Pedal Boat Design

If you use flat panel construction you will get ample strength from 200gsm CF mat either side of a 6mm corecell or nomex core. The corcell would be my choice as it will tolerate imperfect infusion. You can do the sums and a single layer provides sufficient strength but no rigidity. The corecll composite provides amazing rigidity for weight. If you have some of this then make a test panel and see what I mean. Need bulkheads at load bearing points.

If you do not want to make a composite panel then 4 layers of 200gsm CF will provide ample rigidity. Need bulkheads at about 1m intervals and any load bearing.

You could easily cut the four panels of the hard chine hull from a 1m wide panel. Just a matter of stitch and glue the panels into a boat. Neat and fast.

 

BG
Here is a post on weight saving from my Canadian friend:
http://www.adventuresofgreg.com/HPB/2008/08/cp2-goes-on-diet.html

You can go back through the blog to see a lot of detail on the boat build. It is a method that I recommended after playing around with my own boats.

Ultimately for a one-off I am thinking seriously about all corecell flat panel. My deck was so easy to do and was the fairest part of the boat. My original foam boat that was used as a male plug for V11J was hand formed from layers of foam. So more work than Greg but less money.

Overall his boat is quite likely the most efficient long-distance calm water craft ever built. Its slight weight disadvantage over something like a K1 is offset by more efficient propulsion system. Without it being set up for sprinting he is able to do 1000m in 4 minutes. Not bad for 250W.

 

Just for grins, and because I spent time doing it, here's the latest drawing from the Conical Paddle Wheel Design Studio, located somewhere in the basement of the Bureau of Lost Causes. This one has twelve blades, each with a surface area of (gasp!) 16 1/2 square inches!!

I could wave my hands in the water at the sides of the boat and get just about as good a result. What's even more insulting is that (since the greatest point of this exercise was to reduce windage,) the profile of this model is a whopping four inches lower than for a conventional wheel of equal radius.

Some of you may have noticed that I started another thread (human powered surface drive) to explore another idea. I haven't thrown that one in the dustbin yet . . .

Anyway, enjoy.

Oh, yeah . . . the blade resize was an effort to get the immersion depth to within 60 degrees. It's more like 74. At least this one would be lighter -- I think.

 

Rick-- Excellent resource, thanks.

Curtis-- If you can give me the length of the long and short sides that form the angle at the axle I can give you the exact weight in say 1/16" 6061 and in carbon fiber. Also, you didn't post a link to the other thread.

Frame progress-- First version was already at 26 pounds so I have scraped it and started over. According to the COSMOS design analysis 500 pounds of weight did not deflect it 1/4" so I may have over engineered it slightly. Lol.

Will post images when its a bit lighter. New product launching in 2 weeks so my free time has vanished sadly.

 

A good load case for a catamaran is to support it 1m from the bow on one hull and 1m from the stern on the other hull. This will test torsional stiffness. It is often disregarded but is happens often when you quarter waves. The more reserve buoyancy you put in the ends the worse it gets. See what deflection you get with this.

Rick W.

 

Rick--

I am having a few issues with the loading and deflection calculations. As previously established I am mathematically impaired. For the record, I am also a very bad speller =)

SolidWorks can do amazing testing but I am unsure what loads to introduce and where and in what directions. The other issue is that frankly I really can't make heads or tails of some of the numbers it spews out.

Once I am done with what I think is a great structure I was hoping you could suggest some loading tests (like the one mentioned in your last post) and then help me interpret the results.

Next, I burned 2 hours on Greg's Blog. Lots of good information there. Starting with you passing yourself off as some stodgy old grumpy professor type. You guys have to check out pictures of Mr Willoughby lol.

One thing you may want to mention to Greg is that his construction negates about half the benefits of carbon fiber construction. He actually did a wet layup over his plug. Wet layup construction is suitable for decorative non load bearing laminations but is quite bad for structural parts like boat hulls and tubes etc. He has a VERY resin rich laminate with voids I could easily pick out in the pictures.

The wet layup method typically doubles weight and decreases strength and rigidity by as much as 50% over prepreg layups. The weight increases go up a good deal with each added layer and overall part size as well. I KNOW I could shave something like 5 pounds out of his hull--maybe a bit more. The surface would be like glass too =)

I also have to say that your hull form is starting to grow on me a bit. I still wonder why the skulls he was puttering around with all had some curve to the bottom...strake or w/e it is called. I know racing skulls have had something like a century of constant improvement and refinement. They are human powered and operate in our speed and power output range right?

 

Just to mention Greg's layup. All layers were done in one go and he ended up with a good resin ratio but it was a bit lean in patches because I know the vacuum left some pinholes. Overall it was far from optimum because it only requires 1 layer for the required strength so all the rest was simply for rigidity. A CF/Corecell composite would have been much better. With hindsight and a bit more learning I think he could have heat formed Corecell over the plug and done a single layup with CF on the plug, Corecell sandwich and then CF on the outside all pulled together with a vacuum over the plug.

I am only a new player with low density structural composites and my results are ordinary so it is an area I am keen to learn.

From memory the whole CP2 hull is about 5Sq.m. I believe it would be possible to make the entire boat for 7kg. It is an inherently strong structure and relatively simple form. Better than either rowing scull or K1 as it is fully enclosed with fixed load bearing points.

For V11J my initial layup was 2 X 320gsm plus 1 X 100gsm all in one go. The total weight of the hull without deck was 7kg. I was not happy with the rigidity so I hot formed a 3mm layer Klegecell on the inside, epoxied in place and then did a layer of 200gsm on the inside of this. That added another 2kg. By the time I added the deck and bogged it out for fairing it was up to 14kg. So way heavier than it could have been. Saving 7kg on a hard chine flat pack boat would make up for any loss due to sub-optimal shape.

The worst loading case on any cat as far as I have determined is the quartering sea as stated earlier. However more than happy to review your designs and suggest load cases.

You are gradually convincing me I should have a look at Solidworks. My problem is that I do not spend enough time with the things to learn them well. I find Delftship way more intuitive than Autocad so I just use it for all the things I do. I have even produced prop designs for milling using it. There are large overheads in learning to use new tools and I need to be confident that I will get benefit for the effort. I was forced into using Office 2007 with purchase of a new computer last year. It took me months to come to grips with the revised menus - very frustrating. Still takes me an hour to find the second Y-axis on a plot.

 

I am not sure, but it sounds like you think pre impregnated (prepreg) is limited to one layer at a time. This is not the case. You can do as many layers as you want all at once (within reason). Most people would look at correctly laminated CF composites and think they were way too dry. This is mostly because people see decorative CF parts with a gel coated exterior finish. Something I would still do on the boat hull unless I did a lead off layer in silk weight glass.

The obvious problem with parts 20 feet long is an "autoclave" tank. Crushing forces help make a thin light laminate with the correct resin to cloth ratio but drawing 20 inches of mercury in a tank and heating your part to 250 degrees (or so depending on materials) makes a very smooth pinhole free structure. That said, I have found for parts to big for a tank that if I take my prepreg after I have rolled it out and cut it to shape and put it in a little (5 gal bucket) and pull it down for 30 seconds then do my layup as normal with plenty of MANAGED heat I come pretty close to a pinhole free structure.

Building a hot box is a LOT easier then building a pressure tank that big.

Anyone who does "real" composite work, like aerospace or race grade marine, will tell you the same thing. Wet lay up is almost a joke for anything beyond aesthetic class parts.

The flat top on your design would make assembly very easy. Why do you run the flange out instead of in though if you don't mind me asking? Your using a plug and not a female mold likely.

 

I was under the impression you thought Greg did separate layups rather than a single layup. Hence my previous comment. There is no doubt the boat has adequate strength. He can support it at the very ends and sit in the middle. THis is a much worse loading case than it can experience in actual operation.

What it does show is what could be achieved with a proper manufacturing process. It is already considered by some of us the fastest cruiser on the planet and there is no doubt room for improvement.

I have only seen samples of factory produced flat pack and it certainly looks lean.

The thing about pedal powered boats it scratch resistance and panel rigidity rather than anything to do with actual strength. There is detail required to avoid tearing reinforcing from core when colliding with something for example. So local point loading issues rather than actual structural strength is a design factor. It is virtually impossible not to beach the front of a 24ft boat when you are coming ashore. The ends are typically solid foam internally for the first 6" or so.

The flange is because we are building one offs from a male plug. My current boat started out as a female plug but I was happy enough with the external finish to use it as a boat. Also the fellow who wanted a copy got a new job that now occupies what free time he had to put into a boat so another reason to shortcut. The flange also provides a strong mounting point at any location along the hull so provides a lot of freedom in attaching bits.

 

OK, we can scratch the much stronger part but I KNOW you would not pass on the much lighter part. As for strength, there is also a good gain in rigidity--his problems with the boat deforming between the bulkheads.

As for your learning SolidWorks. There is no taciturn way to say this, but if you could see what I can do with it you would be very impressed. Keep in mind that compared to you I am pretty much velcro fasteners on the shoes and helmet material.

There are a good many well done tutorials on the software and all it's many facets available for free online. I went through those and was able to function reasonable well within a few days. I was "functional" to the point I could draw at about 75% of think/doodle speed in about 10 days. I broke down and bought a SolidWorks for dummies book for $20 on Amazon and got to about 90%. The biggest limiting factor truthfully is unlearning the way 2D CAD works.

The ability to make any part your brain can think of in 3D, then make the other 20 parts that bolt together to make your creation is just hard to explain. Imagine being able to actually assemble your boat from the actual individual parts, then actually be able to move any part that is supposed to move--pedals, cranks, sprockets, gears, pulleys, rudders etc. in real time where each part affects any part it touches...turn the pedals and the prop spins...detect any collision or range of motion issues etc.

It also has a very full set of sheet metal fabrication tools. Design your part as you want it, then click one button (flatten) and theres the sheet shape you need to cut in one second.

It also has full molding tools. Add any amount of draft you want in any direction with a few clicks. Add parting lines etc till your happy then let the wizard automatically and easily show you any straddled surfaces etc.

I just finished my first set of product designs for acrylic parts being injection molded in China. A very complicated pump with a full helix and built in reservoir.

I spent 4 days trying to learn AutoCAD with my Dad as a tutor (professional Draftsman in the Navy for many years) and gave up. Used TurboCAD because I just didn't need to use software that could design nuclear subs to draw house plans.

Being able to see how stuff works in 3D is like a tinkerers dream. I am old enough to know now that I am never going to get to build all the amazing (to me) things I think up, but 3D creation software means I will at least get to try most of them.

Besides, I had nobody to ask when I got stuck with SW and you would have me to ask (till you got to all the math analysis junk anyway =)

 

Oh yeah, forgot to mention one of the better things.

Most companies that make products we use, like nuts and bolts, gear boxes, on and on now offer their engineering drawings in Solidworks files so you can download the part and just add it to your boat. Not a 2D with dimensions...the actual part in 3D that you can drag around and actually install into your assembly...

I have the main frame rail that I designed in about an hour as an edrawing that I can show you if your really interested. Download and install the viewer. Download and install are 2 minutes and free. There is a grand total of two buttons you need to completely understand the part...you can rotate it and see it from an angle in real time. Then you will know what an average guy can design in under an hour (with 90% of that as thinking about it time). Will take you 5-10 minutes tops and you will get a much better feel for deigning in 3D.

G

 

Shortly after you first intro'd your conical model, I suggested that there may be some lateral thrust associated with its design, and an easy fix would be to design in dual paddlewheels to eliminate that issue. I believe your response at the time was that any tracking issues of the hull would be minimal, or rather that issues would be due to the hull rather than the propulsion methodology.

 

Do you know of any links or resources about building hot boxes or ovens for composite work.

Thanks

Gareth

 

Well, I had planned on this being a side-wheel design, if that's what you're talking about . . . I wouldn't have considered using this as a single unit, as a stern-wheeler.
And, I still believe that propulsion methods won't fix a tracking problem. I have direct experience here, first with an inflatable (Sevylor Tahiti) which couldn't be made to stay on course for any length of time, then, with my current rowboat. The rowboat was so bad that every oar stroke was a steering stroke, literally. Half or more of my rowing effort was dedicated just to keeping it on course. I built a kick-up stabilizer for it, like a non-turning rudder, and now it tracks just fine -- as long as the stab is down. I'd give a written guarantee that this boat would be all over the place without some such add-on. Hull shape -- fineness ratio, keel, chines, etc., (which this rowboat lacks in sufficiency) -- determine tracking ability to a great extent.

I may be partly wrong on this; this rowboat is made to accept a small outboard, and it might be designed so that it steadies out under that kind of power -- but I can't imagine how or why, and I'd be loath to trust it.

Curtis

 

Grob--

I actually just build my own. I already have a 2' x 2' x 8' box that I made from 3/4" particle board. I went the extra distance and laid up several layers of glass with a high temperature epoxy resin. Its a few years old now so I don't remember the exact resin, but I think it has some sort of ceramic additive...Might Google Master Bond Epoxy as that seems to ring my memory bell.

Anyway I built a very simple box with screws and gorilla glue. Leave off one of the ends. I then added 2" wide strips to the outside edge of that open end. Two layers so I got a total thickness on the open end of 2.25". My end cover is sized to fit that--28.5" x 28.5" square. Then I added 2" strips around the perimeter to give some extra rigidity. Laid up a few layers of glass on the inside.

I couldn't find a gasket that would fit so I used the old standby, pure silicone. Covered the finished lid in saran wrap then added a fat bead to the lip on the box and screwed the lid on and let it dry 24 hours. Take the lid off and remove the plastic wrap and you have a cheap but functional seal.

I had some bulkhead air fittings in my junk can (no idea from where but I am sure McMaster would have em) that I connect my air line too. If you have careless friends or dim employees do NOT use the same fittings you have on your air compressor--I happen to know this particular tank will take 50 PSI--now my vac pump uses a bigger fitting just to avoid a scare.

I use a pump I got off ebay for about a hundred bucks. I can pull 29 on it though it takes a tad to get her there as that box is 32 cu ft. Be sure to get at least a quarter horse and 2 stages. I thought about hooking a shop vack up on a butterfly valve to speed things up but it didn't turn out to be needed. Obviously don't turn on the heat till you have sealed your part/s in.

For heat (you will laugh) but I use 16 gauge nichrome wire on a 20 amp variac. Both are cheap on ebay. Figure a hundred for a good variac and 50 for wire. I cut foam so I already had both on hand. One strand per corner about 1.5" - 2" from each wall. The wire is mounted on bolts with one end on a spring. I think they are about 6' long. I ran 10-32 by 2" screws through the tank sides with epoxy on the threads to seal them (low tech I know but cheap) to connect power to the wires. I also ran several lengths of aluminum duct work tape behind the areas with the nichrome to reflect the heat out. Don't omit this step.

I normally run the box at 250-300 degrees but it has accidentally been run over night at 400 (by the same former employee that accidentally aired it up to 50 PSI).

Hmm. Did I forget anything?

I used 3M 77 spray adhesive on the inside to get the cloth in place long enough to slop the resin on...reaching in is a drag but mounting your brush and squeegee on a broom stick (at about a 45 degree angle) makes it bearable. Its a heat box so it does not have to be sexy inside.

I spaced the screws about 2" apart on the lid. Be sure to mark your lid and top of the box with a marker for alignment. Obviously your screws go outside the gasket line (never hurts to make sure people understand =)

I never got around to insulating the outside of the box...it works fine and doesn't draw a lot of juice though. My thinking is that as rarely as I use the box it would probably take 50 times the energy to actually make the insulation then the electricity I use...

Total build time was an easy weekend which was mostly cure time.

I can air mine up to 40 PSI if I want to crush any bubbles that the vac pump hasn't popped. I can't recommend or guarantee this is safe however so proceed at your own risk. Better yet just stick to pulling the air out instead of pumping it in =)

Hope that helps. I can get pictures or a drawing if any part is unclear.