3d printing

[hikerbob]

Not sure quite where this should sit.

http://www.3dfuture.com.au/2011/06/3d-printing-a-huge-spanner/

Currently out of the league of most ATM work but easy to see the potential for larger budgets (especially long term space missions). Hopefully a time will come when this stuff will be home viable.

I'd seen plenty of 3d printing before not the scanning capabilities nor the ability to print mechanical assemblies.

Bob

[gary]

Hi Bob,

Like many manufacturers, we have been making use of 3D printing for several years,
specifically for rapid prototyping of mechanical assemblies. These take data directly
from our 3D CAD systems. The high-end systems are still expensive and there are
third party 3D print houses that you simply email your data to and they will send
you back a model. However, with the advent of systems such as V-Flash,
RapMan and the BFB-3000, entry level systems are now well within the reach of
smaller enterprises or home machinists.

There are some DIY solutions available as well for those who enjoy tinkering.

Systems that scan an existing object and create, say, an STL file for import into
a 3D printer are already ubiquitous in industry.

Will they accompany astronauts on future voyages to destinations such as Mars?
It would be certain that planners would be giving them serious consideration.

[multiweb]

My understanding is that a scanner will only scan visible surfaces. How can they possibly scan hidden or internal mechanism without pulling it apart?

[gary]

Hi Marc,

The short answer is that they can't scan hidden or internal mechanisms.

Typically 3D scanners are used to import models for organic shapes
where the user does not have an existing CAD model for the part.
For example, scanning a human face or entire head.

For objects like the shifting spanner in the video, the scanned data would
be imported into the 3D CAD environment and then tweaked by hand
to create the hidden parts and to separate the spanner into its
various fixed and moving components. In other words, the scanner
is simply being used as a convenient way of performing the metrology
on the visible surfaces.

However, in many situations, one would not scan such a part but would
create it as a 3D CAD model by hand.

In the hypothetical scenario of a mission to Mars, one could
always transmit the data from a CAD model from Earth to Mars where
it could be printed.

[hikerbob]

gary I've looked hard at the plastic extrusion printers on a number of occasions (and the possibility with printing with some other substances). Very tempting stuff. I saw a couple of them at the Woodworking show in Brisbane recently.

The setup in the clip took it to another level (assuming that it's all legit) and like Marc I'm pondering how a hand held scanner could manage the internals of a metal object.

Z Corporation's website has a link to the video so it should have some legitimacy, although I didn't see mention in the scanner capabilities for scanning the internals of objects.

Bob

[multiweb]

That's what I thought. The whole video presentation is a bit of a publicity stunt. Not clear either how they can spray the powder in multi layers without affecting contiguous granules/powder. I mean you get the concept from the video but there are a lot of holes in how it's actually done. :question:

[gary]

Hi Bob,

I may have been composing my previous post to Marc at the same
time you were composing your question. So my previous post my
address some of your questions.

In the case of the demo in the video, there would have been a segment
which was not shown where a 3D CAD modeler package would have
been used to create, by hand, those components which were hidden
from the scanner. Likewise, components that need to move would
have suitable physical separations modeled in the CAD package.

Broadly speaking, if your primary interest was creating models of what
the 3D CAD community refer to as "organic shapes", then you might
consider using a scanner if there was some pre-existing physical
embodient of what you wanted to replicate.

For "organic shapes" that you don't have an existing embodiment of,
one might consider acquiring a 3D CAD package that specializes in their
creation. There are some good ones out there, both free and commercial.
A good commercial one can be had in the AUD300 to AUD1000 price range.

For parts such as the spanner or many of the parts one might use in
some functional context, then one might consider acquiring a parametric
3D CAD package. Again, there are some good ones about even in the
sub AUD1000 price range that will output in a format that can be
directly interpreted by a 3D printer.

In other words, rather than investing in a scanner, a good parametric
modeler or if modeling your own organic shapes, a good surface
modeler, such as those that create NURBS models, might prove to be
a more productive investment.

For example, using a good parametric 3D model package, one could
create a model of the spanner from scratch within an evening and
an experienced designer in some fraction of that time.

The materials that the various 3D printers use varies from manufacturer
to manufacturer. As one can imagine, the 3D printed copy of an object
may not be suitable to be used as a replacement for many parts.
One could print a model of a hammer, but its performance for hammering in nails
is likely to fall very short compared to the real thing. :)

The resolution of the printers also differs across the range from the low end
to the high end.

Where they can be very useful is for rapid prototyping. A good example
is to create prototypes of parts that one plans on having injection moulded.
As you can appreciate, the tooling for injection moulded parts is typically
very expensive and the ability to create a 3D printed prototype that you can
hold in your hand and check that it will integrate with the rest of your system
prior to getting the tooling done can save lots of time and money.

Depending what your target parts are, you might also want to consider CNC
machining.

[AdrianF]

I recently attended a Robotics in Industry Seminar through work where they demomstrated a 3D printer printing a cog to fit into a machine. I was impressed. It took approx 21 minutes to print a cog about 75mm diameter cog about 10mm thick. The demonstrator got someone to check the printer and media to make sure that there was no trickery involved.
I think he said that the printer would cost around $4000 to purchase for a desktop unit but TCO was expensive and would not be cost effective for home or heavy use, but perfect for prototyping. I will wait for them to come down to $500 before I buy one :P

Adrian

[hikerbob]

Gary I do play with CNC a bit. I converted a little Seig mill to CNC some time back. Lot's of fun but I still need to master the Cad side of things (and the conversion to gcode). At the moment I hand code the instructions for Mach3 which works well enough for simple jobs but is a pain when it gets more complex.

I wasn't thinking of the gear in the video for my use at the moment, more that I found the capability that they are achieving fascinating and liked the angle of spare parts production for off planet trip's.

Of course if someone wanted to give me one of those printers I'm sure I would find some room for it somewhere.

Bob

[supernova1965]

Can you just imagine if you owned one of these you could make any adapter for your camera or your scope you could possibly need. Now that would be cool:thumbsup:. You could print out Bahtinov masks the possibilities are endless:eyepop:

[gary]

I noted that last night our time 3D Systems (http://www.3dsystems.com/), a 3D printer company founded in 1986 and based in South Carolina, announced that they
have acquired Texan headquartered 3D CAD vendor, Alibre.

Press release here -
http://www.3dsystems.com/content/news/releases/pdf/071911%203D%20Systems%20Acquires%20 Alibre.pdf

Their strategy appears to be to help make 3D content-to-print solutions more accessible to a broader group of organizations and individuals.

With NASDAQ ticker symbol DDD, they certainly look like an interesting company to watch -
http://quotes.nasdaq.com/asp/SummaryQuote.asp?symbol=DDD&selected=DDD