Modern Vespa

Anyone ever do any 3D graphic models of Vespas? I was thinking of starting one for the hell of it, no real purpose since I have yet to find a part I would have to make. I would probably export the model to 3DStudioMax for near photo rendering (screen savers, photos, etc). If we got far enough, we could do some cutaways, dynamic assemblies, etc.

I normally use Catia for 3D as it generates all my machine code as well as static and dynamic element models, stress, etc. This software is big bucks, very few people use it outside of industry, and I wouldn't expect anyone to collaborate on a model. So I am playing with Autodesk Inventor Fusion, frustrating as it is limited, but it is nearly free on the mac.

If you have time, send me a PM and we can split-up the work. I'll post screen shots as I get going.

Are you able to work with .eps (vector files)? I would be glad to help in any way. Thanks!

There are a number of software programs that can take a series of photographs and use them to create a 3D model. One that I've been looking at is Autodesk 123D Catch. It's basically a low cost 3D scanner. Just take a few dozen shots from all different angles around your Vespa, feed it into the software and it should work!

Don't know if Sketchup is good enough for you... but if you are mainly after decent renderings, Kerkythea is good and free and works well with it.

And there is Tgi3D, which also "scans" from a series of photographs and exports to SU. http://www.tgi3d.com/index.php?Page=Home

sketchup has a few of them that are decent such as a rally 200, a 90ss, a vbb, px and even an Ape'

I've not played with sketchup. This 'free' inventor for the mac had my hopes up but I found a major flaw in that you can't paramaterize solids. So it is almost worthless for modeling. They must be importing models for all their demos as it would be nearly impossible to draw from scratch those that are used in their training videos.

I was thinking a low-end CAD or modeling system for my macbook air would allow me to play around on planes as I travel extensively. I have another laptop with CATIA on it but that is too large to carry and I can't really work on it unless I'm in first class.

I've been doing 3d CAD/CAM modeling since the early 80's. I use 130k of software to drive my machine tools but thought others would want to join in as somewhere 50yrs from now the Vespa parts will be gone so we might as well start now if we want to preserve the legacy.

I get carried away all the time. My friend's son asked me to help him make a pine wood derby car. I spent 300 hours drawing a Ferrari Enzo over 18 months, long past the derby date that year. When my kids came to age, I was in the position to cut the cars on my mill. I made three. Highly tuned wheels and axles, they ran pretty fast. We killed everyone in the preliminary pre-race. On the day of the race, we brought the cars in along with my son's friend whose dad is our Reverend. Everyone went crazy over the cars, weighed them in at .01oz under the limit, tested to ensure all wheels hit the ground on a surface plate, etc. Then the judge said "you used our block of wood, didn't you." I couldn't lie, we were ejected. Screaming ensued, threats were made, until the Reverend intervened which was long after my wife left in embarrassment (of me). They let us run in 'open' class where if you don't weigh 5lbs you don't have a prayer. Lost horribly. The car that won looked like some kid carved it with a dull pocket knife, like it should.

Hilarious.

Pine wood derby is one of the few times I wish I'd had a kid or two. As a kid I did a lot of work on the axles and wheels. The standard nails sucked. Just recently realized (37 years too late) that jewel bearings are easy to get and perfect for the application.

I do a bunch of design in Ashlar Argon. I've always sucked at trying to recreate organic shapes. Most my work sticks to function and if it happens to look good that's a bonus. Tried modeling the front of the Vespa LX for making radiator ducting. spent way too long on the project and had nothing useful at the end.

picture of my derby car.. removed.. trying to keep thread on track. pardon the interruption.

here's one of them. they made us use their wheels so they aren't scale. Also, we had to tweak the dimensions. Ended-up using redwood (italian, of course). Not to get off track, but I kicked around the whole jeweled bearing as well. SmallParts.com.

Back on track:

If I design the 3D model using my manufacturing software it would be exact but the rendering on that system isn't as good as others. I usually design in Catia and export to 3dstudiomax or Maya and have played with lightwave.

I'm working on the front suspension first. I'll put the pictures up as I go.

Just another distraction.

Nothing like the programs talked about above, but I thought I would share with you the 3D programs from Pixologic. There is a high dollar software called Z-brush, and a free version called Sculptris. You start with a ball of digital clay, and can add, subtract, etc... I think it is mainly used for animation, concepts, and art.

http://www.pixologic.com/

How far are you thinking of taking this? At first, I was thinking just a 3D model of the extents. Now that you're talking about starting on the suspension, I'm thinking you're going all in, down to the component level. That's a ton of work but it would be sweet when you're done.

btw, fricken awesome job on the Enzo!

If you want to go really detailed and want to do a smallframe, I've got the malossi 135 kit done.

Oops, what did you do that in?

Ashlar Argon. It sometimes has issues when things get really complicated, but I love the workflow.

I gave-up on the other software and you can see how lame CATIA is for rendering. I roughed-out the front suspension link, lots of flaws as the top and bottom have somewhat of a curve to them. I was trying to get a cast look in the material and it came out wierd.

Using this software though, I can simulate the mechanisms which is pretty cool. Never had luck with springs, but levers, pivots, sliding joints all work well.

It is raining so hard here I doubt I'll be riding for a week.

I was riding today. What's a little rain?

I'll be we had 4" of rain in the past 24, maybe double that.

in Texas we have not had more than an inch of rain in over 3 months.

make that in metal looks shit hot



rain wind pffftt

But, of all the things to make? That fork piece? What is the point of that?

No point other than that is one of the pieces in the front that would be needed to play with the dynamic simulation mechanism in the software.

Gotta start somewhere and as I was trying out that auto desk software that part had cylinders and pads that looked interesting. Normally if I were to build a model for rendering I would have started with the frame.

I thought I would takle something a little tougher and got a start on the headset cover (or whatever it is called). The software I am using doesn't lend itself to modeling very well but I would be able to 3d surface machine it. It's not quite right and I split one of the surfaces and this software doesn't have a stitch function like others so I might have to start over anyway. The concept is you design one side and then mirror it over to the other.

Hey CNC, don't back away from the fork link.










Anybody that says shit is just jealous.

This shit blows my mind! I know nothing about it, but I like it! TALENT!

I wanted to move away from the front suspension (since all those parts are easy) to something that would tax the modeling and assembly capabilities of my CAD software. 3d Studio Max, my usual modeling software, is much, much easier to model complex shapes in but you wouldn't be able to machine them as the surface isn't parametric unless you use their NURBS surfaces. Once you use their NURBS then their software doesn't handle trims, filets, etc well at all.

So this was my first try at something as complex as the headset cover using subdivision surfaces. You can see that the surface is not going to work as my initial cruciform primitive ended-up with a knot at the intersection of the right side with the slope of both the speedo table and the angled flat panel above it. This knot will prevent me from creating the surface accurately.

When I went to CAD/CAM class in the early 80's I remember the instructor telling me over and over "you can't machine that" which was true then but with 3d printers, today you build about anything if you can settle with the accuracy limitations.

As far as the headset, in our case we have one that has been around 35yrs or so. If I had the equipment, I could scan the part and make a point cloud. The point cloud would be used to make surfaces for a computer model. Those surfaces are limited in that they are very difficult to modify without impacting other adjacent surfaces. You end-up with disjointed surfaces that even though they look like they flow together, they don't mathematically. So your machine program will have subtle discontinuities that might or might not be visible in the finished product.

By stretching and pulling the points, lines and faces in the headset model, I can pretty much copy it and as long as I stay with one mesh, the machine programs will flow perfectly.

I did a couple simple models in 3DS Max for some animation tests (not super realistic):

Very cool Mr.Moi2000! Is there a place that we can watch the animations?

I think I am going to scale the headset cover down and machine it out of nylon. I took another stab at drawing it. This one is a little more accurate on the top surfaces.

The green line is the tool path.

I like what you're doing... just wondering to what end are you doing it?

for the hell of it I guess. I drew cars as a hobby for a long time, switched to 3D modeling, ended-up making some of the models in part.

This 250-GTO I drew was on its way to being a 1/12th aluminum shell. I had machined a few molds and was going to press the body panels but work intruded. I did make some parts out of fiberglass which was a challenge.

I never got that good at modeling but I made-up for it in machining. I had some guy that was badgering me for a while for the enzo machine files. He had me scale them up so he could make knock-off body panels for kit cars. He had a huge 8 foot by 12 foot CNC router that was used to make foam molds. He lost interest when I discouraged him with the price. He cut the hood as a test which looked pretty cool, but I wasn't into it since the last thing I wanted to be part of was a knock-off business, espeically Ferrari's that I love. I wouldn't mind having an aluminum headset cover for my P200 though. Maybe I'll take a shot at one.

Today I designed a small dish for dipping bread in oil. The area at the bottom is the same shape as the snowflake wrench I make. Having it 3D printed in glassed ceramic.

I was going to offer to cut it for you from brass. Glassed ceramic? Is that an expensive process? Contacts?

Not that expensive at all $0.18/cm^2

http://shpws.me/mQK1

This is the first time I've done anything in ceramic, use nylon for most things.

I'm roughing out the frame. Was raining all last week.

An interesting project..!

If someone has detailed views (dimensioned prints if they existed!), it would be fun to model abunch of scoots and make desk weights via SLA or maybe even a chess set! Vespa 50's as pawns, SS90's as knights, Li125's as bishops, etc.!

I use SolidWorks almost daily, would be easy to create a model...The hard part is measuring! But I doubt I am telling you anything you do not already know...!

Hi,
I too, find this 3D modeling very interesting. You stated "If someone has detailed views", how do you mean this? As in pics? Sorry, if this seems all too naive of a question. I have a 66 BB/Sprint 150 in parts and primered ready for final paint. Can pics or measurements help?

That's nice Ken. What color are you going to paint?

Hola amigo! It'll be the OEM silver. I was able lift a sample from an
un-molested section on the top of the tank.

I was about ready to plop down money for the NextEngine 3D scanner but found the scanned output nearly impossible to work with. My expectations were low, just wanted to be able to use the scan as a scaffolding to guide the new model. This would have let me build parts that fit existing objects.