home
aircraft database

Aeronca C3
Aeronca Champ
Aerostar Super 700
Beech Akro Musketer
Bellanca Cruisair
Bellanca Viking
Square Tail Cessna Hawk
Cessna Hawk XP
Cessna 120 and 140
Cessna 170B
Cessna 185
Cub Comparison
Clipped Cub
de Havilland Chipmunk
Fairchild 24R
Fairchild 45
Globe Swift
Luscombe 150hp
Luscombe 8 Series
ME 109 E
Meyers 145
Maule M7235C
Mooney Mite
Mullicoupe
Navion
North American AT6
One Design
Piper J-5 PA-12 Cruiser
Piper J3 Cub
Piper Apache
Ryan SCW
Ryan STA
Samba
Spitfire MK IX LF
Stinson L-5
Stephens Akro (Laser 200)
Stearman YPT-9
Storch (Fieseler Fi-156)
Taylorcraft BC-12
WACO QDC
Zivko Edge 540

 

Mullicoupe

by Budd Davisson, courtesy of www.airbum.com

Mr. Mulligan's Kid

Every pilot has several images stored up, which he or she projects on their mind's eye from time to time as a way of remembering a wonderful moment. The one I replay the most is the image and feeling of dropping the hammer on my first Bearcat takeoff. The acceleration was so sudden and all-encompassing it felt as if it was launching me through some sort of space continuum and I'd emerge on the other side a different person. In fact, I did emerge a different person. It changed my way of looking at airplanes forever.

Every takeoff since that first Bearcat experience has been a comparison, another step in the search for the same exhilaration. Until this week, only one airplane has come even close, the Grumman F3F/G-32A.

And then I flew Jim Younkin's Mullicoupe last week. Now there's another image to rotate through my mental theatre with those of the Bearcat and the F3F. The Mullicoupe definitely left it's mark.

Any who were at Oshkosh undoubtedly came away with stories of the two hulking red and black airplanes that were "...sorta Monocoupes and sorta Howards." A spectator didn't have to know what the airplanes were to know they had just seen magnificent examples of the airplane builder's art. If they hung around long enough to pick up the specifics, they'd realize those were Bud Dake's and Jim Younkin's Mullicoupes, just two more in a long, long line of nearly unbelievable flying machines to bear Younkin's unique touch.

It's probably redundant to once again explain, or try to explain, Jim Younkin who operates Historic Aviation in Springdale, Arkansas, as he's regularly mentioned in these pages. However, it is very necessary to point out that he is far more than simply a builder, or a designer or a creative thinker, even though in each of those categories, he may well stand at, or near, the head of the class. Younkin's combination of talents may well make him unique in our field. There are a number of others who can free form aluminium nearly as well. There are others who are adept at designing and engineering. There are numerous shops out there that build and restore airplanes as well. There are very few, however, who, like Jim, combine it all.

Quite often, when an individual has the above kinds of characteristics, he is likely to be selfish with what he knows. Just the opposite is true with Younkin. Walk in his shop with a question and he's likely to drop what he's doing, step over to the trip hammer or English wheel and show you how it's done. He's almost zealous in his urge to get knowledge and understanding about what he's doing out to other people. His much-modified Piper Pacer is a classic case in point: It's hard to imagine how many other Pacer owners have borrowed ideas from Jim's airplane, most with his help. It's probably the most copied airplane of its type ever, although he often doesn't receive credit for some of the mods.

Jim is blessed with the intellect and the skills that are required to take any idea and make it a reality. For that reason, when he starts musing about a particular restoration project or new design, it's a good idea to sit up and take notice. Jim Younkin's day dreams almost always become reality. That's exactly how the Mullicoupes came to be.

Actually, Younkin blames Bud Dake, he of the familiar menacing black Clipwing Monocoupe and central figure in the on-going Creve Couere aerodrama, for the Mullicoupes coming to be.

Younkin had taken Mr. Mulligan, his 600 hp recreation of the golden age racer, to the antique fly-in at Blakesburg for the first time ("...I didn't really plan on landing, as it was way too short, but there the runway was, so..."). Towering over so many of the other antiques, it was the center of attention. The year was 1982. He and Bud Dake were sitting in the shade admiring the airplane's lines that afternoon when, according to Younkin, Dake said something to the effect of, "...you know what we really need is a 450 hp, two seat version of Mulligan just for personal transportation..." Younkin agreed.

As Younkin remembers it, they talked about it for a couple of days at the fly-in, deciding such an airplane should borrow heavily on the lines of the Monocoupe. Jim says that's when the name "Mullicoupe" came into being.

When Younkin came home he started doodling. Jim Younkin, however, doesn't doodle like the rest of us. His doodles have numbers and dimensions attached. So, in another year or two a totally accurate scale model of the airplane took shape in Younkin's shop. He had the concept. He had the dimensions, so he did the next natural thing and started cutting metal.

At that point he was building a single airplane for himself, but it wasn't long before Bud Dake and Red Lerille, another Monocoupe fanatic, had jumped on board. So, when Jim built a component for his airplane, he'd build one for theirs as well.

For a number of years the Mullicoupe was a "fill-in" project as Younkin's shop was completely immersed in a Staggerwing Beech assembly line. At one time he had four Staggerwings lined up, each receiving massive amounts of restoration, modifications and aluminium work.

Finally, several years ago, the Staggerwings were finished to the point they were ready to be delivered to the owners, and the Mullicoupe project really got serious.

As quickly as Younkin would finish a component, it would be shipped to Dake and Lerille and the race was on to see who would fly first. Dake won.

On the first flight, they discovered the airplanes needed much larger vertical fins. The tiny surface which was desired to maintain the Monocoupe look just wasn't large enough. Younkin produced larger surfaces for his and Lerille's airplanes, but Dake, true to his nature, was beginning to like an airplane that had little or no airborne directional stability so he hasn't modified his.

The final airplanes are a little daunting both in presence and in specifications. From a distance, it would be easy to mistake them for D-145 Monocoupes. As the distance is closed, however, they grow in size until it's realized they are very serious airplanes. The lines start with the flawless bumped aluminum cowls wrapped tightly around a fuel injected P & W R-985. The lines flow back with more Monocoupe than Mulligan in them until they wasp-waist their way down to the tiny tails. The wings are very much Mulligan in both line and execution.#

The 29'3" wings look short for the dense-looking fuselage and in fact are short: The wing loading goes from 24 to 27 pounds/square foot, depending on how much of the 150 gallon gas tanks are filled. Even with the 22 gph fuel burn of the Pratt and Whitney, that gives a solid (and impressive) 7 hours of range at a high altitude cruise speed of 225 mph.

I was more than just a little apprehensive as I backed up to the door to hoist myself up into the cockpit (the door jam is nearly chest high). The airplane has a pugnacious presence about it that just sitting on the ground says "fly me if you can." Bobby Younkin, Jim's son and the well known airshow pilot of both a Twin Beech and Samson, scrambled into the other seat to help make the introductions.

Once inside the airplane, the extreme deck angle, 15°, was clearly evident. The cockpit slanted downhill much steeper than any airplane I'd ever flown. This was going to be a challenge.

When the engine cranked, it caught on only the second or third blade laying down the wonderfully classic opening movement of Symphony de Round Motore.

The airplane doesn't have a steerable tailwheel, so most ground handling is done with brakes. I had only a short, five foot wide stripe of the side of the taxiway visible at an extreme down angle, so I was at first a little overly cautious. I found quickly, however, that the nose tapered fast enough that by leaning against the side of the cockpit, I could actually see a fair distance ahead and only a slight S-turn was necessary to clear the taxiway ahead.

The cockpit has an open cheery feel to it because of the sky light but the feeling is also helped by the antiquey, vaguely triangular instrument panel. If the panel hadn't been scooped out at the sides and had a round top like most instrument panels, the visibility to the quartering sides would have been abysmal. As it was, I had a clear view at an angle of about 40° (a guess) off the nose.

As I glanced around the cockpit I again remembered Younkin's description of the fuel system. The tank selector is between the seats, along with the push-pull tailwheel lock. What is not visible is the float and vent system in the header tank. R-985's with fuel injection, rather than the usual carburetor, take a frighteningly long time to restart once you've run a tank dry. Sometimes the silence lasts as long as 20 seconds. Younkin's fix for that is a low fuel warning light in the header tank that lets you know you've only got a few gallons left before the pilot light goes out. At the same time, the float activates a servo which opens a direct vent into the header tank so, when you switch tanks, it will fill faster. If, for some reason, the servo doesn't activate, there's a tiny spigot over the co-pilot's head that manually opens the vent for faster filling.

As we rolled out and centred ourselves on the runway, I could see it was going to be a real challenge to actually get the airplane into three point position on landing. At that point, I didn't know exactly how much of a challenge it would be.

I also didn't know how exhilarating the takeoff would be. At our weight, the power loading was down around 5.2 pounds per horsepower which is about on a par with the lighter aerobatic specials. But that number doesn't take into account the effect of having nearly 1000 cubic inches feeding that big prop.

I fixated on the furthest point where the edge of the runway hit the nose and started easing the power in. The airplane responded by instantly leaping forward. When I saw the airplane wasn't going to go darting off one way or the other, I finished putting the rest of the power in and hung on. I was just in the process of picking up the tail when the airplane got light on its feet and I was slow to react to the message. The result was we didn't separate cleanly and I let a slight crosswind push us a little.

Gheez! I was thinking. Gimme a break. I hadn't caught up with the airplane yet and it was ready to takeoff long before I was.

I glanced at the airspeed as we left the ground and was amazed! We were rocketing through 110 mph and I had just barely gotten full power in.

I brought the power back to climb settings immediately and held what I knew was a fairly shallow climb angle. This seemed a smart thing to do around an airport. It was obvious the Mullicoupe would sustain any nose angle I wanted, but I'd be blind as a bat in a bucket. With everything all squared up, I again checked the airspeed. We were indicating 160 mph on a very hot, humid day but the VSI was showing 1800 fpm! How's that for a cruise climb, sports fans?

Through out the entire takeoff and climb out, the world didn't exist out the other side of the airplane. The width of the cockpit and the bulk of the nose conspired to shut that part of the world out of my view. As the nose came down, the view improved dramatically, but the very top of the cowling was still slightly over the horizon and it was hard to see out the other side. That, however, is just the way old airplane are and, as far as the Mullicoupe crew is concerned, what they have built is a new, old airplane.

The concept from the very beginning was that they would take up where Benny Howard had left off with the Mulligan and do it they way they would have done it in those days. As we rumbled across Arkansas at 5,000 feet and 205-210 mph TAS, it looked to me as if Benny would be proud to have his name associated with the Mullicoupe.

Jim told me Bud Dake had conducted exhaustive tests at 11,000 and determined the airplane's best cruise speed was indeed 225 mph at 22 gph. This is only 5 mph slower than the Mulligan after which it was patterned. Younkin also said the airplane really needs a 12:1 blower drive rather than the stock 10:1 which would make it much more efficient at higher altitudes.

The controls are much better than any Monocoupe I've ever flown. Actually, they are better than any Howard's (with the possible exception of Mr. Mulligan). There is no way to describe them other than they are "normal." The break-out force around neutral is just about right and the aileron pressure goes up slightly with displacement (positive gradient). The response is better than a modern Cessna and on a par with a new Beechcraft.

Rudder is extremely powerful and it took a while to get my feet toned down so I wasn't slamming the ball around. Younkin has a locking bar for the rudder which locks it straight ahead for cruise which adds it's area to the fin thereby making the airplane more directionally stable in cruise.

I brought the power back and started setting up for a clean stall. I was doing fine until we got down to around 110 (we'd been indicating around 195 mph). Then the airplane didn't want to slow down any more and I was out of trim. I keep pulling, the VSI kept going down, and eventually we started shedding excess speed, but it was a struggle. Finally, down in the low 80 mph range, the stick came against the stop where I held it to see what would happen. The airplane mushed ahead and nothing happened.

Then I started playing with the flaps. The flaps are another Younkin original and are beautiful the way they work. They are true Fowlers, but he's been able to keep the entire track within the wings surface. They track back 5 1/2" before they go down more than a few degrees. He considers half flap to be about 10° or so. That amount of movement takes up about 80% of the track length. The final percentage of travel cranks them down 30° quite quickly.

The flap controls are two switches. One is a traditional three position momentary switch that can be used to select any flap position. It can be paired with a second switch marked "1/2" and "full." With that switch in the "1/2" position, slapping the power switch all the way down gives only half flaps. Then, when you're ready, you can just push the position switch down to "full" and that's what you get.

They hadn't finalized their flap speeds when we flew the airplane but they were using 110 mph for half flaps and 100 mph for full. Getting slow enough to get full flaps was the biggest challenge of the flight. Half flaps gives almost no drag and the airplane doesn't want to slow down. Also, we were nearly out of trim at that point so stick pressures were building. As soon as the rest of the flaps went out, however, all was right with the world and the airplane was perfectly willing to sit at 100 mph with very little help from me.

I had expected the full-flap stalls to be much more pronounced, but even tugging the stick into my lap and holding it there produced little more than a bobbing mush in the low 70's mph. As the airfoil is the tried and true 23012, I'd expected a sharper break. We didn't investigate accelerated stalls which I suspect may have been sharper.

I let Bobby shoot the first landing, which was a tail-low wheely. I made the second approach with no intent of actually landing. I just wanted to get the feel of the airplane in approach, so I flew it into ground effect and added just enough power to keep us skipping along the ground after a brief touchdown while I tried to develop some references. Throughout approach, the runway was clearly in view but it disappeared the second the nose was brought up. If a person isn't used to flying blind airplanes, this airplane is going to be something of a shock. However, the side of the runway is clearly visible because of the way the windshield wraps so far down the side of the fuselage.

On my next approach, I resolved to try for a three-point. I wrestled it down to 100 mph, got the flaps out and made a curving intercept with centerline. Runway disappears behind the nose as I begin pulling and then pulling some more. Then I felt the mains touch long, long before I was ready for them, so I nailed it on in a wheel landing. It probably looked okay from the outside, but it wasn't what I wanted.

With it's running on the mains, only a little more of the runway is visible, but it is so well behaved it doesn't make any difference. The tail came down and the airplane still tracked straight. No sweat. Power up more firmly this time. I was beginning to get comfortable and was enjoying the takeoffs. What a blast!

Okay, I was telling myself. Next time around, I'm getting the three-point position and nailing it on. Curving approach, good speed, good position. I spot the side of the runway and start trying to hold it off. Ground effect wasn't helping me much so I was working hard at rotating fast.

I wasn't fast enough. As I was pulling and we were in what I thought was a three-point attitude, we ricocheted off the mains, this time leading into a healthy skip. It was one of those,"...do I or don't I add power?", marginal situations. I opted not to add power. That was a mistake. As the airplane started back down, I increased the back pressure only to find the stick was already against the stop. The mains touched and bounced, then the tailwheel touched and bounced. Then the mains came down again, then the tailwheel. I've got the stick nailed to my belly and all I can do is sit back and watch a spectacularly ugly crow hop landing.

The good news was that at no time did the airplane want to do anything but roll straight ahead, nor did it unload and actually drop us hard. The only bad result was severe embarrassment for me. If I'd just touched the power on the first hop we'd have been okay. After that, it was too late.

By the time I'd taxied back to the hangar, I'd thought about the entire process enough and was ready to go at it again. This is an airplane I'd really enjoy getting good in. Not to mention the adrenaline rush every time the throttle went in.

The three Mullicoupes now flying are likely to be the only ones ever to be flying. Although Younkin has been approached by several people wanting him to build components for them, he hasn't decided he really wants to do that.

The Mullicoupe is definitely not everyone's airplane. But for those drawn to nostalgia and who love round motors and serious performance, this could very well be the answer to a dream. Who knows? Maybe Jim will relent and open the doors to others wanting a taste of Howard's legacy as interpreted by Jim Younkin. If not, at least the rest of us can admire them from the sidelines.

A Short Hop With Mr. Mulligan:
When Benny Howard designed and built Mr. Mulligan in the late 1930's there would be no way he could have imagined it being replicated 60 years after the fact. When Jim Younkin arrived at Oshkosh with his near-perfect replica it blew minds right and left.

When Younkin, Bud Dake and Red Lirille decided to build the Mullicoupes it was with the goal of building a better Mr. Mulligan in smaller scale. So, when Jim offered me the chance to fly Mulligan, I felt it was important to my scientific research that I force myself to take the time to accept his invitation. Yeah, right! I'd wanted to fly the airplane since the first time I saw it. Besides, it really was important I sample what had set the Mullicoupes in motion.

Both Dake and Younkin have said they feel as if they fell short of their goal to better the Mulligan.

On the first takeoff, the first thing I noticed was how much better I could see out of the Mulligan. Better, however is a relative term as most people would still think it blind as a post hole.

Also, I had no trouble saying up with the airplane and getting it solidly on its mains before it caught me unawares. That may be because I'd just stepped out of the Mullicoupe or because the airplane, with a wing loading in the mid 30 pounds per square foot range, so was solid.

In terms of pure performance, the Mullicoupe appears to have the edge in climb but the Mulligan is much faster at low altitudes. We were indicating 205 mph (230 mph TAS) where the Mullicoupe was showing 190-195 mph. Of course the Mulligan was also burning 33 gallons as opposed to the Mullicoupe's 22 gph.

Where I did notice the difference was the control feel. The Mulligan's ailerons are much quicker and smoother. In fact, they were flat-out lovely. The rudder, however, was one of the most sensitive I've ever touched.

In cruise the Mulligan was a solid as a brick building, a feeling that continued right down into approach. With full flaps and 100 mph showing, the airplane was dead simply to gently hold off before putting it on the mains. Once down, it was solidly down with no tendency to hop or porpoise. Some people have ask whether it was like landing a Howard DGA-15P, which it resembles and the answer is an unequivocal no. The Mulligan's gear is much stiffer and the airplane's easier to land. Any similarity to a DGA-15 is strictly cosmetic, as the Mulligan's character is so much more refined and precise.

After flying the Mullicoupe, when you hear Dake and Younkin express negatives about the airplanes, you have to think they're out of their collective minds. After you've flown Mr. Mulligan, you begin to understand where they're coming from.

The Mullicoupe is a helluva airplane. Mr. Mulligan is the daddy of that airplane and it shows.