Understanding Rhino from a MultiSurf POV

Discussion in 'Software' started by Murdock86, Nov 2, 2012.

  1. Murdock86
    Joined: Oct 2012
    Posts: 11
    Likes: 2, Points: 0, Legacy Rep: 26
    Location: Netherlands

    Murdock86 Junior Member

    Hello to all boat designers on a lovely Friday afternoon!

    I am fairly familiar with AeroHydro's powerful tool Multisurf for WAMIT and have been using it for a few months now. We just got a new Rhinoceres license and my mates plan to use it to read GHS data files.

    My interest is in exploring the other (deeper) features of Rhino. Does it perform a hydrostatic analysis comparable to the level of Multisurf? Some popular opinions I get are that Multisurf is numerically more reliable than Maxsurf, Rhino or any other modelling tool out there. If that is untrue, I stand to be corrected.

    Is there anyone who has used both modelling softwares and can make a comparison of sorts? Plan to dig my teeth into Rhino over the next week and see how it fits in the 'bigger picture'! :cool: Thanks in advance, folks.
     
  2. Joe Petrich
    Joined: Jun 2008
    Posts: 165
    Likes: 10, Points: 18, Legacy Rep: 141
    Location: PNW

    Joe Petrich Designer

    Rhino alone only has rudimentary hydrostatic capabilities. There are plug-ins for Rhino, most notably Orca3d, which provide better hydrostatics and marine functionality. I'm sorry but I have no experience in the other programs you mention so I cannot give you a reliable comparison. What I can say is that our firm has used Rhino and Orca3d for years and have experienced no problems with accuracy.
     
  3. DCockey
    Joined: Oct 2009
    Posts: 5,229
    Likes: 634, Points: 113, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    Rhino uses double precision for floating point calculations which is yields around 14 significant digits for math accuracy. But many geometry calculations require iteration and the accuracy of those calculations will depend on the tolerances set in DoucmentProperties under Units. The default values are usually too coarse for me so I reset them to my preferences. If I'm working on a small boat in inches then I usually set the Absolute tolerance to .001 units. For a small boat in feet I usually use .0001 units. I use .01 or .001 percent for Relative tolerance and .1 or .01 degrees for Angle tolerance.
     
  4. DCockey
    Joined: Oct 2009
    Posts: 5,229
    Likes: 634, Points: 113, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    Claims have been made by John Letcher that the way MultiSurf works with geometry is inherently superior to "NURBS" modelers such as Rhino. For instance Letcher has argued, correctly, that the exact solution to the intersection of two NURBS surfaces in general is not a NURBS curve, and representing it by a NURBS curve is an approximation. But are the results from MultiSurf any better in reality? I have not done a direct comparison but I have gone through the MultiSurf users manual. As far as I can determine from the manual arbitrary surfaces and curves are usually modeled using a form of NURBS. The curve resulting from the intersection of two arbitrary surfaces in MultiSurf appears to be a NURBS type curve which only approximates the intersection rather than fiting it exactly. This is the same situation as in Rhino. I don't recall if the accuracy of that approximation can be set by the user in MultiSurf but it can be set by the user in Rhino.
     
  5. Murdock86
    Joined: Oct 2012
    Posts: 11
    Likes: 2, Points: 0, Legacy Rep: 26
    Location: Netherlands

    Murdock86 Junior Member

    In a broad way, one can. When solving for a MultiSurf modelled object in WAMIT, there is a choice between FAST and ACCURATE modes of evaluation. From what I understand, the accuracy is good enough to opt for a "FAST" mode of evaluation every single time.

    Thanks, David and everyone else for answers. As I explore Rhino more, hopefully, I'd be able to share some of my insights (in terms of comparison with MultiSurf and applications) with you folks on here :)
     
    2 people like this.
  6. CWTeebs
    Joined: Apr 2011
    Posts: 232
    Likes: 15, Points: 0, Legacy Rep: 171
    Location: Maine

    CWTeebs AnomalyGenerator

    Mr. Cockey and I had a robust conversation about the tenets of Rhino versus Multisurf (see link below).

    I will say that if you've mastered Multisurf it'll take you about a half hour to become a zen buddhist in Rhino.

    I'm currently working on a fatigue analysis project which requires both a solid structural FEA model and a hydrodynamic surface model. I've been having trouble creating the latter in Rhino (via lofting the outer contours) and quite literally just loaded it into Multisurf to see if I might have better luck (wish I was well versed in MSurf).

    Here's the link to the discussion I mentioned above (see especially page 2):

    http://www.boatdesign.net/forums/de...terminology-comparison-transitions-42578.html
     
    Last edited: Dec 6, 2012
  7. DCockey
    Joined: Oct 2009
    Posts: 5,229
    Likes: 634, Points: 113, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    One challenge most people face in going from one software package to another is to start "thinking" in terms of how the new software works, not just translating commands, etc between the two. While not a MultiSurf user myself my estimate is that an experienced MultiSurf user could become familar enough with Rhino in a half hour to do useful work in Rhino, but that it would take considerably longer to master Rhino. Rhino does seem to have a reputation for being relatively easy for most people to learn though.

    I'm curious about the surface you are having trouble creating in Rhino by lofting the outer contours. Can you provide any more information?
     
  8. CWTeebs
    Joined: Apr 2011
    Posts: 232
    Likes: 15, Points: 0, Legacy Rep: 171
    Location: Maine

    CWTeebs AnomalyGenerator

    In my case it has more to do with the other tools I'm using and what the surfaces are used for.

    It's a fairly large, high speed craft, the forward sections are sharp V-shaped, then there's the flat transition region, then the aft sections have the upward concavity from the shafts. I had trouble surfacing these transition regions together in a way in which the resulting geometry could be imported into ANSYS-Workbench and then meshed for hydrodynamic analysis in ANSYS-AQWA. The requirements are strict, so your incoming geometry needs to be a Catholic Saint or you don't stand a chance. I'm also severely limited by the fact that I don't know what I'm doing. A real drafter would have no problem with this.

    As I understand it, the 'meshing' phase for WAMIT is straightforward as MultiSurf creates panels for the low order problem directly (Rhino also saves in GDF format, it would be interesting to compare the two). There's also that add-on for high-order WAMIT, but I'm not sure how that works (do you purchase it as a .DLL add-on for MultiSurf or something?)
     
    1 person likes this.
  9. Ad Hoc
    Joined: Oct 2008
    Posts: 7,774
    Likes: 1,679, Points: 113, Legacy Rep: 2488
    Location: Japan

    Ad Hoc Naval Architect

    :D:D:D

    Sadly most are like this.

    As a rule, I manually mesh; takes me longer, but i trust the results more too.

    What are you have trouble with...?
     
  10. CWTeebs
    Joined: Apr 2011
    Posts: 232
    Likes: 15, Points: 0, Legacy Rep: 171
    Location: Maine

    CWTeebs AnomalyGenerator

    I do recall you saying you require your engineers to manually edit the vertices in the AQWA input files. The fluoxetine industry thanks you. Okay, I'm being silly, a bad mesh can yield catastrophic results, especially when you're dealing with very large floating structures which, when they bump into other things, have a tendency to explode.


    The only unresolved issue I have is at the transom. It changes direction in every plane, I'll probably have to mesh it manually. Or hire you to mesh it manually for me :)
     
  11. SukiSolo
    Joined: Dec 2012
    Posts: 1,269
    Likes: 27, Points: 0, Legacy Rep: 271
    Location: Hampshire UK

    SukiSolo Senior Member

    As a long time Rhino user, Pro-E aned SW too, I have had problems with Iges files from some marine hull modellers. On a 4m boat the sheerline has been up to 1.5mm out. Rhino has excellent tools for 'solving' surfaces properly. The curves are king, they are the key element to fairness if modelling inside Rhino.
    I don't know about Multisurf but have used Maxsurf. For more general modelling Rhino is far superior and surface tools will show you tangency problems with even Pro-E exported surfaces. Deviations of 0.001 to 0.0001mm if that tolerance is important to you. Personally I model at 0.0001 or 0.001mm and pretty tight angular tolerances. You can always ease off later. CNC only cuts to 0.1mm.
    Amazed if you do not have the tools in Rhino to 'fix' a file. Nearly always it's just knowing the full capabilities of the tools in it. You also have lots of other file formats - STEP (worlds 2nd best translator) etc in the package. Sometimes these are more reliable ways into other packages, if supported.

    I am suspicious of 'lofting the contours' which would seem to be inocuous.
    Have you checked the curves? - how many control points/knots etc. Have you edited them? Are they out by 0.0001mm? unfair? kinked?. Mostly a very mild edit will give you good results. Is the geometry from single or double precision software? What tolerance ar the original curves?
    Sometimes the first way you try to create something is not the right way and it is worth persevering with one or more further ways. Rhino is a very powerful tool but to get the best from it does take a little learning.
    Curves are the key in Rhino. Good curves give good geometry.
     
    1 person likes this.
  12. CWTeebs
    Joined: Apr 2011
    Posts: 232
    Likes: 15, Points: 0, Legacy Rep: 171
    Location: Maine

    CWTeebs AnomalyGenerator

    I'm not good at doing that in Rhino, any pointers?. I think manually editing the control points so they line up appropriately is the better way to handle the transition regions mentioned above. What I typically do is either connect or explode the curves and then vary the number of control points and tolerance until I get what I want.

    EDIT:
    Boiled down from SolidWorks by an associate firm. The curves are immaculate and highly professional actually.
     
  13. SukiSolo
    Joined: Dec 2012
    Posts: 1,269
    Likes: 27, Points: 0, Legacy Rep: 271
    Location: Hampshire UK

    SukiSolo Senior Member

    Normally I tend to use rebuild curve - look at the number of points and see which type - interpolate through/interpolate etc. Aim for the simplest type ie Type 3 NURB spline or even simpler, Type 2 is a true radius. Simplify to the absolute minimum the number of points. Rhino also requires a point at the end of a lofted surface to create say an end bulge. Sometimes a rail sweep is a better option than lofting. With double chined hulls I have done better with a 2 rail sweep than loft. If you rebuild a curve within less than 0.1mm it really won't affect the integrity of the model. In the past I have reduced some curves by 70% of their control points, surfaces to. Make sure you use preview and examine the variation encountered.

    It is also possible that the curves you need are not in the SW file. They may need to be extensions and then trimmed - it is very hard without seeing the geometry. They may need to be created, Project curve? curve curve intersect? Maybe some untrimmed solid shape from SW for model purposes only would help. Like tangent draft surfaces when modelling mould taper. This type of stuff is needed in Pro-E too.

    Solid Works is fine. BTW Rhino surfaces go into SW parametrically as an operation and can be worked on and deplaced as a feature in SW. Rhino has been a Gold partner to SW since the earliest days as it has always had better surface creation tools. SW is a lot better these days and Rhino has progressed too. The NURB 'engine' in Rhino is from Integrity Ware.
     
  14. DCockey
    Joined: Oct 2009
    Posts: 5,229
    Likes: 634, Points: 113, Legacy Rep: 1485
    Location: Midcoast Maine

    DCockey Senior Member

    So what did I say in this thread which annoyed someone?
     

  15. CWTeebs
    Joined: Apr 2011
    Posts: 232
    Likes: 15, Points: 0, Legacy Rep: 171
    Location: Maine

    CWTeebs AnomalyGenerator

    I personally miss the days when your follow up posts were accompanied with an offer for a no-strings-attached hug.
     
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.