3D Printed Airframe

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

aharonwe

New Member
Joined
Mar 31, 2017
Messages
3
Location
Bonita Springs, Fl. USA
I have been thinking about building a Cozy Mark IV and, also being a bit of a techie, have been noticing the capabilities of 3D printing technology. There are plans for 3D printers with relatively large dimensions (5' high printing capacity, although the width leaves a bit to be desired), for printers which can use carbon fiber and fiberglass. My guess is that it would cost about $2k to build a large acurate printer. It can't be hard to hack one to be larger. They use a simple coordinate system and its just a mater of defining the coordinates to an acceptible deviation and using accurate stepper motors, gears, and cables. Has anyone here had any experience with 3D Printed fiber materials? I can't seem to find any technical specifications on the finished materials. There are also metal printers which can mold aluminum so a BD-4C would also be feasible if the material strength is up to par. I'm not really thinking that this would save build time but for some reason I just have an itch to try it. I would think that the finished product would be more accurate to plans and be a more streamlined aircraft. What are your thoughts or if anyone has any data on the strength of printed vs. hand laid materials I would love input.

Thanks,

Aharon
 

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
8,812
Location
North Carolina
To get good properties from composites, you need a high fibre fraction and long fibres. 3d carbon material has a low fibre fraction and short fibres. It'll be a jump up from unreinforced material, but I doubt it would reach strength of Al, or even the strength to weight of Al.
 

aharonwe

New Member
Joined
Mar 31, 2017
Messages
3
Location
Bonita Springs, Fl. USA
To get good properties from composites, you need a high fibre fraction and long fibres. 3d carbon material has a low fibre fraction and short fibres. It'll be a jump up from unreinforced material, but I doubt it would reach strength of Al, or even the strength to weight of Al.
What about 3D printing with aluminum? Say a BD-4C frame, wings, and skin? No bolts or rivets would be needed so I would think it would have great aerodynamic properties. Again, I'm just looking for pros and cons of my idea. I'm sure that if I built a printer and the materials are strong enough that this would work but I'm having trouble finding any specs on materials that have been printed.
 

aharonwe

New Member
Joined
Mar 31, 2017
Messages
3
Location
Bonita Springs, Fl. USA
To get good properties from composites, you need a high fibre fraction and long fibres. 3d carbon material has a low fibre fraction and short fibres. It'll be a jump up from unreinforced material, but I doubt it would reach strength of Al, or even the strength to weight of Al.
Thanks for the input.

What about 3D printing with aluminum? Say a BD-4C frame, wings, and skin? No bolts or rivets would be needed so I would think it would have great aerodynamic properties. Again, I'm just looking for pros and cons of my idea. I'm sure that if I built a printer and the materials are strong enough that this would work but I'm having trouble finding any specs on materials that have been printed.
 

lr27

Well-Known Member
Joined
Nov 3, 2007
Messages
3,822
I'm sure it would be possible, but more complicated, to make a 3D printer that could lay down carbon and/or fiberglass tow. However, that might make design trickier. OTOH, taking advantage of the flexibility might result in lighter structure and cleaner shapes. Maybe this technology already exists?
 

tspear

Well-Known Member
Joined
Feb 12, 2014
Messages
834
Location
Outside Boston
A boat builder told me about a large format printer used to make exotic one off racing boats.
He said it prints a "single continuous strand" of carbon or kevlar fiber. However, it is super slow. No idea how it works....

Tim
 

SuperCritical

Member
Joined
Mar 1, 2011
Messages
15
Location
Castle Rock CO, USA
Thanks for the input.

What about 3D printing with aluminum? Say a BD-4C frame, wings, and skin? No bolts or rivets would be needed so I would think it would have great aerodynamic properties. Again, I'm just looking for pros and cons of my idea. I'm sure that if I built a printer and the materials are strong enough that this would work but I'm having trouble finding any specs on materials that have been printed.
My former employer (US defense contractor) went through a period where they were investing heavily in 3D printing. We had big machines to print plastics some other machines which sintered aluminum, inconel, titanium, etc. Before long the executives were begging the Engineers for items to 3d print so they could justify the investment. It turned into a huge money drain and the machines sat idle the majority of the time. Reasons:

1. Part to part variability is very high. You could make multiple identical parts and the properties could vary by orders of magnitude with no obvious signs of weakness. This was a deal breaker for any flight critical parts.

2. Non-critical parts (think air nozzles in your car) could be made but the production rate was so slow it wasn't worth the investment in the machine when compared to out sourcing the manufacturing to a traditional shop.

The biggest benefit was in prototyping things to see how they fit together or making jigs to assemble traditionally manufactured parts in the factory. These were single item runs and strength wasn't an issue.

For a home built aircraft #1 is the deal breaker.

A couple last minute thoughts I also had:
- The minimum printed thickness is actually pretty thick by aircraft skin standards
- Even thick parts are relatively porous. Printed Ultem and PEEK parts would not hold air pressure even when 1/8 inch thick.

No experience with composites printing so that may be different.
 
Last edited:

ScaleBirdsScott

Well-Known Member
Joined
Feb 10, 2015
Messages
1,281
Location
Uncasville, CT
Printers like the MarkForged will lay down continuous fibers along the perimiters of the part.

Problem 1: The printers are tiny (you would have to build a massive moving gantry CNC machine, and then adapt the printer's head to it)

Problem 2: The material is super expensive. Where a regular 3D printer spool of ABS might be $30 for a 800cm3 spool of filament, the materials for MarkForged are around $175 for an equivalent 800cm3 spool. This means at least 5x the materials cost. Something anywhere close to the volume of an airplane in any regime will be likely tens of thousands of dollars in materials cost alone if you were trying to use that filament.

Even ABS, and very thin wall, I bet one could spend thousands on the production of a full-scale aircraft shell. You're talking huge surface areas.

Problem 3: Unless one is using a custom spool system that has the kilometers of filament (average spool is about 30m) needed to complete the parts in a single run (I'm sure those exist but aren't common for hobby builders) you'd end up with discontinuity which means the program will be frequently pausing, or stopping, for reloads. Also, if there ever is an error, filament jam, layer seperation, anything, the printer needs software and hardware solutions for being able to restart a print at a given layer, maybe detection for bad extrusions, etc, so that one doesn't get halfway through a massive 100 hour print for a cowling or what-not and it starts making a birds-nest.

Just totally get the aluminum out of mind for the next decade at least. And I'd go out and say you'll almost never see any large scale vehicle hulls or volumes 'printed' as such from alumium, with maybe exceptions for Boeing or Airbus or on some orbital assembly platform making space-stations. It won't be impossible to make a massive aluminum printer but there's no way that tech is going to be cheap at all for a scale the individual could ever justify. The costs just aren't going to scale enough I don't think. For a bracket or bulkhead, a landing gear yolk or a fork or maybe some kind of spar-attach bracket sure. Especially if there's ways to take these printed aluminum parts and then fire them into being a single crystal with near the same properties as if it was billet or forged aluminum, then I see massive applications in the coming years. But again those will be for people with the space and money to have massive autoclaves and massive aluminum printers and do the massive engineering required and the massive coding efforts to ensure perfect parts.

So 3d printing robots making rotating space-stations from harvested asteroids? Perfectly plausible. A guy designing an aircraft hull to be printed in a small shop from aluminum? Plausible, but by the time it's remotely viable there will likely be more advancements with composite printing so it'll be a bit moot.

I don't want to be that person that poo-poo's the ideas and I personally am hopeful that someday soon I'll be able to fire up a custom 3D printer and make at the very least a structural core that I can use in conjunction with hand-laid CF and then vac bag the whole thing for a even resin infusion structure. But what's put me off is the many small problems above which apply to any attempt at large-format 3D printing. If it was any one issue, and the rest were solved, then it's not so bad. But there's a bunch of issues all coming together so its like, where's the benefit?

As it is so far, I'm going to stick with using a large CNC router to 3D carve foam into plugs for composite layup. The machine is known, subtractive milling is well understood and software exists to readily program and operate the machines, foam is relatively inexpensive, and the result is a composite shell with well-understood properties and ones proven for aircraft applications.

I will say for certain details and features, I might consider 3D printing an insert to add detail to a large-scale part or in corners or pockets where milling isn't the best approach. Those would have to then be left in palace or melted out with acetone, or some combination of the two. But in general for one-off aircraft I see a lot of benefit to this method.

Until the printer can yield faster, cheaper 'plugs' then it's going to be a novelty, and it will be some time before whatever it prints is better than a base on which to apply a real structural material.
 

tspear

Well-Known Member
Joined
Feb 12, 2014
Messages
834
Location
Outside Boston
Now I wonder about a half/half solution. If you have ever seen videos on how Beech (now Textron) made the fuselage for the Premier, or Boeing making parts of the 787, or Airbus the wing spars for the A380. All each case, they are winding around the a base structure to apply CF.
I am sure I am missing something, but I am wondering if using CNC to cut foam cores and then wind CF around it would be a viable "home" solution. Probably easier to hand apply mats, but now I am curious....

Tim
 

cheapracer

Well-Known Member
Log Member
Joined
Sep 8, 2013
Messages
6,312
Location
Australian
Now I wonder about a half/half solution. If you have ever seen videos on how Beech (now Textron) made the fuselage for the Premier, or Boeing making parts of the 787, or Airbus the wing spars for the A380. All each case, they are winding around the a base structure to apply CF.
I am sure I am missing something, but I am wondering if using CNC to cut foam cores and then wind CF around it would be a viable "home" solution. Probably easier to hand apply mats, but now I am curious....

Tim
Why not, I have built car panels on a number of occasions by laying steel tube where required and winding rope around the parts to form a shape, then applying resin to the rope. These were only patterns for molds, but very successful technique.

My computer has some pictures of some of that process, but darned if I can find them, have another look later.
 

aviast

Well-Known Member
Joined
Jun 4, 2011
Messages
153
Location
Brisbane, Australia
3D printed submarines for $60k... https://youtu.be/8UY15kDVUek

They've chosen an interesting cross-section. Why wouldn't you just go with a cylinder to minimise the structure required to deal with the loads?

Reading back over the thread.... so the printed material may not have the right properties for an aircraft, but could you print a mold to use for a traditional composite aircraft?
 

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
4,302
Location
Warren, VT USA
I'm sure it would be possible, but more complicated, to make a 3D printer that could lay down carbon and/or fiberglass tow. However, that might make design trickier. OTOH, taking advantage of the flexibility might result in lighter structure and cleaner shapes. Maybe this technology already exists?
It's called "Direct Fiber Placement" and Boeing, Northrup, Airbus, etc... have been doing it for a long time. But it is cubic dollars. Lookup mTorres.
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,765
Location
Rotterdam, Netherlands
Reading back over the thread.... so the printed material may not have the right properties for an aircraft, but could you print a mold to use for a traditional composite aircraft?
That is one application where it makes a lot of sense. You'd have to machine it back anyway though for sufficient surface quality, so that decreases the advantage a lot.

Better to use subtractive technology. For big molds: Machine big XPS blocks and undersize them. Cover them in CFRP and put a thick layer of tooling gelcoat on the surface. Machine again, but this time with a very high resolution.
 

Victor Bravo

Well-Known Member
HBA Supporter
Joined
Jul 30, 2014
Messages
8,957
Location
KWHP, Los Angeles CA, USA
Once again (everyone now humming in perfect harmony...) what is the mission you are trying to achieve? Fastest build for one prototype aircraft? Fastest production for 100 aircraft? Cheapest production? Win award for maximum and most complex use of technology in aircraft prototyping?

Filament winding a fuselage for a "normal" small airplane is great, and is indeed being used. but the setup and programming of the machine that lays down the continuous fiber is huge. The machine is expensive to rent.

If this is a "reality based" discussion rather than a theoretical one, then I believe I'm safe in saying that the best use of 3D printing for your example (Cozy) is making molds in which you can vacuum bag the skins for the wings, strakes, fuselage shells etc. Those molds (even if they were not perfectly smooth gel coat) would allow you to build a sandwich construction Cozy instead of a solid core moldless Cozy, and there is possibly significant weight to be saved and stiffness to be gained by doing this. The airfoil is probably thick enough to where hollow core sandwich skins will save weight. The strakes are definitely thick enough, and since they are used as fuel tanks you stand to get better control of your inner skin and fuel barrier by bagging the strakes. Using 3D printed molds you can also much more easily "camber" the strakes like some of the Long-EZ guys did, and with hollow sandwich your cambered strakes will hold a little more fuel and provide a little lower drag, as well as getting cleaner air into the propeller.

We have at least one HBA participant (Marc Zeitlin) who is very knowledgeable on the Cozy, and he will surely be able to verify or de-bunk the ideas that I have mentioned above about the Cozy.
 
Last edited:

cheapracer

Well-Known Member
Log Member
Joined
Sep 8, 2013
Messages
6,312
Location
Australian
Filament winding a fuselage for a "normal" small airplane is great, and is indeed being used. but the setup and programming of the machine that lays down the continuous fiber is huge. The machine is expensive to rent.

.
I actually have some related experience in this methodology.

The picture here, is of a Cold Isostatic Press and pump (100,000 psi).

The large blue double ended arch is made as the new one on the turntable is being done in this picture, 2 end caps and steel columns for sides are being built up by having high tensile flat wire wound around them. This method is stronger and half the weight of just welding steel parts together.

CIP winding.jpg



The blue cylinder vessel inside the arch is made by starting with a steel bobbin on the rotisserie (red arrow) and again winding high tensile wire around it, rows going back and forth just like cotton on a bobbin.

The secret is in the (patented, but simple) mechanism setup holding the correct tension as the arch increases and decreases in it's swing, as does the bobbin as it grows larger with each wind.

I am mentioning this because the same principle could be applied to winding on fiberglass using a polystyrene dummy for fuse and the wings. As the strand, or strands wind on they might go through a wetter. From my experience with the system above, I don't see it's that difficult to set up for a medium level venture.
 

tspear

Well-Known Member
Joined
Feb 12, 2014
Messages
834
Location
Outside Boston
I would think stated goals would be critical.
For me, I think the following goal applies:
-- Reduce time and/or costs for the builder.

And for the kit company, you have the following constraints:
-- Low volume
-- Minimal capital available
-- Possible some aversion to risk

Therefore, at this point I think 3D printers are really useful for trim and non-structural parts; especially interior parts.

Tim
 
Top