Copyright © Custom Flight Ltd. 2013
Thinking through the Wing Structural Design for the 4000 lb gross weight
North Star 6 Galaxie
There are a number of goals to be considered when designing anything, including aircraft wings. An efficient structure will be light, strong, simple to produce, and have a low parts count to keep labor and material costs low. Another goal could be ease of installing fuel tanks in the wing. Simple right?
Let's take a look at the history of some wing designs. Since we are designing a STOL wing, to fly slow and carry a heavy load, and are using the US35Bm airfoil, like the Cub used, we will look at the cub wing. The Cub evolved from the Taylor Cub. Wood spars were used with ribs slid over the spars and fixed in place. Wood is one of the best spar materials. It is light, strong, easily cut, and it is resilient. That resiliency makes it virtually immune to fatigue. Aluminum structures inevitably will fatigue, this has to be accounted for in the design. Some very expensive commercial airplanes have flown enough to reach the end of their wing’s safe service life! A very expensive fix for the aircraft owner. Luckily, due to lower use, private aircraft, properly designed, will never approach the end of their designed service life. Wood has somewhat of a bad reputation in the market place. I believe this is undeserved. It is not uncommon to hear of cracked wood spars in Champs, Cubs, and other similar classic type airplanes. So the market runs to get aluminum spars to replace the poor wood spars. What is not understood is that the cracks were usually caused by excess load, often a wing tip hitting the ground. A similar load would bend an aluminum spar. So why don’t the manufacturers tell the market this? There are a number of reasons that will become understood, keep reading. Rot is another problem, but if the structure is designed so as to not hold water, and it is sealed properly with enough coats of good varnish, it will not rot. The resilience is a major reason wood spars are used in high performance aircraft like the Pitts Special. I have built about 6 sets of Pitts wings for customers in the past, so I know them well. When I built Pitts wings, with truss wood ribs, after attaching the ribs to the spars, I cut-out the rib cap strip above the spars and laminated another span wise spruce strip. This increased the height of the spar. Height of a spar (beam) is a MAJOR contributor to its strength. Double the height and you get about 8 times the strength, this is a generalization but it illustrates the value of making a spar as tall as possible. When I built the prototype North Star in 1988-1990 I built it with wood wings, with the spars full height, and well sealed. I just inspected that airplane in 2104, the wood inside the wing is as clean and as good as the day I built them.
So if wood is so good, why did the next 30 sets of North Star wings have aluminum spars?
A few reasons. It’s getting hard to find aircraft grade Sitka Spruce in the quantities needed for production, and it is much more expensive than aluminum. The labor is much more expensive with wood, I estimate it takes 5 times longer to build with wood than it does with aluminum. The person building a wood wing needs to be skilled in inspecting and grading wood, every piece must be inspected for various defects, some obvious, some subtle. This can be learned, but is expensive in business. Aluminum on the other hand is much more uniform, although don’t assume it’s all perfect, I have found a few surprises over the years.
If you look at an aluminum spar Cub wing, you will see that there is about a 5/8" space between the spar and the fabric top and bottom, that means there is room to make the spar 1 ¼” taller! That’s a huge increase in strength. To take advantage of that on a certified aircraft would mean a very expensive change to the certification. But we are not certified so we can easily investigate it.
I like fabric wings, but fabric wings usually have drag wires, another expense. It’s also very labor intensive to route drag wires through gas tanks. And more labour to cover and finish a fabric wing than aluminum.
So if we make the spars full height of the wing then we can easily rivet aluminum skin top and bottom, no more drag wires, easier to install tanks too.
The full height spars will require 3 piece ribs, but with the accuracy of CNC machining, it’s easy to make temporary lineup jigs that position the 3 rib segments accurately on the spars.
Ok, so It looks like aluminum spars, ribs, and skin might be the way to go. We have the luxury of a clean slate here, lets evolve the concept some more.
As a lifelong fabric wing guy, going through the wing stress analysis was enlightening. It probably helped that over the years I have looked long and hard at many wing structures.
With a good aluminum wing the wing skins carry most or all of the loads. But wait, the spar must be taking a good share of the load. Not as much as you would think, and that's a foreign idea to a fabric wing, where the spar is everything (almost). Think back to I beam theory, or roof trusses. The loads are carried in tension or compression in the top or bottom of the structure, the middle, the web, only serves to keep the top and bottom pieces apart and in position relative to each other. Down the centerline, is the neutral axis, where there is very little or no load. So think of the cross section of a wing as the end view of an elaborate I beam. The spar keeps the top part (skin) and the bottom part (skin) apart and in position. So do the ribs, redundancy? Yes to some extent, If you have a chance to look in a Navion wing you will not find a spar! Well there is a spar, but it only extends from the fuselage centerline out to support the landing gear loads, and stops there. The wing is hollow, with ribs, and top and bottom skins with stringers. The Navion was designed by North American, the same guys that designed the P51 Mustang, they knew what they were doing!
But we are not eliminating the spars, they act as self jigging parts that make wing building easier. If they are eliminated then elaborate jigs are needed, we don’t want that.
The point is, the spars can be light since the skins carry a lot of the load. An added bonus is that these same spars, because they are light, can be used in the wings of lighter airplanes, and then used to carry a greater share of the load, or all the load if fabric covered. This potential for the spars to be used in a number of designs makes the high initial expense of custom extrusion dies and minimum production runs a little easier to justify.
Many spars are built up of sheet aluminum webs with angle extrusions riveted to the top and bottom to form the I beam. This requires a number of extrusions since the top and bottom and front and rear are 4 different angles. It also takes a lot of labor to build up a spar that way. And you have multiple layers, you might want to corrosion proof in between. If not using extruded angles top and bottom it will be necessary to perform some accurate bending to form the angles. Bending is not a simple task, especially on an 18’ long bend as we need for our wings. Bending is one area where variations in aluminum show up. It’s common for two pieces of aluminum with the same alloy and heat treatment, to bend differently, with different spring back due to slight variations in properties.
Extrusions have much less dimensional variation, and are much easier and faster to build with. So we decided to bite the bullet and pay for custom extrusion dies to make one piece spars, eliminating the labour required to build up a riveted spar.
But what shape? We want to optimize the spar design with a number of requirements in mind.
The Cub spar is shaped the way it is so that it can be inserted where the wood spar used to be, that dictated its overall dimensions. That shape won’t work here.
We want to have easy access to rivet the leading edge skin on. Top and bottom caps extending forward like an I beam would be difficult to reach inside a leading edge skin. The “I” shape is not critical, the important detail is to keep enough metal as far from the neutral axis as possible. Don’t forget the big picture, when the skins are riveted to the spars, the skins become part of the spar caps, making a very wide I beam. If we make the spar a C section with the caps or flanges extending aft it will be easy to rivet the leading edge skin on. In addition a wide, aft pointing bottom flange on the front spar will make a great shelf to rest a gas tank on.
The same goes for the rear spar, if we extend the lower flange forward. If the top flange of the rear spar also extends forward then it becomes hard to drop the tank into the wing, but if the top rear flange faces aft, now the tank can easily drop into place, and the rear spar becomes a Z section.
Wide top and bottom flanges not only put more metal farther from the neutral axis, they also give lots of room to rivet skins, without needing to overlap, makes a nice smooth wing.
As we discussed the spars can be light, that means thin. But the thinner the cross section the more difficult it is to produce a straight extrusion, so we decided on .090 inches thick.
This puts more metal, and weight in the spar web than we need, but we can cut lightening holes in the web. The other advantage is the holes provide access to the leading and trailing sections of the wing. It was interesting working with my engineer on the design, he loves lightening holes in spars, but isn’t allowed to use them in the jets he designs. The reason is that if the jet has a bird strike, the bird might pass through a lightening hole and do damage deeper in the wing. Without the holes the spar web would stop the bird! We aren’t going that fast.
This new spar design allows many options. The wing can be all aluminum, or just the bottom or maybe top aluminum, to take the drag loads and the rest fabric to save weight. It’s much faster to build an all metal wing than a fabric covered one, and it costs less than expensive fabric paint and labor. As I said before, I have been a diehard fabric guy, but I am starting to see advantages with aluminum. I always felt aluminum wings had to be much heavier, but look at the Luscome, Maule, and especially the Aeronca Sedan wing. Believe it or not, a 4 place Sedan is very close to the weight of the average PA18!
So what we ended up with is an economical, easy and fast to build, strong, versatile wing design system.