Pedal Powered Boats

[Jeremy Harris]

Originally Posted by I57

Andreas
Any details yet of the drivetrain you plan to use.
Quote:"I plan to use a HTD belt with a 90° twist, as this method is absolutely maintanence and corrosion free and very cheap, as it will use only one custom built part, which can be lasercut very cheap. The whole drivetrain with the best and only new premium parts will cost me only 150€ (but can be built for half or less)."
Do you know what distance is needed between the crank and the shaft to get the 90deg twist. Is the distance less than with a bike chain? Also what sort of gearing ratio are you looking at.

Ian

I tried this with a 3mm pitch, 15mm wide HTD belt and it didn't work well. The belt constantly wants to ride up over the side retainers of the pulleys and gets chewed up unless it's very tight. The only way I could get it to sort of work was with a very high belt tension, and then the losses were high from friction at the belt edges. The belt length I tried was around 400 mm between centres, with a pair of 26t pulleys. It might work with some sort of guide to maintain the twist in the belt, perhaps.

The nice thing about twisted chain drive is that the twist forces only act at the centre of the chain, plus chain has a lot more "give" torsionally than toothed belts. I believe V belts work OK like this, but again would have high frictional losses.

[Coach Dave]

Originally Posted by spidennis

Ok, here's a close up of coach dave's peddle setup. I have a bike that I got sitting in my garage, it's upside down and looks just about the same! Did he not even cut and weld this? It does look a bit high though. What would be the proper geometries that I'd be looking for? Feet that far up looks uncomfortable for very long.

No welding was required. A local bike shop had an old frame sitting outside that they let me have. I cut off the chainstays and removed everything I didn't need. All that was left was the main triangle of the chrome moly frame. I put in a new bottom bracket because the old one was damaged. In retrospect I could have used a smaller frame and gotten my feet a little lower. The ideal height IMHO is so your heels just barely have clearance at the bottom of the rotation. The current setup is comfortable to pedal 2 or 3 hours at a time. Here is a video with 4 people in the boat, food, water, camping gear and two solo kayaks bungeed on. http://youtu.be/c_l86SA3i5E
The boat uses dual stabilizers in that video vs the proa (i.e., single outrigger) that I used initially.

Dave

[Coach Dave]

Originally Posted by spidennis

Ok, here's a close up of coach dave's peddle setup. I have a bike that I got sitting in my garage, it's upside down and looks just about the same! Did he not even cut and weld this? It does look a bit high though. What would be the proper geometries that I'd be looking for? Feet that far up looks uncomfortable for very long.

After hauling the two solo kayaks around the lake for awhile we launched them from the mother ship in this video. http://youtu.be/oxwTNn_7OKk
One thing that is striking on this video is how much ripple the short solo kayaks make. In comparison the 28' mother ship cuts through the water with very little ripple.

Just for grins here is another video of us kayaking in the beautiful clear Silver River. We went upstream the entire length of the Silver River - from where it joins the Ocklawaha to its source at Silver Springs. The video is when we are heading back downstream and enjoying the sunshine. http://youtu.be/JzFYcxY4WQc

Dave

[Scheny]

@Submarine Tom: Only the upper hull is planing, which has to take only a little amount of the weight. So I hope to achieve at least 10kph @150W. As no one has tried this approach before, it is a bit of gambling if the project will be a success or a loose. 150W is my design point.

@I57: For the drivetrain, I will use a HTD 5M with 9mm width (9mm should be enough, as I only use the 12x12 prop requiring less torque but higher speed). Jeremy is right, that it will want to jump of if not constucted properly, but radio controlled helicopters all use twisted belt since many years, you can see how it works here http://www.matthias-kluge.de/trex450se/PA080006.JPG. The trick is the space between the belts. Imagine that the belt consists of a single rope located in the middle of the belt. If you construct the drivetrain the same way as a twisted chain, it will jump off for sure, no matter what length you use. If you design the imagined rope to always be within the pulleys, it will work also without force. The mechanics call this "natural bending". See Nauticraft as an example, who have only 100% natural bending. A length of a few centimeters would be enough in this case.

I will try a ratio of 1:8.5, what will give me 12kph at pace. Even if the boat will have the "sonic barrier" at 10kph, the water flow will have around 12kph then. I have watched many small pedal boats and saw, that also catamarans with slim hulls were bound by hullspeed mainly. So I hope, that interfering with the waves might help me.

I already had a conversation with Giuseppe from openwaterbike.com. If my drive works fine, my next project will be to convert it into a modular drivetrain using a twisted belt and consisting of non-corrosive materials only which will be available as a driveleg and/or internal drive and cost only 100€ if manufactured in China. It should be able to be customized to any length and hulltype at a price change of 20€ only.

Currently I am moving into a new flat, having to work hard to develop a next generation car for a major car company until end of this year and working on my next course's material (I am training airline pilots as a freelancer), thats why I have to put the project for pause for about 6 weeks.

Andreas

[Jeremy Harris]

Originally Posted by Scheny

Jeremy is right, that it will want to jump of if not constucted properly, but radio controlled helicopters all use twisted belt since many years, you can see how it works here http://www.matthias-kluge.de/trex450se/PA080006.JPG. The trick is the space between the belts. Imagine that the belt consists of a single rope located in the middle of the belt. If you construct the drivetrain the same way as a twisted chain, it will jump off for sure, no matter what length you use. If you design the imagined rope to always be within the pulleys, it will work also without force. The mechanics call this "natural bending". See Nauticraft as an example, who have only 100% natural bending. A length of a few centimeters would be enough in this case.

Andreas

Narrow belts are OK, and it was the model helo drives that gave me the idea to try one. The problem is that you'll need either a fairly wide belt to take the torque of the large diameter prop, or one with a fairly big tooth pitch. The 15mm wide 3mm pitch HTD belt I used was very marginal with 26t pulleys in terms of torque capability, it would probably have been fairly unreliable at 100W for any length of time, but when I tried a 5mm pitch 15mm wide belt the losses from the much higher belt stiffness were very high. As I mentioned, I was running this at about 400 mm centres, plenty long enough to allow the belt to twist OK.

You could probably get the twisted belt drive to work using something like the narrow width, bigger tooth pitch, Gates Carbon Drive system, rather than HTD, but the bigger tooth pitch means a bigger lower unit pulley diameter. I didn't want to go over about 30 mm overall diameter on the lower unit, which meant a 26t or maybe 28t 3 mm pitch pulley was as big as I could go, and even then I was on the limit in terms of belt tooth stress.

I did a lot of experimenting with HTD belt drives (some of those experiments are documented here on the electric boat thread) and found that to get acceptably low losses you need to use the smallest pitch you can with the largest diameter pulleys you can, because belt flexure loss is very high at these low power levels. For example, my first 5 mm pitch, 15 mm wide drive used around 20% of the available power just to overcome belt bending losses, even with the least amount of belt tension that was still reliable. Switching to 3 mm pitch massively reduced these losses, because 3 mm pitch HTD belt is much more flexible than 5 mm pitch.

What the belt manufacturers don't tell you is that the high efficiencies they quote are only obtainable at the higher rated speeds, at low speed and high torque toothed belt efficiency is poor, no better than maybe 80% at the 100 W at a few hundred RPM I was working at.

[portacruise]

Jeremy;
Thanks for some good info.

Do the flexing losses only apply to belt drives?

Pardon if this is a daft question, but does it make an efficiency difference particularly for large gear ratios if different diameter GEARS or CHAIN are used to obtain that ratio? For example, say I am interested in a 6 to 1 one stage gear ratio, which diameters are more efficient at medium speeds (400 rpm output)? Suppose the choice in a pedal application going UP is using 36 cm diameter drive and 6cm output vs. 18 cm and 3 cm? What about going the other way, in an electric boat motor going DOWN reduction of the same 1 to 6?

Always wanted to know but was afraid to ask....


Thanks,

Porta

Originally Posted by Jeremy Harris

Narrow belts are OK, and it was the model helo drives that gave me the idea to try one. The problem is that you'll need either a fairly wide belt to take the torque of the large diameter prop, or one with a fairly big tooth pitch. The 15mm wide 3mm pitch HTD belt I used was very marginal with 26t pulleys in terms of torque capability, it would probably have been fairly unreliable at 100W for any length of time, but when I tried a 5mm pitch 15mm wide belt the losses from the much higher belt stiffness were very high. As I mentioned, I was running this at about 400 mm centres, plenty long enough to allow the belt to twist OK.

You could probably get the twisted belt drive to work using something like the narrow width, bigger tooth pitch, Gates Carbon Drive system, rather than HTD, but the bigger tooth pitch means a bigger lower unit pulley diameter. I didn't want to go over about 30 mm overall diameter on the lower unit, which meant a 26t or maybe 28t 3 mm pitch pulley was as big as I could go, and even then I was on the limit in terms of belt tooth stress.

I did a lot of experimenting with HTD belt drives (some of those experiments are documented here on the electric boat thread) and found that to get acceptably low losses you need to use the smallest pitch you can with the largest diameter pulleys you can, because belt flexure loss is very high at these low power levels. For example, my first 5 mm pitch, 15 mm wide drive used around 20% of the available power just to overcome belt bending losses, even with the least amount of belt tension that was still reliable. Switching to 3 mm pitch massively reduced these losses, because 3 mm pitch HTD belt is much more flexible than 5 mm pitch.

What the belt manufacturers don't tell you is that the high efficiencies they quote are only obtainable at the higher rated speeds, at low speed and high torque toothed belt efficiency is poor, no better than maybe 80% at the 100 W at a few hundred RPM I was working at.

[Jeremy Harris]

My experience was that chain losses are low pretty much at any torque/rpm, but belt losses tend to be dominated by flexure of the belt around the pulleys, so had higher losses (in percentage terms) at low power levels. As the belt thickness increases (as it does with increased tooth pitch) the force needed to bend the belt around the pulleys increases, increasing these fixed losses. In my case I found that a 15 mm wide 3 mm pitch HTD belt had around 1/4 the losses at 400 rpm than a 15 mm wide 5 mm pitch HTD belt.

The losses seem to be dependent on pulley diameter - the bigger the pulley the lower the losses, so using the biggest available set of pulleys for the ratio you want seems to be the best bet. This also tends to increase the maximum torque rating a bit, by allowing more teeth on the belt to take the load.

My only experience with chain was with small (6 mm, slightly smaller than #25) pitch chain. This had near-negligible losses, even when running around small (10t) sprockets. Based on this experience I'd suggest that chain losses are generally likely to be a lot lower at these low power levels than any belt drive, although the belts do have the advantage of low noise level and cleanliness.

[portacruise]

Originally Posted by Jeremy Harris

My experience was that chain losses are low pretty much at any torque/rpm, but belt losses tend to be dominated by flexure of the belt around the pulleys, so had higher losses (in percentage terms) at low power levels. As the belt thickness increases (as it does with increased tooth pitch) the force needed to bend the belt around the pulleys increases, increasing these fixed losses. In my case I found that a 15 mm wide 3 mm pitch HTD belt had around 1/4 the losses at 400 rpm than a 15 mm wide 5 mm pitch HTD belt.

The losses seem to be dependent on pulley diameter - the bigger the pulley the lower the losses, so using the biggest available set of pulleys for the ratio you want seems to be the best bet. This also tends to increase the maximum torque rating a bit, by allowing more teeth on the belt to take the load.

My only experience with chain was with small (6 mm, slightly smaller than #25) pitch chain. This had near-negligible losses, even when running around small (10t) sprockets. Based on this experience I'd suggest that chain losses are generally likely to be a lot lower at these low power levels than any belt drive, although the belts do have the advantage of low noise level and cleanliness.

Thanks, Jeremy.

A couple of thoughts on the belt drive. You can increase contact surface on the tiny pulleys by using a spring loaded (or fixed) idler positioned near the tiny pulley end, so that it wraps around to give more contact with thin, flexible belts. I liked the spring load better in my gadgets because it always adjusted to the proper tension with changes in temperature, humidity, wear, etc. and it allowed for almost instant repair/exchange by manually pulling against the spring. It was possible to run 50%+ thinner, under rated belts for awhile as they failed at a predictable time. I just carried spares that could be threaded very quickly without tools in a very open, accessible system. All before the carbon belts came around... This was with non-twisted belts, but maybe the twist could be done in 2 stages of 45 degrees, assuming there's any advantage....

One disadvantage of open, unshielded belts is that debris like water weed, mud, etc. can get caught in the grooves between the belt and pulley where it remains in a mush, causing tension/vibration, etc. issues....

P.

Porta

[Scheny]

Rick was so kind to analyse the wave interference hull data and it seems, that I will stop the project.

The estimated cadence speed is about 8kph with a top speed of 12kph. The net drag is the same as for a catamaran of equal length, but the upper hull seems to cause slightly more total drag than a cat. The hullform has advantage to others at a length of +6m. As long, as I cannot push speed to at least 10kph, I will stop the project in favour for another boat.

[Scheny]

This boat here http://www.youtube.com/watch?v=4uCXmdNYp_M has a flat stern.

How would such a boat compete compared to one with a sharp stern at the same length, same width, but higher draft?

[Submarine Tom]

Good exploration Scheny!

-Tom

[portacruise]

Originally Posted by Scheny

This boat here http://www.youtube.com/watch?v=4uCXmdNYp_M has a flat stern.

How would such a boat compete compared to one with a sharp stern at the same length, same width, but higher draft?

Square sterns in general are not as efficient from my experience. Notice the turbulence right behind the flat stern when looking from behind the boat in the clip. This would appear to cause a drain in energy (suction) compared to a boat with two pointed ends. The pointed but elongated football shape is supposed to be best from what I have read. Even the outriggers on Rick's boat use this basic 2 point form.

After McDenny's experiments and measurements, I have even gone to putting cone shaped fairings in front and behind my flex shaft prop to increase efficiency! See my previous post to you on model RC props for MdDenny reference.

P.

[Coach Dave]

Originally Posted by Scheny

This boat here http://www.youtube.com/watch?v=4uCXmdNYp_M has a flat stern.

How would such a boat compete compared to one with a sharp stern at the same length, same width, but higher draft?

If you have some "what if" questions about hull shape and its impact on hydrodynamics you can analyze the alternatives with Michlet. If you are after an optimal low drag hull you can run Godzilla with your design constraints.

http://www.cyberiad.net/michlet.htm

[spidennis]

coach dave,
got any more info on your drive system?
prop, shaft, chain, gearbox, etc.
Those videos show you moving along pretty nicely with that load!

[portacruise]

Originally Posted by portacruise

Square sterns in general are not as efficient from my experience. Notice the turbulence right behind the flat stern when looking from behind the boat in the clip. This would appear to cause a drain in energy (suction) compared to a boat with two pointed ends. The pointed but elongated football shape is supposed to be best from what I have read. Even the outriggers on Rick's boat use this basic 2 point form.

After McDenny's experiments and measurements, I have even gone to putting cone shaped fairings in front and behind my flex shaft prop to increase efficiency! See my previous post to you on model RC props for MdDenny reference.

P.

Oops, I was going on the assumption that the pointy boat would not be completely submerged. Still, I think the record for pointy, one man, human power subs (Omer?) is around 7 knots.

Thanks for the link, Coach Dave, much better answer.

P.