What would you do if you were invited to fly a one-of-a-kind, hand built, world-record-setting airplane? You would probably do what I did: think about it, research it, and accept the invitation.

 The call came from Mike Arnold, who spent 10 years designing and building his AR-5 tail dragger that set a world speed record in 1992 in the FAI C1aO class for airplanes weighing less that 661 pounds at takeoff. The 2-kilometer closed-course AR-5 record, set in August, 1992, was 213.18 mph on AR-5's 65-hp Rotax 582 two-stroke engine. The record stands to this day.

 At the time of his call last spring, Arnold and a friend of mine, writer Peter Lert (who has written about other subjects in KITPLANES), were the only two people to have flownthe AR-5. "I'm thinking of selling the AR-5," Arnold said, "and I'd like you to have a chance to fly it before I do."

Research Time:
 When in doubt, check it out. I reread aerodynamicist Bruce Carmichael's October 1993 KITPLANES article on the AR-5. The airplane is built entirely of fiberglass and foam using the moldless composite technique pioneered by Burt Rutan on the VariEze.

 The AR-5 was intended originally as a fast, personal sport plane that Arnold could use to keep up with his father in his Thorp T-18 and friends like Kit Sodergren in his Midget Mustang--without the cost of buying and operating a Lycoming or a Continental.

 Rather early in the design process, Arnold realized that he could set a world record with the AR-5 (his fifth design but first construction, by the way); the previous C1aO record was near 120 mph.

 Visiting the project, Carmichael was amazed to discover incredible attention to detail both in design and construction. Laminar flow over much of the wing's surface would be critical to drag reduction. The 8:1 aspect ratio wing features a NACA 65-418 airfoil at the root tapering to a NACA 65-212 at the tip. This wing would promote superior laminar flow if smoothness and lack of waviness were good enough.

 Using reflected light, Carmichael looked for ripples and saw little waviness. He them rocked a steel straight edge fore and aft on the wing, listening and feeling for unevenness. Finding little, he measured waviness by moving a dial indicator along a piece of masking tape placed in a chordwise direction on a wing. The indicator was centered between two pegs 2 inches apart. His largest peak-to-peak reading was 0.0036 inches over a 3.6-inch length, revealing a waviness of one part in 1000. Farther forward on this chord line, waviness was found to be only one part in 4000. Remarkable!

 The fixed landing gear,engine cowling, cabin air intake, empennage treatment, and the wing/fuselage junction were all found to maximize best drag-reduction techniques.

 Using an estimated fixed-pitch propeller efficiency of 82% and the record-setting speed on 65 hp (which was purposely flown at an allowed descent angle of 1 degree 18'), Carmichael determined that the AR-5 has a drag area of only 0.88 square feet. Formula1 air racers come close to this figure, but it is possible that AR-5 has the lowest drag and the lowest drag coefficient of any man-carrying, propeller-driven airplane in the world. Wow! 

I called Peter Lert about flight characteristics.He noted the left-turning propeller, the sensitive rudder, and the somewhat disconcerting loading of the high-pitch cruise propeller late in the takeoff run. The cruise prop unstalls and loads the engine, dragging rpm down about the time you get airborne. Arnold had said I'd fly with a climb prop, which doesn't have this annoying problem. Lert assured me that I'd like the airplane.

Putting It Together:
 A few days later I called Arnold, and we set up a schedule to get together near Sacramento, California. Thirty minutes later, Bruce Carmichael called to sell me on an article idea, and he mentioned the AR-5. We hadn't talked in a year. "Did Mike Arnold tell you I'm about to fly it?" I asked. "No," he said. "I haven't talked with Mike in a year and a half." Talk about a coincidence!

 On the appointed morning I arrived at Sacramento International Airport on schedule and was met by Harry Arnold, Mike's father. We drove to Yolo County Airport, where the AR-5 is based.I'd arranged with Ray Ferrell of Sky Dance Sky Diving on the airport to photograph the AR-5 in the air. Strapped to the right front floor of the Cessna 180 and facing backward to shoot out the open door, I motioned Mike and the AR-5 into various positions as we circled. You see the result here. {We aren't able to reproduce Dave's beautiful pictures here, so we've substituted some of our own shots for the web page. MA}

Covering the Quirks:
 Back on the ground after 20 minutes,Arnold prepared me to fly the AR-5. As we looked into the cockpit, he described the controls. They seemed conventional until he got to the throttle, which is hooked up backward. Pull for more power and push for idle. What?! "I know. Peter Lert screamed too. But he found it not to be a problem," Arnold said. He says he rigged the throttle this way because it was lighter and simpler...and because he intended to be the only pilot.

 There's a vernier on the throttle that helps set power precisely. In this case it also slows throttle application,which is helpful with a nonstandard installation. My experience, it turns out, was like Lert's; the screwy throttle wasn't a problem.

 Next Arnold covered an AR-5 powerplant characteristic that could be a problem. Because of the propeller's high pitch and the airplane's high speed, reducing power to midrange during a descent can result in overheating the engine enough to seize it. How's that?

 Pulling power (which in the AR-5's case means pushing the throttle) in most fixed-pitch airplanes means reducing engine rpm, even in a descent. But because of the high pitch of the AR-5's propeller--even the climb propeller--the engine tends to maintain its cruise rpm in a descent with the throttle back some (err, forward; you know what I mean). The result is reduced fuel and the same amount of air: a leaner mixture that results in instantly higher EGTs. Keep the EGT to 1200 degrees F or less, Arnold cautioned. Let it reach 1450 degrees F and you stick aluminum parts together. The solution in a descent is to leave full power set or to reduce it to idle. The technique works.
 

One other unusual feature of the AR-5 would be anticipated by anyone who has flown a composite taildragger with a tiny tailwheel: extremely noisy taxi. "You're the dog listening to His Master's Voice," Arnold said, referring to the old mechanical RCA Victrola record player. The pilot sits at the end of a megaphone attached to the needle-size (2-inch) tailwheel scratching along the concrete. "When the tailwheel contacts the runway after your wheels landing," Arnold predicted,"you'll think it came off." Fair warning.

Flying It:
 After watching Arnold take off, fly the pattern, and land using his recommended wheels-landing technique, I checked the airplane externally and scrambled aboard. There's not much to check in the cockpit. Controls free. Flaps retracted for takeoff. Ignition switch on (there's a single electronic CD ignition that sparks a pair of plugs in each cylinder). Clear the area and start. Canopy latched and locked. Toe-operated wheelbrakes are moderately effective.

 The first surprise was how good visibility is while taxiing. Cushions below and behind achieves about the eyeball position for me as for Arnold without them, and I could see the taxiway everywhere except directly ahead. Little S-turning is needed to clear thepath. I set 3000 engine rpm (the prop turns half as fast) for taxi to minimize clatter and wear on the speed-reduction gearbox.

 On the runway, I slowly dialed in some power and eased the stick forward to raise the tail as the airspeed indicator came alive. Because the prop turns counterclockwise, raising the tail requires a little right rudder because P-Factor is momentarily reduced. On getting airborne, a touch of steady left rudder is needed to counteract P-Factor in the climb. Immediately apparent during the takeoff run is that the rudder is highly sensitive. A bit more practice would be required to achieve a smooth, straight takeoff.

 I aimed for 110 knots to keep the Rotax's coolant temperature within limits, and that results in a less-than-spectacular climb angle, although the rate is acceptable. Leveling initially at 2500 feet west of the airport, the AR-5 confirmed Arnold's note that no rudder is needed for rolling maneuvers at cruise airspeed. Setting a moderate 5500 rpm achieved close to 140 knots level, and I set up for my standard 45 degree-to-45 degree roll rate check.
For an airplane with a 21-foot wing with relatively high aspect ratio (8:1), the AR-5 rolls remarkably fast: not much more than 1 second from 45 degrees to 45 degrees. The roll is equally rapid both directions, and little or no rudder is needed. "You can roll or loop at 120 knots," Arnold had said, and for a change I took him upon it and tried an aileron roll each way. Really nice!

Next came the stick-free pitch stability check. For this, we need hands-off, trimmed flight, and the AR-5 has no trim control. But with my weight and this loading, the plane flies level without help at 135 knots. After easing the stick forward a bit, I let go and waited. The lazy pitchdown slowed, reversed upward and passed through horizon going up. Pitchup slowed and reversed, and this time the pitch downangle was less than before: Damping indicated positive pitch stability at this configuration. The rate was long --40 seconds per cycle-- but stability was there.

 Slow flight and a first use of the three-position flaps followed. Maximum flap speed is 95 knots, and I simulated flying a landing pattern at 3000 feet using a road for orientation, remembering to "push" the throttle to idle before descending. Fly downwind at 110 knots, turn base slowing to 100, apply flaps and continue slowing to 80 knots.The descent with full flaps is steep, and the intent is to flare low to a level-attitude wheels touchdown.

 Approach and departure stalls revealed that the AR-5 is docile at this edge of the envelope. A gentle stall break occurred at 56 knots with flaps up and 52 knots with flaps down. Arnold says the ASI indicates somewhat too high at the low-speed end of the dial.

 A last check before the first real landing was a full-throttle level run. Wearing its climb prop, the AR-5's Rotax can exceed the marked 6500 rpm limit, and I made it a point to reduce power as the tach needle wound up toward the engine limit. Indicated airspeed (which is quite accurate at cruise and faster speeds, Arnold says) was near the top of the yellow arc: about 165 knots.

 Lacking an outside air temp gauge, I can only estimate temperature at 3000 feet msl. If the guess of 78 degrees F is close, true airspeed was 178 knots (207 mph) on the climb prop.

 Entering the left 45 degree leg for Runway 34, I tried to duplicate the higher-altitude dry run: 110knots on downwind, slowing to 80 with full flaps on final, flattening the final approach with a bit of power. Touchdown in a level attitude was soft and looked normal and I relaxed, which was mistake.

 Failing to ease the stick forward upon touchdown, and also leaving the flaps down, I found myself back in the air a foot or so after a 5- or 6-second roll. A couple of ungraceful runway contacts followed, and the tailwheel megaphone let me know when it was firmly on the runway.

 Fortunately, none of the several runway contacts scratched the rear of the wheelpants, which is one reason Arnold makes wheel landings in the AR-5. Missing the midfield turnoff, I slunk to the far end of the runway having bruised nothing but ego.

 Followup back at the hangar:
 Arnold reminded me of some of the details required to achieve the AR-5. After a 10-year career as a filmmaker, Arnold turned his interests and talent to aviation, where he worked on the composite Amsoil Racer designed by Burt Rutan. An apprenticeship with composites guru Fred Jiran explains AR-5's gorgeous, low-drag conception.

 Engine cooling was initially a problem, and Arnold spent countless hours on baffling. A special oval exhaust expansion chamber was designed to fit under the slim cowl. When a Rotax 582 replaced the original Rotax 532, the 582's oil injection system had to be removed as it would not fit.

 Arnold has fine-tuned the AR-5 in the years since the world record. He changed stick force ratios and made other modifications including lightening aileron forces and increasing rudder force a little.

 Before leaving for home, I told Arnold that the AR-5 is one of the handful of airplanes sampled in this job that are memorable because of their excellent handling. I compared it to the feel of the Stelio Frati-designed F.8L Falco (available as a kit from Sequoia Aircraft).  Mike Arnold understood the compliment.

-Dave Martin, Kitplanes Magazine