Pedal Powered Boats

[Guest625101138]

Leo
I have attached the Michlet file I used for my latest hull.

Actual performance is within 2% of my prediction at design speed of 3.3m/s and measuring power at the crank. This is very accurately measured using an SRM power monitor and includes all aspects of the hull, propulsion and rudder.

Will be interested in what you come up with as I am still trying to track down that last 2%. I am still under predicting power. My prop model uses analytical data for lift and drag coefficients and an empirical model for induced drag so it is not ideal - still waiting on some clever person to develop a pure analytical method for prop optimising.

Rick W.

[Leo Lazauskas]

Originally Posted by Rick Willoughby

Leo
I have attached the Michlet file I used for my latest hull.

Actual performance is within 2% of my prediction at design speed of 3.3m/s and measuring power at the crank. This is very accurately measured using an SRM power monitor and includes all aspects of the hull, propulsion and rudder.

Will be interested in what you come up with as I am still trying to track down that last 2%. I am still under predicting power. My prop model uses analytical data for lift and drag coefficients and an empirical model for induced drag so it is not ideal - still waiting on some clever person to develop a pure analytical method for prop optimising.

Rick W.

Rick,
My predictions for the V11J hull are attached.

Calculation of dynamic effects requires offsets above the waterline,
so I just added a few extra waterlines to your data.

As you can see, the dynamic resistance, i.e. with squat included, is
a little higher than the drag assuming a static attitude for all speeds.
Around 3.3 m/sec is close to where the hull starts to squat significantly,
so maybe that's where some of the missing 2% is going.

Cheers,
Leo.

[Guest625101138]

Originally Posted by Leo Lazauskas

Rick,
My predictions for the V11J hull are attached.

Calculation of dynamic effects requires offsets above the waterline,
so I just added a few extra waterlines to your data.

As you can see, the dynamic resistance, i.e. with squat included, is
a little higher than the drag assuming a static attitude for all speeds.
Around 3.3 m/sec is close to where the hull starts to squat significantly,
so maybe that's where some of the missing 2% is going.

Cheers,
Leo.

Leo
I would appreciate the dynamic ship_output_by_speed.mlt file if you can provide it. I will use it to repredict performance. It looks like you may have found my lost 2%.

Rick W.

[Leo Lazauskas]

Originally Posted by Rick Willoughby

Leo
I would appreciate the dynamic ship_output_by_speed.mlt file if you can provide it. I will use it to repredict performance. It looks like you may have found my lost 2%.

Rick W.

I used "Flotilla" to predict dynamic effects, so the output file is a bit different to Michlet's. I renamed the files as .mlt because this board doesn't accept .csv files.

Have fun!
Leo.

[Guest625101138]

Leo
There is something here that I do not understand. The value of drag at 3.3m/s from the files you provided is:
Static 31.45N
Dynamic 32.13N
They make relative sense.

Now when I run my version of Michlet using the V11.mlt file I sent you I get drag of 33.1N at 3.3m/s.

Not sure what is going on here as I would expect my figure to be the same as your static result.

Rick

[Leo Lazauskas]

Originally Posted by Rick Willoughby

Leo
There is something here that I do not understand. The value of drag at 3.3m/s from the files you provided is:
Static 31.45N
Dynamic 32.13N
They make relative sense.

Now when I run my version of Michlet using the V11.mlt file I sent you I get drag of 33.1N at 3.3m/s.

Not sure what is going on here as I would expect my figure to be the same as your static result.

Rick

The differences between the two sets of results are mostly due to the inclusion of boundary layer displacement effects in Michlet. In the results I posted, I used Flotilla and I did not include boundary layer effects.

I'm very pleased that you are getting good results from Michlet, but be a little careful when interpreting predictions. Michlet and Flotilla are only mathematical models and they do not include many real world effects. (If we included a second order effect like squat, shouldn't we also include other second order effects?)

Your Michlet results and my Flotilla predictions used the ITTC line to estimate skin-friction. That line is theoretically very suspect. I prefer Grigson's line, but that would probably give results that under-predict your experiments. Some people might then use a "form factor" on the skin-friction to improve correlations; others might also use a form factor on the wave resistance. I.e., there are a lot of ways to massage data to get better looking agreements with experiments, but that doesn't mean we have found the best model of "reality".

As it happens though, the inclusion of squat moves the results in the right direction for your problem. If you wanted to improve correlations of your Michlet drag predictions, you should multiply them by the ratio of the total resistance with squat to that without squat from my results. That should give you close to the 2% you are missing.

I noticed that your V11J offsets are given to 3 decimal places. For small hulls, it is sometimes better to use 4 decimal places in the offset table.

If I get some time I'll see if I can do a bit better than the V11J. (But first I have to defeat the evil Mongolians and Persians in another project of great importance.)

All the best,
Leo.

[Guest625101138]

Leo
The testing that is being done on the hull is quite accurate. Speed is repeatable to 0.1kph in calm conditions. We account for everything and have done power testing and comparison on every aspect as the boat will be used for a water record attempt for 24 hour distance.

Every thing has been refined to the limit and then adjusted for practicalities. The last bit of work now will be to add some fairing as he cannot find a windless lake and the wind factor becomes significant as soon as there is any head wind.

Rick W

[Squidly-Diddly]

Is that what those two across seater 'pedal boats' have?

I'm toying with attaching a long 'whip' to my kayak rudder, so when I work the rudder back and forth the whip should give me some propulsion, and steering at the same time.

That might be handy for kayak fishing-where you want to be able to maneuver a little bit once you got 'fish on!' but still keep both hands free.

[Guest625101138]

how about a one piece pedal/paddle wheel across a canoe?

Here is one for sale:
http://snorlax.lampi.us/mike/dakanu.html
You need quite large blades to get efficiency and then wind resistance becomes a factor. Also they suffer the same problem as all paddle wheelers in that they do not operate well in rough water.

Rick W.

[beppe]

Hi ASM
Rick's posts are quite interesting and detailed, as ever.
I'd like to address what he says: "the problem with the cycle drive shaft is that you need a large spinner on the prop to shroud the gear. This comes at a significant cost in extra drag. There are other more efficient ways to eliminate the chain."

I agree. But you don't need a regular cycle chain for a pedal boat drive unit.
I hope it could be interesting to have a look at this drive unit, developed within the Open Waterbike Project (http://www.openwaterbike.com/)
The thickness of the leg is 32 mm and the (external) diameter of the propeller shaft case is 40 mm.
It is a twisted-chain unit, it's very efficient (details at the Open Waterbike website). Lamar, the small firm that developed it, has been testing it for a couple of years and it never broke nor jammed in operation.

Best regards and thanks to Rick for starting the pedal boats thread!

Beppe Carignani
founder, The Open Waterbike Project

[Guest625101138]

Originally Posted by beppe

Hi ASM
Rick's posts are quite interesting and detailed, as ever.
I'd like to address what he says: "the problem with the cycle drive shaft is that you need a large spinner on the prop to shroud the gear. This comes at a significant cost in extra drag. There are other more efficient ways to eliminate the chain."

I agree. But you don't need a regular cycle chain for a pedal boat drive unit.
I hope it could be interesting to have a look at this drive unit, developed within the Open Waterbike Project (http://www.openwaterbike.com/)
The thickness of the leg is 32 mm and the (external) diameter of the propeller shaft case is 40 mm.
It is a twisted-chain unit, it's very efficient (details at the Open Waterbike website). Lamar, the small firm that developed it, has been testing it for a couple of years and it never broke nor jammed in operation.

Best regards and thanks to Rick for starting the pedal boats thread!

Beppe Carignani
founder, The Open Waterbike Project

Beppe
My comment to ASM was with regard to this:
http://dekrabike.com/ddrive.htm
I felt the necessary shrouding for this gearbox would be power sapping.

I have issues with twisted chain drives on reliability and efficiency grounds. I regard their appendage drag to be fair if nicely streamlined.

Have you determined the mechanical and hydrodynamic losses of the leg pictured.

Rick W

[I57]

Its seems to me the reason pedal boats have not really taken off is the lack of a reliable cost effective drive unit. I have had problems with chains coming off and prefer gears but gears can be expensive. Maybe with a belt drive cost could be kept down as well as weight, water getting in would not matter as much if all components were plastic and fibreglass. A drive leg that pivots up could be fitted to almost any canoe or kayak, for a receational boat speed is not so important. A lot of the touring canoes are beamy and very stable and I think would be much easier to move with a drive leg than with a paddle.
Having said that I have gone back to the drive in the boat, drive legs worked but boat was slower and the drive leg I made was large and heavy. I have gone about as far as I can with the current boat and this is its fifth configuration.

[Guest625101138]

Ian
Have you got any pictures of the latest configuration. I think what you have now is close to ideal. How does it perform?

The gearbox is heavy but will take giant force - see attached. This is the application that the box was built for.

I recommended that Greg K use a universal joint as I had had success with them at high rpm. With his boat at 4:1 cadence the joint felt clunky. I set him up a curved shaft when I was over there and he has since got hold of some 1/4" spring steel. He likes it a lot now.

I am going to use the 1/4" shaft on my latest boat. It is a mould off my old boat used as a male plug.

So a possible improvement would be the curved shaft. Maybe rev 6.

Rick W.

[I57]

No I don't have photos of this latest revision as I am still working on it. Next few weeks it should be ready, the prop is 450mm with a 3:1 gear ratio. It has two universal joints with the thrust being taken at the skeg, I tried the curved shaft in version 2 but the gearing let me down. This layout is about as efficient as I can get and the most reliable, have gone for the fixed skeg so I can reverse easily. Keep you posted.

[Guest625101138]

Ian
You can use a skeg with the curved shaft it simply avoids the universals.

With the two universals they should be set 90 degrees out of phase and ideally both have the same set angle. This means they will give steady angular velocity at the prop. I think this was the problem with Greg's as it only had one universal.

I hope to have my latest version of V11 finished in a week or two. I am going slow at the moment because the epoxy won't cure. I am having to heat my garage.

Rick