PROJECT: Scanning, Editing & 3-D Printing

[AnonyMouse]

Hi folks. Would very much like some informed input on a project I'm considering.


I'm a DIY'er and hobbyist, and I'm also retired, so costs are important. To keep costs down I realise I'm going to have to learn to do certain things myself.


The project I'm contemplating at the moment is to clone certain automotive interior trim parts. The vehicle is 30+ years old so replacements for damaged parts are not that easy to come by.


My plan is to remove the parts from the car, get them scanned by a 3-D service bureau, edit the resulting image to correct the damage, then have it 3-D printed. Finally the item can be painted or covered, though at the moment I'm leaning towards flocking.


So a few questions:

1. Is it a feasible plan overall?

2. I understand that the scanning will provide me with a .STL file. What simple (and preferably free) application can I use to display the STL file and to edit the image to correct any damaged areas?

Thanks for your assistance.

[Hugues]

Hi there,

Yes it's feasible.

If you plan to do many parts like this, consider buying yourself a 3d scanner, it will be paid fast. The new Einscan1 from Shining3d is below 1000 USD. Search this forum for more details.

To modify/repair an STL file, Meshmixer is doing the basic stuff and is free. But you'll want to do more and not be constrained to STL files, then 3dcoat becomes interesting.

[Nihonddd]

Feasible yes, but not with out a pretty steep learning curve. I started out about 5 years ago with the same intentions to scan, edit and then 3d print.
You are going to need a decent PC with a fair bit of ram.
You will find that the scans are large files, unwieldy and difficult to edit because of the number of vertices. 100 of thousands if not millions.

If you want to go ahead I would recommend that you learn Blender. It is the most versatile and powerful modelling tool available for free. You can import your scans as Stl's and convert them to blend files that you can then retopologize and edit on a manageable scale. This way you will be editing 10's if not hundreds only of vertices.
Have a look at blender and if you have any questions I am happy to help further.

[curious aardvark]

If you get it scanned by a bureau I would expect all the crappy tidying up to be done by them.

You should end up with a fairly clean useable file - or what the hell are you paying for ?

Now, depending on the size of parts you want to replicate you might consider an all-in-one scanner and 3d printer.
there are a few around, cheapest coming in at around $800
The xyz Davinci aio - by all accounts does a passable job of being a fully functional 3d copier.

A more expensive all in one is the Zeus - runs at around $2500.

I will say that getting parts printed by a service is bloody expensive, add in the scanning costs and I reckon you could easily justify a davinci aio.

[awerby]

I don't see one of those little all-in-one scanner/printers being very useful for scanning car parts. The sort of trim pieces the OP is talking about are not going to fit in the small part envelope they've got. You're going to need a different sort of scanner, or to use a service bureau (which can be costly). Once you've got a scan, even a cleaned-up scan, it's not likely to be perfect; there are usually some artifacts from the scanning process that are introduced.

If you're trying to reproduce mechanical sorts of parts, there's a reverse-engineering step that's advisable, where you use the scan data to create a clean CAD model that retrieves the original design intent. That way, holes will be truly round instead of irregular, and other features will be similarly accurate. Software that does this is available, but it's not cheap.

Conditions inside a car, particularly in areas subject to extremes of climate, are going to be hard on 3D printed parts, which tend to be based on low-melting-point thermoplastics. Temperatures in excess of 150F, which can occur in cars parked in the summer sun, will degrade many plastics, and freeze-thaw cycles aren't good for them either. There are also some issues to research when considering the adhesion between the plastic surface of your printed parts and the paints and/or flocking you plan to apply, particularly if these surfaces are subject to wear.

So yes, this plan is feasible, but it might not be economical.

Andrew Werby
www.computersculpture.com

[Nihonddd]

If you want to keep cost down, avoid much of the advice suggested on this thread your project is going to spiral in cost. Avoid investing large amounts of money. Invest your time instead.

Don't buy expensive proprietry scanners. Avoid the service bureaus they charge big bucks. If you have to get your object scanned by a burea avoid the in house post production work it is going to cost you.

There are other ways. There are open source solutions.

Check out 'Davidlaserscanner' if you don't know it already. There is a large forum and someone may be able to help you there.
Learn blender or a CAD program. As Awerby above suggests reverse engineer this, learn to retopologize and clean up the scan yourself. It's the only economical way to go.

[Hugues]

If you want to keep cost down, avoid much of the advice suggested on this thread ..

...Check out 'Davidlaserscanner' if you don't know it already. There is a large forum and someone may be able to help you there.

??
David laser scanner is more expensive than Einscan and is less accurate and flexible. A little research before advising others and challenging others proposals.

[dklassen]

The David SLS-2 is a much more proven and reliable product.

[raysspl]

The AIO Robotics Zeus all-in-one 3DP has served us fairly well.