2d pattern to 3d hull

Discussion in 'Software' started by Alball, Sep 7, 2014.

  1. rwatson
    Joined: Aug 2007
    Posts: 6,163
    Likes: 495, Points: 83, Legacy Rep: 1749
    Location: Tasmania,Australia

    rwatson Senior Member

    You dont have to worry about that any more than you do the turn of the sides to a pointy bow.

    Once you have the exact dimensions of both "Rhomboids", the lengths of the smaller sides direct the Z axis ( up and down) as well as the other two axis. They form a non-ambiguous triangulation on all axis.

    If you have ever 'stitched' together a set of plywood panels, and seen them pull themselves in the right axis, that is a similar process, but with much smaller panels
     
  2. DCockey
    Joined: Oct 2009
    Posts: 5,229
    Likes: 634, Points: 113, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    rwatson, have you tried the method you described or is it "theory"?
     
  3. rwatson
    Joined: Aug 2007
    Posts: 6,163
    Likes: 495, Points: 83, Legacy Rep: 1749
    Location: Tasmania,Australia

    rwatson Senior Member

    The Recreate Flat Panels to 3D method in Rhino

    I have tried it to the point that I am satisfied it will work.

    The illustrations show the projection done to the second segment of the floor and side panel from the transom.

    In the first experiments, I found manually matching the "meeting points" a great problem.

    So, following in on Tansl's comments, " spherical trigonometry, not flat, to be accurate.", I ended up creating transparent Spheres that were an exact match to the three "legs".

    Then, by creating intersecting arcs, I was able to triangulate the "reference point".

    edit: Each of these "Reference Points" is the Point in Space where both the Side and Floor Segments will meet on their common edge, when their long edges are placed against the "known" edges for the preceding object. The "Reference Point" Location is the Diagonal of the two segments (red and yellow) , and the length of the common short segment edge ( green).

    The accuracy of the orientation is perfect. I was very surprised when I "Cut" the first Arc created by the Intersect2Sets, and it did it without protest. I fully expected it to come up with an error.


    I have put some illustrations below to try and explain.

    I have also attached the latest version of the drawing in case anyone else wants to carry them on.
     

    Attached Files:

    Last edited: Sep 10, 2014
  4. TANSL
    Joined: Sep 2011
    Posts: 7,369
    Likes: 699, Points: 123, Legacy Rep: 300
    Location: Spain

    TANSL Senior Member

    rwatson, seems a great job but my limitations with the English prevented me to follow the reasoning that you expose. However my logic leads me to believe that this is not possible, in practical terms. If we would know the true position of the transom, that would be simple, but if you do not know the transom´s angle with the vertical plane, for example, I don´t see a way to place in the space the starting point you take as the center of the spheres. On the other hand, you're assuming that the various sections of equal length you take on the edges of the plates are straight, which is not correct, unless these sections are very, very small. It thus leads us to a process of testing / bug with lots of points with little accuracy that, in my opinion, makes it impracticable.
     
  5. rwatson
    Joined: Aug 2007
    Posts: 6,163
    Likes: 495, Points: 83, Legacy Rep: 1749
    Location: Tasmania,Australia

    rwatson Senior Member

    Its true that the process of 'slicing up' the flat sections will produce an imperfect curve along the chines and gunnels, but these are easy to correct. You could go to very small lengths of sections, to make the steps even smaller, while making a lot of work.

    But as far as assembling the boat, out of small sections, its as logical as physically building it this way..

    Let me see if I can explain the process better - not just for your English, but also for the sake of a simpler explanation for any new reader.

    The 'true' position of the transom is irrelevant. It could be lying on the floor, nailed to the wall, or hanging from the ceiling.

    The problem starts as say, the first 2 centimeters of the floor and side of the flat panels. ( see the Green and Brown sections in the attached illustration) - and how they fit to the transom.

    The REAL Problem, is how can we place the 2 centimetre section accurately in RELATION TO the transom.

    2/3rds of the problem is solved. The stern most section of the panels MUST fit the transom. There is no problem there. I show them sticky taped as Green and Brown, but flapping loosely in space.

    They they can flap around at will, but we MUST get the Blue Dot and the Red Dot on the panels to meet.

    They meet at the Orange dot.

    The ORANGE DOT is the ONE Point possible for them to meet. No other place will do.

    I have shown wire frames where the two panels must ultimately meet.

    This is the only place where they will fit IN SPACE in relation to the transom ( no matter where the transom is located )

    Problem solved - as long as the slices of the panels are the same proportion to each other.

    I hope that helps
     

    Attached Files:

  6. TANSL
    Joined: Sep 2011
    Posts: 7,369
    Likes: 699, Points: 123, Legacy Rep: 300
    Location: Spain

    TANSL Senior Member

    Sorry I can not express myself properly, so I will not try to do it because it might get to say things totally incomprehensible.
    I understand perfectly what you want to do and think it's not accurate. That's why I always talk about a trial / error procedure that can consume anyone's patience.
    I think that even if your procedure is correct, in practice it can not be carried out.
    Good job, rwatson, imo and thanks for sharing your experiences and procedures on how to work in 3D.
     
  7. rwatson
    Joined: Aug 2007
    Posts: 6,163
    Likes: 495, Points: 83, Legacy Rep: 1749
    Location: Tasmania,Australia

    rwatson Senior Member

    Never one to ignore Tansls "intuition", I added a few more "segments", and superimposed them over the original hull.

    The keel line has "sagged", if you can see that in the illustration,

    Because the orginal hull isnt guaranteed Developable, maybe the Unfolded Panel is deformed ?

    I would have to cut out the patterns and build them to see where the error lay.

    In the process, I speeded the process up a lot - using the Copy Option with Orient3Point - it made the job quicker
     

    Attached Files:

  8. Alball
    Joined: Sep 2014
    Posts: 7
    Likes: 0, Points: 0, Legacy Rep: 10
    Location: Somerville, ma

    Alball Junior Member

    Op here. I have been way from internet for a few days but just caught up with thread. This is great stuff you are showing, I have been thinking along the same lines but have not had a chance to try it in rhino. I will try tonight.

    It seems a bit intensive, hence my original inquiry about software that could do it automagically.

    This approach is a sound alternative.

    Al
     

  9. rwatson
    Joined: Aug 2007
    Posts: 6,163
    Likes: 495, Points: 83, Legacy Rep: 1749
    Location: Tasmania,Australia

    rwatson Senior Member

    Discovery time again.

    I think I have discovered the reason why the keel line is 'drooping'.

    I did a few segments on the right hand side of the hull, and you can see a gap along the keel developing.

    The problem is that the segments arnt being Oriented to the centreline.

    You would have to draw a vertical surface along the centreline, and distort each segment to touch the vertical surface.

    If you have a look at the Zebra stripe of the bottom, you can see quite a bit of curvature - which is causing the problem.

    There is no such thing as 4 way best fit - so it would mean tricky panel bending stuff.

    Time for bed now - I might be able to figure out a solution.
     

    Attached Files:

Loading...
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.