Because of the power loss, this new airplane would either have to be smaller to use the very fine ST 46, or about Bearcat size with a larger engine. In general, Bearcat size airplanes have a slight advantage over smaller airplanes because of slightly more favorable Reynolds number and usually a somewhat better visual impression. I therefore decided to build the Bearcat size airplane and use the lightest 60 available having the necessary conservative porting and reasonably long stroke. This narrowed the choice to the ST 60 at 12 ounces (with venturi, not carburetor).
The extra weight of a heavier engine, larger tank, extra fuel, muffler and stronger nose structure dictated that my new stunt ship should have a short nose and could possibly profit from a longer than normal tail moment. After spending a few hours looking at Green's Famous Fighters of World War 11, I found several airplanes which would make excellent semi-scale stunt ships by profiting heavily from the above-mentioned nose and tail moment changes. The Spitfire and Hawker Tempest V seemed the most ideal with the Tempest V having the edge because it offered the possibility of enclosing the muffler within a large "chin" radiator. When I drew up the Tempest V, the nose was so large that the ST 60 cylinder head didn't even extend into the dachints radiator area. As a matter of fact, at that scale, the engine and muffler would go into the radial cowling of the more attractive Tempest II. From there I couldn't help but notice the Hawker Sea Fury which has the same wing and cowling as the Tempest II, better lines, and a colorful service paint scheme. To the disadvantage of the Hawker airplanes were their wings which have outboard dihedral breaks. After a couple of weeks of stewing over the immensity of the project I decided to "go for broke." Win or lose, I was going to build my muffled airplane and would somehow manage to make it more realistic than previous semiscale designs.
Building a heavier, more realisUc airplane would be possible only if ways could be found to significantly increase the lift of its wing. As it happens, I have been using an airfoil test rig for about three years, and based upon experiments with different airfoils I was sure that extra lift was attainable. (Fig. 1)
Next, I designed my first "super" airfoil. It was a built-in, "simple" flap type where the flap forms the aft contour of. the wing much the same as Keith Trostle used on his Nats-winning Focke-Wuif Ta 152. I tried to go Keith one better and use a full-scale aircraft practice of moving the flap hinge line slightly aft. When this flap moves down, the nose of the flap moves slightly upward, closing the hinge gap and projecting a slight "bump" at the hinge line. This "bump" in full-scale aircraft improves maximum lift by reattaching the separating boundary airflow as the airfoil nears stall. I was so convinced this wing would offer an improvement over the Nobler type that I designed my first super semi-scale stunt ship around it. It was to be a T-28B with an exact scale fuselage from North American lofting data obtained from Dave Platt. Imagine my surprise. and disappointment when tests showed that, in spite of the more sophisticated hinge location, my "super" airfoil performed about 10% poorer than the smaller, thinner Nobler airfoil plus its sheet flap. (Fig. 3)
To investigate UC stunt airfoils, one must begin by understanding the best available "classic" airfoil - the Nobler.
A "simple" type flap. Harder to build, nearly impossible to properly hang and 10 to 15% less efficient than Nobler airfoil with sheet flap.
In my NACA studies I also found flaps work better on thicker airfoils than thin airfoils, I suppose because the airflow is better directed across the hinge line and flaps. My own tests showed that a flapped stunt airfoil lifts about half again more weight than a no-flap airfoil of the same total chord length. (Fig. 4)
Next, I found an NACA graph of CL Max of symmetrical airfoils of various thicknesses, with and without flaps (Fig. 6). This chart shows. what many stunt fliers have read for themselves in various publications - a 12 to 15% symmetrical sectibn will give maximum lift. But that is true only without flaps. This graph also shows that flaps work better on thicker airfoiIs. Highest CL max were obtained at approximately 28 to 32% thick sections with flaps. To test this at model size I built a 25% test section and found that it did indeed produce substantially more lift. Using this big improvement in lifting capabiltiy, I designed the Mustunts I and II for the novice stunt fliers to help them lick problems with weight. (Fig. 7)
Figure 6: Variation of maximum lift coefficient for an
Now I was ready to design the Sea Fury wing. I knew that to carry the Sea Fury's weight I was going to have to use large flaps and a 25% section. Take your choice, the wing would have to be either big or thick. Wait a minute! The AMA stunt pattern itself is asymmetrical requiring far more lift for the lower right triangle and hourglass corners than anywhere else in the pattern. Since these are both inside corners I should be able to get away with an asymmetrical airfoil with a 25% curvature on the top of the wing and a flatter 20 to 22% curvature on the bottom for the less demanding outside squares. By tailoring the airfoil to the lift requirements of the pattern, I could have my high lift characteristics and still slightly reduce the bulk of the wing. Again, I decided to "go for broke" and use the completely untried asymmetric airfoil concept on my Sea Fury to improve the appearance of the wing. At this point I built two new asymmetric test sections and waited six weeks for a calm wind night to test them (I live in Texas, you know). Finally, in desperation, I ran the tests in an eight-knot wind-I just had to get started building the Sea Fury wing.
While tests under windy conditions must be inaccurate, I still felt they would be useful indicators of relative performance. Sure enough, the airfoils tended to group on the graphs into families related by thickness. The 25% sections all performed 35 to 40% better than the best of the 18 to 20% Nobler and Bearcat airfoils. Clearly, thickness is far more important to airfoil lifting capability than any other characteristic such as profile or leading edge radius (Figs. 7 & 8).
The Sea Fury 25-20 test section lifted better inverted than the inverted Sea Fury 25-22 which was contrary to what I expected but probably accounted for by the 25-20's blunter leading edge radius. (Fig. 8) Had I conducted these tests in a calm wind, I would probably have used the 25-20 on my Sea Fury. As it happened, however, the airfoils with slightly sharper leading edges tended to run more smoothly, buffeting less in the wind, making it possible to gather suffficiently good plotting data with fewer automobile runs in each direction. This characteristic of smooth operation in the wind was, I thought, more important than the slight loss of lift inverted, so I selected the 25-22 and finally began to build my Sea Fury wing.
I am through testing airfoils, for now, at least. I think the point of diminishing returns has been reached and look for little additional improvement in practical stunt wings. I say "practical" because: 25% is about as thick as a wing can be made and still look reasonably attractive (though thicker wings would, no doubt, lift more); stunt flaps are near the limit of development; practical leading edge devices could be used but would hurt appearance; and the only remaining area of possible significant improvement is boundary layer control which, at present, looks too gimmicky and unreliable for heavy competition use. It seems now that optimum practical results will be obtained by a thick airfoil with moderate leading edge radius, large flaps, and a profiled trailing edge. Every eff6rt should be made to keep the point of maximum thickness as far forward as smooth transitioning of airfoil curvature will permit.
While I used the asymmetrical 25-22 on both Sea Furys, I am not recommending asymmetrical airfoils for general use. They are a "special" solution to a. "special" problem. Tests would seem to indicate that I am getting about half again more lift now from these thick wings than would normally be obtained from conventional stunt wings of the same area. Over a thousand flights with my Sea Furys and almost 500 more on Mustunts seem to verify this approximation.
Since "going for broke" was the order of the day, I decided to use a shock absorbirig landing gear to overcome the conventional gear airplane's tendency to bounce that cost me the 1970 Nats. They only add 1/2 oz. each and work well. As a result, the Sea Fury seldom bounces, even on a hard landing. Also, spectators seem to enjoy watching the Sea Fury "float" over a rqugh surface with the gear working to follow the contour.
The Sea Fury's shock absorbing gears and Rabe's method of lacing them to their mounts with 032 brass safety wire. Will never loosen at all.
Method of assembling molded fuselages.Second half is still on mold. Four aft bulkheads laminated 1/32" ply 3/32 balsa. Note tailwheel mount installed.
What a thrill that first flight was! The sight of a Sea Fury out there on the end of the lines was terrific! Line tension, however, was only fair and seemed to disappear during maneuvering flight. The controls were so sluggish that round loops were difficult. When the engine cut, the glide was good and the Sea Fury settled gently onto its shock absorbing gear to end a highly satisfactory first flight,
I called to Linda for a sandwich and a Coke while I removed the wing. To correct the sluggishness of control, I changed the flap/elevator ratio from 30 degrees flap - 30 degrees elevator to 30 degrees flap - 45 degrees elevator, and added 1/2 oz. of tipweight to improve line tension,
On its second flight the Sea Fury was very responsive to control. Line tension, while improved, was still insufficient and it turned tighter inside than outside. To balance inside and outside turn rates, I removed the wing again to adjust the elevators downward slightly with neutral flap and more tipweight was added to further improve tension.
On the third flight the Sea Fury turned well, both inside and outside, and the general improvement in flying characteristics permitted closer evaluation of specific areas which needed improvement. For example, it was very light on the lines overhead and now that I had time to look, the Sea Fury was flying banked into the circle both upright and inverted. So that's why I had so little line tension! The Sea Fury was my first airplane built with exactly equal span inboard and outboard wing panels and the extra lift of the faster outboard wing was causing it to fly banked into the circle. While I had anticipated a need for adjustable tipweight, I was surprised at the bank angle and a little unhappy when I saw that the tipweight box wasn't big enough. That first night I had to settle for taping extra weight onto the outside of the outboard wing tip.
To adjust the tipweight, I added 1/2 oz. onto the tip each flight. With each addition of tipweight, the airplane flew flatter and line tension inproved until, at three oz total tipweight, the Sea Fury started "hinging" in the squares. With a reduction of 1/2 oz. of tipweight, the tendency to "hinge" was gone and the Sea Fury flew flat with good tension.
To improve the overhead tension, I moved the leadouts forward three times, in 1/4 in. increments until no further improvement in tension was noted and the Sea Fury was beginning to feel "doggy." Moving the leadouts back one 1/4 in. increment resulted in nearly optimum leadout location.
By this time it was beginning to get dark and time remained for only one final evaluation flight This would be a full pattern flown at five feet except for the triangle and hourglass bottoms where I would pull out high and extra tight to evaluate the turning (lifting) capability. The Sea Fury flew beautifully except for a noticeable stalling tendency in the lower right triangle and hourglass corners. Well, I thought, you can't have everything, and all things considered, I was pretty proud of my 71 oz, airplane at this point.
That night I thought about the stalling tendency and decided that it might be improved if I could only hang the flaps out a little farther for more lift in the corners. To accomplish this, I changed the flap-elevator ratio to 30 degrees flap - 37 degrees elevator. The change of ratio would definitely use more flap for any given rate of turn, but the extra flap would also reduce the effectiveness of the elevators making the airplane appear to turn sluggishly again. To compensate for the aerodynamically reduced sensitivity, I modified my small E-Z-Just handle for wider line spacing by cutting the plastic, moving the lines to the extreme top and bottom ot the handle, and epoxying small pieces of plywood into the slots under the relocated lines. The next day's flying proved the combination of ratio change and handle modification did reduce the stalling tendency to the point that the triangle and hourglass corners could be tightened normally with no sign of buffet and no apparent change of sensitivity.
When I tried flying the Sea Fury with its spinner, I found that it ran smooth and true but the nose cap would fly off each time the engine stopped. It seemed the precessional effects of high pitch rates were flexing the spinner backplate along the propeller axis where the backplate was weakened by the prop blade cutouts. This flexing would loosen the nose cap. After stiffening the backplate with a fitting machined by Bob Wilder, the problem was completely cured. In fact, considering spinner vibration problems that I have had on smaller airplanes, I am amazed at the rather unbelievable smoothness of this very large spinner. (This flexing has since been corrected by Williams.)
To summarize the Sea Fury's flying characteristics at this point, overall I thought it flew very well and was definitely competitive. It turned as well as my Bearcat, but felt and looked much smoother due to the increased tail moment. Line tension was good except for the top of the vertical eight In wind over 12 knots. In a strong wind I would occasionally run out of elevator in the vertical eight as I started down from overhead. Also, the shock gear seemed to cause bouncy landings but that turned out to be a simple matter of improper location. Bending the gear back slightly fixed the landings completely.
Finally, however exciting, the Sea Fury was difficult to fly. It could fly any maneuver competitively, but only nine out of ten times. I found that no matter how much I practiced, I would usually bounce a corner or miss a pullout or an intersection on nearly every flight. Still, if I could put it together at the Nats, the combination of smooth corners, good shapes and impressive appearance could very well win.
At the '71 Nats, neither my flying nor the Sea Fury's appearance seemed to make much impression on the judges in the first round of finals. I wound up "in the pack" in the mid 440s. In the second round, the Sea Fury nosed over on takeoff after being released downwind in a strong, gusty wind condition. All right, so an honest conventional gear is still a competitive disadvantage, even with shocks. A month later, I found this type of takeoff accident could be prevented by launching into the wind. A 73 oz. airplane seldom becomes airborne accidentally.
I think most Nats Stunt fliers were impressed by Gene Schaffer and his "Stunt Machine" performing in the wind. Gene's pattern seemed typical New York style except for his rather blinding corners. Chuck Hora thought Gene's corners looked as though the airplane had been "nailed," then "swiveled." While I also thought Gene's corners were unnecessarily tight, they were impressive and attention-gathering in their own right, and like beauty or semi-scale, served to attract the attention of the judges for "out of the pack" scores. I went home with the realization that the Sea Fury's corners should be further improved before Cleveland's FAI Finals to improve my chances of making the FAI team.
As a matter of fact, I had two problems to overcome. First, the Sea Fury left a heavy smoke trail which distracted the judged by drifting in the rounds and being blown to the ground by the flaps in the bottoms of the squares. This caused the Sea Fury to appear to bobble when, in fact, it hadn't. Since the ST 60 would tolerate much less oil than Superfuel's 29%, I mixed my own brew; 5% nitro, 20% Ucon oil, and 75% methanol. The smoke problem was cured, but the extra 9% combustibles increased the engine run from 6:15 to 7:00 which resulted in overruns in the FAI pattern. I decreased the run by enlarging the venturi from .305 to .315 ID. Burning the extra 9% combustibles in 6:15 instead of 7:00 released extra power from the engine which "perked up" the Sea Fury's performance all around.
Second, I began retrimming. I wanted to increase the crispness of my maneuvers by tightening all of the corners somewbat. To accomplish this, I added nose weight, knowing that more stability would result from a more forward CG location. Now more control deflection would be required to maneuver, The extra flap deflection increased lift again permitting sharper corners and a troublefree vertical eight. Admittedly, it also took more deflection of the handle too, but I soon became used to it. The extra stability from the increased nose weight also cured the Sea Fury of being difficult to fly. It no longer had a mind of its own. It drove smooth and tight to crisp, accurate corners. At last the Sea Fury had arrived as an unlimited competition stunt ship. Five flights on that snarling, pulling, groovy son of a gun would spoil for anyone, forever, the put-put of a Fox 35.
On the day l was to leave for the FAI Finals, I ran out of fuel in cloverleaf on my last practice flight. Rather than accept a safe inverted landing with minimal damage, I "went for broke" again and tried to whip it through to save the airplane, undamaged, for Cleveland. It didn't work. Looking down at the mess, I knew my "go for broke" year had ended.
As tar as I was concerned, the Sea Fury had proven itself a competition stunt ship even though it had never won a contest. Excepting Bob McKinney. no one had seen it fly well. Rather than let it end there, I decided to rebuild the old Sea Fury and to start immediately on a new Sea Fury incorporating improvements based upon experience gained from 303 flights on the original.
First, ground handling could be improved by shortening the landing gear a little to provide a flatter sitting attitude. The old gear had turned out over scale length anyway. Shortening the main gear 1/2 in. would actually improve upon scaleness. I would, of course, retain the shock absorbers as the guaranteed landings and visual effect were certainly worth the small extra weight.
Second, it seemed that slightly larger stabilizer and elevators would further improve the Sea Fury's "groove." Also, the larger elevators should allow changing the flap-elevator ratio of 30 degrees flap-37 degrees elevator used on the Sea Fury I, to 30 degrees fIap-30 degrees elevator, thereby wringing more lift from the wing's available area. This would permit still tighter corners.
While designing the new stabilizer, it seemed a good idea to incorporate some "direct lift". "Direct lift" is simply rendering the elevators aerodynamically ineffective or insensitive around neutral. With "direct lift," small control handle inputs have little or no effect at the elevators but the accompanying small movements of the flaps cause the airplane to rise or descend smoothly, without any change of pitch attitude. By making most small flight path adjustments with the flaps only, the apparent smoothness of the flight is greatly improved.
There are several easy methods of obtaining "direct lift." The first, and most common, is simply to drill out the elevator pushrod bushing in the elevator horn to put "slop" in the elevators. Most fliers who use "slop" can move their elevators up or down about 3/16 in. at the trailing edge without moving the flaps, or, more properly, can move the flaps without moving the elevators. Another method used by World Champion Bill Werwage on his "Pacemaker" is to make the stab thicker than the elevators. Also, at one time or another, RC fliers Phil Kraft, Jim Kirkland and Art Schroeder have related their experiences with stab types and generally agree that an airfoiled stab arrangment will have less elevator sensitivity around neutral than a flat stab setup. Since I practice 800 to 1000 flights each year, I already have problems enough with wear, so I decided to avoid "Slop" and combine Bill's thick stab with as much airfoil as my construction method permits to obtain "direct lift" characteristics for the Sea Fury II.
Third, just in case the ratio change for increased lift was not quite enough by itself to obtain really outstanding corners, I would add one in. of span to each wing panel for a 4% increase in wing area.
Finally, I would try again to reduce weight, particularly in the tail. Because of the Sea Fury's unusual moments, a fraction of an oz. saved in the tail would be greatly magnified in the reduction of nose weight required. For example, the wing fillets were already molded from 1/16 in. sheet balsa but the Epoxolite fillets at the fin, stab and fuselage junction could be replaced with 1/32 in. balsa sheet fillets. A plastic hub Williams tail-wheel would replace the metal hub Perfect. Additional weight savings could be realized by skeletonizing the main gear platforms and spar webs. In addition to the normal hollowing of the wing tips and cowl ring, the rudder and elevator spars were hollowed as were the specially cut wing trailing edge pieces. I'm sure you get the idea. In fact the Sea Fury II did weigh nearly two oz. less than the Sea Fury I. I had hoped for a greater savings. It looks like the larger wing and stab offset most of my additional effort.
The new Sea Fury and repairs to the old were begun simultaneously. The Sea Fury I needed a completely new forward fuselage assembly beginning just ahead of the cockpit. Molding the new fuselage half-shells fitted nicely with the molding of the new "II" fuselage shells and much time was saved by constructing duplicate bulkheads, firewalls, tank compartments, cooling air bypasses, cowl rings, etc. By the time the Sea Fury l's structure was repaired, the basic fuselage for the Sea Fury II was complete.
I'm certainly glad that I repaired the broken Sea Fury because work progressed slowly on Sea Fury II due mainly, I guess, to watching too much TV in my workshop. By the time I finished Sea Fury II on July 4th, I had been able to put in nearly 300 practice flights and two contest wins on the repaired Sea Fury I. Also, the repaired Sea Fury was used to break in several new and overhauled engines and at least 100 flights were made using Sea Fury Il's tank, engine mounts, and muffler before Sea Fury II was finished.
My final change to improve the overall appearance. of the Sea Fury's pattern was to increase the control-line length from 65 ft. to 70 ft. People have remarked before that a large airplane like the Sea Fury tends to look cramped for space while maneuvering on my old lines. With the new, longer lines this cramped appearance was completely eliminated. Now people complain about the size of my maneuvers. Anyway, the net effect was definitely beneficial so the Sea Fury II was trimmed to fly best on the longer lines.
I did notice one curious effect of flying on longer lines. To maintain the necessary five sec. lap time for a 6:30 pattern, I had to lean the engine slightly. The leaner mixture brought unexpected fuel economy and my engine run went up to 8:00. Further enlarging the venturi led to instability of my run, so instead, I shortened the duration to 6:30 by increasing the nitro content of my fuel to 10%.
By working a less desirable schedule and losing a lot of sleep to early morning practice, I was able to get 150 flights on the Sea Fury II before leaving for the 1972 Nats in Chicago and another 45 flights in Chicago before Open finals. With a good airplane and a luck draw in flying order, my Sea Fury and I were able to pull off a win in Open Stunt. This was particularly satisfying as a few short years ago separate Stunt events were sometimes held for semi-scale stunt ships because of their supposed inability to compete effectively with "classic" stunt ships. This year, realistic semi-scale stunt ships took first, third and fourth in Open Stunt and first in Senior Stunt. Maybe now I've heard the last of those cracks about semi-stunt scale ships.