Pedal Boat Design

The shaft for the control wheel is not supported off the large wheel. It has to be carried an a separate shaft supported independently on the boat structure. It could alternatively be a fixed cam that is large enough to encompass the main wheel shaft.

And yes, the water force on the immersed blades will cause their linkage points on the control wheel to be in line with the shaft of the control wheel and the connection point on the blades. Once the blades are unloaded the point takes up a position controlled by the next immersed blade.

 

Man, that video confused the heck out of me. Check out the PDFs below to see simple diagrams (as well as all kinds of specs right down to thickness of the paddles). There are also some patented ideas for alternative methods of feathering. I like #526,533 - you could make one mold and then four identical composite wheels, and overal height is reduced.

Heh - I don't have a copy of the original feathered paddlewheel patent. I think it was British.

 

conical paddle wheel

This drawing merely shows the basic geometry of what I've got so far. That's the approximate profile of part of my rowboat on the right; the 16" dimension is from the gunwale to the bottom of the keel.

On the plus side, this still gives much lower windage than a conventional wheel of the same radius.

Problems . . .

The first thing that pops out at me, at least, is that the waterline is too close to the outside end of the wheel (vertically); this would result in uselessly stirring a lot of water without adding propulsion. I'm thinking I need to remove some of the paddle surface from the outboard end; the central shaft can likely be shortened by half.

The second thing that looks -- uh, less than optimal, is that the wheel is immersed to considerably more than 60 degrees, which would likely cause more wasted effort.

I want to refine this somewhat before I attempt a physical model. I already have a 1:10 scale model of the hull of my rowboat, and I want to build this system for a tub-test powered by rubber band.

How can I improve this?

Curtis

 

OK, I said I would stop rambling for a while, but I couldn't resist...

Rick's mention of "cam" got me thinking about the camshaft on an engine, and I scratched out this design. It is not to scale, and I'm not sure what the diameter of the inner wheel should be (I used a CD and a soda can). Essentially, you have each paddle immersed inside the wheel and they are spring-loaded. Once they hit the inner wheel, they are forced out into the water to do their duty. It is simply another way of 'feathering'.

One would think there would be excessive wear and/or binding between the inner wheel and the paddle edge. If this is a problem, it could be somewhat alleviated with rollers (think 'roller cams' on performance engines). I think the greater issue would be the loading on the paddle vs. its guides on the rim of the paddlewheel.

Now, on a completely different tangent, has anyone considered putting shrouding around, say, the lower 10 arc degrees of the paddlewheel to achieve more thrust? If it works for props, why wouldn't it work for paddles? Would the increased drag through the water negate any performance gains? Obviously, it could create problems in extremely shallow water...

 

Interesting thought, V.I. Eccentric motion brings up a lot of other possibilities.

Anyway, here was my next iteration. I cut into the middle from the outboard end. On this one, I also added detail to indicate the paddle blades separately from their frames. More framework is needed, of course.

. . . anyone remember the chariot race scenes from "Ben Hur" and "Gladiator"?


And, it still didn't fix the problem of stirring a bunch of water for nothing . . .

 

. . . so, I then cut off all those nasty points. Notice that this results (as did the previous one) in losing a corner of the paddle. The lost area is almost exactly the same for this and the last one.

Sorry about having to look at it sideways; I thought I'd gotten it rotated. Damn scanner software is glitchy.

Any better? Let me know, please.

Curtis

 

Well, you haven't really feathered the paddles. You want them as perpendicular as possible to the surface of the water. Your design would be useful if the entire wheel was submerged (such as early submerged paddlewheels used for steering) but your paddles are still angled as they enter and exit the water.

 

Are you still convinced there is an advantage to angled wheels as opposed to upright sidewheels? Seems they might be making the boat wider than old school wheels would.

 

I think the angled wheels though not "feathered" give much better entry and exit vectors then a traditional wheel would. I read through all the information you posted Tin Horn and then started researching specific terms in the document (always better search results when you refine what your looking for) and the angled approach might lessen one of the bigger problems.

They called it "paddle slap" and apparently it cause a pretty nasty vibration throughout the boat--not to mention a huge loss in efficiency of course.

I was finally able to get the feathering wheel working (thanks Rick =) in Solid Works, though several things are less then perfect in the concept as a whole. On designs similar to the original English patent they very often experienced "rocker lock" if changes in speed or direction were not managed very carefully. The modeling really has this tenancy because I can't simulate the load on submerged paddles.

The smaller inner wheel starts (and changes) its rotation independent of the the larger larger wheel and there is lag and sometimes "bounce". On the big boats that meant one rocker arm/connecting arm hyper extended and locked. I am betting a lot of cursing usually followed lol. In subsequent variations it looks like they geared the smaller wheel to the larger one and this took care of the problem...unless the skipper had to rapidly stop by throwing her into reverse.

Attached is an image of the wheel I modeled and if possible a small (4mb) animation of it in motion that shows the inner wheel bounce. Sorry the video is so wide--not sure how to crop them. For that run I applied a motor to the main wheel at 10 rpm. If I could apply loading forces from the submerged paddles the bounce would not be so bad I am sure. On a later version with the pivot point centered on the paddles it was a little better also.

My lightest design added about 5 pounds to the weight of a regular 7 spoke wheel. All the bearings (24) even if you use UHMW, and the arms etc really adds up. That isn't counting structural changes needed to accommodate the extra wheel. And thats on ONE side of the wheel only. Not positive a 30" wide wheel could get by on one side.

I am going to look elsewhere as I just don't think feathering, or at least my interpretation of this method works on our scale.

 

Curtis-

I really like the angled approach you taking. I have already started imagining the structure needed to lock down the far end of your tilt-o-wheel lol. I rotated the image for you.

 

BG
I like the model. I would have the control linkages in tension rather than compression. I also think you could tweak the length of the linkages and the location of the centre of the control wheel.

 

If you think about it, these things sticking out from the sides are still much narrower than my boat oars, and I'd rather have 'em wider than higher. I don't need to worry about sideways clearance, but my rowboat already blows around too much in the breeze as it is.

They probably would, but I don't know if that's a worthy trade-off for whatever other losses they might cause. I'm hoping for more input from others on that.

As to the feathered wheels, I think you're better off just staying with fixed blades; lighter, simpler. There is a way to make them work without the 'rocker lock' and 'bounce.' It takes lots of chain, and one sprocket for each blade, plus two more. You have two chains; one is sprocketed from one of the blades to the wheel hub, where that sprocket is fixed to the frame, non-rotating. That one sets the angle of the paddles relative to the frame, and it's at a 1:1 ratio. The other chain goes around sprockets on each of the paddles, and synchronizes them all at the same angle. All this stuff has to be synchronized at assembly, and I can only imagine what a chore that could become. Also, all this chain will get wet, may catch weeds, and will add friction in addition to its weight. I know that plastic roller chain exists, but I just wouldn't bother.

 

Curtis-

Heres your latest with 7 blades in motion if you want to know what it looks like...I didn't see there was a box to set the animation to be reciprocal so you could see it both rotation directions till after it rendered. Theres also one for how many frames per second so it doesn't look so choppy...ah well, I will try those on the next one lol. Theres a bottom view too--under water so you can see whats going on.

Hope it helps your process at least a little.

Rick, I tried them (the connecting arms) in both directions, and about a dozen different arm lengths and little wheel diameters. There are little gains and losses either way. If you manage to decrease the locking angle on the "long side" it just makes the angle too acute on the short side. There is likely some magic balance point but I would wager it was only in one direction. Forward might be fine, but reverse would suffer.

 

BG,
Uh, I don't see a clicky thingie for your picture . . . and I couldn't make your other animation play; it tried to open in QuickTime, said it needed a plugin. I took my best guess and downloaded AutoDesk FBX, and it still won't play . . . any suggestions?

Btw, thanks for righting my capsized wheel ;-)

 

Well, for w/e reason it wont let me upload the files. So here is linkage:

http://geno.boxgods.com/curtis_angled_wheel.avi

http://geno.boxgods.com/curtis_angled_bottom_view.avi

Interesting to see how the paddles strike the water.