An attempt to model a ship hull in software.....

[Kevin-the-lubber]

Try this;

https://forums.autodesk.com/t5/fusion-360-design-validate/how-to-draw-a-boat-hull/td-p/6428751. I always listen to what Peter Doering says, he knows his stuff. You'll find examples in the thread which might be useful - I haven't looked but some of the lofts on my stern follow the same principles.

 

Personally I wouldn't go near forms for this, and I only use surfaces when lofting or sweeping aren't possible.

[Egilman]

From some of what I read on that thread, lofting is still something they are working on trying to expand, right now it is hit or miss on it's usefulness... Like I said it can be done, but is the time it takes really worth it?

 

I'll give it another shot.... and see what I can suss out....

[Kevin-the-lubber]

I've used it extensively - bear in mind that thread is several years old.

[Haze Gray]

Hi Egilman,  Kevin probably has the best idea for attempting this in fusion 360  - using 3d sketch to draw a spline and manipulate the control points to get the right angle and curve but that could be little messy/time consuming but could be made to work.  Since I really only design ships based off frame/hull lines I don't have any meaningful experience doing a hull without them.  

[Kevin-the-lubber]

Just the basic lofting process here, sample file attached. (with so few steps I haven't bothered to name things....!)

 

1.       Create sketch, select top

2.      Draw one half of the deck shape using a fit point spline. This is going to be the guide rail for lofting so you only need the outer deck edge. Note, you need a continuous line, otherwise you’ll have problems when lofting, so run that spline point to point and use the handles to get a straight side shape. Don’t mirror, not needed. Close sketch

3.       Construct an offset plane 30mm below the deck plane

4.       Create a sketch on the new plane. Project the deck rail from the deck plane but don’t link. Offset the projected rail by (10mm in sample) to get the mid-level hull profile. Tip – trim the overshoots (T on the keyboard). This will break the offset constraint but that’s not a problem. Close sketch

5.   Create a side view sketch using the front plane (this will be on the hull centreline). Draw a keel line, whatever you want. Make sketches 1 & 2 visible and join the bow point to mid-level bow point with a line. Draw another line to the keel line. Close in the top and stern end, to create a closed profile. Close the sketch

6.   Construct an offset plane in the left view. Offset by 100mm. This is the stern plane. Draw a sketch joining the deck to mid-plane to keel. Ensure it’s a closed profile.

7.   Loft the stern to side profiles using the deck rail as a guide. Mirror if you wish.

 

 

   

chine hull v0.f3d

[Egilman]

Thanks Kevin, that will create a fairly generic hull, how does it work for a modeled hull of a specific boat/ship......

 

I've sussed out the projected curve....

The purple line represents the projected curve....

Edited October 26, 2021 by Egilman

[Egilman]

And here is the first lofted surface, the upper hull side....

 

[Kevin-the-lubber]

The principle remains the same. https://forums.autodesk.com/t5/fusion-360-design-validate/introduction-to-loft-using-rail-curves-in-fusion/td-p/7296238 

 

To be honest, I need to work through this methodology myself as I still struggle with multiple rails (I didn't use project intersect in the example and couldn't successfully use both rails as intended). Everything is fine when it just works and behaves exactly as expected. The problem is when it doesn't, invariably because it would cause the loft to intersect itself. That's about all F360 tells you, but is thin on fixing the apparent error. I fully understand the concept but it can be frustrating when the sketches themselves seem fine. I expect that, like everything else F360, I just need to read more and learn what I do wrong.

 

However, that is what you would need to do for your hull, as you want an explicit shape at mid-level. So one rail for the deck, another for the 'waterline'. Does that make sense?

[Egilman]

3 hours ago, Kevin-the-lubber said:

However, that is what you would need to do for your hull, as you want an explicit shape at mid-level. So one rail for the deck, another for the 'waterline'. Does that make sense?

Yes Kevin, it makes perfect sense...

 

It's exactly how I lofted the hull side above except the bottom of the surface is the projected chine line....

What you need to do for it to work is make each of the side and top profiles one single solid spline... Then the end lines being one single solid spline as well... AND! turn on 3D sketch... So you can attach the ends with coincident relationships....

I still have to tweak the sketches a bit cause I made an error, I accidently put the stern spline on the same sketch as the chine line... It worked, but it's not the way it's supposed to be done...

 

Where SW only uses two sketches to project the line, F360 uses three... (the projected line has to go on it's own sketch)

I'm going to correct my error, then attempt the hull bottom panel loft.... After that it should be pretty straight forward...

 

I'll post a detailed step by step when I get it finally sussed out....

Edited October 26, 2021 by Egilman

[Egilman]

Been playing with it, I've gotten it to the point of being a solid body....

 

The whole thing.....

One piece, that's where we want to be.... Onwards.... (although I might take a bit of a break, this last trip was kinda rough figuring out how to get this done)

[Kevin-the-lubber]

Well done! Are you happy to share the file? I'm interested to see how you've got there. Quick tip for all those unconstrained sketches, in case you don't know it - 'Fix' is your friend, especially for splines. Open the sketch, select everything, hit 'fix', it all turns green and you get a padlock sign as a reward.

[Egilman]

Thanks Kevin,

 

It was a ten hour marathon to get it to that point, lots of research, lots of trial and error... Right now, it won't shell, in any way I've found to try and that is the next step....

 

It's not ready for prime time yet... There are some errors in the time line way back at the first lofts that needs to be corrected.... So I'm going to take another shot at creating it with the experience gained, Hopefully, I'll get it back to the solid state and be able to shell it, or figure out why not.... Not completely sure of how I got to this point yet...

 

The path of discovery continues... Not going to say I'm the first to do it this way, but I'll tell you what, there are no mentions of modeling an existing hull, (as opposed to simple building a generic hull) this way in F360 I've been able to find on the net nor in AD's forums.... Their tut's are all just an example of what they think "Could" be done.... Every one actually modeling hulls in F360 are going down the cad route rather than the much faster solid modeling route...

 

The one thing I am going to do is comment on the non-helpful warnings they offer when F360 thinks you did something wrong... It's gets very old very quickly when all they tell you is to re-draw your sketch or re-model your model for every single issue... Cryptic does not build confidence in the product..... {chuckle}

The feel of it is like your using cad to build your model or at least they have taken a cad approach to the tools they offer solid modeling... And if the input isn't perfect, it either automatically rejects it or will fail without explanation... I've also had it bail out and crash on me four different times... Seriously reminds me of the early blender days...

 

Anyway I'm taking a break for the day, and will hit it again tomorrow... will keep everyone posted....

[Richard Dunn]

I know I have said this before but if you can stretch the budget and plan on doing this a bit you really are better off with Rhino ,in so many ways I can't sum up here.
I know I am a professional 3d artist and yes I have years experience doing boats and ships but I believe anybody can do it with a couple of good tutorials in Rhino and knock out a basic hull in 4-5 hours, Rhino is also a lot less fussy about the input curves and very rarely fails.
Not to mention the reason Naval Architects and Boat /Yacht designers use it should say it all.

If anyone does try it I will gladly make a tutorial, I am a qualified teacher of 3d modelling and taught it at Uni for 5 years.
I know it is not free but really it is a huge investment and you can do soooo much with it.

Much easier program to use, more intuitive.

[Egilman]

Richard, I have access to Rhino, You make it I will come.... I've yet to find a decent tut on it yet though and if you write one it will be the first.... And yes, Marine modeling/design has been using Rhino and a version of Dassault's proprietary system for decades to model ship hulls... They are the go to's of the ship design world.... What Autocad is to engineering and structures, Rhino is to Marine Design... Many people could use it most don't have the resources to pay for it...

 

So yes, a heartfelt and sincere statement, you build it, (rhino tutorial) I will come... Right now SW from Dassault just plain works for solid modeling it's reliable and simple to learn, Fusion 360 has issues IMHO..... Good for simple stuff a bear when you get into anything larger or compound curvy... Very capable but a mountain to climb and not reliable in it's operations... works in some cases not in others and the unbelievable part is it does this on the Same Geometry one day to the next... Seems like it works when it wants to.... The direct comparison to SW shows that it is overly complicated for what it does, which I have seen all too often when the basis of the software is a cad background.... (all Inventor/Fusion products stem from Autocad as it's original base point)

 

What I'm going to do is get the process down and repeatable for the basic hull shape solid, then let it go there... Since I've gotten this far, and the process does work, might as well as document it for what it's worth... But I'm not going to put a lot more energy into it... 

 

McNeel, Dassault and Autodesk have been competing for several decades now and each have developed their own little playground, Rhino is best at Marine Applications, Dassault is best at Aeronautical Applications, (and is pushing hard into general design) and Autodesk is trying to push into those areas but is best at BIM and mechanical/structural design.......

 

They are all capable of doing what the others do, but each is best in the area they were originally designed to perform...

 

It's been 30 years and that still hasn't changed....

[Richard Dunn]

Yeah its funny isn't it because marine and aero are not that dissimilar in terms of surface quality and the needs for fairness etc but you are right.

BTW Rhino is $995 US in case anyone is wondering.
But you have a deal I will do a video on a hull, deck, superstructure.
I might do it for my store as I have training on Artstation already

Edited October 27, 2021 by Richard Dunn

[Kevin-the-lubber]

This is the bit I’ve been waiting for, the comparison. For some strange reason I feel slightly defensive about F360, probably because I’ve cut my teeth on it, but so much of what you say, Egilman, is so true. I opened my now finished stern master file today, intending to make a simple tweak. Throughout the week it’s been in perfect condition, not a warning in sight. This evening, on opening, it presents me with a sea of red and yellow fails. How can that be? Surely software is still binary, it can either carry out the instruction or it can’t, it shouldn’t depend on whether it’s a Tuesday or Wednesday. So I face many, many hours getting it sorted again, 90% of that time being spent waiting for the programme to process the fixes. I know it’s something I’ve done, that it’s my fault, but I am fairly fed up with fighting it every step of the way. I don’t think it’s an exaggeration to say most of the hundreds and hundreds of hours I’ve spent on the stern have been spent just waiting. I’m sure this is true because when I make something without any issues, Ieven when it’s still a bit involved, I’m done and dusted in the same kind of time it’d take if I was drafting by hand. And you’re absolutely right about the lack of help messages and crashing. I want the software to simply refuse an illegitimate command, not spend 20 minutes thinking about it mysterious, heartsinking silence and then crash. I’m ranting, I’ll stop. 

 

I’m at a natural break point now. I can finalise all my F360 work into stl’s and change horses without losing too much ground. You’ve convinced me to give both Rhino and SW a go, invest to save. Hopefully, most of what I’ve learned on F360 will make Rhino or SW easier to learn. Richard, unless I’ve missed something Rhino doesn’t seem very expensive, think it’s about £100, quite bearable for me anyway. Though I’m still a bit nervous about it not being solid body.

 

[Richard Dunn]

Rhino does it all, it does Nurbs, Solid, mesh and SubD.
Please don't take me wrong I am not on a mission to convert, but I know from my experience as a teacher and my work life Rhino is easier to use and more versatile, I have used 360 to and hated it but I am not here to try to convert, I am just saying if you want to have access to the proper tools needed to do hulls and for that matter the work I am doing on My CNC model you are better off with it, heck I taught Rhino to myself in 4 weeks and then got a job using it!

And yes the skills translate across more or less

Edited October 27, 2021 by Richard Dunn

[Richard Dunn]

This is a screengrab of my Rhino screen right now and the model I have in my other build blog being cut on cnc, this would take ages in 360 to so , this also has the ability to do true plate development of sheets, unroll developable surfaces so making hull panels etc in true shape is easy, that's another example.
Guys give me a hull you can agree on to learn with and I will do the video/videos for you.
Lets see what moderators say but I might even do it as a live lesson/course over zoom, that's easier for me and you can ask questions then

Edited October 27, 2021 by Richard Dunn

[Kevin-the-lubber]

The type of hulls I'm going to be most interested in would be classic age of sail types i.e. plank on frame rather than chine, but a better description is that I'll need to be able to model compound curves with not a flat line in sight.

 

You're selling it to me Richard, if it can do all that. I guess I'll just have to make the probably painful journey of learning the basics until it all falls into place.

[Richard Dunn]

Well why don't I show you how to do Cutty Sark from the real data from ship?

Edited October 27, 2021 by Richard Dunn

[Kevin-the-lubber]

That would be fantastic. Yes please!

[Richard Dunn]

OK then , lets see how to go about getting an online event happening and I will do a reconstruction of the ships form from scan data and the NVM plans and take it to the point where we plank the hull.

 

[Kevin-the-lubber]

Cripes, I'd better get signed up to Rhino then.

[Richard Dunn]

I have messaged one of the moderators to see what's what.
It wont be for a few weeks mate no hurry.

[Egilman]

1 hour ago, Kevin-the-lubber said:

This is the bit I’ve been waiting for, the comparison. For some strange reason I feel slightly defensive about F360, probably because I’ve cut my teeth on it, but so much of what you say, Egilman, is so true.

Well thank you for the validation... I didn't start out for this to be a comparison of SW & F360... But it has become just such a disparity in what they actually are they aren't even in the same class.... I personally believe that F360 trying the multiple compound curved surfaces of a Predator 108 will put it in it's grave... It can't seem to handle a simple two panel chine hull.... What I drew in an hour in SW, was next to impossible in F360 and then is locked up, you can't go any further....

 

And there are people that pay for that crap? (yes according to autodesk)

 

I think since your willing to try, I'll just ditch F360, I don't have the patience I once did with things that are less than expected.... And I no longer accept beating myself up trying to make less than adequate tools work the way I want them to... I know what the tool is designed for and know how to use it so I don't have to spend my abilities and time making it do what it is supposed to be doing itself without issue...

 

Enough...

 

Thank you for the honesty Kevin, I know it can be painful when you have invested so much into something... (especially when someone from outside confirms your suspicions) I invested a lot into Sketchup, until it went commercial and lost a lot of what made it great..... Same with Blender, there was a movement to make it more flexible to use then they went along and at one of their congresses decided that they were aiming directly at the CGI markets and would concentrate their efforts there.... It's now a wonderful do it all tool, if you like CGI and gaming modeling... (ultra high poly videos or reasonably high poly game models) It's not worth much for anything real world...

 

The work flow is the same, the model should be the same once it goes thru the workflow, SW will do it and I'm sure Rhino will as well...

 

Thanks for tuning in and your commentary....

[mtaylor]

51 minutes ago, Egilman said:

And there are people that pay for that crap? (yes according to autodesk)

 

 

Go back in history a bit.  Autodesk was one of the first drawing packages and had a big impact on CAD.   They sealed their reputation back then with lots of companies jumping up and using it.   Their core market has been hardware design used by manufacturers.   I suspect that ship building and the special features needed really never was a core part of their business.  Shipbuilders might use their programs for ship equipment but I doubt very much if the hull is designed in it.   

[Egilman]

24 minutes ago, mtaylor said:

 

Go back in history a bit.  Autodesk was one of the first drawing packages and had a big impact on CAD.   They sealed their reputation back then with lots of companies jumping up and using it.   Their core market has been hardware design used by manufacturers.   I suspect that ship building and the special features needed really never was a core part of their business.  Shipbuilders might use their programs for ship equipment but I doubt very much if the hull is designed in it.   

Yes it is and yes they did brother.... Autocad is cad and is great software never said it wasn't brother.... (it's why I use it myself)

 

I was specifically referring to Fusion 360....

 

Marine needs have always been different than general drawing/drafting... Can you design a ship in Autocad? yes, do you want to no.... {chuckle} Can you design a building in Autocad yes, do you want to? no Autodesk themselves make a product called Revit exactly for that...

 

Autodesk themselves know the benefits of specialization.....

 

My first brush with autocad came 30 plus years ago....

[mtaylor]

Yeah.. I see where you're coming from on this.  I was given Autocad, a computer and some other software when it first came out and told "use it... it'll be great for the art in the tech manuals".     Err... yes and no.  The mechanical things it was great.  A PITA in many ways to get final art for a manual but schematics?  Nope.  On the bright side, I, one other tech writer and our manager lead the company kicking and screaming into the computer age.   Fun times.... mostly.

 

There's still a lot of "crap" out there sadly.   Seems sometimes there's more crap than good stuff but I guess it depends on one's needs.  As for ships, I suspect there's not much out there that's fit for purpose but probably a lot of custom "fixes" by some creative types.   Problem is, finding those "fixes".

[Egilman]

3 hours ago, Richard Dunn said:

Guys give me a hull you can agree on to learn with and I will do the video/videos for you.
Lets see what moderators say but I might even do it as a live lesson/course over zoom, that's easier for me and you can ask questions then

 

3 hours ago, Richard Dunn said:

Well why don't I show you how to do Cutty Sark from the real data from ship?

Well I'll follow along as best I can, live online stuff may be the thing today, but it shuts a lot of people out.....

 

The CS, (a beautiful ship btw) being a composite hull made of wood and steel wouldn't have been my first choice... Something steel with at least two screws, a steamer bow and cruiser stern.... Just my preferences....

 

Anything you offer Richard will be good....

[Richard Dunn]

Ok here is the thing, would you rather live and have access to recording , or just videos?.

Live you get to ask questions

Videos you still can I guess but after the fact.
Lets see what interest their is, not sure how people realise its the way of the future even for model builders, because as you guys know if you can make it in 3d you can build, CNC, resin print and even do photo etch although that's vector work.

I was intending to do a course on her anyway, here was the initial idea on my artstation page.

 

https://www.artstation.com/artwork/qArwVe

Edited October 28, 2021 by Richard Dunn