When your project needs ID type work, how do you handle it? Does the ID come from another software like Rhino or Shapr3D or Illustrator, or do you do it directly in your engineering CAD tool?
And when you get ID data from some non-engineering software source, do you just work with the imported data or do you rebuild the imported data into a live CAD engineering model?
You means those Artssy Fartssy projects? From signage, logo, industrial art, furniture etc
Start with a hand drawn illustration or a photo.
Sales Project Manager then add their illustration which look nothing like customer’s.
Model it in CAD with manufacturable material/shape.
A few back and forth to get what they want to what they need.
Made sheetmetal “body armor” from a model by a college student.
Only used outside surface of the model.
DXF from Illustrator or other art software will require hours/days/weeks of “cleanup” to reduce line segments so the G-code file is small enough to run.
Autotrace on photo might have similar result.
Got an artist to check the finished weldment. Things are off according to him and we need to fix it.
Left it for a week and it’s perfect when he check it again.
I think I’m way off topic …
Basically redraw everything in CAD.
To illustrate the way we do it here, I will quote the letter written by one of the engineers about ten years ago, recommending one of the CAD monkeys for an award: “His most impressive ability is turning engineering sketches, hand-waving, and verbal descriptions into 3d working prototypes of new inventions and designs.”
I’m not called upon to do ID. When I do projects involving ID, I’ll almost always get a CAD model from the industrial designer, and I’ll incorporate what I can within the confines of what the optical/mechanical stuff will allow.
BTW, almost all the industrial designers I work with use SW.
It varies on how we get the models. Sometimes we get SolidWorks models. Sometimes, we get surfaces to import. I usually remodel them so that I have something that is stable in CAD and adjusts in a way that we need.
If I do the ID then I go straight into Solidworks after sketching/Illustrator, the output of which might be used as backdrops to model over. Occasionally I have to use Fusion 360’s sculpting mode, exporting the geometry as STEP to Solidworks.
I receive external data as scanned sketches, Illustrator, STEP, Rhino, Solidworks, STL and sometimes a Maya OBJ (architects doing product design or furniture). The quality of the 3D data determines the need to remodel. STLs and OBJ are used as backdrops, often after a trip through Modo for tidying up.
I did some injection molding project in the past and the imported surfaces are the bible. at least for aesthetic components for vehicles.
you can thicken them add ribs on the hidden side and propose a design modification with a well argumented reason.
Having a dummy body was easier to identify our modifications as they were always appended below.
We never rolled back the tree with 3rd party data to avoid accidental modifications downstream and to be sure we had always a controlled allowance of machinable steel in the mold. making easyer to judge if the modification required adding or removing steel from current mold design.
we were using NX which is a ton more flexible than SW thanks to Wave geometry linker and other more robust tools to swap parts, and handle non parametric geometry operations.
Lot of brute force modelling.
Being a Model Maker i’ve had to be a mediator between both ID and Engineering thruout the years, first turning 2D into 3D surfaces using Surfcam CAM software because solid modeling surfacings sucked, to now using SW to surface art’s farts surfaces and models. Now when SW cannot get what I need I use a range of softwares to get the shape i need, Fusion SUBD, 3D Experience XShape SUBD, Surfcam still though it’s like AutoCad 12 UI, imported from whoever needs the solids (students since 2008), I also will hand shape and laser scan using DezignWorks inside SW which can extract all sorts of surfaces with geometry you can modify in SW. Then I need to make the prototype aka model via CNC, 3D printer, laser cut, vacuum form whatever is needed for ethier side … you guys are welcome 
So Matt to your question my answer is: YES, NO, MAYBE 
lenny
EDIT: Matt we we need the Stirring the Pot emoj and the S H I T Hitting the fan on too like Practical Machinist Forum has.
Probably not the same icons, but here they are: ![[stir]](/uploads/default/original/2X/4/44ae3c6936b6990a699b38c89605e9d862893bef.gif "Stir-the-pot")
and ![[fan]](/uploads/default/original/2X/9/9d53b4f10805081771e8ba6364742c56a850402a.gif "hit the fan"){kind=small}
the S hitting the fan then face is to good!
I have a lot of leash regarding how much I’m technically allowed to “adapt” the ID stuff, but there are sound reasons to put in effort to maintain the industrial designer’s vision… I might be getting my next gig via an industrial designer who appreciated said effort on a past project.
in any case, the workflow you describe with reference body and freezing the ID stuff is exactly what I also use. Since I work mostly in NX, sometimes I’ll create an extraction of the ID model, cut out relevant bits of this and unite them into my working model. (The SW feature trees which the ID folks create are often metastable and fall apart at the slightest parametric change anyway.)
If there is ANY type of vector artwork that is involved in your design and manufacturing process then you will 100% benefit for purchasing a seat of Rhino3D. The goal here is to knock down a heavily segmented (.DXF) or (.SVG)…etc, file into something that is much more manageable for 2D Sketches.
@Cadmonkeychris: You might want to take a look at Power Surfacing for Solidworks. Really great at SubD modeling inside of Solidworks and works with mesh data very well.
There are plenty of software’s that work great with DXF\DWG, SVG and vectored PDF’s.
I use Corel alot.
CorelDraw
Illistrator
InkScape (FREE99)
Another software I use to create SVG or DXF from PNG, JPG is called Convertio (https://convertio.co) so i can use in SW to add to solids to either cut logos into part for the CNC or combination or CNC\Laser.
I also have Surfcam that hadles large 2D aswsome and it links to SW.
I tried Power Surfacing early on but couldn’t quite get my head around it back then. I have been on Fusion360 for a long time, so it was more convenient and cheaper to use that than buy yet more software. Maybe I will have another go when I next need the capability.
There ia also another option to get Sub-D surfaces for $48 bucks a year, take this at a grain of salt, but an option anyway.
3D Experience Maker gives you X Shape for Sub-D that can be exported the sameway you’d have to via Fusion.
Technically speaking, F360 cost more over time than Power Surfacing’s one time fee… 
Yes but I do use F360 for other stuff, though I won’t be renewing my license later this year. After that…
Maybe. I am supposed to have been fully employed by one of my clients but funding for the project has been inconsistent. Ideally they will be supplying all my software and equipment soon and this all becomes someone else’s problem…
My experience working with real Industrial Designers—a combination of marketing analysis needs, engineering design constraints/requirements, and artistic abilities—is a bit different. The ID people I worked with were very talented and great communicators. They were the intermediates between Sales/Marketing and Engineering/Manufacturing. My experience was a give-and-take between the ID group and design. Normally we would start projects as a platform of products, and define market needs and engineering requirements. Then engineering could provide dimensions such as wheelbase and height and visual cones to the ID person. They would then start producing artwork at scale. This was generally done on a Wacom board and the images were then passed back to the CAD jockey. The image (jpg or png) was then imported into CAD, scaled to 1-1, and then the primary surfaces were created. Depending on the complexity of the project, those surfaces were used to feed the CAD assemblies and part-level designs. (using top-down design aka skeletal design principles). This used to be all done with class-A surfacing tools. Today, tools like Sub-Division modeling and lattice design significantly speed the process up. I hope that helps.
It’s very sad that’s often how it works. 
But the real answer should be “end-user needs fulfillment”