 |
|
 |
|
|
|
|
          |
            |
 |
This item is discontinued. Designing The Great Planes Giant Extra
GREAT PLANES 1/3 GIANT EXTRA 330L
I've been asked many times what it's like when my husband, Mike Cross, designs an aircraft. "How do you decide what plane to do?" "How long does it take?" "Do you test-fly the airplane?" These are just a few of the questions asked. Since I was actively involved in the design of the 1/3 scale Giant Extra 330L, I thought it would be fun to tell you about how this design came to be. "How Did You Decide What Plane To Do?" Before the project was approved by Great Planes, we had to recommend several full-scale aircraft. The research for this was my domain ... it was fun and challenging for me! We looked carefully at the recommendations from our product surveys and Web site. We knew Extra was working on a new aircraft, but knew little else; we looked at the CAP 222 (a production Giles), Edge 540, Giles G300, and other top-performance aerobats. But when I dug deep enough to come up with photos of the prototype Extra, dubbed the 330L, we knew we'd found our new project. (Extra has since renamed the planned aircraft the 330LX; however, this model is designed from the prototype, the 330L. At the time of this writing, we do not know what differences there may be between the 330L and LX.) Extra had given us everything we could hope for - the great handling of the 300L's wing and fuselage, mated to the incredible 3D-capable performance of a large, CAP-style aerobatic tail. We were ready to begin! I made numerous contacts with Flying (a British magazine) and others in Europe. I also contacted Bob Banka's Scale Model Research who, to my amazement, already had a Foto Pak on the second prototype! Now we had documentation and were ready to work.
"What Next? You've Picked The Airplane, Where Do You Go From Here?" The aircraft's size was the next key parameter. Before we can work on anything else, we need a scale outline in the intended size. The plane's scale was selected based upon wood sizes. Since we would be building wood wings (more on that to follow), we selected the wingspan based upon each wing panel being sheeted with 48" wood, allowing for an extra inch from each tip to be trimmed as needed. Staying within 48" allowed us to use readily-available wood, and to avoid the potential weak/fracture points (stress risers) and building complexity caused by assembling long sheeting from shorter pieces. Before any AutoCAD was done a list of parameters was written. We thought a lot about the feedback we hear from modelers, and our own experiences in building a broad variety of aircraft. We wanted handling and assembly to be as easy as possible, so we made the wing and stab plug-in. We also made them removable with just a single bolt each, both using the same wrench. The model's gear remains on the belly so the fuselage can be rolled across the ground, not needing to be carried prior to assembly. The scale shape of the Extra's nose allowed us a one-piece (fiberglass!) cowl, rather than a two-piece cowl (which is heavier and has higher maintenance needs). We wanted interlocking pre-cut formers rather than asking the modeler to stick-build these critical structural members. We knew we would reuse the enormously popular interlocking web construction of the Great Planes Giles wing, but needed to modify it to accommodate the wing and stab tubes. We also had to allow for the airfoil fin construction separate from the rudder. Maintaining fidelity to scale was important, including keeping its mixed-skin appearance, but this was secondary to the flight characteristics. We wanted easy access to the fuel/smoke tanks and radio gear, strong and secure cowl mounting, pre-planned and engineered-in radio gear mounting, and radio options that allow modelers to use what they have rather than requiring 100+ in./oz. rated servos everywhere. The aircraft had to be easy to build using familiar tools and techniques, guided by a detailed manual and plans, so that a modeler who's built a few sport planes could assemble this model as his/her first giant scale aircraft with confidence.
But for Mike, first and foremost in designing any aircraft is flight performance. Every single step of the design process, from start to finish, shared a common focus of creating an aircraft that is easy and enjoyable to fly, then balancing all the other concerns. We knew we could trust Extra's basic design, so the passion for flight handling focused around eliminating rudder coupling (many aerobatic aircraft pitch or roll when rudder is applied, such as in knife-edge), as well as providing excellent ground handling, and overall confidence-inspiring ease of flight. The key ingredient to achieving these goals, plus allowing exceptional slow speed and 3D handling, was light weight. This was a critical factor in this plane's design. Every square inch was carefully considered for light weight first, and all other parameters second. "Why Wood Wings?" Well, for many of the reasons stated above ... light weight, familiar tools and techniques, readily-available materials, pre-engineered servo installation, and reasonable cost.
Many giant scale aerobatic models have foam wings and stabs. Structurally, a solid foam wing is unnecessarily heavy. Since most of the stress and load a wing takes is near the root, that is where the structure needs to be the beefiest. A conventional foam wing is identical structurally from root to tip, so if it is strong enough inboard, it is overbuilt at the tips. On an aerobatic aircraft where roll inertia (the willingness to start and stop a roll) is critical, excessive weight out at the wingtips is a problem to be avoided as much as possible. (Remember when you were a kid and you sat on a seesaw with somebody twice your size? The other person had to sit way in on the arm of the seesaw for you to be able to move it, right? Similar thought here - the less weight out at the tips, the easier it is for the aerodynamic forces to get that wing to start or stop moving. On the other end of the spectrum, high roll inertia [heavy wingtips] is one of the main reasons why many aircraft over-rotate snaps and spins - once they get that weight moving, it's hard to stop.) A foam wing also requires new tools and new materials in many builders' already-crowded work area. She (or he) needs to cut foam precisely for servo placement, spars, and maybe even wing tubes. The modeler then needs to assemble very large sections of sheeting, work with large quantities of epoxy, find ways to weight down the wing properly and evenly, and then let this weighted wing sit for long periods of time to dry. Then start over again with wing #2, then each stab panel, and perhaps again for a vertical fin. While a wood wing requires a bit more labor time to build, it doesn't tie up a workbench, drying for long periods of time. An interlocking design also requires significantly less precision to assemble and offers opportunities to spot and correct mistakes. Add to that the issues of trying to repair a damaged foam wing. For many modelers, working in small areas with limited time frames, these are all issues that can make a difference in whether or not an aircraft ever gets completed.
The three wood wings on this model (main wing panels, stab panels, and vertical fin) are structurally designed with full-span webs. This interlocking construction provides several benefits: easy assembly which self-aligns rather than carefully placing key parts and trying to hold everything in place and square during gluing; incredible strength courtesy of D-tubes backed by I-beams; and light weight thanks to the lessened structure needed because of the strength of the load-bearing webs. Designing them was a challenge! It took many hours with AutoCAD to design the complex, notched-out webs that fully interlock with the ribs. We began with accounting for the change in rib size, then allowing clearance for the angle at which the web meets the ribs, then supporting the ribs properly during assembly to avoid twists. We added interlocking the servo trays and control horn mounts into the webs for greater structural integrity. Forget a variable, and what looked perfect on AutoCAD won't be! The structure of each of the three wing types is slightly different from the other two, taking into account the need to support the stab and wing tubes, give the ailerons spars for rigidity, and assembly techniques.
What About Building The First Prototype? When Does That Start? Before prototype assembly could begin, basic design issues had to be answered, even after the scaling and completion of an AutoCAD outline (no small task!). The inner fuselage box (from firewall to tail gear) needed to be sketched into the outline. This provided, among other things, the angle for the root rib of the bolt-on stab, rib positioning at the ends of the tubes, and the positioning of the fin base. Wing and stab tube sizes were determined and availability confirmed. Then we immediately began designing, then building a stab half. Three stab halves were built before the concept, parts, and assembly all came together. In the mean time, when the stab needed more thought, we worked on the wing's design. Mike paid special attention to the feedback from Giles' builders that they didn't like trimming off a jig. This time when we supported the wing at the trailing edge to ensure straightness, we did so with a stick placed in the aileron spar slots, with nothing to cut free and risk breaking. Servo mounting was designed into the wing, including the optional second servo for each aileron. The vertical fin was also taking shape: for simplicity the rudder's counterbalance was built as a part of the vertical fin, to be added to the flat-built rudder near the end of its assembly.
So, to get back to the question, as soon as some basic drafting was completed, we started drawing parts and then test-fitting assemblies to be sure the concept and the execution were both satisfactory. The wing, stab, and fin assemblies don't require plans to build them (but the production kits have them anyway for familiarity), so we didn't have to worry about detailed plans to begin the prototype. Construction did require basic outlines and a lot of thought into the big picture to be sure everything properly interlocks and to maintain the central focus of light weight and performance.
When the wing, stab, and fin were satisfactory (including building several complete wings - more on that later), work continued on the fuselage. We finished the fuselage sides first (die-cut plywood in areas of high structural load, and stringer construction between for light weight), allowing for the positioning of interlocking formers, a tank floor, landing gear mounting, doublers to support the tube area, rudder servos, and the outer shell formers that would give the Extra its scale shape. To be sure things were working properly before the design got too complex and would be difficult to correct, we cut and test-built an inner fuselage box with just the tube cut outs, doublers, and formers. When they worked as planned (woohoo!), the additional changes were made to accommodate the other structural issues. We built the prototype's inner box, complete with tail gear mounting, aluminum angle main gear mounting, tank/servo tray, and antirotation dowels for the wing and stab. Then we fitted the wing and stab to the fuselage. Next, we attached the outer shell formers that support stringers and sheeting to create the plane's scale shape, making sure everything interlocked properly. Finally, the front and rear decks were designed and test-built onto the aircraft.
How Do You Test An Airplane Like This? Every aircraft Great Planes produces goes through an extensive series of tests to check its ground handling, stall characteristics, performance capabilities, structural integrity, pilot skill range, control throws, and C.G. range. In the case of a high performance aircraft such as the Extra, there are many additional demands. For starters, it must sit still at an idle and take off cleanly without any yaw under throttle application. It must not only pass the "no tip stall" tests, it must exhibit a predictable break into upright and inverted spins. It must climb and dive vertically without significant pitching toward the canopy or gear. It must snap crisply, starting and stopping precisely enough to allow for 1/4 and 1/2 snaps to be performed with confidence. Inverted flight must require little or no elevator to keep the plane from diving. There must be no rudder coupling (model rolling or pitching when rudder is applied). It must also perform predictably when flown below stall speeds for 3D aerobatics.
These are just some of the many things we asked of this airplane. Like designing the plane, the test flying required a specific plan in a set order to account for items that interact. Starting with simple tests such as going from full throttle to idle on a straight line to check thrust angles, we worked through the specific routine, flight after flight. We made minor adjustments as needed, until we were really pleased with the end result. The aircraft was then loaded with 5 lbs. of ballast and put through extensive stress and speed testing, including the very maneuvers that are specifically cautioned against in the manual. As the TV announcer says, "Don't try this at home, folks," but it's still our job to test it, if possible, to failure. We were unable to fail the aircraft. When the airplane completed the 1999 Unlimited IMAC competition routine and Mike's full range of 3D maneuvers fluidly, we were satisfied! We tested the multiple wings with varied leading edge shapes (leading edge shape is important in determining how easily an aircraft will stall into a snap - a sharp leading edge is much "snappier," whereas a blunt, round leading edge is stall resistant) and determined the pair of templates to provide in the kit. We selected the second sharpest one for optimum aerobatic performance and "snapability," and the bluntest for pure 3D as well as sport pilots and first time giant scale modelers who want the aircraft to be less willing to snap and generally more stable. When the final sport wing was on the plane, I enjoyed the thrill of taking this gigantic kitten through the sportsman sequence and other "sport flying." Numerous others within the company flew it and provided feedback regarding preferences in set up and throws as well as overall flight performance. Finally, the flight testing was complete and the model was prepared for the artwork photography. So Is That It? Is It All Set? Boy, we wish! Once the test flying was completed, minor changes were made as needed to set final thrust angles, and so on. Then the instruction manual needed to be written (requiring building a complete framed model), as well as the plans completed, the parts list generated, box size determined, advertising copy and photography planned and prepared, and so on. We collaborated to write the instructions. Mike told me what he was going to do, I wrote how to do it, then I read it to him and he did what I said, checking for accuracy. This is an exciting challenge we enjoy together. We put literally hours on end into being sure the manual was as complete as possible. This provided the modeler all the things we learned during the design/assembly/flights. After all, once we've worked out all the radio gear mounting, tank placement, safety equipment, engine selections, and so on, why ask the modeler to work out all those details again? Last, when the manual was completed and the instruction aircraft finished, it was time to complete all the labeling and detail on the plans. Wow, is that an undertaking! I can't begin to estimate the number of hours spent placing all the pieces, adjusting the line types, labeling parts, and double and triple-checking for accuracy. This is so that the plans can speak to the builder and be a guideline during building just as much as the manual. And Now It's Time To Let Go Well, this is undoubtedly the hardest part of the entire process, for me at least. Now it's time for all the constructive criticism and feedback. First came the proofreading of the plans by the rest of the Great Planes R&D team. These boys can be brutal! But it is an "initiation" every kit's plans go through to ensure accuracy, consistency, and completeness. Next, the manual was proofread, corrected, then professionally prepared by the "desktopping" team. I have to take a moment here to specifically thank Garry Morenz for all his efforts in pulling together such a huge, complex manual. And then, as if the manual itself wasn't enough, when the decision was made to create and include the Look At Aerobatics booklet with the kit, Garry and Neil Liptak (the sketch artist) stepped up to the plate and did a stellar job! (Thanks, guys!) More proofreading, more corrections, more questions, and minor changes. The die drawings, likewise, were proofread as well as checked for best use of wood and for compliance to set-up guidelines. Advertising and box photography was completed; advertising copy was drafted, proofread, and approved; and many other key decisions were made regarding marketing and production. Finally, after the bill of materials, dies, box size, draft manual, and plans were prepared, two check kits were produced from the exact materials that would later be used in kit production. One came to Mike and I to double check the parts fit, plans, and instructions. The other was provided, as always, to a modeler not involved in the kit's design process to get a separate set of eyes and skills. (A huge thank you to Greg Duitsman, an exceptional modeler, who provided us invaluable insight and feedback on the final product.) At long last, when the check builds were all done, the final manual and plans sent to print, the dies approved for production, and all materials in house, it was time for the kits to be packaged. There's just nothing quite so satisfying, I have to say, as lifting the lid off the very first kit to be checked by Quality Control for completeness, die-cutting quality and accuracy. There they are! Wow. Two huge boxes proving that all the hard work and hours were worthwhile. If you don't mind, I think I'm just going to sit here a while and enjoy the view. Photos by AnnMarie Cross. Reprinted with permission.December, 2000 R/C Modeler Magazine Editor: Dick Kidd
|
|
 |
