Open Source Aircraft Design for CNC?

[Jeffrey Meyer]

Hi Folks,
I also want to join this thread - so here's my first 2 cent salvo:
* Like jim.bo I will gladly contribute files from SolidWorks with which I have considerable experience.
* In addition I also have some aeronautical/mechanical engineering experience and will gladly attempt to answer questions in that area if I can.
* It seems to me that the baseline for this group focuses on 2D CNC machines powered by stepper motors & drivers - routers, laser cutters, punches, etc.. It's important to emphasize the 2D (as opposed to 3D) because the price of 2D equipment and software is an order of magnitude cheaper and less complicated than 3D equipment. Also, the home CNC machines/kits almost universally can accept DXF files which are essentially 2D. There is an abundance of cheap/free software which can generate DXF files. 5 minutes on Google will find you stepper motors, stepper motor drivers and software to translate DXF into CNC G-code.
* SolidWorks is a great tool for creating geometry, but IMHO the big issue is where to get the geometry to translate into DXF. I seriousely doubt that the scratch-from-plans/kit manufacturers will cooperate in supplying CAD geometry (2D or 3D) or allowing us to digitze their plans or reverse engineer their parts for our own group use. It's a copyright issue. Nobody can stop me from making a copy of a product for my own personal, non-profit use - but it's a different story if we as a group make several copies thereby reducing the manufacturer's profits on his intellectual property - maybe I'm wrong but at the very best it doesn't seem ethical and is probably not legal. I don't know - I'm not a lawyer.
* Sorry to be a party-pooper but I do have 2 suggeted ways out: 1. Rethink the option of design-from-scratch suitable for home CNC, or 2. Negociate a deal with one of the scratch-from-plans/kit manufacturers. Number 2 is a subject of give-and-take (both sides have to feel that they gained something) - so what do we have to give that would be of benefit to the manufacturer? On the other hand number 1 is not as daunting or as crazy as one might think. Based on the adage that comes from the R/C modeling world: Attatch a big enough engine to your kitchen table and it will also fly. Yes, it won't be the best flyer - but it will be quick to build and cheap. The disadvantage comes from another adage: A camel is defined as horse designed by a committee.

My 2 cents seems to have expanded into $2 - so I'll stop here.

Jeffrey

[Sonex1517]

I can't add much of value to the conversation except to state THIS IS AN AWESOME IDEA!

[Dana]

We had a discussion like this over on homebuiltairplanes.com a while ago. The general consensus was that the OS model makes a lot less sense for hardware than it does for software, where anybody can test the latest build without any significant investment. With an aircraft design, it's much more difficult... who approves the latest "official version? And how do you approve a design change without building and verifying a prototype? But to wait for a prototype to be verified could take a long time...

It might make sense for an "orphaned" design that's static. The OpenEZ project is one such attempt. The VPs might be another; I'm less familiar with that. A good choice might be a Quicksilver ultralight... many have cloned them, and the Quicksilver company no longer makes a true 103 legal version.

[Brian E. Evans]

Before I retired I used to program sheet metal files for CNC routers, Both stepper and servomotor kind. As somebody has already said, it is quicker to lay one out by hand if you are only making one part, on the other hand, parts like ribs and gussets on such a machine can save hours and be a godsend. I think the part of this that has to be looked at closely is the 51% rule. If you program the machine, run it and the parts are yours, then I think there is no problem. If somebody runs the machine, and makes these parts , then sells them or gives them to you, then I think you have just stepped into the area the FAA is looking at very closely right now.
Brian Evans.

[JNicol]

Hi Folks,
I also want to join this thread - so here's my first 2 cent salvo:
* Like jim.bo I will gladly contribute files from SolidWorks with which I have considerable experience.
* In addition I also have some aeronautical/mechanical engineering experience and will gladly attempt to answer questions in that area if I can.

Hi Jeffrey,

Thanks for that, I will send a PM with my contact details.

* It seems to me that the baseline for this group focuses on 2D CNC machines powered by stepper motors & drivers - routers, laser cutters, punches, etc.. It's important to emphasize the 2D (as opposed to 3D) because the price of 2D equipment and software is an order of magnitude cheaper and less complicated than 3D equipment. Also, the home CNC machines/kits almost universally can accept DXF files which are essentially 2D. There is an abundance of cheap/free software which can generate DXF files. 5 minutes on Google will find you stepper motors, stepper motor drivers and software to translate DXF into CNC G-code.

You got it, with the abundance of open source hardware projects dedicated to CNC, from gantry systems through to controllers as well as software, this is a great time to jump in and see how we can leverage it for homebuilder use. Now just to add a little to the conversation, there are software applications out there that can slice 3D models that can be cut and carved in 2D slices and then re-assembled into 3D objects. (www.vectric.com) Think of now cutting out a plug or mold for composite building.... (cowls, fuel tanks, wings, fuselage etc). Also maybe a windshield form that is assembled from foam slices and then the acrylic/lexan heated and molded over that form... now think of using that same cnc machine to build a chair and ottoman out of wood to sit in and admire all that work....

* SolidWorks is a great tool for creating geometry, but IMHO the big issue is where to get the geometry to translate into DXF. I seriousely doubt that the scratch-from-plans/kit manufacturers will cooperate in supplying CAD geometry (2D or 3D) or allowing us to digitze their plans or reverse engineer their parts for our own group use. It's a copyright issue. Nobody can stop me from making a copy of a product for my own personal, non-profit use - but it's a different story if we as a group make several copies thereby reducing the manufacturer's profits on his intellectual property - maybe I'm wrong but at the very best it doesn't seem ethical and is probably not legal. I don't know - I'm not a lawyer.

Yes, that is why the plans would likely need to be designed from scratch with the open source model in mind from the start.

* Sorry to be a party-pooper but I do have 2 suggeted ways out: 1. Rethink the option of design-from-scratch suitable for home CNC, or 2. Negociate a deal with one of the scratch-from-plans/kit manufacturers. Number 2 is a subject of give-and-take (both sides have to feel that they gained something) - so what do we have to give that would be of benefit to the manufacturer? On the other hand number 1 is not as daunting or as crazy as one might think. Based on the adage that comes from the R/C modeling world: Attatch a big enough engine to your kitchen table and it will also fly. Yes, it won't be the best flyer - but it will be quick to build and cheap. The disadvantage comes from another adage: A camel is defined as horse designed by a committee.

I would say that we should probably proceed with number 1 and see where we get to. If a plans supplier wants to donate for number 2, then that would be great, but again, even if we go with number 2, there would still need to be re-design for CNC manufacture and assembly as there are few designs out there if any that would comply with the overall vision that we have to make this as quick and simple as possible with associated instructions, video and content we are ultimately trying to achieve. This will become apparent as we begin to go public with this..... more to follow.... :-)

[JNicol]

Before I retired I used to program sheet metal files for CNC routers, Both stepper and servomotor kind. As somebody has already said, it is quicker to lay one out by hand if you are only making one part, on the other hand, parts like ribs and gussets on such a machine can save hours and be a godsend. I think the part of this that has to be looked at closely is the 51% rule. If you program the machine, run it and the parts are yours, then I think there is no problem. If somebody runs the machine, and makes these parts , then sells them or gives them to you, then I think you have just stepped into the area the FAA is looking at very closely right now.
Brian Evans.

That is a good point. In Canada, you can "build your own aircraft, provided you construct, assemble or oversee the construction, by other persons, of at least 51% of the total number of items assembled during the project". By overseeing the construction of these parts, or assembly, you are in the clear. In the US, the major portion rule would apply.... I am not a lawyer and far from expert in this area so don't know. I would see that supplying the components as part of a "quick-build" option following the lead of other kit manufacturers might be the rule of thumb if anyone wanted to supply the parts for the open source aircraft.

[Jeffrey Meyer]

That is a good point. In Canada, you can "build your own aircraft, provided you construct, assemble or oversee the construction, by other persons, of at least 51% of the total number of items assembled during the project". By overseeing the construction of these parts, or assembly, you are in the clear. In the US, the major portion rule would apply.... I am not a lawyer and far from expert in this area so don't know. I would see that supplying the components as part of a "quick-build" option following the lead of other kit manufacturers might be the rule of thumb if anyone wanted to supply the parts for the open source aircraft.

I don't think this is an issue and I'll illustrate by taking it to an extreme: It's conceivable that a future homebuilt kit will consist of a package containing raw materials and a CNC machine. All the builder has to do is set up the machine in his living room, feed in the raw materials and press a button marked "build". Both the Canadian "number of items" rule and the US "major portion" rule have been complied to. What I'm trying to say is that the CNC machine and its programs are simply tools, and I don't think the regulators have placed any restrictions on the tools we may use. I suggest we simply go ahead and make our CNC machines, program them, enjoy building our aircraft and leave the knit-picking to the law makers.

[jim.bo12345]

Hi all,

To figure out the most efficient use of the CNC router will be decided in practice. Since CNC routers have 3 axis, intuitively it would make sense to start to think in 3D. I have personally already ordered a plan and this book http://www.amazon.com/Build-Your-Mac.../dp/1430224894 to become familiar with it's capability. As far as what is the most efficient software combination to use with it, again we will be able to tell over time.

As far as I am concerned, a CNC foam cutter also in order, since foam ribs are lighter the wood (presumably). But again, during the first phase of the project, those things will come out.

Now both of those tools are expensive, but I can see a scenario, that once there is a proven, standardized design agreed, each EAA chapters could build one for their members use, so they can be leveraged. Or in parts of the world where there are no EAA chapters, interested group of people/communities can invest in shared CNC tools.

As far as design to build, I tend to agree that a proven design that CAN be built with CNC tools (even though it is not designed for it) a good starting point. There is so many other things that needs to be done in an early phase of an open source project, that the design is the last thing one wants to worry about.

Once the group gets over the initial hurdles, than it makes sense to create a design optimized for CNC tools specifically and for cost generally.

Dana bought up a good point that questions <<who approves the design, and changes in the design?>>. I think the strength of a open group method is that there is a lot of people will join hopefully. Among those there will be some qualified aeronautical and structural engineers, some master craftsman. I expect that some of the group members will have access eventually to some of the sophisticated FEA http://www.sv.vt.edu/classes/MSE2094...s/history.html and CFD http://en.wikipedia.org/wiki/Computa...fluid_dynamics tools so we can verify structures and aerodynamic changes. Although traditional configuration airplane design is well understood, optimization is always difficult for any single designer. But in a group setting we can work in a divide and conquer mode, i.e., each of the group member can specialize on area that are well within their competency. So one of the main group "charter" could be to develop those sound PROCESSES that lead to sound decisions (the other to store the design, performance, test and manufacturing information). In other word, if a member of a group does know how similar tests were done before, does know, what test report format he/she needs to use to report the result, and there are qualified members willing to to do a "design review", I see no obstacle, why the format wouldn't work. Of course the whole thing depends on a competent leadership, but I am positive about it that it CAN be done, but of course time will tell how well.

Brian mentions 51% kit related issues. In my mind, the group should not supply parts (only design and process related info). If the CNC files (g-code) is available, anyone can go to a local CNC shop or hopefully an EAA chapter and make their own parts. I don't think the FAA requires us to build our own CNC machines.

In summary, I think it COULD work and if it does work, it has the potential to seriously to reduce the entry cost barrier of flying. Again in my personal opinion, if we can't make an airplane that beats the average car both in price and performance, why bother?

jim.bo

[Jeffrey Meyer]

You got it, with the abundance of open source hardware projects dedicated to CNC, from gantry systems through to controllers as well as software, this is a great time to jump in and see how we can leverage it for homebuilder use. Now just to add a little to the conversation, there are software applications out there that can slice 3D models that can be cut and carved in 2D slices and then re-assembled into 3D objects. (www.vectric.com) Think of now cutting out a plug or mold for composite building.... (cowls, fuel tanks, wings, fuselage etc). Also maybe a windshield form that is assembled from foam slices and then the acrylic/lexan heated and molded over that form... now think of using that same cnc machine to build a chair and ottoman out of wood to sit in and admire all that work....
:-)

Yes, SolidWorks can also slice up the 3D geometry to make such plugs. Now all we need is the geometry...

I would say that we should probably proceed with number 1 and see where we get to. If a plans supplier wants to donate for number 2, then that would be great, but again, even if we go with number 2, there would still need to be re-design for CNC manufacture and assembly as there are few designs out there if any that would comply with the overall vision that we have to make this as quick and simple as possible with associated instructions, video and content we are ultimately trying to achieve. This will become apparent as we begin to go public with this..... more to follow.... :-)

Glad to see that we're on the same wavelength. So, let's start pushing the cart. The way I see it this is a team effort - I can't do it on my own and I suspect that nobody else will claim that they can. But we have strength in numbers - of the 4000 odd forum members we only need 20 or 30 enthusiasts to identify with the idea. As such, we have to lay down some ground rules to prevent the horse from looking like a camel. Some comments/suggestions:
1. This is not just a team - it's a cyber team. We all have to practice restraint, flexibility and tolerance of eachothers' ideas and opinions, but we can't please everybody all the time.
2. This forum is the main line of communication and we are an international team so we don't have the advantage of frequent face-to-face meetings. Again flexibility and tolerance is the name of the game.
3. We need to work to a predefined plan with well defined activities and milestones. For example, I suggest the following:
* Through this forum specify the type and mission of the aircraft we want. The milestone is freezing of the spec.
* Preliminary layout and design through sketches and pictures. The milestone is a preliminary layout of the aircraft and specification of the required CNC machine to produce it.
* More detailed general layout as well as specification of the aircraft powerplant. Milestone is exact 3D CAD model of the external geometry.
* Division into sub-groups to deal with various activities - CNC machine, Wings, Undercarriage, Fuel system, Controls, Cockpit, Fuselage, Empennage, engine, electricals, documentation, etc.
* Design of main structural elements and layout of equipment. Milestone is reports and analyses required by the regulatory authoroties.
* Detail design of all parts including DXF files for CNC. Milestone is the first CNC part produced on our machine.
* It may be prudent to build a 1:4 scale R/C model just to give everybody the confidence that the thing will fly.

I've probably left out a great deal but hopefully it's a start.
Am I crazy to think that this will work?

[jim.bo12345]

Hi all,

I just noticed this plane, designed for CNC. Building hours claimed to be 160-250 hours (see the bottom of the page). http://www.skykits.com/Comparison.html Now that is closer to my liking and shows the true potential of CNC based plans...

jim.bo