Many thanks for the welcome, Tony.
What you describe is pretty much exactly what I want with the electric Winsome, a boat that the two of us can use to glide silently along our waterways, with as little disturbance as possible to the environment. I also want a boat where the technology is hidden, inasmuch as I want it to just do the job it was designed to do, without being obviously high tech.
I wonder how much interest there would be in an electric version of Winsome? At the moment, I've just asked Nick and Matt to build me a Winsome hull (I will fit the propulsion system) but I have wondered from time to time whether a commercially produced electric Winsome, at a sensible price, would be viable.
Anyway, answering your questions in order, as best I can:
1. The present battery pack is made up from 32 off, 10 amp hour cells, connected as 4 series banks of 8 parallel connected cell sub-packs, so has a nominal voltage of 12.8 volts and a capacity of 80 amp hours. Unlike lead acid, you can use pretty much all the stored energy in this type of cell, so its rated capacity is also the usable capacity.
I bought the cells directly from China, from a company called Headway, but there are are now cheaper and lighter versions of the same type of cell available. If I were to buy another pack then I would probably get 16 off 20 amp hour A123 pouch cells, at about £30 each. For the same capacity, these would only weigh about 8.5 to 9kg (including the weight of a suitable box) and would be smaller than my cylindrical cell pack as well.
I have built a small plug in charger that will fully charge the battery pack in about 6 hours. Both this and the battery pack are easily portable, the battery pack box has a sealed connector on the end so it can be unplugged and lifted out of the boat for external charging.
2. I have used the experience of the human powered boat developers when designing the prop, so am reasonably sure that it will shed weed OK. Many have already discovered that using big model aircraft propellers on boats gives a significant efficiency gain. Cedric Lynch used them very successfully years ago on his record breaking solar powered canoe and many of the human power boat developers have used props derived from those used on aircraft, with slim, long blades turning at a relatively slow rpm. The best source of design information I had was from a chap called Rick Willoughby in Australia. He has cruised long distances on weed-infested lakes, and undertaken a marathon trip down the Murray River, using his folding prop design. There is a photo of his prop here:
http://www.boatdesign.net/forums/boat-design/pedal-powered-boats-23345-25.html#post295095 and a report from Rick on the Murray Marathon here:
http://www.boatdesign.net/forums/boat-design/pedal-powered-boats-23345-41.html#post3300963. The sealed boxes for the motor, electronics and battery pack are all off-the-shelf enclosures purchased from Farnell. The plastic battery box was the most expensive, at around £25, the others were around £12 to £15 each. Most of the boxes I've used are rated at IP67 (1 metre immersion) but some, like the battery box, are only IP65 (hose proof), because they are located in a relatively dry location.
The leg that holds the gearbox and prop bearing is made from an epoxy/glass moulding. I made this in the standard way, by making a buck then creating a split mould.
4. I remember the EL84, too, although in my case I was more familiar with banks of glowing EL34s in guitar amps...............
5. 6kW is probably bigger than needed, even for a big boat. One thing with electric propulsion is to get away from thinking in terms of traditional, very inefficient, engine ratings and go back to basics. My starting point was the resistance curves for the hull. Nick had calculated these for Winsome and I cross checked them using a very nice hull analysis programme called Freeship. These gave the power needed to overcome still water resistance, the major part of total drag, and then it was just a matter of estimating windage for the range of wind conditions likely to be encountered. The result for Winsome was a surprisingly low power figure; the same would be true of other boats I suspect.
The key is efficiency. A traditional internal combustion engine will probably deliver less than a quarter of its rated power as propulsive effort. Because it is cheap to just fit a bigger engine to overcome the losses, that is what we have become used to doing. If the losses can be reduced, then the motor power requirement reduces too, by a great deal. A boat that might need a 5hp outboard, with a fast turning, small diameter, inefficient prop, may only need a 1.5hp electric motor to give the same performance.
Battery charging is a problem for big, heavy batteries, but less so if you can make then to be around the same size and weight as an outboard fuel tank. You can then take the battery ashore to be charged, or have more than one, so one can be ashore charging whilst the other is in use. You could even have a very small portable generator for emergency charging on board.
6. A 1kW system would do pretty much as you say and would be roughly equivalent to the biggest Seagull. I used to have a tiny Seagull Featherweight (around 2hp I think) and that would happily push my old 1 1/2 ton, 22ft gaffer around, even in a bit of a chop. Torqeedo have shown that a small electric motor, turning a big prop, slowly, can give remarkably good performance on modest power. They are doing pretty much the same as I am, in many ways, although I'm aiming for higher efficiency.
The off-the-shelf solar cells I have bought (from here:
http://www.everbrightsolar.net/diy-pretabbed-solar-panel-kits.html ) are around 16 to 17% efficient, so would deliver around 160 to 170 watts per m² at full sun (1000 W/²). In practice, in the UK, with the cells mounted horizontal, I can expect maybe a third of that figure. Even so, that would be enough to drive the boat independent of any power source provided that I do not use the motor for more than about half of available daylight hours each day. I think I can get a bit more than 1m² of cells on the boat, so may well find that I never need to use the charger. The battery will give a whole day of use, so I am hopeful that this will be enough of a buffer to allow for cloudy days or days when the boat gets lots of use with little charge time. It will be charging all the time that it is motoring, so the power drawn from the battery will be a lot lower than the motor power.
7. Whilst it is perfectly possible to put in a lot of time to do something like this as a hobby, the development time to turn it in to a product is the killer. I am sure that this is the reason that the Torqeedo units are so expensive. They are using the same very cheap RC model motors as I am, with similar controller technology, yet they sell for five to ten times the cost of the raw components. They have undoubtedly put hundreds of hours into development, hence the need to charge to recover this cost.
Some parts of the overall system are fairly straightforward to put together. Motors, batteries, battery management systems, chargers and controllers are pretty much off-the-shelf components, that only need housing. The challenge is in making an efficient drive leg and prop. It is surprisingly difficult to reduce losses and make something that is robust enough for boat use. The Sillette leg is probably pretty inefficient, as it is really one designed for an internal combustion engine that has just been converted to accept a big electric motor. It is proven technology though, and reliability counts for a lot.
8. For an off-the-shelf small boat system then I think it would be hard to beat the Torqeedo range, although they are noisy (they use a high-speed motor and gearbox, which whines a fair bit). You can just buy the motor and use your own battery packs to reduce the cost a bit. This would allow the use of cheap lead acid batteries, to reduce initial cost, and then explore the use of lighter LiFePO4 batteries once some experience of real battery capacity requirements are known, remembering that a lead acid battery will only deliver about 80% of its rated capacity.
Once I have worked out how to post photos here I shall have a go at showing what I have done so far with all the various experiments!
Jeremy