PHOTO BY BONNIE BARTEL

compromiseand

40

DECEMBER 2005

composites Playing into the strengths of composites while avoiding the pitfalls LAST MONTH FEATURED an introduction to designing composite aircraft by examining the advantages of using these materials, and then reviewing the fabrics, core materials, and adhesives typically used in this construction method. Now it’s time to dig a little deeper into how these materials compare to aluminum.

>

EAA Sport Aviation

41

compromiseandcomposites Wet layup with epoxy

Longitudinal Tension

RA7725 Bidirectional

RA7715 Unidirectional

Strength (KSI)

Modulus (MSI)

Strength (KSI)

Modulus (MSI)

32.1

2.0

70.8

4.3

Transverse Tension

33.6

2.2

6.2

0.59

Longitudinal Compression

23.8

2.6

43.5

4.1

Transverse Compression

26.8

2.4

13.7

1.1

In-plane Shear

7.3

-

6.2

0.43

Table 1. “Room Temperature” Ply strength for wet layup “Rutan” style fiberglass fabric in epoxy.

“A basis” values 250º cure prepreg

3M E-glass 7781 (70º F)

3M E-glass 7781 (150º F) & 85% Relative Humidity

Strength (KSI)

Modulus (MSI)

Strength (KSI)

Modulus (MSI)

Longitudinal Tension

52.1

3.61

42.6

3.31

Transverse Tension

-

-

-

-

Longitudinal Compression

63.4

3.54

48.0

3.54

Transverse Compression

-

-

-

-

In-plane Shear

16.35

0.59

12.3

0.45

Table 2. Impact of temperature and humidity on ply strength for 7781 E-glass bidirectional prepreg fiberglass.

A designer needs reliable material strength data in order to properly size an airplane’s structure. There are two strength values commonly given for materials. The first are “A” basis values, which means that there is a 95 percent probability that 99 percent of the material will be at least this strong. Use the “A” basis design parameters when sizing structures that are single load path, which means that if they fail the whole structure fails. Examples include a single, one piece strut or a single main spar that takes all the bending loads. “B” basis design allowables are higher than A basis values, but there is a 95 percent probability that only 90 percent of the materials will be at least this strong. These values are used to size secondary structure or primary structure where there is a second load path if the first member fails. In some ways, aluminum aircraft designers have it made. A and B basis allowables for the various alloys typically used are readily available in Mil Handbook 5. The composite designer, however, doesn’t have this luxury. A composites designer has to use reliable test data from other sources, or generate his or her own material design allowables—and this can be costly and time-consuming. Fortunately, there is some data available that can be used as a starting 42

DECEMBER 2005

point. Table 1 (from Reference 1) provides some test results for epoxy wet laid up unidirectional and bidirectional cloth developed for the Rutan VariEze. Additional composite material properties can be found in References 2 and 3. Reference 3 in particular is a goldmine of information for those interested in prepreg allowables for both fiberglass and carbon fiber. This data was generated as part of the NASA AGATE

Aluminum is an “isotropic” material, which is a fancy way of saying that its strength is the same in any direction.

program in the 1990s, and is now public domain information. Aluminum is an “isotropic” material, which is a fancy way of saying that its strength is the same in any direction. This is not the case with composites. Not that this is bad though, as long as the designer properly orients the fabric to handle the loads. Bidirectional fabric’s properties are roughly the same in the longitudinal and transverse

directions. As you would expect, unidirectional is much stronger in the longitudinal direction (as can be seen in Table 1). The table also shows a column labeled “Modulus” for each material. This refers to the modulus of elasticity, or E, which is an indication of a material’s stiffness. The data shows that the unidirectional material’s E is about twice the value for bidirectional cloth, and therefore it is twice as stiff. This is because the bidirectional’s fibers are not completely straight (like the unidirectional fibers) because they have to weave around the transverse fibers. The low E for fiberglass is one of its drawbacks. Aluminum’s E is about 10 million, which means that all things being equal, a fiberglass part will deflect over twice as much under load. While not shown in the table, carbon fiber material properties are much better. Unidirectional carbon is much stiffer and stronger than either fiberglass or aluminum. It’s also lighter, but unfortunately it is much more expensive. Another drawback of using hand laid up unidirectional carbon is that it is difficult to get the fibers straight enough to take full advantage of its superior strength. The good news is that there is at least one company, Avia Sport Composites, that makes pre-cured

carbon fiber rods in various sizes. Test data indicates that these rods have a maximum strength of 320,000 psi in tension and 270,000 psi in compression. See Reference 4 for information on how it has been used by Jim Marske for spars in his sailplanes. The data in Table 1 are for loads acting at 0 or 90 degrees to the fibers. However, if the loads are at some other angle, the strength properties are different. The properties will also be different if the laminate is made up of several plies at different angles. Calculating the stresses and strains (how much a ply “stretches” under load) for the different plies in a layup is not a simple process. There are various methods to do this and Reference 2 provides a computer program that makes this analysis much easier. Every material used for aircraft construction has manufacturing and structural strengths and weaknesses. One of the weaknesses of composites is their reduced strength at higher temperatures. Often a higher safety factor of 2 is used (compared to 1.5

for aluminum) in the design process to account for this. Table 2 (from Reference 6) shows the impact of moisture and temperature on prepreg 7781 fiberglass cured at 250º F. The data indicates a 25 percent loss in strength. Room temperature layups would likely have a worse reduction, though this can be offset somewhat by post curing room temperature laid up parts. The best way to avoid excessive temperatures is to paint composite parts white, which is why almost all composite airplanes are white. Another weakness is bearing strength. Fiberglass in particular is roughly 50 percent weaker in bearing than aluminum. This makes sizing highly loaded joints (like for wing attach fittings) difficult and heavy, and is the likely reason a few production composite airplanes have one-piece wings. For composite sailplanes this is not an option, so they use a tongue and fork arrangement to join the wings together. This allows the wing bending moments to be spread over a larger distance. Designing composite aircraft

requires a good understanding of the materials used and how it has been successfully used in other aircraft. It is well worth the time spent studying other successful designs, particularly how they dealt with concentrated loads and attach points. Like any new design, analysis is done to reduce the risk of failure during structural testing, and not to eliminate the need for it. References: 1. “Reducing the Cost of Composite Materials”, Neubert, Hans, EAA Sport Aviation, January 1978. 2.Composite Aircraft Design, Hollmann, Martin, 1993. Available from www. aircraftdesigns.com 3.Advanced General Aviation Transport Experiments (AGATE) material design allowable reports for a variety of composite materials available at: www.niar. twsu.edu/agate/ 4. “Graphlite Carbon Rod”, Markse, Jim, available at: www. continuo.com/marske/ARTICLES/ Carbon%20rods/carbon.htm

EAA Sport Aviation

43

compromiseandcomposites

expertchat expert O

ften designers tend to stick to a construction material they are familiar with, but Tom Hamilton is not one of them. One of the pioneers in composite kitplane design with his Glasair series, he also was responsible for the Glastar, which uses a mix of construction materials. After selling the StoddardHamilton to the employees many years ago, Tom started Aerocet to produce composite floats. He has also been heavily involved with the development of the aluminum Quest Kodiak aircraft. Because of his extensive experience with both composite and aluminum design, I posed the following questions to him. Q: What do you consider are the strengths of composites as a construction material versus aluminum? A: The ability to mold complex shapes, which tend to be aerodynamic, is a piece of cake. Eliminating the very expensive stretch forming process for shaping aluminum is a real value to

Glasair designer Tom Hamilton: “If the composite structure is designed properly, fatigue life can be pretty much a non issue with a very light structure.”

the homebuilder especially. That is why we see somewhat squared up designs for aircraft built of aluminum coupled with composite cowlings or tops for the fuselage like the RV-10 working the more complex curves, insets for doors, windows, etc. Another nice feature is the ability to beef up structure by adding more laminates or using different fiber types. The stresses are also reduced from point attachments like rivets to lots of square inches of resin bonding. This kind of dovetails with the above statement but warrants mention by itself. If the composite structure is designed properly, fatigue life can be pretty much a non-issue with a very light structure. In aluminum, one must keep the materials from working too hard in order to achieve good fatigue life. Part count is a big advantage for composites by reducing items involved in structure by a whole magnitude. When factoring inventory control, vendor count, quality control, purchasing, etc., the costs for making a structure with multitudes of components can become significant. Labor can be significantly reduced because of the low part count and the thousands of fasteners required in putting together a metal structure. This is especially true if one goes back and deburs the holes after drilling and/or countersinks for flush fasteners. Processes can be reduced dramatically over metal. Etch and alodine treatment, heat treating if the metal is formed in “O” condition, fay sealing if pressurized, and priming the internal parts are eliminated. Fixtures for mass production of a metal airplane for proper and speedy locating of the multitude of parts, is typically very expensive. Corrosion is a non-issue with composites, unlike with aluminum. That’s a big advantage around saltwater. Also the handling or mishandling of sheet aluminum, its availability and costs can vary dependent upon demand. Lots of aluminum is going

to the war for protection of vehicles. As we view the new Boeing 787, a filament wound fuselage controlled by computer is an incredible program. Weight savings, fatigue, less sweating as the plane goes to altitude, part count, costs, toughness against damage, etc. are showing the composite strengths making their choice. Q: What do you think are composites’ weaknesses versus aluminum? A: Apart from very expensive computer tape laying machines, composites require hand labor for placing plies and core material. In other words it is difficult to mechanize the process and utilize the computer power available that we have for typical metal fabrication. Composites have very low peel strength, so parts need to be critically designed to avoid allowing a structure to exact a force that would allow delamination. Aluminum comes pre-made. In other words you buy certified material with known strengths proven over time. One just has to design to its strengths, watch for fatigue and crack propagation potential and it will do the job. In composites we manufacture the structural material from more of the raw ingredients. For certification we go through a complete material validation report, testing coupons for strength to establish allowables. In contrast, one only has to pick up Mil-Handbook 5 for aluminum. Aluminum machines well and has good bearing strength. Composites are hard on cutters and one needs to give extra care for the dust, especially graphite. It will easily start shorting out electrical circuits in a grinding room. Aluminum has inherent lightning protection and by following the zone requirements for thickness in certain areas provided by the FAA, it will pass certification. Composites need to have lightning paths introduced through a typical metallic mesh put into the outside laminates. This

mesh doesn’t need to cover 100% but needs to be applied strategically to achieve proper protection. Difficulties with composites also come from bonding mesh lightning strips to other components of the aircraft. Aluminum, if designed correctly can be self-jigging. It’s difficult to automate composite production, Pilot or even holes to full size meaning hours of laborious work. can, by computer, line up perfectly with a mating part, allowing adequate placement. This only comes with secondary a composite plane overweight. The operations on composites. typical homebuilder can take on the Composite structures can become mindset that if a little is good, a lot is heavy if one doesn’t design properly better when doing laminates. When from the beginning. Sandwich the plane hits the scales, reality rears structure, unidirectional fibers, its ugly head. Metal aircraft vary correct resin selection, finish on the little from plane to plane because the fabric, surface finish prep, etc., if builder has little option to vary the not properly done can easily make ingredients on the basic structure.

RV BUILDERS SHEET METAL INSTRUCTION LEARN THE BASICS OF SHEET METAL CONSTRUCTION WHILE BUILDING YOUR OWN RV TAIL ASSEMBLY!

• Complete your entire tail assembly in our climate-controlled sheet metal shop in less than a week. The RV-10 takes a little longer. • Learn basic sheet metal construction under the supervision of professional instructors. • Work with our tools so you’ll know what you need before you waste a lot of money on unnecessary equipment. • Gain a level of confidence working with our instructors that will carry over as you build the rest of the airplane in your workshop.

CALL TODAY AND GET THE FULL STORY!

Composites don’t like heat. That is why most planes are painted white. It really isn’t a problem if that rule is adhered to. Paint a composite plane dark gray, red, etc., and then you start checking the charts about the “heat distortion temperature” of the resin and core used. Remember that a dark colored plane will easily have surface temperatures over 200 degrees in a hot sunny climate whereas a white one will remain ambient. UV is not really a problem with proper primers or surface coats today. An unprotected resin will break down over time. Not so with aluminum. Preimpregnated composites, where resin is introduced by the vendor into the fabric, require expensive ovens or autoclaves for curing, refrigeration for the materials and costly high-

ELECTRIC TRIM SYSTEMS Install this small, 14 volt servo motor to control Elevator, Aileron or Rudder trim.

Servo

Trim Systems include:

Position Indicator

Rocker Switch

Clevis/Pushrod kit

T2-7A Trim System (.7 inch travel)......$235 T2-10A Trim System (1.0 in. travel).......235 T3-12A Trim System (1.2 in. travel).......255 Position Sensor

Use this sensor to measure wing flap, cowl flap position. 1.2" travel.......$30

STICK GRIPS NEW! G3 Grips feature 4-way toggle switch, multi-color faceplates.

Speed Control

Adjusts trim sensitivity......$35

G101

G307

G207

866-967-5740 770-467-9490

buildtofly.com

[email protected]

2525-8 Pioneer Avenue, Vista, CA 92081 USA Phone 760 599 4720 FAX 760 599 4383 see more details at:

www.rayallencompany.com EAA Sport Aviation

45

It’s difficult to automate composite production, meaning hours of laborious work.

temperature tooling. For volume production and consistency, this may prove necessary but comes with a price. Aluminum typically doesn’t need finishing. This means that when you are done with construction, little prep work is required prior to paint. In composites, dependent on the process, there may be very little or a great amount of finish work needed. Q: Why did you use wet layup construction instead of prepreg for construction on the Glasair? A: The Glasair was basically the first successfully molded kit aircraft out there 25 years ago. The Windecker Eagle and gliders were molded but not kits, and everybody was on a steep learning curve. As mentioned, the costs associated with the prepreg process and the amount of finish work because of the relatively porous surface finish from prepregs was a turn-off. The other problems that we encountered were fuel and water ingestion into the honeycomb structures because of the very thin laminates we needed to achieve low 46

DECEMBER 2005

weights. We put prepreg composite panels up on our roof in the rainy Northwest and even with epoxy surface finishes the honeycomb structures became partially filled with water. Back to the freezer and the part would delaminate. Also there was the potential for weight gain – a major concern on balanced control surfaces going over 300 knots indicated. There were not any good high temperature closed cell foam core materials available at the time. This has changed and one can note that Cirrus uses foam core rather than honeycomb. With extra laminates and maybe better surface coatings today, people appear to get away with honeycomb core on the exterior but I felt very nervous back in those early days. I did purchase one of the bigger ovens in the Northwest for prepregs in the early eighties and it operates basically every working day making parts for airline interiors by my brother’s composite company. It should be noted that the Glasair is and always was made using vacuum bag procedures for the surfaces of the airplane in order to control the resin to glass ratios and compress the

laminates. Q: When designing the Glastar, why didn’t you make the fuselage and wings all composite? A: In essence the Glastar is the epitome of a composite airplane. It’s a combination of aluminum sheet metal, composite fuselage shell (stressed in the empennage), and a welded tube center fuselage frame. The idea behind the construction was to make the building time low for the homebuilder, hence as many of the items as possible are pre-jigged. With a welded fuselage cabin frame all the essential hard points could be done for the builder in our weld fixtures – wing attach, gear attach, engine mount attach, etc. This would have been awkward to do without pre-bonding the composite shells together and fitting the attachments in with pretty elaborate tooling. Today they are bonding the composite shell halves together if certain options are exercised but when the plane was first introduced we felt we were pushing the factory versus the builder ratios. The metal wings, horizontal and control surfaces are all built in aluminum. By doing a weight study, it was very hard to compete with aluminum without going to expensive graphite. Remember, this airplane was intended to niche into the market with speeds of a 182 and yet not be far off a Super Cub on the short end. We had enough wing area that aluminum proved attractive. We also started pioneering CNC procedures for matching up holes for quick assembly. This was prior to the RV’s doing it. Dick VanGrunsven initiated the practice a little later and that revolutionized his product. Q: Do I see any breakthroughs in composite manufacturing that will help reduce material and labor costs? A: Yes, and we are in the process of getting a new composite process certified for aviation use. Pioneering again, I suppose. We have been able to save 13 percent in weight and 25 percent in labor with a new process that works with graphite, glass, Kevlar and/or hybrids. It doesn’t require a high temperature manufacturing process. I will remain a little elusive here until we have confirmed a few more items in the process.