3D printing

[H-viken]

Hi

Anyone dabbled with printing spare parts for their old cars? Or printing their own upgraded parts?

Attended a workshop today where we looked at parts made out of among other things Titanium and Aluminum/Scandium alloys. Pretty cool stuff!

This will clearly take off in the next couple of years and I think it would be fun to collect what we come up with in one location.

/Johan

[304065]

Never done it. Application #1 in my mind would be MFI space cams-- with a software package to translate AFR curve into the bumps on the cam. Bosch did this 50 years ago with slide rules so there's no reason why a simple algorithm can't do the same.

[Olivier]

Great topic!
I am also very interested in 3d printing, and we used a couple of these machine in a previous job (plastics in both powder and resin forms).
We made some functional parts for engine application, like intake manifolds, electronics devices casing, cooler shrouds...
I know formula one teams are using some 3d printed titanium bits with hydraulic circuits in them.

The problem, for classics application, is that most of the parts were fashioned in an old school way, castings and so on. They are difficult to redesign (as you have to produce a 3d cad model first), and 3d printed parts usually have a rough surface finish. Will post some picts of pattern I had made. And it is even more true with metallic parts.

For the space came, you would need to have some finishing machining to do. It would be just more cost effective to do it the way it was done. I think it was done like the camshaft were made: a master profile, and a copying grinding machine. In this specific case, what could be used would be 3d scanning, and then cnc machining.

One thing that could be easilly done for 911s would be plastic stacks, and maybe throtle bodies. But that wouldn t be cheap!

Olivier

[Silverbullit]

Never personally experienced 3D printing, but witnessed several examples at SEMA last month. Utterly unbelievable.

A miniature, smooth-finish motor with 8-32 threads "printed" in…not machined. Awesome

[37yrold911]

...

The problem, for classics application, is that most of the parts were fashioned in an old school way, castings and so on. They are difficult to redesign (as you have to produce a 3d cad model first), and 3d printed parts usually have a rough surface finish. Will post some picts of pattern I had made. And it is even more true with metallic parts.


Olivier

Is there any way to have an existing part scanned to automatically create the 3D model?

[Flieger]

The cheap plastic DIY ones I've experienced are absolute crap. Same for their laser scanners. The high-end ones are amazing though. They use direct metal laser sintered (DMLS) Titanium 6Al-4V parts in aerospace applications. Appears the bulk material properties are just as good as billet but the surface needs to be electropolished or otherwise smoothed to provide a good fatigue life. All that is great, but the end part is super expensive so I prefer to design for simple billet parts since I'm not in aerospace where mass is everything and safety factors are cut closer.

[Olivier]

Is there any way to have an existing part scanned to automatically create the 3D model?

To my knowledge, this is not yet possible, but, this step is fonction of differents factors, like scan quality, complexity and surface rougnhness of the original parts...
It requires some some expensives softwares, and a certain amount of know how to post process the scans datas.

FYI, I was struggling with my scans to produce the 3d models. I ended with using an India based company to process my files. It was quick, efficient and cheap! Was very pleased with the results.
If you have a precise project in mind, do not hesitate to contact me.

And to aswer the previous post, yes, it is an expensive process, and no, materials have not exaxctly the same caracteristics, but with specialised heat treatments, you can get very close, or even better than castings with metal parts.
As an exemple, Eurocopter is using plastics printed parts for aircon ducting.
This is yet to be developped further, but we will see more and more parts in the future, as it opens a whole new world of possibilities.
Another stunning thing with this techno, is that its now possible to produce directly sand castings molds, it is cold direct casting or rapid castings. Wish I could try it!

Olivier

[jpnovak]

I have had a 3D printer for about 1.5 years. They are great for certain things and have lots of limitations. You must understand these limitations to take advantage of the process.

There are three main types of 3D printers. Extrusion, liquid polymerization, and DMLS (mentioned by Flieger). Each of these takes a small form factor raw material and uses a positioning system to organize into a 3D object. They are also ordered in increasing complexity and cost.

Extrusion process melts a plastic filament and squirts it out like a tube of toothpaste. This is what is actually affordable to the average user for a few hundred or thousand depending on print volume. An XYZ positioning system moves the nozzle around and puts the plastic layer where you want it. The part is assembled in layer slices like a stack of Pringles Potato chips. The layers need to be somewhat contiguous and connected so the part has to be designed and oriented in an effort to enable the printing process.

I would not use parts made in this process for any structural application. they are just made with ABS plastic (or other similar type including Nylon, PLA, etc). I use my printer to prototype parts and check for fit. They amount of time savings compared to prototype runs at a machine shop is tremendous. I have made twin plug adapters, fuel injection adapters, Intake adapters, guage mounts, clock block offs, and a host of other things. I particularly find it useful for sizing jigs. Here is an example I posted last year for MFI linkage jigs.

http://www.early911sregistry.org/for...hlight=printer

My printer is now at my office doing the same thing. prototyping small parts, layout jigs, assembly jigs, etc. I work for a startup building advanced Li battery packs doing application development. Even printing a PCB layout footprint saves us a lot of mechanical design time by being able to quickly have a tangible object to check the fit of a CAD program.

Once any of these parts has acceptable fit then you move to a machined or cast version. There are water soluble plastics and polymers that can be printed and these can be used for cast molds. However, the surface finish is generally not where you need it for a finished part.

the Sintererd Polymer machines and DMLS are significantly higher cost and complexity. I have not had the chance to use one.

3D laser scanning does exist and is in the cost realm for home use. The resolution is not great and they do need some smoothing in CAD to be able to achieve great results. A direct laser scan to 3D stereolith (.stl) surface map file is highly routine at this point.

[37yrold911]

I'd like someone to 3D print some rear turn signals that are lightwt., have LEDs in them, and look like stock - not R type.

[Eminence Gris]

I'd like someone to 3D print some rear turn signals that are lightwt., have LEDs in them, and look like stock - not R type.

I've been thinking about this for a few years - self mounting; bypassing the need for any housing... Could be done with LED arrays epoxied to the lens itself.