What Is the high-level process flow for designing a plywood boat?

Hi There,

I’d like to discuss the usability of Rhino 5 for Mac at a high level, to design small watercraft – including how to derive flat patterns and cutting files, fabrication and assembly drawings, and also visual renderings. We’re pretty sure we want this path but I don’t want to blunder in without understanding how to get to the goal.

Our specific project is to enter a design for a small oar-driven boat made of mostly plywood, to massage the design to perfection, then create CAM output for automated cutting of all the wood parts.

We’d also like to know if it’s feasible to use the Mac version to get this done and even do some visual renderings without buying more software. We’re a startup so budget is tight.

Is this project feasible as described, in Rhino 5 for Mac, and if any additional software is needed, what would be best? Costs?

Can I move the part and assy. line geometry to DraftSight (from Dassault) software and do the drawings there? I have that software and like it. It is DWG file compatible.

Also, I do not have the recommended graphics processor on my MacBook Pro. I have the Intel HD Graphics 3000 512 MB card that is standard for the late 2011 model. Will it work?

What effect will my hardware have and can I still get the job done? I have a lot of open disk space and a 16 gig memory.

Is it even possible to upgrade a MacBook Pro 2011 to one of the processors Rhino likes?

What would be the high-level steps to get the whole task done assuming I learn how to use Rhino using the Tutorials? I did achieve guru status in Solid Edge but that’s a history-based solid modeler. Surfacing is all different in the way you model things and edit the models.

Is anyone here able, and motivated, to speak to these issues?

Your help is greatly appreciated.

Everything asked for above is possible using Rhino for Mac through creating a Rhino file with the profiles to be cut.

Hi J,

David is correct in his answer that you project is well within the capabilities of Rhino for Mac. You might lose some naval arch functions that would be available via Orca3D which is a plugin that is used for yacht and boat design but only availble for Rhino on WIndows. I have done a number of small plywood boat and kayak designs and can help you establish a workflow and would be interested in hearing more about the nature of your project. You can PM with details or post here. 

Jody

David and Jody,

Thanks so much for your reassurances. Based on the new confidence you’ve given me, I’ve downloaded Rhino 5 for Mac and installed it, and the manual and some tutorial files. I’ll start learning how to use Rhino and, as soon as I can, I’ll run through the workflow from beginning to end with a small design. After many years of using parametric solid modelers, I’m quite taken with the power of this surfacing system and all it can do so well. What a stunning system for the price.

I’ve found that the flattening of developable surfaces is a built-in function and that is probably the end step on the design phase. From there, the CAM toolpath can be made by the shop crew to suit their equipment. It will have a lot of specifics matching the details of whatever machines we pick, and that will vary based on whether it’s a router, laser cutter, or possibly even waterjet.

One aspect to consider is organization of your design, including managing changes. I see Rhino as a hobby, and use it for designs for my other hobby, designing/building/flying model airplanes and -sailing the odd model boat. But I work closely with a small company who does this for a living, and uses cad/cam to manufacture kits.
What we found is that when the parts count is high, it is very easy to lose track of the status of all the flattened surfaces. The workflow we follow is design the model in 3D to a reasonable state of completeness, extract the parts as flattened surfaces, build the prototype and test it, change the 3D model as necessary in multiple steps (i.e. correct errors during building and make changes resulting from testing) and extract/replace the flattened surfaces as often as necessary.
We use numbered sub-layers for each part and its flattened equivalent, this way it is easy to check whether all parts are done. If you are not sure if a certain part has been extracted, just look at the layer it is on and choose “select all objects on this layer”, and check if the relevant part is selected in your collection of flattened surfaces. I also developed a simple script to label all flattened surfaces with their layer number as an easy reference, and another script to label the parts in the 3D design in the same way, creating a visual cross reference…

Max.

Hi, Max,

Thanks for joining the conversation. I comment interspersed with your
observations, below.

One aspect to consider is organization of your design, including managing

changes.

With any sort of CAD, the first issue you create by adopting the tools, is
data proliferation. I have a full career as a mechanical engineer and
corporate CAD manager behind me. It’s interesting that in your experience,
you came upon all the classic issues, and found a way to organize your
proliferating data that works for a business. In any CAD environment,
success in design ultimately depends largely on organization of process
flow and data. This is why I became a radical supporter of PLM, without
which a large design operation soon comes to its knees, expensively.

In small operations like hobbies, a good, consistent process, coupled with
a file naming convention that tracks versions, can cover the needs. It
seems that this is what you saw a need for, and created.

I see Rhino as a hobby, and use it for designs for my other hobby,

designing/building/flying model airplanes and -sailing the odd model boat.
But I work closely with a small company who does this for a living, and
uses cad/cam to manufacture kits.

This is exactly what I am embarking upon now, kits of parts for real-world
products, and helping launch the company that owns the designs that I will
document in CAD. We will begin as usual, following the smaller-scale,
hobby-style approach, and as the operation grows, products will
proliferate, and with them, the CAD data. Soon it will become necessary to
consider PLM and hopefully, by that time, sales will support that cost.

What we found is that when the parts count is high, it is very easy to lose

track of the status of all the flattened surfaces.

Oh yeah . . . So here is a key point. You see the flattened surfaces as the
base units of the design – the “parts.” So is all the CAD data that
generates the flats, including the surfaces that will be organized into the
assembly representation of the end “product.” At this point, though I am
familiar with how this occurs in parametric solid modeling, I see the
surfacing approach is different enough for me to wonder if the same
organization works for a Rhino-based design. A prime consideration is
whether there is a top level assembly that comprises all the Parts in the
Product and serves to prove the fit and function of everything in the
“Design” (top level CAD Model.) I naturally think in terms of this
organization now, based on experience with, say, Solid Edge or SolidWorks,
and other PSM’s.

The workflow we follow is design the model in 3D to a reasonable state of

completeness, extract the parts as flattened surfaces, build the prototype
and test it, change the 3D model as necessary in multiple steps (i.e.
correct errors during building and make changes resulting from testing) and
extract/replace the flattened surfaces as often as necessary.

Again, this is an analog of my usual process flow with a few notable
differences, mentioned in-line.

We use numbered sub-layers for each part and its flattened equivalent,

Within the top assembly, this seems a good way to isolate the parts into
separate, manageable entities and that’s a key need. This proves the fits
and I immediately wonder if there is associativity of the part profiles to
the part (assumed, because the profile generates the part surface,) and how
the parts are associated to to the assembly. I think that last question
will become apparent as I delve into Rhino.

In the PSM world, Parts have 2D profile sketches that generate them, so to
edit a base part, you edit the profile and extrude along an axis. Further
operations then modify that base solid until you have something that
represents the physical item to be manufactured, in this case, a flattened
surface that is cut from sheet materials. The details of this Part
generation differ in the surfacing world but the top level result seems
analogous.

Along the way, modifying a part may involve rolling back it’s history to
the point where a specific operation created a part feature such as a hole,
which may assume subtractive manufacturing. Both additive and subtractive
final processes can be supported. However, they may occur at different
times and in different processes, of course. Such as making a casting and
then machining it. However it’s made, a Part’s controlling document is
integral to the Part itself and is tracked as a collection of CAD
operations.

this way it is easy to check whether all parts are done.

Of course completion is something to track.

If you are not sure if a certain part has been extracted, just look at the
layer it is on and choose “select all objects on this layer”, and check if
the relevant part is selected in your collection of flattened surfaces.

I think you are working within the top assembly as you do this . . .

I would go so far as to assign part numbers on the fly and actually mark
the physical Parts, as they are made, with their part number. When cutting
from flat sheet stock, the available processes allow this to be done. In
additive manufacturing the lettering can also be implemented.

I also developed a simple script to label all flattened surfaces with
their layer number as an easy reference,

Or perhaps, by hand, this could be done part by part as some notation
that’s visible inside the file. I’m inclined to use a Part file naming
convention that not only numbers all the parts but also tracks their
revision status. This way, every part carries a clue to it’s right place in
the world within it’s very name. The Part number/file name is visible in
computers’ directories and facilitates identification and management
without opening the part file to see what it is.

and another script to label the parts in the 3D design in the same way,

creating a visual cross reference…

Not sure yet about this detail. What impresses me is how a “hobbyist” has
found ways to serve the needs of a business and built a process that works
to deal with the data monster that comes in the door that is opened to the
power of 3D CAD. Well done!

Now my burning questions are about the nature of assemblies in Rhino and
what sticks them together. Also, if the Parts have histories and if the
Assembly is a separate file type. In other words, how is associativity
implemented and leveraged for good design practices?

Today I plan to dive into the tutorials and see how much progress I can
make going through them thoughtfully, resolving concerns and making
comparisons to my past practices. In the end, I’m sure there are ways to
achieve the needed revision control and organize process flow rationally
and professionally.

Thanks so much for sharing your knowledge. It really helps for articulating
how to work in Rhino for best results.

Joseph

Max.

Visit Topic
http://discourse.mcneel.com/t/what-is-the-high-level-process-flow-for-designing-a-plywood-boat/34683/5
or reply to this email to respond.

To unsubscribe from these emails, click here
http://discourse.mcneel.com/email/unsubscribe/ff6d4b3d6821c725eff1de7f219f5344d19c5e8653494a4e7df8ee65271c8d26
.

Joseph Riden

EMAIL: joseph@josephriden.com

Max,

When I responded before, I did not get the image because I was working from email. Now I see it here in the forum online. I like your fuselage model highlighting all the bulkheads where the wings will attach. I see the part numbering you had mentioned.

How do you make this assembly coherent, so the parts all have specific location and defined spacial relationship to each either? In a solid modeler, there are various relationships one can apply between any 2 parts, such as Mate, Ground, Offset, etc. When an assembly is coherent it’s easy to achieve and prove fits and confidently use existing geometry to define new parts that fit right in. In any products that have multiple parts this is of course a key aspect.

Also, after my email reply earlier, I watched the first two tutorials for Mac. Real eye opener, the Water Bottle modeling. I was impressed by a tangential remark that at McNeel ships product when it is ready, not on a predetermined schedule. Also, I saw a lot get done very rapidly and powerfully using some seriously advanced modeling power. And I had to laugh when the presented used history for a bit, then used it to make a change, then threw it away because it was hampering progress. Then he made downstream changes without concern about history. For me this is a new way to work. But it fits right in with the power of Rhino to accomplish unfettered styling-related development right along with creating a strict part definition.

I look forward to eventually settling my mind about creating assemblies of parts.

Joseph, I leave it to users far better versed on the subject of interrelations that you are raising, but as far as my knowledge goes there aren’t many relational links in Rhino. There are single parent-child object relations like curve/subcurve and polysurface/surface, and there is grouping of objects and blocks. Other than that, as far as I know you have to create your own relations like I showed with the layer example (which has the advantage that there is an interface with possibilities to show these relations), or maybe use object names (where you have to use scripting to bring up the relation). But as I said others know the insides of Rhino far better than I do, and probably understand more of your issues too.

Max.

All will come clear as I keep exploring, I’m sure. When I make an assembly,
for sure.

Thanks again.

Going back through this, I see that what I would tend to call an “assembly” could be the collection of parts in your aircraft, or any collection of parts grouped together in their final orientation, as in the image, which could be a subassembly. It appears that you simply draw your whole end-assembly in the same model with all the parts positioned and oriented as they are when the assembly goes together (Top Assembly.) You isolate each part on a separate layer so you can turn them on and off. Maybe you make them into Blocks. It appears that a Block is the closest thing to a Part File there is in Rhino.

I’ve discovered that Rhino seems to have no object relations like the ones I asked about in this conversation, just models, collections of models (Assemblies), and Blocks, if you want generalized Parts, like a washer, that might be used in many different products. This seems to work. I’ll have to see what options are available when I bring a library Block into a model and place it. The positioning and orientation would have to be specified and achievable if a Block is used in multiple Products (Top Assemblies.)

For a company with more than one product, this puts a lot of weight on the file naming and part numbering conventions. One would want to re-use Blocks as Parts rather than re-drawing them repeatedly.

So I think the organizational scheme issue is resolved. I see there is a tutorial for making a boat hull. So I’m eager to get to that and divide it up into Blocks and see if I can recreate a duplicate model by bringing in and positioning them all. If that works as hoped, my product organization concerns is answered in a very simple way.

I’m an infrequent user of Rhino, it serves my needs well, but one thing I have learned over the last 12 months is to keep things as simple as possible, keep your layers well ordered, named and in some kind sequence, curves on one layer, surfaces on another, solids on another, that kind of thing, and I’m guessing you will start your models from curves, keep them as simple as possible, i.e. only have as many control points as you need, the general rule of thumb for control points is ‘less is more’, much easier to edit with minimum control points. As for boat/yacht design, I know nothing, but it seems a lot in your industry do choose Rhino as their go to modelling software :smile:

Thanks for joining in. I have found the User Manual for Rhino 5 Mac quite valuable.

I went back there and into the McNeel Wiki and found good, relevant info on design organization. However, I had t do some translation from CAD-ese to engineer-speak before it made sense. Once I got terms defined it fell into place. Turns out, you can develop complex assemblies as your products with the right organization and control the design and components of the product well.

I’m a pro in using CAD so I have a company-oriented POV on organization. It turns out that Rhino can be organized with a product PART per layer (all entities) as a BLOCK. That works for business-like organization because one wants to exert revision control at the part level because a PART of an ASSEMBLY is a unit of manufacturing and sales – a product. The part blocks can then be organized into libraries that can be shared among users and used in any product’s assembly.

So I will make parts on layers of a product’s assembly and block them for the library at a given revision in the filename. Other products can accept these parts and if a change occurs at the part level, when the block is updated, that updates all the instances of that block, everywhere. You don’t want to do that unless the new part is backward-compatible with all historical usages or it won’t be interchangeable.

Alternatively, a given block can be exploded and modified, then saved as a new part (block) or with simply a revision roll if it remains backward compatible. A new part number breaks the history and re-starts it and going forward, all products that use the new part will require the new style, which should become a new block with a new part number to keep manufacturing and repairs straight. Products that have shipped or belong to an old branch of history will require the old style part.

Why do you think this level and compleixity of organization is necessary or even desirable to "design small watercraft " which will be built out of plywood? How much re-use of parts you design between designs do you expect? It sounds like you have experience in designing complex devices. You may want to acquire some experience in designing plywood boats before committing to the extensive use of blocks for the parts you design.

I recently visited a company which cuts plywood boat kits for a number of designers and who uses Rhino for CAD work. They requested no blocks in any Rhino files submitted.

I do have many years of experience in product design for companies. So as a design team leader it has been necessary to work with good document control. Companies organize with a scheme that accommodates any and all products they may want to produce and support in the field. The documentation strategy is applied wholesale to everything produced. The process is documented and published and everyone from the buyer to the shipping clerk is bound by it. I didn’t invent this. It’s how things happen in the business world.

The watercraft may be small, but this is a real manufacturing company that is starting up. Many different designs may be produced. Some of the parts may be interchangeable. So there’s a need to manage all products professionally through a complete product lifecycle. Otherwise it’s only a matter of time until chaos strikes.

I can’t account for what other companies do. We will be outsourcing some cutting work to vendors like the one you describe. They will follow our organization. We impose it, not the vendor who does the cutting. Maybe the folks you’ve worked with are thinking of their kits as a unitary product and each kit is entirely unique? I’m guessing . . .

Without using blocks for parts, how would you keep track of parts and assure interchangeability across designs and coordinate multiple designers working on different projects? I’m really open to whatever works.

This typical thing that happens is that companies adopt CAD and don’t see the data monster they are creating. Then chaos strikes and it gets really expensive and then someone who understands PLM and the need for it is called in to fix it. My approach is to avoid rather that to ignore it and deal with consequences later.

I love simplicity and want to reduce everything to the most basic form. Show a better way and I’ll take it. What I’m suggesting is a more formalized and universal version of what Maxz suggested for his model aircraft above. All companies that are staying ahead of the curve have some sort of PLM in place. What I am suggesting may not work for everyone. But it’s what we need. It’s like what my product engineering work has demanded for decades.

Did you try drawing a boat hull and were you then successful in duplicating the model as you suggested you would?

Chris