how do i make curved surfaces into flat ones i can cut?

Hi there! I'm only 15 years old, so bear with me. I have made a 3d model of a small 10ft v-hull boat in AutoCAD Architectural Desktop. I've used the 'ruled surface' command to make my curved surfaces. The problem is, I can't figure out how to 'uncurve' these surfaces and draw them as flat pieces. I need to do this so I can cut these pieces, of either plywood or aluminum, accurately. Then I'll be able to bend each piece, and they can join up perfectly. I've downloaded an evaluation of Rhino. It seems to be a pretty cool, and diverse program. Will it help? Any advise or information that can help me would be greatly appreciated. I'd love to have this thing done before the end of summer! Thanx.

 

Cant help you on AutoCad (but I'm sure someone here can) but it's a piece of cake in Rhino!

This from the help file:

Have fun !

Paul

 

When you create a surface in Rhino, use the loft command, or only use other surface creating commands if you are happy that the shape they are creating can be unrolled or "developed".
Eg a cone can obvioulsy be unrolled, but a surface made with the sweep 2 rails command often cannot be. If yours cannot be developed then try tweaking some of the dinghy lines. A very sharp curve to the profile veiw at the bow is difficult to develop (looking sideways on to the boat, it should run in a reasonably gentle curve from keel to stem). Same goes for the stern though you might have a transom which makes live easier.
(I like comands like extrude -> along curve for creating transoms decks etc, and these can normally be unrolled fine).
When using the loft command for hard chine hulls the simplest way of creating them is to loft from one chine to the next, (ie for a single chine hull, when it prompts for curves to loft thru, select the sheer line and the chine line, this should give you the top panel. For the bottom panel I always split the keel line where the chine line intersects it. Then you can select the chine line and the keel line to create the bottom panel.
Make sure you click at the same ends (near the bow, or near the stern) for each line you select, or the panel will have a lovely but undesirable kink in it.
Before confirming the loft comand dialog box, be sure to select the "developable" option in the drop down box. As soon as you have created the surface, you might want to unroll it straight away to check. (surface ->unroll developable srf).
You can then export these developed panels, print them on card, and stick your model together using sticky tape.
3D cad is awesome,but theres nothing like having a real model in your hands to get a feel for the full size thing.
This method (lofting across chines) is ideal for small boats (<20ft) for its simpicity and speed, no need for cross sections to be drawn. It also seems to be pretty accurate in all the models I have made in card, and one in full size (ply).
I have also taken photos of real boats, from the side, end and plan views, and used the jpg as a background image in rhino, to trace the lines manually. Lofting these surfaces and building card or thin ply model has given me a very close replica of the real thing.
Rhino is the dogs knackers, i've only just started using it, so Im sure there are loads of other tips I dont know, and hopefully others will add to my points. One last tip though, when creating curves in Rhino (like chines, and sheers) use as few a control points as possible (control points are the ones that dont sit on the curve - rather than edit points which do -I never use them). You should only need 4 maximum (one at either end and two in the middle) for a little dinghy, and if you can get away with three, so much the better.

 

I have done some Rhino modelling for a flat panel construction design. A lot of the surfaces are cylindrical and very easy to do with a single rail sweep or extrusion.

Near the bow I wanted to replicate the manual drafting technique called conical projection. I did this by a two rail sweep but with one of the rails extremely short. This gave me the shape I wanted. I checked afterwafds on the computer (guassian curvature) and it was developable. Finally I unrolled the surface and built my model. All OK!

Have fun...........

____________________
Brian

 

Big hammer. Beat it flat.

 

There is an old program called "Plyboats" available from Ray Clark Enterprises that specializes in designing boats from sheet material. It is dirt cheap (less than fifty bucks) and very easy to use. The manual is packed with usefull design information that will enable you to pick apropriate sail areas etc. The program also makes it easy to design sailing rigs and will do the sail area and rudder and centerboard area calculations for you. Also it will tell you the stability of your boat design at varying angles of heel. There is a modeling utility that makes it easy to create cardboard models of your design. This program will show you the "unrolled" surfaces superimposed on standard 4x8 sheets of plywood as a quick check on material utilization. You can also print out offsets of the panels and use these to lay out the full sized panels. You are limited to a sheer line, a upper chine, a lower chine and a horizontal keel panel. This gives you three surfaces a side plus the keel. I have designed and built kayaks and dinghies with this program. Google "Plyboats" and "Ray Clark" and you will find it. Boat design programs that handle developable surfaces generaly cost several hundred dollars and on up.

 

Back to the drawing board!!!!

This is a nice one!
Clearly everybody knows how to use the computer programs, but where is the basic knowledge from the drawingboard?
Provided the surfaces are of the "developable" type (so, curved in only one direction) you can draw the 'uncurved' surface by determining the 'true length'. This is the technique that was uses in the pre-computer type when boats were drawn by hand.

Basiccally, you draw a waterline (name: x) on the plate, perpendicular to that, equally spaced, you draw lines (name: y) on the surface of the plate.
Now you measure the length of the x-line between the y-lines (name: z). And you put all these lengths in a straight line. All these lengths are separated by the lengths y, with their own intersection point.
When connecting the ends of all the y-lines, you have your developed plate!

See the sketch, that will make things a bit more clear!



BTW: surfaces that are not of the 'developable type' can also be developped with this technique, but is more difficult to do!

 

AutoCAD ruled surfaces are not necessarily developable, they may be warped.

However:

1) copy the ruled surface somewhere else or in another drawing (block it out) and explode it.
2) set a ucs with the X axis down the first ruling in the surface and the next ruling in Y using UCS 3P
3) rotate the UCS in Y 90. This makes a UCS with Z down the ruling.
4) rotate the remainder of the surface about 0,0 so the next ruling is in the XY plane.
5) go to step 2 for the next ruling and repeat until you have rotated the whole surface flat about each ruling.

You may be able to find ALISP routines or whatever to do this. Look on the Baseline site.

 

Shoot me an e-mail. info@jemwatercraft.com . You could send me the file and I'll import it and help you with the panels.

 

CD Barry...

CD - same from boatbuilder.com ?

If so - well - hello again and thanks for all your help in the past - I decided not to go to Florida to Regal (or other - cannot remember) but that is life...

Hope all is well...

I am fine - see ABD - http://www.abdaluminumyachts.com/ ...

As for the kid - 15 year old - glad to see in AC but - perhaps try www.carlsondesign.com - DL his hulls program and muck about...

If you transfer files from proggie to proggie it will never be what you envisioned - what you know you can do with computers and software may not be within your financial reach...

My advice to you - 'kid' - goto a local machine shop or college or - 'big hope' - a 5 axis milling shop and just tell them what you have told us - most shop guys will be happy to show you how to translate what you have into usable data that their machines can use - IGES - Mastercam Mill - etc...

Also Email me if you want to pursue Rhino 1.0...

Offer liqour and cigarettes..


SH.

 

Like Nathanael Herreshof (it seems that he was an able naval architect...)and nowadays Nigel Irens (same comment but at present...), I make a scaled whole or half hull.

Yes in the years of computer 3D modelling it seems retrograde, stupid etc. But truly I do not want to spend months learning a complex 3D program, which not always guarantee that the surface is a ruled one (it seems that lot of 3D programs have difficulties with twisted surfaces) or obliges me to use always NURBS surfaces. I prefer to refine my calculations, not to read hundred pages manuals.

I do love computers as fantastic tools but I'm not slave of them. I prefer building.

With some birch aviation GLII plywood, wood or aluminium and cyanoacrylate glue you'll see in a few hours if your hull is a piece of junk or a nice one, and you'll have your panels shape. After, knowing the true shape of your panel and having a "crude offset", you can sweat on the PC and get the panels with a critic eye (and checking it) at the millimeter for a CNC cutting if you can afford it.

The Plywood software works very well within its range of use. An important item, it uses splines and simulates in a kind of way the behaviour of plywood. It's worth the price.

There are also old geometrical methods, not always easy to use.

A lot of boat plans designed by computer show curves that is impossible to make in the plywood or metal of the intented thickness, mainly at the bow area. NURBS are not often compatible with the "pressure curves" or "torque applied bending curves" taken by a plywood panel (or metal) and you may fight during hours because the does not want to lay evenly on the deck beams for example.

With thin softwood plywood it's possible to get an acceptable result torturing it. With thick hardwood plywood like a 30 mm doucier or a thick steel plate it's impossible. You'll spend hours to solve the problem and your benefits are gone with the wind.

Some designers should spend a few days in a shipyard trying to resolve the practical problems they raised. A lot also should learn welding.

 

Yipee - ILAN...

ILAN...

I agree - totally - I weld up 142 foot aluminum beasts - got kids forming panels from 'explosions' from the damned PC's - slavish - with no common sense to see if IT JUST DAMNED FITS...

It is a good base - and great for the designers - and not to slag them at all - but representing 3D on a 2 dimensional CRT screen does not supply me with much faith while 'on the floor'...

I got this new kid who laid on a clear vellum sheet to my frames - ON THE BOAT BEING BUILT - he traced off the panel - cut it - beat it to curves AND bevelled all the edges for my weld tacks - I am putting in for his pre probation raise and MORE - like I WANT him assisting me directly - rather than the 'street or college kids' who have no 'gump' or common sense or sense of form - I (and him) can go home early on fridays - hehe...

I think the trouble with people who have not seen the shop floor fitting is that they believe the software will solve all their problems rather than providing them with a starting point - but hell - I am just a Sean...

What you do is NOT retrograde but common sense and common practice on any shop floor - for Gods sake - most of what is out there is built on Microsoft 'pick and poin't development' crap software - VERY SCARY...

SH

 

Is it just me or did this thread take a sudden right turn? I thought it was just a guy looking to build a boat for himself. Or am I missing something?

 

...and he wanted it to be done before the end of summer 2002!
That does NOT mean that I think you shouldn't post whatever relevant comments you have, but the guy may have turned on to other things by now. Maybe girls - they are just as expensive and (potentially) dangerous as boats, but boats don't talk back at you!

 

lol....WHOOPS!

I didn't notice the original post date. I'm going to get the shoe taste out of my mouth now!!!