Any formulae for calculating dimensions of plywood surfaces ?

Stu
I have done my own mathematical development and it is a tedious process. Basically all I did was a long-winded method that is already contained in a development module of the 3-D software.

I did the hull in the attached photo by calculation and it is not much more than two cones attached to a central cylinder. The outriggers are fibreglass but they wre formed over stiff polyester fabric that were also developed mathematically. These are two truncated cones placed base-to-base.

Rick W

 

Yes, that is close to the way I would do it but you don't just want the length of each edge, you have to do it incrementally to get the proper shape. This can compound errors. The model is better for many reasons. I have done models even for a couple of boats where the panels were generated on software.

 

You will need to divide the surface into triangles. If you use 4-sided figures that follow the plate lines then you will need a diagonal on each one so you get the proper angle of the sides with respect to each other - so effectively triangle anyhow. The smaller the triangles the smaller the error. This is all the development software is doing anyhow.

Rick W

 

The radial line and parallel line method are more accurate than triangulation. Triangulation is just more adaptable for a computer program.

 

The canonical paper on multiconic surface development was written by S.S. Rabl in the 1950s and published in a very obscure journal. It describes the techniques I was taught in high school in the 1950s. Kaiser Aluminum reprinted it and I reprinted it again in the newsletter of the U.S. Vintage Model Yacht Group. If you PM me I'll email you a copy. Once you read it I think you'll be convinced that what your Dutchmen were doing with sticks in 3D utilizes the same geometry that Rabl does with lines on paper. It's not complex, but quite tedious.

Cheers,

Earl

 

Jim Michalak has a series of essays that I liked. Look at "Designing Boats," #6 shows triangulation for twisted panels.

http://marina.fortunecity.com/breakwater/274/michalak/alphabetical.htm

 

Every tool needs learning time, and the simpler they are the less they do.

My first hull design program was Carene2003 once distributed free by UK Epoxy Resins. It is about as easy as it can be: you just enter the major dimensions for the bottom which it then shows you, you then specify the plank angles and heights which it then shows you. A mouse click gives you the developments, and you can get stations, skin areas and weights and do simple hydrostatics.

BUT it's limited, only orthogonal views, only on or two levels of planks, no plank twisting allowed except in the stem foot, plank shapes are straight beyond the stem foot, no double-enders allowed. because it is so fast and intuitive it's the one I use whenever I can, for my current design I had a design to look at within 5 minutes. It took about a half-hour to learn.

I also use Free!Ship, it is weird, you don't have direct control over the hull dimensions other than length. I can get a basic hull shape close to my objective in about a half-hour, view it from different angles, do hydrostatics, get plank developments and try fitting them together for minimum waste, scale the hull, estimate resistance, measure metacentric height, and add buttocks, stations and waterlines. It is not at all intuitive, it took me a couple of days to figure out Free!Ship and I still can't make everything work but it is far more flexible than Carene2003.

There is still a need to invest many hours to complete a design. Free and cheap software isn't going to get out all the details for me. But as an amateur I find these simple programs invaluable and they save me hours of drawing and calculating and many dollars making models of things that are never going to work. The plank developments are a lot more accurate than you can get from a half-model.

You have to be prepared to put in the learning time. That is a problem for the professional, his time is money and it is annoying to invest time learning a program if you then discover it doesn't do the job or save time. I know, been there, done that (different industry) and was glad to retire. But once you have mastered a tool of the appropriate level it's full speed ahead.

 

The attached pdf file is the hull plate development for a yacht that took a few minutes to draw up. It is nominally 10m length with 3m beam.

The development in Delftship took about 2 minutes once I had the hull drawn, which took about 5 minutes although I fiddled with it a bit after the initial drawing.

I normally spend a bit more time aligning the plates on the sheet so I have fewer dimensions to deal with.

I used to fiddle with the maths until I got Freeship (pre Delftship). Freeship gave such a good result that I stopped wasting my time with the maths.

I agree with AK

Rick W

 

You are 'on' Rick. I will do the offsets over the next week, and get you to convert it to Delftship - and I will be happy to pay you for your time.

I got so enthused by your discussion on Delftship, I downloaded the free copy.

Then I remembered ... I have tried this all before with horrible results.

I started a new project (see attached), and fed in the three basic dimensions

Length - 8.5 metres, Beam 2.1 metres, Draft 1.5.

Then I stopped - I had been here before. Any program that asks the the DRAFT (bottom of boat to the waterline) instead of the HEIGHT of the boat is easy to remember. And sure enough, when I pressed GO, this is what I got.

All with no intervention from me (see attached.) it created this mess

No wonder I hate this ^&*&*&%^$ software.

You get a much better result if you enter the draft required ...say .3, but then why cant you specify the height (freeboard) as well. When you get to try to manipulate the lines - all hell breaks loose.

What - you want me to read the instructions !!!! - no way!

 

Ray
I will take you as far as I can at no cost. I can certainly get you to a good looking 3D image and cut files. You could make a nice scale model of the design without further engineering input.

If you have Delftship you can open the attached file and look at the hull. I have broken the hull into plates for development and each plate is on a separate layer. This allows easy manipulation of plates. I am not sure if this is the best way but it only takes a few minutes to make the separate layers from the original hull.

Your side of it will be to review what I do so you will need Delftship or Freeship to review it and modify it. I am reasonably confident I can recover anything you might mess up.

Delftship assumes there will be some of the boat above the water - not a bad assumption. Hence it makes the hull higher than the waterline.

If you put in a draft of 0.35m then you would immediately get something that could be useful. Once you play around for a while and understand how the surface modelling works it is quite powerful. It is the best 3D program I have used. Way better than Autocad for rapid rendering and I am still struggling with a couple of others. The only problem I have with Delftship is that it bogs down once you get large files.

If you get to something that you really like and want to take it to the next phase I know a naval architect who can design to standards/survey. He will take work from wherever you leave off (if he has the time). From what I have seen him do he is very proficient at finalising designs. He tries to keep it as close to your design as the method of construction will allow. He is reluctant to make suggestions that change the look unless you press him for his thoughts.

You could build it direct from Delfship but there is a significant learning step from a good 3D rendering and cut files to having plating, framing and joining details that will meet standards for survey.

Rick W

 

Unfortunately the free version of DelftShip has the plank development facility disabled, but Rick’s version is not doubt a properly licensed one. Free!Ship is almost identical and if you can find a source, it will do plank developments.

"All with no intervention from me (see attached.) it created this mess"

- actually, you were already well on the way to creating a nice looking, fair hull design, if you but knew it. I recreated your odd looking hull and got it back to something sensible, all in about 15 minutes. You made things unnecessarily difficult for yourself by entering too many control points when you started the new project; the number of control points has nothing to do with the number of stations or frames. The minimum number is 3 which concentrates the sheer plank bend at midships with sharp ends (like some kayaks), 4 will allow you to distribute the bend more evenly, approximating a circular arc. Think of how many places you would have to push on a flexible batten to get the curve you want, and that will be how many control points you will need.

However if you have the file and want to try fixing it, here are the steps to follow, and they will help you find your way around this software:

Transform, Scale, OK, Vertical Axis, insert 0.333, Transverse Axis, insert 1.5, OK.
That will give you a more realistic hull shape which of course you can try to optimize by moving the control points, but you have an awful lot to manage. Note the draft also scaled by the same amount so it is now 0.5m, and in the profile view the displacement is shown, it was about 4T.

In the bodyplan view you can close the transom as follows:
Select the aft edge with the mouse Cntrl-Rt Click, Edit, Edge, Extrude, insert 0.25 in the Transverse Direction, OK. You now have to extend the incomplete transom to the centerline: select one of the new points, enter Y-coord = 0, click on any other coord to have the software move the point. Do the same for the other points.

Now verify that you have a watertight hull; click Tools, Check Model which will display the leak points and note they are all above the waterline.

The hull is still rounded; to give it chines, Cntrl-Rt Click on all the chine lines, Display Inner Edges (this allows the program to “think” of the hull as a collection of plates rather than a single deep dish), you will see many new lines in the views, Edit, Layer, Auto Group which will separate the planks into different layers that can now be separately accessed. You will note the stations in the Body Plan View now have corners and straight edges.

At this point you would do View, Plate Developments, but in the free version you can’t. The instructions are slightly different for Free!Ship because the menus are organized differently but it is easy to make the transition once you know how to use either program.

Here's a pic of your boat so far.

Good luck! Life is about learning.

 

Well, Rick and AK, I think you have some good advice there. Last week I was caught between finishing off the scale model from the current drawings, and playing with the free version of DelftShip.

Just reading the capabilities, assigning weights, adjusting bouyancy, printing fact sheets etc - one really has to either go for it hook line and sinker, or leave it to the experienced. I was also playing around with pencil drawing of the existing sections to develop a modified hull, and realised what a lot of work has already gone into the design so far.

At this stage I will have to concentrate on the work to hand - and maybe I will get a 'second-wind' and decide to tackle the ogre of design software.

 

Daniel kindly gave me some rep points for a post made way back last year, and it led me back to this thread. Any more progress on this boat design yet?

 

I still cant make any sense of Freeship or Delftship - especially where the bow is anything but a classic style. Even the explanation of the logic in its basic methodology reads like a cryptic crossword. from Boundary Edges to Internal Edges to Control Nets, its a logicians nightmare. The help is so basic as to be confusing, with very little practical examples presented in a logical style. They need to look at Rhino or Sketchup of decent documentation.

BUT - I did find a great use for Delftship. I managed to perfect the technique of specifying the frame shapes for hard chine hulls, and then importing them to become proper hull shapes.

Once this is done, its easy to print the plans for developed panels, which was the prime aim.

I find RHINO a far better product for creating hulls - but I still have a lot of study to do to become proficient.

At this stage, I am still totally reliant on expert users, and probably willl be for the forseeabkle future.

 

couldn't a clay model be made with a straight edge and curves sculpted by pivoting the straight edge around one point?