Find us on Google+ Proving a Concept - My 3D Printing Test ~ Inventor Tales

Wednesday, August 05, 2015

Proving a Concept - My 3D Printing Test

A few months ago, I had an amazing opportunity to try something with a colleague and fellow aircraft enthusiast, Justin Nishitsuji.

He's a member of a team restoring a Spanish variant of a German BF-109 fighter at Planes of Fame Air Museum, where I'm also a volunteer.  And he posed a question to me.

An example of a restored Bushon. 

The Planes of Fame Bushon under restoration to flight
"Can you 3D print this part?"  Justin shows me an aileron horn.  "I want to see if we could prototype it before making it. The bearing surfaces are worn out."

The original part we started with.  The bearing surfaces were shot. 

Before going further, I'll provide a little background.  A new part had already been machined, so This wasn't going to "revolutionize" the process.  There are plenty of posts and videos that talk about that.  This isn't one of them.

What I was presented with, was the opportunity to prove that 3D printing might be a viable option in in this application:  Aircraft restoration.

It was research and development.  And R&D is often times more the tortoise than the hare.

Looking at the part, the modeling wouldn't be hard.  Its the measurements that would be hard.  There were some compound angles that would have to be dealt with.

I couldn't 3D scan the part, I'm not proficient at it, and I didn't have time to learn it.  No great mystery on why I made that decision!

Machining was also limited to the old school manual knee mill.  There was no making that part in a cloud of chips on the Blastomatic 5000! CNC machine.

Even though we weren't going to machine the part for our test, we had to approach it as if we were.

So I resorted to the old school way of doing things.  Scales and micrometers.

Fortunately, one of the guys had a micrometer that read in metric, because.... German based design!  And I measured away.

Soon, I had the 3D part modeled in Autodesk Inventor.

The part on my Laptop, Autodesk Inventor running on my laptop.
The Bushon is in the background.
Next, I exported a stereo-lithography model and sent it to Stratasys Direct Manufacturing for printing.

In a day or two, I had a quote.  I chose the cheaper option , at about 120 U.S. Dollars, since we were only proving an idea.

After about a week, we had our part!  And while the part came out exactly the way I modeled it, I did screw up one measurement.  The standoff was too long.  Nobody to blame but the guy taking the measurements for that!

Sigh.  Measure twice, cut once.  

The new and the old.  Yes. I know the standoff is too long.  
But here's what we learned from the experience.  And that was the real goal.

For this application, form and fit testing before manufacture still appears to be the strongest case. 

     Not much explanation there.  I know great strides are being made in printing metal, but I'm not sure I want a printed part holding a control surface on just yet.  But for testing to make sure the control surface has proper range of motion?  I'm on board!

Consider leaving some details out. 

      I built the mounting holes into the printed model.  Next time, I'd think about transferring them from the mating part.  In essence, treating the 3D printed part as a blank. The machinist could then take the 3D model and use that to create his part.  That approach might make sure that the new part fits exactly the way they should.

I might have left the mounting holes in the base plate out, and transferred them instead.
Cut the part down to size.

Seeing how I accidentally made the standoff too long, I would probably consider cutting the 3D printed part down.  I know some would consider that sacrilege.  The 3D part should be made right! 
But the point is to validate fit and function.  And at 120 USD a pop, I'm going to think twice about getting another.

Build what you need, not one bit more.

I'd consider building only portions of the part next time.  Perhaps only build the base plate and mounting plate, use a wooden dowel for the standoff.  That would allow us to test the compound angles, and "tweak" the part in.  Remember, we were after fit and function.

I might have left of the standoff, and "tweaked" the part into the right dimensions.
You don't need all the "cool gadgets".

I'd love to have had access to a 3D scanner and a CNC machine, along with the expertise to use them. But it wasn't necessary, and it was no reason not to integrate the technology we could employ.

We were able to combine old school tools (micrometers, calipers, and the Mark 1 eyeball) with the new tools (3D CAD and 3D printing) to reach our intended goal. 

There we go.  It's my small venture into the very real world of 3D printing.  

In this test, we answered some questions, and raised more questions.  

Research begets development.  Development begets further research. 

And the journey continues.  I'm looking forward to where it takes me.

This is a step in the journey.  Not the destination.
Special thanks to Planes of Fame Air Museum and Volunteer Justin Nishitsuji for the opportunity.  I would have never had this chance without them!

Photo Attributes

By Kogo (Own work) [GFDL], via Wikimedia Commons


  1. Very interesting Jonathan. This is where the technology we are using now fits in to "The Future of Making Things". (Nice plug,huh Mr. Bass?)
    It definitely would be interesting to see how well the part would have turned out from a 3D scan. Isn't there an iPhone app that takes multiple pictures and then creates an FBX file to print from? (Autodesk shows a parent taking several pics of his kid and then shows the kid admiring his 3D printed image on their 2014 Showreel.)
    Guess you have now coined a new phrase Jonathan........measure twice, extrude once! Thanks for all your hard work.

    1. MIke,

      I'd love to try a 3D scan. But it was just a case of given what we had at the time, 3D scanning wasn't an easy option.

      It's definitely next on my list though!