Community
Please share your model. I can already tell you that you are using the wrong approach to model that boat.
37 profiles will never connect smoothly with a loft!!!
I am assuming you imported the sketch geometry, because the splines/curves don't have any of the usual controls (control points or spline handles) . Geometry imported from vector graphics applications causes a lot of headaches when working with them it in Fusion 360.
The curvature of these curves is also not exactly smooth. Here's only one example:
No Peter, unfortunately all I got was the mesh, no sketches or anything like that. I converted the mesh to solid, then split the bodies to get these somewhat angular station sketches. I've done that in a separate file as it creaks and groans it's way through operations due to the very large numbers of triangles. Then copied the sketches across into this file, which is where the misalignment arises.
The reason there are no handles is because, for now, I'm just exploring how to do this and, rather than deal with the hundreds of spline points arising from projecting the solids to the station sketches, I've traced the outer line to the profile sketches, offset by 20mm, deleted the original line and then offset back from the 2nd line by -20mm to get a pair of lines without spline points. This is why I then did a trial loft at profile 4, because when I had a few spline points the loft was misconnecting and failing. In fact that was no better so I instead tried surfacing and thicknessing, which did work and may be the way to go.
I know that good practice in CAD ship modelling is to have the same, small number of spline points on each station sketch to allow fairing, connecting these with rails. Down the line that would still be the aim but for now I'm trying to establish the best workflow in F360 and that F360 is as good a tool as any for this purpose.
I'd noticed that flat spot on the upper part of the profiles; I tested addressing it on profile 4 by tracing the spline to just selected points and shaping via the handles before the offset routine, so am not too bothered by that at the moment. Once I get to a point where the workflow is clear and the lofting works, I'll decide which stations to keep for lofting, go back and re-do the splines to allow manipulation. I realise it would be easier to decide now, but I'm trying to avoid creating a boat that's similar to the original over one that is a near exact replica and suspect I'm going to need use most of those at the prow and stern. But I'd want to do them all in one go so I can check that the hull is also faired laterally.
FWIW, the end product is supposed to replicate this, and if I can do that, the principles should be readily transferable to the big ships I'm aiming to model (Victory, Cutty Sark etc);
BTW, this isn't the real thing - this is the 3D model
Can you share that 3D model ?
I am wondering what kind of mesh this is. Often textured models are quad meshes.
laughingcreek, thanks, that explanation of the planes refs is priceless. I would never have figured that out myself.
bigger things... to a large extent I am only using the mesh as a visual reference. I've created some 'good enough' standalone sketches from tracing the slices of the brep, copied these across into a separate file and there's no connection at all between that and the mesh file. My intention is that the 'ships boats' file will be a pure F360 product as though I had scratch-built the boats. However, I suppose what I'm really doing here is taking a shortcut. I'm using those 'good enough' sketches to try to create a close enough model, from which I could, if I wanted, create a normal ship lines plan. If you're not familiar with that, it looks like the pic below. So we're on the same page in that respect. The tricky bit will probably be getting it to loft. I'll be trying to get away with the smallest possible number of profiles but think it could become six of one and half dozen of the other - more profiles will tend to yield the correct stern shape, less probably means I'd need more rails. I often struggle with when lofting goes yellow, any advice or nuggets for dealing with that would be very welcome.
Peter, I'd love to share the mesh but feel it was given to me in the expectation that I would only use it for reference and not share any further. It's a beautiful bit of very high quality work and I'm always conscious that some creators make a living through this and similar projects.
In case you're wondering why I don't just use this lines plan, it's because the lines are never fair; working off a faired model gives me a better chance of ending up with what I want, but is also something of an experiment i.e. is this a way to go, or is it just not suitable for F360.
Thanks again for the help yesterday, all very valuable advice which helped get me to where I wanted to get to and gives me a good steer for the rest of the project. There are a couple of minor wrinkles (literally) in this interim result, one when lofting the return to the prow, another which is just a minor fairing issue. Both will be easy to deal with. From here I can now better see and understand where the stations really need to be to get the prow and stern planking correctly.
laughingcreek, you mention the sketches need to be pristine. Can I ask what you would call pristine? On other similar lofting projects (curved both ways, including S-shapes) I've taken great pains to ensure the curves are smooth, continuous, constrained, railed, everything I can think of, but still sometimes get the yellow hell. By and large I've got around it by surfacing and thickening (just as I've done here) but would really like to understand what to look for to avoid the problem in the first place and fix it when it happens.
@TrippyLighting schrieb:
@wersy wrote:... Fusion is not the right program for that...
Do you have experience with a tool that "is right" for this work and can you recommend one ?
We've had this topic before, you said:
„Looking at the NURBS patches in hat model I agree that this cannot be done that way in Fusion 360 and most likely the majority of midstream CAD software. The hull is mostly one contiguous NURBS surface. To create a surface like that requires control over individual NURBS CVs, something we don't have in Fusion 360.“
It would be a milestone if there were such a script for Fusion as well, although the smoothing process can take several hours.
But even hard chined boats according to shipyard plans (with tenths of a milimeter accuracy) do not succeed without extensive reworking of the frames and longitudinal lines. As with my already well-known sailing dinghy "Pirate"
In the end, I just used the longitudinal lines (rails) approximated to the formers for the loft. You don't really need the formers for hard chined hulls. So I created the ribs afterwards.
All my round frame boats that I have published are made up of 3 splines. The keel line, the deck line and the stern profile.
Today, if I wanted to draw a hull that was as close as possible to the original, I would use a program that was developed ONLY for ships. I am thinking, for example, of "Delftship".
In the video it immediately becomes clear why this method creates smooth hulls.
https://www.youtube.com/watch?v=TCOmPcBZv58
As a cautionary example - how NOT to do it - a 74-gun ship from which I had very good framing drawings.
This was my first attempt in 2000 🙂
Forums Links
Can't find what you're looking for? Ask the community or share your knowledge.
Post to forums
