). I opted for a much smaller motor of the same type, one rated at 2800 watts, then altered it by fitting internal position sensors and changing the wiring from delta to star (these are three phase machines) so that I could run it on a better type of speed controller. The big advantage of under-running one of these motors is that it's efficiency improves, as the copper losses are very low when it's run at low current. My motor now has losses that range from just over 1 watt at idle speed to around 12 watts at full power (200 watts), so is always running at better than 94% efficiency.
To drive the big prop slowly (for good efficiency) I needed to reduce the motor speed. Changing the wiring from delta to star dropped it's peak rpm to 56% of the original, so that the motor now runs at about 1500 rpm at full speed. This was still too fast, so, after playing with toothed belt drives, chains etc I ended up buying an expensive 2:1 reduction right angle gearbox. To my horror, I found that the gearbox absorbed around 14 watts on it's own, so I had to modify it. I removed the top oil seal (oil seals soak up around 4 or 5 watts each) and then removed all the internal grease. The gearbox is now filled with oil and is sealed to the drive leg, like an outboard, so has no need of the upper seal. This reduced the gearbox loss down to 6 watts.
The speed controller is a cheap ($22) one intended for an electric bicycle. It's been modified to make it work on 13V and to reduce it's internal losses. The output FETs were changed to very low on-resistance types, so the controller loss is now between 0.8 watt and 2 watts, depending on power setting.
All the electrical parts are sealed into watertight alloy or polycarbonate housings, with neoprene seals. All cables are fitted inside watertight flexible nylon trunking, so I'm hopeful that it should be reliable in service.
I still have a great deal left to do, including getting the hull and fitting all this stuff, plus sort out the solar charging system. I've purchased some raw silicon photovoltaic cells (108 of them), plus a very neat, high efficiency, maximum power point tracker. I'm either going to make a canopy, with the solar cells on top, or I may see how many I can get on the deck area.
During this lengthy development process I have learned a great deal, so if anyone else wants to have a go at designing a similar low power propulsion system I may at least be able to advise on things I've tried that have failed!
Jeremy