Canoe Hull Design

I have standardised all my in.mlt files for 1 to 6 m/s and 51 steps. This means all files have 0.1m/s intervals. For higher speed vessels I use 1 to 11.

I always import results into Excel because it is easier to manipulate for comparison purposes.

Knots are very nearly twice the m/s value so a rough conversion is easy.

Rick W.

 

Rick,

Your insight is duly noted. I greatly appreciate your help on this. I'm sorry to keep nagging you with questions.

My name is Reid and I'm a CE student a the University of North Florida (UNF) in Jacksonville. This project is a function of the American Society of Civil Engineers (ASCE). They encourage students to get involved in the conferences where the competitions take place.

The two big competitions are the Steel Bridge and the Concrete Canoe. In these competitions the students are responsible for generating the money and materials to bring these projects to fruition. The canoe can especially become very time and money intensive. I expect this project to cost around $8000 and we would like to raise around $12000 to purchase equipment that would serve us better in this project. Some of the larger engineering schools spend tens of thousands on this project. However, we are a young and relatively small engineering school which requires more work from fewer students to compete at the same level.

As you have noted the Canoe race is only 25% of your overall score. However, among the CE students it carries more clout because it is really the engineering part of the project. Although, the other aspects of the project are important, the canoe and its performance draws the most attention. This mainly due to the race as it really brings out everyone’s competitive nature; everyone wants to win the race!

 

Reid
No point winning the race in the zone if you do not get enough points to make the nationals. It seems you also need a finance director in your team. The challenge as PM is to organise the team and get them motivated.

The individual responsible for the presentation should be afforded similar time and attention as the other 3 aspects - 4 or 5 counting finance and whatever other areas of specialty exist.

The muscle you have in the boat on race day and their coordination as a team will be the most important feature for getting to the finish line first. Given the length constraint they will all be hitting wave resistance. The innovative way of reducing the beam while maintaining stability is a distinct advantage but even here a strong team will outperform in a crappy hull.

I am not sure but I think the UT missed the nationals by one place.

When you are putting out a resume it will be very impressive to see that you were the PM for a successive team achieving well above your weight.

Rick W.

 

Maybe I'm missing something here, but given the fact that the canoe will be made of a thin layer of concrete, I think soft chines will result in a stronger shape -- because it's more like an egg, distributing stresses throughout the curved surface rather than concentrating them at the chines.

 

Perhaps a very stiff cradle made of plywood and timber with a custom-shaped styrofoam liner will prevent the same problem this year. Are there any restrictions on the materials used in the cradle or crate that contains the boat during transport?

Will you build the practice canoe in concrete like the final canoe?

 

Ken,

I intend to build the practice canoe out of 1/8" plywood and/or fiberglass depending on the application and the cost.

The sharp radiuses in the hard chine design will concentrate forces at the bend due to the discontinuality. This typically would be no cause for concern, but considering our construction material, it’s something to consider.

Last year we built a large coffin to store the canoe in during transportation. We tried using Styrofoam and peanuts to support the canoe within the coffin, but the coffin's volume was just to great to completely fill. During the trip to the conference the foam shifted, which exerted tensional forces in the wrong locations, and it cracked.

For this year I have two different ideas to solve this problem.

One, I have a company that is willing to donate a large foam block to us. We could then cut it down so that it fit the canoe and place the canoe within it. Two, build a second fiberglass mold off the plug, but cut the sides down some to make the canoe more accessible within the mold. Build a frame to support the mold and place the canoe in the mold.

If you guys have any ideas on how I could improve this please feel free to make any suggestions. The canoe is pretty fragile and this is a vital part of the project.

 

the plywood is way cheaper than fiberglass, or you should consider skin-on-frame. You can build a skin-on-frame canoe for about $100, it will be much lighter but you can get a pretty good shape to test real cheap and fast to build. Seach the internet for "skin-on-frame sea kayaks" and you will find lots of free plans, pictures and hobby sites.

As far a transport goes, a good stiff crate or wood base with 6 in thick sheet foam cradles (cut to the shape of the hull) about every foot or so is easy to build and will support the hull just fine.

Weight is your nemisis, keep the wieght down it will paddle faster, be easier to transport and have less chance of cracking. I would design for an optimished steel tube truss/frame, welded together. Then fine wire mesh wrapped over the steel frame with a thin a mixture of concrete as possible over it, using as many admixtures as the rules allow. Then for good measure put several coats of hypalon paint over the outside, or some similar rubbery paint for waterproofing (it is sold for waterproofing concrete and mobile home roofs).

 

Rick,

I have noticed that the resistance values from the Godzilla ship out files are different from the ship out files from Michlet on the same hull. I'm importing the surfaces into Delftship and then exporting into Michlet with out change. I have been keeping my stations to 19 and waterlines at 11. Have you ever noticed this? Which set of data do you use in your analysis?

 

I always do my own fairing and trimming in Delftship and then recheck with Michlet.

What amount of variation are you getting? Unless you are making gross changes they should be very close. Godzila is just a switch in Michlet. The actual drag calculation uses the same code. Any variation you get will be because of what Delftship is doing.

There is an offset.mlt file in Godzila that has the hull offsets in Michlet form so you could compare this with the offset table in the in.mlt file you are exporting from Delftship to see where the difference occurs.

Rick

 

It is good to have an understanding of what is going on but with a fixed maximum length and fixed minimum beam a lot of what you are trying to achieve could be described as polishing the turd.

If you are prepared to go for the FT concept with narrow central hull and stabilisers there is a range of solutions that will all have similar performance. This will be irrespective of hard chine or full semi-circular hull. At the speeds of interest it is still wave drag that becomes the constraint and the 20ft length is the control here. The FT hull will be superior to any other standard type canoe hull.

So the main hull is not going to make much difference.

You now have to get creative in how you can get the best out of stabilisers that work for both 2 and 4 man crews. You also need to keep in mind that the thing needs to be built in concrete.

By the way I am surprised that the last year's hull cracked. I saw the photos of pretensioning in the UT hull and it should have had good compressive stress to avoid any cracking. Was your 2007 hull prestressed?

Rick W.

 

Rick,

If you don't mind, would you take a look at the attached Excel spreadsheet. It has the ship out data for a few different hulls along with a Power vs. Speed graph comparing them. I must be doing something wrong in my analysis of the data because the tri-canoe is not shown to be performing as it should.

I found when exporting out of Delftship to Michlet that the displacement was changing. That was the reason for the error in the Godzilla & Michlet ship out files.

Concerning the cracking, I would say that about 50% of the canoes at the '08 conference were cracked in some fashion. Some cracks were more sever then others, but neither the less they were cracked. The concrete mix design that is used for this project is totally unorthodox. In normal concrete the aggregate is much stronger then the binder. However, in this situation we are using a higher than normal strength binder with a much weaker aggregate.

Within the next two weeks we are going to begin our testing of over 170 different binder combinations. These results combined with test data from the last three years should give us a nice array of data to work with. After we establish our binder combination we will then test it with different aggregate volumes and gradations to find the best balance between strength and density. Then we will use these results to begin flexure tests on panels. I plan to include some prestressed panels in our testing to determine the significance of its impact.

We did not use a prestressing system last year in our build. There are allot of complications you face when trying to prestress an object of this shape with only 1/4" to play with. I have seen different schools approach this in different manner. From my observation, the schools that use a prestressing system are in the minority.

 

A few things.

I believe there are 2 man and 4 man races. If so then you need to consider two load cases on one hull.

The hull I sent you was set for the two man case. If you increased draft to get to the required 4 man case then the export from Delftship to Michlet will screw up the shape. It will connect at the lowest waterline rather than actual separate hulls. You can see if this is happening by looking at the hull geometry tab in the export window.

If you are going to graph knots then use the correct conversion. What you need to know is 3.28ft/m 6080ft/nm, 3600s/hr and knot=1nm/hr.

I find the best graph option is scatter because you get to set the x-axis for each data set.

On the topic of polishing the turd you should compare the really complicated hull form attached at both load conditions against the best that you can create. The intention with a hull like this is that it lends itself to prestressed flat panels. It would need two transverse bulkheads. You could play around with the tapers to get an optimum but I am not sure how much you gain.

You could simply make up pre-stressed flat panels that are connected together and the seams grouted over. You have total control over the panel manufacture.

You also need to have some idea of the power output of the paddlers. You are, after all, using an excellent design tool with Godzilla but it is not very useful if you have no idea of the engine performance. I expect reasonably fit young males to be able to sustain 200W for a while. An Olympic paddler is more like 500W for 3 minutes or more to give you an idea of range.

The design case for 2 paddlers should be the hull weight plus 2 people at say 400W on paddles. Going to 4 paddlers you have hull plus 4 people and 800W.

Now to get more engineering focus you really need to measure the engine output and paddle efficiency. This will enable you to be designing for the right speed. Two of your hulls are good at low end say 200W but that is not where they will be when racing with 4 on board.

The value of Michlet is that it allows you to accurately predict performance taking all the hull factors into account apart from surface finish. You can try the simplest possible build shapes and see how they go. There is a lot of nonsense in boating about nice curves and good surface finish. The important things in order of significance are maximum length (set), minimum beam (set unless you are creative), crew power, weight (need light, simple, strong hull; good power to weight team), ability to apply power efficiently (stability, paddling position). Things like fairness and surface finish are debatable what is best. You might actually do best with fur attached to the skin.

You could make up a Panel boat in a few hours with some ply, screws, glue and edging timber. See how it compares with other canoes.

Rick W



Rick W

 

Rick,

As you mentioned, I now see how Delftship is cutting the tri-canoe off at the water line. Initally, I was just adjusting the displacement in the input file. However, I also noticed that after doing so Michlet was just showing the center sponson in its window and the data is incorrect. So, what is the best way in adjust the displacement when exporting out of Delftship and into Michlet?

Also, at what block coefficient do you think a canoe would attain good tracking? From what I have read I get the impression that 0.45 - 0.5 is acceptable. I know its a subjective question, but what has been your experience?

 

You set the draft in the hull parameters. Set it to give the required displacement each time you export. The file I sent you is set for 350kg I believe. This is same as your 770lb.

Have you got anything better than the Panel hull at both the lower power/weight and higher power/weight cases?

Rick

 

Rick,

I have not made it that far yet. I'm presently optmizing some of my better hulls with rocker in Godzilla to see what the efficiency costs are. I will run that for you a little later and report back the results.