Has anyone sucessfully built a boat out of unfolded panels from this hard
chine design program?

I set up stations and made a cardboard scale model of a boat by wrapping
the cardboard around some frames, marking, unfolding, and cutting. Then I
put the offsets into the Carlson program and used the "Patterns/Nesting"
feature to arrange the panels on sheets of plywood and print out points
for hand plotting. I plotted and cut the panels from cardboard, same scale
as the model, and taped the cutouts togehter sticth-and-tape style. The
result isn't the same as the model. There is a big gap at the stem, the
topside panels are 25% wider, and the it just doesn't fit the frames.

(I've been back seeing if I can alter the offests to get a better fit and
find the auto spline makign S-curves in the keel at the stem. Very strange.)

Am wondering if others have got good unfolded panels from the program.

(The boat is the 15ft solo cruiser design I've been documenting under
"Boats" on my website.)


--
------------------------------------------------------------------------------
William R Watt National Capital FreeNet Ottawa's free community network
homepage: www.ncf.ca/~ag384/top.htm
warning: non-freenet email must have "notspam" in subject or it's returned

I have not built a boat from this program but I did build a small model
once. I printed the expanded panels on my printer onto stiff paper and
cut them out and taped them together and everything seemed to fit
together pretty much the way it should have.
Never built anything from the offsets table though.
I have noticed some weirdness up by the bow when fooling around with
this program.
Are you sure you locked in the position of your bulkheads before doing
the patterns/nesting? Numbers semed to shift around otherwise.
Let's face it, it almost works and the price is right but I never could
get beyond the model stage though. But I think it could be done.

William R. Watt wrote:
Has anyone sucessfully built a boat out of unfolded panels from this hard
chine design program?

I set up stations and made a cardboard scale model of a boat by wrapping
the cardboard around some frames, marking, unfolding, and cutting. Then I
put the offsets into the Carlson program and used the "Patterns/Nesting"
feature to arrange the panels on sheets of plywood and print out points
for hand plotting. I plotted and cut the panels from cardboard, same scale
as the model, and taped the cutouts togehter sticth-and-tape style. The
result isn't the same as the model. There is a big gap at the stem, the
topside panels are 25% wider, and the it just doesn't fit the frames.

(I've been back seeing if I can alter the offests to get a better fit and
find the auto spline makign S-curves in the keel at the stem. Very strange.)

Am wondering if others have got good unfolded panels from the program.

(The boat is the 15ft solo cruiser design I've been documenting under
"Boats" on my website.)


--
------------------------------------------------------------------------------
William R Watt National Capital FreeNet Ottawa's free community network
homepage: www.ncf.ca/~ag384/top.htm
warning: non-freenet email must have "notspam" in subject or it's returned

Keep in mind that shell plate expansion (what you are doing by hand and with
software) is one of the more challenging parts of a hull design program.
Even programs produced for more professional work, such as Rhino 3D and
ProSurf, do not do a perfect job until you learn the ins and outs and tricks
of the trade to make it work right ...a key one being tolerance management.
It's very easy to create an issue with tolerance stacking, especially in an
iterative calculation like what shell plate expansion uses. You can nearly
always tell which designers actually built the boat they sell plans to or
not by how large the errors are in the panels. I've heard of errors as
large as 5" in a 20' boat for example. Another key is management of curve
complexity. In a developable panel, this primarily refers to the
combination of rate of change of curvature and also the tightness (radius)
of the curves. To be accurate in such areas, the triangulation (what the
software is doing) either has to be very tight across the board or vary as
it goes. You'll find that every program is 'pretty good' to a point, then
once beyond that particular constraint, the accuracy drops off. Try
designing a boat with more gentle curves and see how it works out. If the
software allows you to define a measurement tolerance, then lean towards
making it tighter, not looser. You can loosen the specs after you have a
finished panel that works, but don't do it in the calculation stage (kind of
like not rounding off in precision until you report the final answer with
the right number of significant digits.)

So, the bottom line is: take heart, your experience is not out of the
ordinary. Look into the settings that Carlson makes available and continue
to try different approaches until it all comes together.

Brian

"William R. Watt" wrote in message
...
Has anyone sucessfully built a boat out of unfolded panels from this hard
chine design program?

I set up stations and made a cardboard scale model of a boat by wrapping
the cardboard around some frames, marking, unfolding, and cutting. Then I
put the offsets into the Carlson program and used the "Patterns/Nesting"
feature to arrange the panels on sheets of plywood and print out points
for hand plotting. I plotted and cut the panels from cardboard, same scale
as the model, and taped the cutouts togehter sticth-and-tape style. The
result isn't the same as the model. There is a big gap at the stem, the
topside panels are 25% wider, and the it just doesn't fit the frames.

(I've been back seeing if I can alter the offests to get a better fit and
find the auto spline makign S-curves in the keel at the stem. Very
strange.)

Am wondering if others have got good unfolded panels from the program.

(The boat is the 15ft solo cruiser design I've been documenting under
"Boats" on my website.)


--
--------------------------------------------------------------------------
----
William R Watt National Capital FreeNet Ottawa's free community
network
homepage: www.ncf.ca/~ag384/top.htm
warning: non-freenet email must have "notspam" in subject or it's returned

I have used this for a couple of years for my students in an Intro to Nav
Arch class. They use it to develop a 3-view (plan, profile, body plan)
drawing, as well as small models. My observation is that the unwrapping is
based on a geodesic approach (i.e., attached triangles), not a developable
surface (i.e., unwrapped conic sections). As you get more curvature, the
geodesic can underpredict the real unwrapped length of the surface. Having
said that, I've never seen more than a small gap at the stem. We usually
plot these via the DXF - have you tried comparing plots from the nesting and
DXF through a CAD program? The DXF/CAD approach has the benefit of showing
the BHEAD locations on the CHINE and DECK plots. This is very valuable with
the proper alignment of pieces.

One other FYI - the DXF format exported by this program is not compatible
with all CAD programs. I've had good success with TurboCAD, Rhino and the
Voloview Express viewer. I'm sure there are others.


Regards,

Don

Donald M. MacPherson
VP Technical Director
HydroComp, Inc.
http://www.hydrocompinc.com

[2004 Propeller Seminar - January 16th in Tampa, Florida.]


"William R. Watt" wrote in message
...
Has anyone sucessfully built a boat out of unfolded panels from this hard
chine design program?

I set up stations and made a cardboard scale model of a boat by wrapping
the cardboard around some frames, marking, unfolding, and cutting. Then I
put the offsets into the Carlson program and used the "Patterns/Nesting"
feature to arrange the panels on sheets of plywood and print out points
for hand plotting. I plotted and cut the panels from cardboard, same scale
as the model, and taped the cutouts togehter sticth-and-tape style. The
result isn't the same as the model. There is a big gap at the stem, the
topside panels are 25% wider, and the it just doesn't fit the frames.

(I've been back seeing if I can alter the offests to get a better fit and
find the auto spline makign S-curves in the keel at the stem. Very
strange.)

Am wondering if others have got good unfolded panels from the program.

(The boat is the 15ft solo cruiser design I've been documenting under
"Boats" on my website.)


--
--------------------------------------------------------------------------
----
William R Watt National Capital FreeNet Ottawa's free community
network
homepage: www.ncf.ca/~ag384/top.htm
warning: non-freenet email must have "notspam" in subject or it's returned

"Brian D" wrote in message news:m9NAb.460002$Fm2.453789@attbi_s04...
Keep in mind that shell plate expansion (what you are doing by hand and with
software) is one of the more challenging parts of a hull design program.
Even programs produced for more professional work, such as Rhino 3D and
ProSurf, do not do a perfect job until you learn the ins and outs and tricks
of the trade to make it work right ...a key one being tolerance management.
It's very easy to create an issue with tolerance stacking, especially in an
iterative calculation like what shell plate expansion uses. You can nearly
always tell which designers actually built the boat they sell plans to or
not by how large the errors are in the panels. I've heard of errors as
large as 5" in a 20' boat for example. Another key is management of curve
complexity. In a developable panel, this primarily refers to the
combination of rate of change of curvature and also the tightness (radius)
of the curves. To be accurate in such areas, the triangulation (what the
software is doing) either has to be very tight across the board or vary as
it goes. You'll find that every program is 'pretty good' to a point, then
once beyond that particular constraint, the accuracy drops off. Try
designing a boat with more gentle curves and see how it works out. If the
software allows you to define a measurement tolerance, then lean towards
making it tighter, not looser. You can loosen the specs after you have a
finished panel that works, but don't do it in the calculation stage (kind of
like not rounding off in precision until you report the final answer with
the right number of significant digits.)

So, the bottom line is: take heart, your experience is not out of the
ordinary. Look into the settings that Carlson makes available and continue
to try different approaches until it all comes together.

Brian

"William R. Watt" wrote in message
...
Has anyone sucessfully built a boat out of unfolded panels from this hard
chine design program?

I just got my design for a 20 footer I am building back from the
engineer (who I had go over the design a one time). I designed the
boat in Carlson and was able to shape the bulkheads there. I was
thinking about expanding the panels out and building that way but the
more I read, the more I think I might just get them out the old
fashioned way.
Scotty from SmallBoats.com

"D MacPherson" ) writes:

..My observation is that the unwrapping is
based on a geodesic approach (i.e., attached triangles), not a developable
surface (i.e., unwrapped conic sections).

yes, I made a silly mistake in my first attempt at this boat of having the
lowest point midships adn the widest point somewhat further aft, a shape
to which plywood would not confrom. Neither the BluePeter nor the Carlson
program complained. I knew better but was not thinking. So then I made the
carboard model the old fashioned way to check before keying the offsets
into the two programs to do the calculations and get the unfolded plotting
points for the panels.

I know designers use computers to calculate the shape of unfolded plywood
panels and transfer the points to a computer controlled cutting board.
I've seen advertisments on the Internet from companies like Chesepeak(?)
Ligth Craft who sell kayak kits made this way. Since Greg Carlson sells
cutter/plotters and his program produces a file for his cutter/plotters
then I figure there should be some way I can get the program to produce
accurate plotting points.


Having
said that, I've never seen more than a small gap at the stem.

I've given myself the challenge of attempting a constant bevel which is
making the stem a bit tricky on the small scale drawing on the computer
screen.

We usually
plot these via the DXF - have you tried comparing plots from the nesting and
DXF through a CAD program?

oh no, I have to learn how to use anoother computer program?

BHW: I've found it easier to use the Patterns/Nesting output because all
the files have negative values for plotting points which I haven't been
able to figure out.

One other FYI - the DXF format exported by this program is not compatible
with all CAD programs. I've had good success with TurboCAD, Rhino and the
Voloview Express viewer. I'm sure there are others.

I can display the DXF images with the program that came with the flat bed
scanner I use. They look okay, just like the images displayed by the
program itself. Havent' figured out yet how those images relate to the
problems I've had with the plotting points.

I tried again last night producign a new file of plotting points, drawign
and cutting the panels, and taping them together. I still have more work
to do on this. I could use teh old fashioned way but am determined to
learn how to do it on the computer.

BTW2: I make the panels pretty quickly by taping 2 letter sized sheets of
paper together which gives 16" for the 16' of two sheets of plywood,
plotting and joing the points (straight lines with a ruler is okay for
this), putting duct tape over the back of ach panel outline for some
stiffness, then cutting out the panel with scissors, and taping
teh panels together on the duct tape side with small pieces of cello tape.
The duct tape makes it easier to move the cello tape, no tearing of paper.

Thanks to everyone for the advice.
I'll keep at it.
--
------------------------------------------------------------------------------
William R Watt National Capital FreeNet Ottawa's free community network
homepage: www.ncf.ca/~ag384/top.htm
warning: non-freenet email must have "notspam" in subject or it's returned

Scotty,

Don't be scared off. Buy some cheap 1/8" door skin and build a 1/4 scale
model. Use duct tape as your 'adhesive'. You'll find most errors right
off. Have some fun...

Brian

"Backyard Renegade" wrote in message
m...
"Brian D" wrote in message
news:m9NAb.460002$Fm2.453789@attbi_s04...
Keep in mind that shell plate expansion (what you are doing by hand and
with
software) is one of the more challenging parts of a hull design program.
Even programs produced for more professional work, such as Rhino 3D and
ProSurf, do not do a perfect job until you learn the ins and outs and
tricks
of the trade to make it work right ...a key one being tolerance
management.
It's very easy to create an issue with tolerance stacking, especially in
an
iterative calculation like what shell plate expansion uses. You can
nearly
always tell which designers actually built the boat they sell plans to
or
not by how large the errors are in the panels. I've heard of errors as
large as 5" in a 20' boat for example. Another key is management of
curve
complexity. In a developable panel, this primarily refers to the
combination of rate of change of curvature and also the tightness
(radius)
of the curves. To be accurate in such areas, the triangulation (what
the
software is doing) either has to be very tight across the board or vary
as
it goes. You'll find that every program is 'pretty good' to a point,
then
once beyond that particular constraint, the accuracy drops off. Try
designing a boat with more gentle curves and see how it works out. If
the
software allows you to define a measurement tolerance, then lean towards
making it tighter, not looser. You can loosen the specs after you have
a
finished panel that works, but don't do it in the calculation stage
(kind of
like not rounding off in precision until you report the final answer
with
the right number of significant digits.)

So, the bottom line is: take heart, your experience is not out of the
ordinary. Look into the settings that Carlson makes available and
continue
to try different approaches until it all comes together.

Brian

"William R. Watt" wrote in message
...
Has anyone sucessfully built a boat out of unfolded panels from this
hard
chine design program?

I just got my design for a 20 footer I am building back from the
engineer (who I had go over the design a one time). I designed the
boat in Carlson and was able to shape the bulkheads there. I was
thinking about expanding the panels out and building that way but the
more I read, the more I think I might just get them out the old
fashioned way.
Scotty from SmallBoats.com

Try printing onto trimmed manila folders. They give enough stiffness and
bend nicely. It helps to have a straight-feed printer, though. (The students
use a Laserjet for this.)

Regards,

Don

Donald M. MacPherson
VP Technical Director
HydroComp, Inc.
http://www.hydrocompinc.com
tel (603)868-3344
fax (603)868-3366

2004 Propeller Seminar - January 16th in Tampa, Florida.
http://www.hydrocompinc.com/support/...lerSeminar.htm



"William R. Watt" wrote in message
...
"D MacPherson" ) writes:

snip...


BTW2: I make the panels pretty quickly by taping 2 letter sized sheets of
paper together which gives 16" for the 16' of two sheets of plywood

The Carlson program develops panels correctly but that doesn't mean that
those panels are developable.
Let's explain: the development method is simple, the program divides a panel
in a bunch of parallelograms and then unroll them BUT that doesn't mean that
the surface is developable.
A developable surface must fulfill some conditions: a cone is developable, a
panel made of cones is developable but the Carlson program does not check
that.
There is another program with that flaw, Plyboats.

Good programs like Rhino, Prolines, the old Nautilus and many others will
create a developable surface that fills some conditions and can be developed
within certain limits. All the ones I know are based on the Kilgore
algorithm. Through an iteration process they check for ruling lines:
straight lines that are included on that surface and run from one edge to
the other without intersecting. There is more to it but that is the basic
problem. You must create a developable surface first. It is a much more
complicated task than to develop the panel.
A test is to design a hull with some nice curvature at the bow then cut
stations through it. If the sections close to the bow show stations with
straight sides, then the program does not do a proper job because that part
of the panels should be generated by cones.
Over the years I wrote about that in this group several times: the Carlson
program is good if you start with a hull that is developable. It is a valid
tool to scale up and down an existing boat, create patterns etc. It can even
be used to design a very simple developable hull like one with cylindrical
panels, "a la Bolger", all station sides parallel.
Who said you get what you pay for?

--
Jacques
http://www.bateau.com



"William R. Watt" wrote in message
...
Has anyone sucessfully built a boat out of unfolded panels from this hard
chine design program?

I set up stations and made a cardboard scale model of a boat by wrapping
the cardboard around some frames, marking, unfolding, and cutting. Then I
put the offsets into the Carlson program and used the "Patterns/Nesting"
feature to arrange the panels on sheets of plywood and print out points
for hand plotting. I plotted and cut the panels from cardboard, same scale
as the model, and taped the cutouts togehter sticth-and-tape style. The
result isn't the same as the model. There is a big gap at the stem, the
topside panels are 25% wider, and the it just doesn't fit the frames.

(I've been back seeing if I can alter the offests to get a better fit and
find the auto spline makign S-curves in the keel at the stem. Very
strange.)

Am wondering if others have got good unfolded panels from the program.

(The boat is the 15ft solo cruiser design I've been documenting under
"Boats" on my website.)


--
--------------------------------------------------------------------------
----
William R Watt National Capital FreeNet Ottawa's free community
network
homepage: www.ncf.ca/~ag384/top.htm
warning: non-freenet email must have "notspam" in subject or it's returned

"Jacques Mertens" ) writes:
The Carlson program develops panels correctly but that doesn't mean that
those panels are developable.

I've seen the paper and pencil method of looking for the apex of curvature
in a couple of boatbuilding books, TF Jones for example. For the design
I'm working with constant bevels restrict the bend to one plane. As you
mentioned below I eyeball the pictures displayed on the computer screen to
make sure the station lines are parallel (no twist). Avoiding twist in the
panels at the stem shortens the waterline length but that's a (small)
price I pay for simplicity in design and construction. All I have to worry
about is that the radius of curvature is not to tight for the thickness of
the plywood to bend to. Admittedly I haven't done any radius calcutations.
With a bow half angle of 32 deg I think its okay. One more thing to put on
the "todo" list.

All the ones I know are based on the Kilgore
algorithm. Through an iteration process they check for ruling lines:
straight lines that are included on that surface and run from one edge to
the other without intersecting. There is more to it but that is the basic
problem.

Thanks. I've written down the name and will look for more info as I would
like to learn how it's done.

Over the years I wrote about that in this group several times: the Carlson
program is good if you start with a hull that is developable. It is a valid
tool to scale up and down an existing boat, create patterns etc.

Yes, all I want from the Carlson program are the plotting points for the
plywood panels on the sheets of plywood. I used the Blue Peter program
initially to flesh out the hull with stations (the program rounds the
chines as you ask it to insert more stations) and to do the design
calculations, then printed out the offsets and typed them into the Carslon
program. I naively expected to get accurate plotting points with a couple
of clicks of the mouse.


--
------------------------------------------------------------------------------
William R Watt National Capital FreeNet Ottawa's free community network
homepage: www.ncf.ca/~ag384/top.htm
warning: non-freenet email must have "notspam" in subject or it's returned