2D with Illustrator or 3D program?

**stickers11** · #1 09-24-2012, 01:14 PM

This is only my second day on this forum, but I am already getting the design bug. I've read a few threads where people use 3D modeling programs to design, then "flatten" the surfaces for paper modeling. Seems easy enough...lol.

What about forgoing the 3D modeling and just drawing the surfaces in a 2D program such as Illustrator? I know this is simple enough for anything with right angles, or extrusions, but I imagine you would run into problems with complex curved surfaces?

So what is the right answer? As much as I would like to learn a 3D program, I'd rather tinker with paper and scissors. I imagine with enough geometry knowledge you could recreate any surface in 2D.

**thorst** · #2 09-24-2012, 02:40 PM

Hi to the forum!

To your question: Yes, you can also develop all shapes by hand which can be developed with 3D-Software. For example, I did the parts for my yet in development Space Shuttle all manually in a 2D CAD, with tools just as a compass and a ruler. (1/48 Space Shuttle orbiter under construction).

Therefore I first draw a 3 view drawing and then divide the surfaces into triangles, just like 3D software does. Then the triangles are drawn in their plan view and put together... also, just like the software does. It works very well as you can see on the pictures, with the advantages that you learn a lot about geometry and that you are able to smooth the outlines to avoid the pepakura-like corners on the finished model. The big disadvantage is that this method needs a lot of time. The parts for the nose of the shuttle took me about two full working days to draw (and that's just the last step of the process!).

If you need more information, please ask me. That's how I do it, but there may be different approaches...

Best regards,
Thorsten

**stickers11** · #3 09-24-2012, 02:53 PM

Thanks Thorsten. Great to have found some resources already. I look forward to reading the thread on your orbiter...

**stickers11** · #4 09-25-2012, 02:43 PM

Hi Thorsten,
Let's take for example the gas tank from this motorcycle: Ariel For Sale: 1935 VH Red Hunter 500cc
It's probably the most rounded piece on this bike. As you can see, this site as good top, and side views. Still, this looks challenging to "map" in 2D. Sounds like you would still draw it in 3D first, then slice out the triangles?

**thorst** · #5 09-25-2012, 05:01 PM

No, I wouldn't draw it in 3D (simply because I don't have the software to do so). One would start to draw a side view and then some cross-sections, that means views perpendicular to the side view, at some different lengths. These define the borders of the parts. Then you can choose how exactly you want to approximate the part and
1) add or reject some cross sections for accuracy in side view and
2) devide the curves of each cross section into straight lines (the more edges, the better).
The edges of two neighbouring cross sections are then connected in a way that they form the triangles.

Thereby the main work is drawing the cross sections as you have to study your references very carefully to achieve the correct impression, but that also holds for 3D-modeling.

Does this answer your question?

Best regards,
Thorsten

**stickers11** · #6 09-25-2012, 05:11 PM

It really does - thank you.
I followed your thread on the shuttle, and saw how you mapped the nose, etc...
Either way, there's probably a learning curve involved. I do like the idea of choosing how the part flattens myself...and like you said, avoiding some of the sharp angles you get with some software solutions.

**thorst** · #7 09-25-2012, 05:36 PM

Yes, it surely has a learning courve. That's mostly at two points; first you will get a lot of experience at how to set the edges to get a smooth result, and second you'll find your own way to quickly construct the triangle developments. For example how to choose the viewing directions of your drawing, so that you have many sides already known and don't have to reconstruct every vector afterwards.

If you once know how to construct one triangle, the rest is totally boring, stupid work...

**stickers11** · #8 09-28-2012, 03:19 PM

Me again.

I promise not to turn this into a building thread, but if I can get the tank down, I may just start a build thread. I've got front, side and top views of the tank. What you probably want me to do is get a few more angles of the top of the tank, right?

I'd have no trouble building a 3D model based on these three views, but this is almost doing the reverse? I can see how the edge of one view will give me the curvature of the other...but still a little stumped on how do this for a paper model.

Do I now divide the top section into straight lines? You've been a great help thus far...if I can lick this part I'm good to go.

**thorst** · #9 09-28-2012, 04:43 PM

Ok, you have now the basic 3 side view, that's fine. If you don't want to make angled cross sections/part borders, you don't need to draw more views.

The next step is to create cross sections. For a first approximation, you can take the front view and scale it in both height and width separately to match the top and side view at some stations (You won't get a nice result at this particular object with under 5 stations I guess. I have indicated possible positions in the picture at the bottom). For a more realistic result, you'll have to carefully look at the photos to see where the cross sections (let's call them CSs now) need to be tweaked then, and if more CSs are needed.

When this is done, you'll have to split the surfaces into the triangles. Therefore you can have a look at Cami's new thread, he describes a method quite similar to how I work:
Designing Curves The Old Way

Thorsten