rhino unroll prob's!?

Discussion in 'Software' started by tugboat, Sep 20, 2013.

  1. DCockey
    Joined: Oct 2009
    Posts: 4,497
    Likes: 237, Points: 63, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    Attached is a file with a bottom surface based on tugboat's original surfaces.

    I duplicated the edges of the original surfaces, then simplified them using Match and RebuildCrvNonUniform. A bottom surface was created using DevSrf, the pointed ends were trimmed at Isocurves. New end surfaces were created using EdgeSrf, and those surfaces and the middle section of surface were joined. Finally the bottom was unrolled using UnrollSrf with a tolerance of 0.0001.

    The curves for the upper surface need to be extended at the bow.
     

    Attached Files:

  2. DCockey
    Joined: Oct 2009
    Posts: 4,497
    Likes: 237, Points: 63, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    Don't know when service release 6 will be officially released. The candidate can be downloaded at http://www.rhino3d.com/download/rhino/5.0/commercial/rc or click on the Help tab at the top of the Rhino screen, select Service Release Candidate for frequency, and then check for updates.
     
  3. tugboat

    tugboat Previous Member

    sure- just scroll up about three posts and you can see the model posted- just use DupEdge to get the lines...!!:):)
     
  4. tugboat

    tugboat Previous Member

    Hi Mike- thanks.

    in the case of the vessel posted it is being designed in steel...you make some good points- I figured I could just template the hull for plate surfaces...or use strips in any area that might have slight compound curves...my Gaussian analysis shows pretty good curvature pretty flat and little curvature at all. Maybe im over anal-yzing it?

    but one section aft is a bit flat...but I am not even sure Im reading the graph correctly or that my settings are correct.
     
  5. tugboat

    tugboat Previous Member

    That's Great!- I am watching ufc right now- but Ill take a closer look tomorrow. Like I said I owe you one...have you built any boats you've designed? and if so what materials?
     
  6. DCockey
    Joined: Oct 2009
    Posts: 4,497
    Likes: 237, Points: 63, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    Attached file based on faired curves, top of stem moved rearward. Top surface can be improved. I may try altering the planview shape of the shear near the bow slightly. The stem is not completely fair were the sides and bottom meet. Enough for tonight.
     

    Attached Files:

  7. tugboat

    tugboat Previous Member

    if the bow section stations are arc's- doesn't that make this compound curvature? if you see- the curvature is going both towards the bow and also turning downwards?(upwards if looking at the stations right side up)?please see attached...I figured if it were a true developable hull? wouldn't they be straight line contours?
     

    Attached Files:

  8. DCockey
    Joined: Oct 2009
    Posts: 4,497
    Likes: 237, Points: 63, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    The surfaces in the file are developable without compound curvature or twist.

    It's a very common misconception that developable surfaces mean straight frames and stems, and vice-versa. Usually developable hull surfaces result in curved frames and stems, and straight frames and stems result in non-developable surfaces. The exception is if the flare is constant along the sides and the deadrise is constant along the bottom.

    Consider a truncated cone which is a developable surface. Cut sections through it and they will generally be curved. See the attached example.
     

    Attached Files:

  9. SukiSolo
    Joined: Dec 2012
    Posts: 1,270
    Likes: 25, Points: 0, Legacy Rep: 271
    Location: Hampshire UK

    SukiSolo Senior Member

    If you can get a straight line in any direction on the surface there is a chance it is developable. The rule is the straight lines (think of a straight edge or ruke) must not intersect themselves inside the area of the surface. If you imagine a whole load of straight lines radiating from the centre of a cone, that is as tight as you can get. If there is space between each 'cross linking' line along the sheet you should be able to develop it.

    Conics are developable so if you are crafty and use a part conic through to definitely a 2D curved plate it will be developable. BTW the stations (sections) are curved in the bow to give more buoyancy and straighter (straight?) waterlines. It also obviously changes the curve of areas. Very common problem with single and double chine hulls of any size.

    Rhino has the tools, if you can't build it one way you must be prepared to try 3 or 4 different methods. Other modellers are similar even if a lot more expensive. It is a learning curve all 3D modellers have to go through to get the desired results not just 'what the computer' can do.
    I have found the curves are the most important thing in Rhino, keeping these very simple and fair gives good results.

    DCockeys help to you has been excellent, thanks DCockey.
     
  10. DCockey
    Joined: Oct 2009
    Posts: 4,497
    Likes: 237, Points: 63, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    Not every surface which is made up of non-intersecting straight "ruling" lines is developable. A helix is an example of a non-developable surface made up of straight ruling lines. See the attachments for a picture of a helix with the straight ruling lines. Also included is the Rhino file if someone wants to experiment with it. UnRollSrf will unroll the helix, but the area increases by about 72%.

    The problem with the helix is it is twisted. A developable surface can not be twisted anywhere. This requirement can be described as the normal to the surface must point in the same direction everywhere along a ruling line. In physical terms it means a narrow, untwisted board will lie flat against the surface for the entire length of each ruling line.

    Twist is the usual reason a surface made by simply connecting two curves with a series of straight lines is not developable; for example surfaces made using the Straight or Normal styles in the Loft command.
     

    Attached Files:

  11. tugboat

    tugboat Previous Member

    ahh ha the light has dawned!( a line from the famous Sam Rabl postcard anecdote!)

    ok...because the plane is tangent with the ruling lines? makes sense if so.

    There is a guy on glen-l's site- he cut his forward frames im properly- I assume he saw the curved frames in his plans and thought it must have been a mistake and when he lofted the frames drew straight lines from chine to keel. when he plated it up - it left a huge gap between the plates and the frames...see here: http://boatbuilders.glen-l.com/union-jack-design/?album=139&gallery=689 Im guessing this is what you are taking about.



    - btw just for reference sake - the tugboat "Fred Murphy" a la Hankinson designed, is a chined hull with straight line contours for the frames.
    It also has a flare throughout- a flare- the only reason I didn't buy the plans.

    what if I simply used those plans but made the sides plumb simply by making a 90 degree angle between the lower frame section? Im just theorizing here that its possible. but I wouldn't do it as it increases displacement and god knows what other probs!? thanks Dcockey! you've been a great help and I'd add a million rep point if it would let me!
     
  12. tugboat

    tugboat Previous Member

    ditto for you too Sukisolo! your input has been appreciated !
    as mentioned in the above post- the Hankinson Fred Murphy has all straight lines for the frames.. its a beautiful boat too! http://www.glen-l.com/designs/hankinson/steel-shrink.html
     

    Attached Files:

  13. tugboat

    tugboat Previous Member

    so to define this: basically if you have a two directional panel- if one plane has straight ruling lines which run along the flat part of the plate, then its ok to have a continuous section running in a different direction such as would be in a conically or cylindrically developed object.?
    but the bow is an "apex" and it comes to a point - is this still conic? it is not a true cone?
     
  14. DCockey
    Joined: Oct 2009
    Posts: 4,497
    Likes: 237, Points: 63, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    Generally there is a choice - straight frames and compound curvature in the plating, or curved frames and developable plating.

    We visited a builder of steel tugs in Maine a few years ago and was told they used straight frames and compound curvature in the plating. They have a source nearby which formed the plates for them.

    At IBEX last week we went to a talk on aluminum duckboats. A naval architect designs the boats with developable surfaces and the aluminum is CNC plasma cut.
     

  15. DCockey
    Joined: Oct 2009
    Posts: 4,497
    Likes: 237, Points: 63, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    Not sure what you are describing. :confused:

    All conics are developable, but not all developable surfaces are conics. The bottom panel has pointed ends but those areas are not conic or even close to conic. The straight ruling lines run across the panels, not through the pointed ends.
     
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.