CNC Wing Ribs

At our monthly meeting of the Experimental Aircraft Association, I was telling the members that I had access to a CNC router. Quickly, one member asked if I could cut out wooden wing ribs. The traditional approach is to make a jig, take hundreds of small sticks and glue and staple them together. Considering his project is a bi-wing aircraft, and has 32 wing ribs, that’s a lot of tedious hours. I said I’d give a go. I have a background in fabricating Museum, Zoo and Nature Center exhibits. For 25 years I was asked to build the most amazing things. This project would be a fine experiment.

Step 1: Your STUFF

Things you need / Materials and equipment list:
Large scanner
Computer
Vector Drawing Program
CAM generator
Mach3, EMC, or similar (http://www.machsupport.com/) (http://www.linuxcnc.org/)
CNC Router   (www.100Kgarages.com)
Tooling Bits
Baltic birch plywood – ¼” 5 ply or similar
Staple gun
X-acto, utility knife, something sharp
Sandpaper or foam blocks

Step 2: My STUFF

(Things I need)
Patience
Dust collection system
Writing skills
Photography Skills
More tools
Bigger tools
Faster tools

Step 3: Scan Original Blueprints

I was given the sheet of blueprints (actually white zeroxed copies) to work from.  I taped the relevant portions together and traced the image to be cut on white butcher paper. Then took it to a local architectural copy shop (FedX-Kinkos would do) and had them make a digital file. They also sized it a bit to match the dimension given on the plans. I was enlarged 1.6%.

Step 4: Drawing Programs

(CAD – Computer Aided Design)
There are so many design programs out there (http://crunkish.com/top-ten-cad-software/) (http://www.freecad.com/). Some are free, some are not. Try the free ones, or the trial downloads,  and spend some time assessing their capabilities. Make sure you look at what they can import and export. This will be important in moving your drawing files to the CAM generator (more about that later). The higher end ones have a steep learning curve and take a while to be comfortable with.  The lower end and free are getting better all the time.  Check out Sketchup (free) with the ruby script that allows output in .stl and .dxf formats. Sweet! (http://www.guitar-list.com/download-software/convert-sketchup-skp-files-dxf-or-stl)
I loaded the file into my CAD program and with node editing, smoothed the exterior curves, or as known in boat making, fairing the lines. Using created circles at each intersection within the internal trussing, I smoothed and cleaned up the intersections, thus replacing gussets that would have been present if built the traditional way.

Step 5: Output to CAM

(Computer Aided Manufacturing)
Some drawing (CAD) programs have no way to get g-code from them. G-code is a language defining where the router must go to perform the task you have in mind and is generated by the CAM. Some CADs have CAM with them but they tend to be the pay-for types.  For those of you who are using a CAD program that does not have this function, you’re going to have to find a separate one to do this. Here is some links. (http://www.rhino3d.com/resources/display.asp?language=en&listing=545)
(http://www.freecad.com/CAM_Programs/))  For those of you who want to learn about g-code there are sites that will show you what it all means. (http://www.cncezpro.com/gcodes.cfm)
(http://www.cnccookbook.com/CCCNCGCodeCourse.htm)
My program has CAM output, so the g-code was created, and ready for the homemade CNC machine.

Step 6: Selecting the Bit

On our homemade CNC machine, is a Porter Cable Trim Router  As I had selected Baltic birch plywood, I chose a ¼”shank 1/8” end mill. This means it is strong (thicker shank), and will cut 1/8” wide with a square bottom.
Our router only goes one speed in rotation and maxes out at 25 linear inches per minute, so the only control we have is how deep it cuts per pass, and the inches per minute speed it travels thru the material.  Before any machinery is turned on, install the bit into the router you have. Be VERY careful, they are quite sharp. I wouldn’t know that, but, there are a wide range of band aids and compresses readily available in our shop, just in case.

Step 7: Load the Material

I had some cardboard to use for the test cut. Make sure that the piece you load on the machine is bigger than the project. Seems obvious, doesn’t it?  Also that it is oriented correctly in the X and Y plane. I never make that mistake.
Stabilize the cardboard (later the plywood). Many treacherous opportunities lie here. Cut the cardboard/plywood  larger than the project.  You want extra room for clamping down the outer edges with screws, cam locking devices, or even double-sided tape. I used a staple gun with ¾” staples around the outer edges of the cardboard and the baltic birch, out of the way of where the router bit would travel. Regardless of the method, one wouldn’t want the highly rotating router bit or the router itself crashing into something hard and unyielding. Some very interesting results WILL occur and you really don’t want that experience. The least is you may lose some steps in the g-code and get the remainder of the cut all cockety-wampuss, and the worst is flying bits of high speed steel flying around. Which brings us to the all-time favorite subject of OSHA,   da-da… SAFETY.

Step 8: SAFETY

Here we go, ignore this at your own peril. Ok, the condensed version:
---Ears - CNC machines are loud. I like to put on noise cancelling earphones with my favorite music playing in the background. Not        too loud as you want to lightly hear the machine as it runs, giving you a heads-up to any problems.
---Eyes and face – Flying objects do sometimes come from the machine, best to protect them. I use a face shield.
---Breathing – All materials have some dust created during the cut. Use at least a filter mask for non-toxic woods and cardboard something better for MDF and plastics.
---(If at all possible make or buy a dust collection system to keep the fumes and dust to a minimum. That’s the next project for us after this contest. Wish I had done it earlier.) (http://solsylva.com/cnc/vacuum.shtml)

Step 9: Power the Machine and Computer

At last the fun began. I powered the machine and computer and opened up Mach3. This is the interfaces that actually translates the g-code into movement by the router. Mach3 works with windows. EMC is the open-source Linux version, Next I loaded the g-code using a flash drive. Use that or wireless or whatever you move files with, from your cozy office to the cold and manly shop. I then  clicked the Reset button. Now I could move the non-running router around to position the 3D coordinates(X, Y and Z) of the bit by using the number pad on the keyboard or an separate number pad.

Step 10: Zero, Zero, Zero

You really want to fire that thing up. You have been at it for a while. You are impatient (well that, at least, describes me at times). But we have to get off on the right foot, or inch, or thousandths of an inch. X, Y and Z must be zeroed. Most of the time, this point will be the top-lower-left corner of the drawing.  Be slow and careful to NOT run the bit into the material or hold-down clamps. The tooling bits, I’ve found, do break, and are pricey. Once over the zero, lower the bit ever so gently down to just above the top of the piece. Then I place a feeler gauge below the bit and, extremely slowly, move it down to just touch the gauge. Using .002 gets the bit located in space just half-a-hair above the material. You do remember NOT to have the router on during this step. Right?

Step 11: The Magic

It’s finally time to turn the router on and start the sequence of cutting from Mach3.  Just like your mother used to say “don’t forget your mittens”, remember the safety stuff.  Turn the router on first.  Then I clicked the green button on the screen labeled Cycle Start. Off and cutting. Yeah!!!  I prepared to stop the cycle if it ran off in some bizarre direction. Remember the adage that computers, (along with dangerous machines attached to them) only do what you tell them to do, not what you want them to do. If all is going well, just watch and enjoy the moment. To me, this is amazing stuff, I used to hand cut all of these projects. It’s a wonderful world!
When the cycle ended, and the router came back to zero, I turned it off, then removed all the staples. I remove the waste sections, and vacuumed or blew off the dust so the machine is ready for the next experiment.
It almost worked perfectly. Several of the interior waste pieces moved around in an uncomfortable way, and two of the trusses bent out (got too thin) as the router bit pushed them sideways.  Lastly, if the project is cut completely from the waste material, I have a template that will prove useful. So, I went back the CAD and placed a number of tabs around each part. (Tabs are small bits of the wood that hold all the various parts together. They are less thick than the plywood, short in length, and are easily cut with an x-acto knife.)
For this run I had the cardboard template to trace on a piece of Birch plywood. I could place staples more accurately to secure the material safely. Plan B was run thru successfully.

Step 12: Examine

Now is the time to really look at the product.  Test it, measure it, look it over closely. If it needs improving, return to Step 4, and go for Plan-C, or Plan-D, etc.  Most of the time, it takes several tests and iterations to get it right.

Step 13: Finishing

I used foam sanding pads, and smoothed all the surfaces, then cleaned with.
I then met with the aircraft builder for his approval.
Now, only 31 more.

Step 14: Why We Want to Win

Last summer Mike and I began getting together at least once a week, or more, for what we called guy time. We would work on some idea, hack, project, or repair of some item with what we had in the shop. This resulted in some clever solutions using rather odd materials. The interplay between us was so fun we began to invite others to our “club”.  Thus a Maker Space has taken over my shop. I love it. About this time Mike introduced me to Instructables.com and I was hooked. He also brought over his homemade CNC Router.  We got it up and running and have been creating a number of interesting projects that are being submitted for this contest.
Our club of makers would continue to explore the vast world of CNC routing. We have in mind a large backlog of ideas that would be furthered by winning the PRS Standard 96-48-6 ShopBot .  Our homemade CNC can cut 31” x 21” and cuts at a maximum of 25”/ min.  The 4’ x 8’ table and 300”/min will vastly improve our ability to produce prototypes, develop and improve designs, create larger projects and test out 2.5D, then move to full 3D sculpting.
Here is a small selection of my personal list of projects:
2D
                    Mazes for more than one player
                    Multi-level mazes
                    Giant mazes on a half sphere to be played by young children
                    Interchangeable mazes of increasing difficulty
                    Picture frame mazes
                    Spirographs – small 8.5 x 11
                    Giant Spirographs using chalk for outside
                    Patterns for Cabinet doors
                    Wood Blocks for printing from ancient lithographs
2.5D
                    Bas-reliefs
                    Cameos
                    Negatives for molds
                    Negative busts
3D
                    Sculpture
                    Full Busts
                    Art prototypes

Thank you for considering this submission
Rick Shore

Comments

[wally6800]

I would like to know more about how you faired the rib profile before you created the "G" code. I have never been able to make a rib that I could use right off the CNC table without doing a lot of hand work to get the profile fair. By that time I have lost the exact shape and the rib is worthless.

I hope you can help me. I have two Marske Flying wings that I want to build. 70% of the work is in making the ribs and making them to the correct size.

Thanks.

Wally

[ultralajt]

I suggest that a comparative load test should be done between that CNC-ed rib and ordinarry build up rib from spruce sticks. Ordinary rib will be superior as grain orientation and strength of the spruce (or douglas fir..) is way higher than of the plywood. The backdraw of the plywood is also the grain orientation, and it can not be used to its best strenght.
Pure plywood ribs should be designed for that lower strength of the material. Caps and diagonals should be much wider. And this will resulted in heaviest ribs, than odf classic design.

Othervise it is clever idea to simplify ribs manufacturing and to speed up the process. The newest design from team AeroMax use many CNC done wooden plywood parts in the aeroplane structure.

[rickharris]

Just a CNC tip - We often for complicated cuts used to print out the drawing - full sized - and the lightly tack glue to the material. that way we could see a) that the material is big enough - b) where to put the hold down clamps that is going to be out of the way.

We always used a piece of sacrificial MDF under the material so we could cut all the way through. this also allowed us to pin the material - usually wood - down with panel pins. Much better than double sided tape and easier than various clamping systems.

[Wroger-Wroger]

One could make a wooden vacuum table, with countersunk holes underneath the spar parts and waste bits, while avoiding the tool path.

A 30mm hole grid, 5mm holes, counter sunk to about 10 - 15mm in diameter?

[rickharris]

We tried a vacuum table but couldn't get it to work - although they are made commercially.

The hole size is critical not too big but not too small.

[Wroger-Wroger]

Naaa the hole size if BS...

Drill your thick plywood wood where the router cuts don't go close by and then counter sink them out.. to about 20mm in diameter.

Holding power = force x area.

And maybe route away one end and then apply a clamp to that, and perhaps a couple more as you go.

If I had to make any more than one or two of the spars like this, I'd make a table to suit the spar.

Make the table the exact same size and shape as the spar, perhaps add a couple of square "tabs" to clamp the sheet down and then lay the sheets on it.

The other thing is too, that you can make them in the stencil pattern.

Cut all of the shape out - except for a few connectors, that hold all of the pieces in place.

That way you can scrap the vacuum table and not have your timber and bits flying around everywhere around the router bit, when the cuts are made.

Once all of the cutting is made, then take a small hand saw of some description and cut the tabs.

Look up stuff on stencils.

And design it so the long thin bits are restrained as well as the thick bits and keep an eye on the cutting speeds... not too high.

[CameronSS]

I'd be curious to run a stress test to failure on this and a traditional built-up rib. In standard rib building (as I understand it) the wood grain runs parallel to each member, whereas with plywood the grain is going every which way. Then again, plywood would have increased strength in other directions, and wouldn't have failure-prone joints. Have you looked into this?

Please don't take this as a disparaging comment, it's a great use for a CNC router, especially if it gets another airplane in the air. Well done!

[Wroger-Wroger]

I'd be thinking that a UNIFORM construction from one sheet, if properly designed would be better.

Most of the traditional construction techniques, while they are good, there may also be other considerations that come under cost, wartime and efficient use of the resources - such as timber.

There was also the technical angle of the limitations of the equipment and the speed of production.....

It was deemed to be better to use a little glue and all the offcuts etc.. and to really minimise waste, than to be throwing out huge amounts of material, and it probably would have been far easier to cut lots of little straight bits from the one sheet, to assemble a spar from, than to try and cut out one as a single piece.

Not that they could not be pressed out with a knife press, but the smart people really factor in ALL of the considerations to get the most suitable product out the door at the best price relative to the market and circumstances.

[CameronSS]

Oh yes, it's absolutely much faster and very likely better use of materials. But if you're making something like a Pitts that puts heavy loads on the wing ribs, ultimate failure strength is by far the top priority. Mostly I'm just curious. :)

[Wroger-Wroger]

Yeah, but plywood can be a generalist product or a specialist product.

There is some really great informaiton about designing and creating plywood and products from it.

The types of glues, the types of timber, they direction of the grain in the layers, in some regards plywood is the ultimate in wood porn.....

I guess that premium aircraft grade plywood is expesive, and if one were to knife presss and cut out, or route out or laser out whole spars from one single piece of ply, while I really like the idea totally - the amount of timber left from the sort of rectangular shape, would I guess, be about 5 to 8 times the amount of timber in the actual spar.

So I know that really GOOD aircraft grade plywood, that is free from knots and defects etc., IS really expensive, so with wages etc., it may actually be enormously cheaper, to cut that sheet up into lots of little straight bits and gussets, and to lick and stick them into a jig and then profile the finished component, than to throw away heaps of expensive timber.

[Toxictom]

Having dealt with wooden wing ribs I can see where this will save the builder lots of time and money. Good job.