On Sun, 14 Aug 2005 18:20:15 -0500, Reynaud said:
I would like to build an replica of an old Chestnut canoe. I have the
full size boat to refer to but am not sure how to take off proper
measurements . Could someone suggest where to find information or clue me
in. Help would be much appreciated.

I've often wondered at how casually John Gardner, for instance,
mentions he "took down the lines" from some boat he'd come across,
as though it were a thing you can just do in a moment. I believe
that he could do just that, and the world is poorer for his passing.

I also would be very interested to see replies to this question.
I do suspect that watching an experienced boatbuilder do this is
probably the only way to learn, and it's an art as much as a skill.
I have seen it done once, and not done ideally at that.

A quick summary of my understanding:

1) You need a safe, level floor or other surface to work on,
and it's a bonus if you are able to scribble all over it.

2) Setting up the boat is important, but the only critical
part is that it must be level athwartships (relative to
the floor). Fore and aft trim is only important if you
want your measured waterlines to be real waterlines. Then
you have the problem of working out where the real DWL is
and probably the easiest way to do that is put it in the
water.

3) You need to make your own jigs and large squares from
scrap lumber. These do not need to be precision gear,
for the most part you can work out a way to set up
so that errors either cancel out, or are consistent
across all measurements and as such can be ignored.

4) You take down buttocks as distances up to the outside
of the hull from lines on the floor at fixed intervals,
if the boat is set up right-way-up, or else you do some
more complicated arrangement with a jig if it's upside
down. This is where drawing a grid on the floor helps.

5) You take waterlines with a jig that can be set to a fixed
height and either moved or adjusted to give the distance
from points on the hull to a reference line. Again, a floor
you can draw on is an asset. You would draw your reference
line parallel to the centreline of the boat, and just outside
the point of maximum beam, then draw a series of lines that
represent stations, at right angles to it. Your jig, then,
would resemble large calipers: a set square with an extra,
sliding arm that can be set to intervals marked on the upright.
All you'd need to do for each waterline, would be to set the
height, then move the jig from station to station with the end
of the top arm butting against the hull, quickly scribing
half-breadths away from the reference line directly on the
floor.

Like I said, I'd be interested to see discussion on this subject,
especially from the old hands. I've only seen it done once and while
obviously I've got my own ideas, this is the sort of thing where the
voice of experience is called for.

--Damian

I also would be very interested to see replies to this question.
I do suspect that watching an experienced boatbuilder do this is
probably the only way to learn, and it's an art as much as a skill.
I have seen it done once, and not done ideally at that.

Just to keep this thread alive for a while, until I finish building a
computer model of recreational rowing shell I built recently. The idea was
to use "trilateration" - a technique used by GPS. Basically it involves
measuring distances of certain points on a hull from three different
referent points. It is then possible to calculate xyz coordinates of those
measured points on the hull. My boat is a single-chine ply hull, so I just
measured keel, chine and sheer. Waterlines and stations will be created in
SW later. I'll keep you posted.

"mislav" wrote in message
...
I also would be very interested to see replies to this question.
I do suspect that watching an experienced boatbuilder do this is
probably the only way to learn, and it's an art as much as a skill.
I have seen it done once, and not done ideally at that.

I finally transfered the lines of my rowing shell into the computer and
wanted to share the experience. Like I said previously it involves
trilateration measuring technique used by GPS. I set 3 separate reference
points (put some small nails through the wooden fence in my backyard to
which I could attach the end of my 10m long tape measure). Previously I
supported the shell with sawing horses not caring to level anything. The
shell was turned upside down. One reference point was designated 0,0,0, the
other lied on y=0,z=0 axis, the third only on z=0. It was only necessary to
measure all distances between nails, in fact the sides of a triangle
(a,b,c), and then using Heron's formula find coordinates of the third point:

y|
|
|
y3 |--------P3
| a . | . b
| . | .
| . | .
P1-------|-----------------P2-----
/ x3 x2=c x
/
/
z/

P1= 0,0,0
P2= x2,0,0
P3= x3,y3,0

where y3 = (2*A)/c
and A = sqrt(S*(S-a)*(S-b)*(S-c)), S=(a+b+c)/2
then x3 = sqrt(a*a - y3*y3)

Once you have coordinates you can use
http://en.wikipedia.org/wiki/Trilateration formulas to calculate coordinates
of any point in the surrounding space by measuring distances from that point
to all three referent points. However this technique is not without it's
drawbacks. Basically any measurement error will multiply in the results
because of the second and fourth power in formulas, so you have to be very
careful.

I marked and measured around 120 points in total on keel, chine and shear. I
set up a spreadsheet from trilateration formulas that instantly calculates
coordinates from my measurement data. Since this is a plywood boat made from
panels with straight edges it is reasonable to assume that all points on a
certain edge lie in the same plane. Therefore after importing points in CAD
I draw 3 planes through all points taken from keel, chine and shear, then I
projected all points on their respective plane (projection means creating a
new point that lies on the intersection between a plane and the pependicular
that goes through the source point), then draw splines through all new
points that lie on the plane. Note that some final fairing of the individual
lines (keel, chine and sheer) was also necessary.

This technique could be used on round bottom boats too. It would only be
necessary to mark and measure keel and waterlines, since all points on the
waterline naturally lie on the same plane.

As a final note, this is not meant to be a manual on how to cheat a boat
designer by not buying the plans and build a replica of their boat. The time
and effort spent in setting all this up will probably worth more then the
money you spend on the plans, plus with the plans you'll probably get much
more useful info then just line offsets. The reason I did this is because I
wanted to investigate hydrodynamic properties of my shell with Michlet SW,
but the plans I used to build the boat didn't include any offsets.

You, in fact, did use the best way to take the lines off of a boat. I've
got several articles on the topic and some are somewhat ad hoc, others
require setting up an X-Y Cartesian coordinate system from which you take
measurements to the boat, but the best of all is the one where you do as
described ...triangles. If you know the 3 sides and define the reference,
then you can get the most accuracy out of your measurements. Boats being
what they are, any other system can result in trying to take measurements
from very low angles and that can result in a lot of error. You *still*
have to fair your results when done though.

And BTW, nothing illegal or unethical about documenting your boat
....designed by someone else or not. I would however say that the most
ethical thing to do with the lines data is that you ask the designer for
permission before you share them in any way. The lines are the designer's,
not the construction method. Using the data for your own personal analysis
of a boat is something I would highly recommend and something I've done.
But I don't share the data with anyone else without written permission.
Just my way, my 2 bits worth of a way of living and let living.

Brian D



"mislav" wrote in message
...
As a final note, this is not meant to be a manual on how to cheat a boat
designer by not buying the plans and build a replica of their boat. The
time and effort spent in setting all this up will probably worth more then
the money you spend on the plans, plus with the plans you'll probably get
much more useful info then just line offsets. The reason I did this is
because I wanted to investigate hydrodynamic properties of my shell with
Michlet SW, but the plans I used to build the boat didn't include any
offsets.