Hughes XF-11 Article
Ever since the first time I saw a picture of the Hughes XF-11 several years ago I have wanted to build a flying model of it. But, as so often happens, that thought got filed away in the back of my mind like so many other projects. Of course, this is a common affliction for model airplane enthusiasts the world over. With the recent release of the movie ‘The Aviator’ there has been a resurgence of interest in the planes of Howard Hughes. This rekindled my interest in the Hughes XF-11.
The XF-11 would evolve from a 1939 proposal made by Hughes to build a bomber, designated the DX-2, using the Duramold process. The Duramold process involved molding resin-impregnated plywood into the desired shapes. It had been demonstrated to have strength and rigidity comparable to metal. In late 1941 the USAAF turned down the DX-2. The Army Air Forces just couldn’t picture itself flying something as outdated as ‘wooden’ airplanes.
After the U.S. entered the war, the USAAF decided to reconsider the Hughes aircraft built from the Duramold process. In May of 1942 the Hughes Duramold plane was ordered with the experimental attack designation XA-37. Hughes would not sell his plane to the Army until he had test flown it himself, though. This occurred in June of 1943. The plane was a twin engine design and was believed to have a top speed of over 430 mph. However, Hughes was the only person to ever fly the plane. The plane was developed in secrecy; and Hughes wouldn’t even allow photographs to be taken of the aircraft. The life of the DX-2 ended prematurely on November 11, 1943 when a fire destroyed the hangar housing the aircraft.
Ironically, during the same time that Hughes was developing his ‘wooden’ aircraft the USAAF has used some of the Royal Air Force’s deHavilland Mosquito. They used it in a high-altitude reconnaissance mode. The deHavilland Mosquito had remarkable performance, and was also made of ‘wood’. Hughes was already developing a high-altitude recon aircraft. In October 1943 the USAAF issued a purchase order for 100 D-5s. The order, though, insisted that the new aircraft be made of metal.
Previous to this order all reconnaissance planes had been modified versions of some other type of aircraft. The XF-11 was designed as a high speed recon plane from the outset. This is reflected in its sleek and uncluttered appearance. This sleekness of design no doubt contributed to the XF-11’s impressive performance. Powered by two Pratt & Whitney R-4360 radial engines generating 3,000 hp each the XF-11 had a top speed of 450 mph with a ceiling of 44,000 ft. and a range of 5,000 miles.
The XF-11 was to be ready in March 1945, but several problems during its development, including a large walkout by Hughes employees, delayed the project. In May of 1945 the USAAF canceled 98 of the 100 planes ordered. Hughes was paid to finish the first two aircraft. Hughes was determined to finish and fly the XF-11.
The first XF-11 flew on April 24, 1946. It had a wing span of 101 ft. 5 in. and a length of 65 ft. 6 in. Full flight testing began on July 7, 1946 with Hughes at the controls. After about one hour in the air a hydraulic problem caused the right propeller to reverse pitch. Hughes attempted an emergency landing on the Los Angeles Country Club golf course, but went down about 100 yards short crashing into a home in Beverly Hills. Both the plane and the house were destroyed. Hughes suffered severe injuries, but lived another 30 years. The second XF-11 was completed in November in 1947 and was test flown in 1948. The plane was then parked at Shepard AFB in Texas, effectively ending the saga of the XF-11. This was the last plane Howard Hughes ever built.
The genesis of my model of the XF-11 goes back to the beginning of my involvement in Control Line Stunt flying. In mid-1990 my brother Doug and I decided to give control line flying a try. After quickly destroying a couple of airplanes my Dad had hanging in the garage from about 25 or so years before we met local CL guru Bill Bischoff at Mike’s Hobby Shop in the Dallas area. Bill showed us an easy to build, good flying design using Gotcha 460 foam wing cores and low heat iron-on covering. Bill had designed a Zero and a Mustang using this method. We quickly built these planes and they were exactly as advertised.
We built, flew, and crashed many of these planes; all the while improving both our building and flying skills. We went on to adapt the design many times over into other aircraft: Messerschmitt 262, Hawker Typhoon, Gloster Meteor, Bernard HV 220 and others that I can’t even remember.
About five years ago I decided that I wanted to try something a little different for the Dallas Model Aircraft Association’s annual Profile-40 contest. This is the first contest of the year, and serves as a good warm up for the upcoming flying season. I had decided to try my hand at a twin engine aircraft. I took a look again at Bill’s design and decided to adapt it once again. Using a Gotcha 500 wing core, adding large wing tips, and cutting the cores where the engine nacelles would go I designed a B-25 Mitchell.
This plane turned out to be more than I could have hoped for. The plane flew great, sounded great with two OS 20 FPs humming along, and, best of all, took only three days to build; flying on the fourth. The plane was quite competitive; I just missed out on winning the contest.
When looking at a three-view drawing of the Hughes XF-11 it was quickly apparent that I would not be able to use a Gotcha wing core if I wanted the plane to have any semblance of a scale appearance. The XF-11 has an equal taper on the leading and trailing edges out to the wingtip. The Gotcha series of wings have much more taper on the trailing edge. I still wanted to build the plane using this method, though, so I called Crist Rigotti. Crist cuts foam wing cores for many stunt flyers. I sent Crist a drawing of the wing I had in mind, and a couple of weeks later I had the wing cores in my hands. They are beautifully done, and exactly what I wanted.
The remainder of the plane is done in balsa with ply doublers where the engines are mounted. My method of design is admittedly not very scientific. I thoroughly study the 3-view drawings of the plane in question, put a big piece of vellum paper on my drawing table, and start drawing. I mark off the desired ‘numbers’ (nose moment, wing root, tail moment) I wish to attain and begin drawing the aircraft to fit these parameters. I do my best to retain the original proportions of the plane while fitting it to my desired numbers. I drew the fuselage booms, center section, stabilizer, elevators, and rudders and sent this plan off to Kyle Tankersley at Tank’s-Hangar for laser cutting. Since discovering Kyle and his laser cutting abilities several years ago I do very little manual cutting of parts. The fuselage booms, center section, and elevator are all cut from ¼ balsa and then sheeted. This method of building yields a stronger and lighter aircraft than one simply built from sheet balsa. The wing cores are glued to the fuselage booms and cockpit center section with the spars passing through these sections.
CONSTRUCTION & FINISH
As I am sure you can gather from the subtitle here, this plane is built and finished in unison. This makes it much easier to prepare the wood parts for painting. The first step of the construction process is to glue together the ¼ balsa pieces that make up the fuselage booms and cockpit center section. Use thin CyA glue for this process with wax paper underneath. Next, sheet the fuselage booms and cockpit section with contest grade 1/8 balsa. This is going to turn out to be a fairly large plane, so it is necessary to keep all the pieces as light as possible. Next, epoxy the 1/16 ply doublers and ½ x 3/8 maple motor mounts into place. I used 5-minute epoxy for this, as the OS FP motors vibrate very little. The stabilizer can be sheeted with 1/16 balsa at this time, also. The elevator is ¼ balsa, and will not be sheeted.
Once all these parts have been built and sanded to shape, add the necessary hinge slots to the tail surfaces and begin the finishing process. For many years now I have used basically the same finishing method for my painted aircraft. I will provide a quick summary here.
The first step is to seal the wooden surfaces with two coats of brushed on Sig Nitrate dope; thinned 50/50. Once this has dried thoroughly, sand with 280 -320 grit sandpaper. Now, brush on two more coats of dope and sand once again. The next step is to spray all these parts with sandable auto primer. Note: for a full bodied competition stunt ship you may wish to add silkspan over the wood surfaces at this stage. I do not, and have been quite happy with the results. Spray on one thorough coat of primer, and sand with 240 – 280 grit paper. Spray on a second coat of primer and sand with 320 paper, and then sand again with 400 grit paper. These parts are now ready for paint; which will be done once the plane is in one piece.
Now that the wood parts have been built and prepped the wing and flaps can be prepared. The wing cores are actually four pieces. With the cores still in their cradles mark the flaps on the top of the cradles. Using a bandsaw, cut the flaps out of the wing cores. Now, with the flaps still in their cradles cut ½ inch off the front of the flaps. Sheet the flaps with 1/16 balsa and add a ¼ balsa leading edge. With the cores still in their cradles cut two recesses for the fuel tanks. Notice that the inboard recess is cut out of the left inboard panel and the outboard recess is cut out of the right outboard panel.
To make this design a little sleeker I decided to put the bellcrank assembly inside the wing. On all the previous planes built using this construction method we have mounted the bellcranks outside the wing on a 1/8 ply ‘tongue’ that extends through the fuselage. This method is very sturdy and easy to build. It could be done on this plane underneath the wing, mounting the bellcrank on a tongue extending through the inboard fuselage boom. To mount the bellcrank inside the wing relieve enough area out of the wing cores to allow free movement. I do this by wrapping some coarse ‘sticky back’ sandpaper around a dowel and using this to remove the necessary foam. Take your time, and check your work repeatedly to insure easy movement of the bellcrank assembly. A slot must also be cut to allow the pushrod to exit the wing.
With the wing cores and wood parts prepared the plane can be assembled. Before gluing any of the pieces together be sure to mark centerlines on both sides of the center fuse section and both fuselage booms. Also, make the 1/8 x 3/8 spar slots in the fuselage pieces. Use the wing cores to mark the slots, then drill 1/8 holes and use a small X-acto saw blade to make the slots. I’ll describe what I feel is the best way to go about this. I’ve built three planes using this system, and after long thought I feel this is the best method. Start by gluing; I use 5-minute epoxy, the left inboard panel to the center fuselage/canopy piece. Do this with the fuse piece laying flat on the workbench. Once this is dry, glue the right inboard panel to the center fuse piece. Now glue this entire assembly to the inboard fuse boom with the boom lying flat on the workbench. Turn the whole assembly over and glue to the outboard fuse boom. Also, at this time insert the stab into the slots in the booms; do not glue. To insure the whole assembly is square fashion a long triangle out of some scrap wood by taping a straight piece to a triangle and using this to square up the two booms as the epoxy sets.
At this stage the plane will look like a large rectangle. To complete this stage of the construction the outboard panels will be added. First, epoxy on the right outboard panel. Once it is dry turn the plane over and suspend it between two tables, stools, etc. With a Dremel Moto-tool relieve two small areas in the inboard fuse boom to fit the bellcrank into. Epoxy the bellcrank into these slots. For added strength relieve a 1” x 1” area out of the top and bottom of the left outboard panel and epoxy 1/16 ply pieces into place. Don’t forget to install the short bellcrank to flap pushrod at this time. Now the left outboard panel can be epoxied into place. Epoxy a ¼ balsa trailing edge into place. Square up the stabilizer to the leading edge of the wing and glue it into place using thin CyA glue. The final step of this stage of the construction is to add the wing spars. Use 1/8 x 3/8 x 48” spruce for the spars. Test fit the spars first by sliding them a few inches through each of the six slots they must pass through. To allow for a little more working time I use 15-minute epoxy for gluing in the spars. Fill one of the wing spar slots with epoxy and begin sliding the spar through. Work slowly until the spar has passed through all the fuselage pieces. The spar doesn’t cover the full span of the wing, so even it up and press into place. Add small spruce pieces to fill the spar slot. Once this is dry, turn the plane over and repeat. For added strength I glue .007 x ½ carbon fiber on top of the spruce spars. If you have CyA glue that does not attack foam, use this. Carbon fiber glues very well with CyA glue. Otherwise, use 5-minute epoxy.
The final step is to install the wing tips. The inboard wingtip has an adjustable leadout guide built into it. This is done by taking a piece of ¼ x ½ x 2.5” basswood and drilling and tapping a series of 4-40 holes in it. After drilling and tapping the 4-40 holes run thin CyA glue into the holes and retap them. This will give the threads some extra strength. Now, stand the basswood piece up vertically and cut through the series of holes with a bandsaw, connecting them. This will allow the leadout lines to be moved from one hole to another for adjustment. Using a pen vice, carefully drill holes into two 4-40 nylon screws. The leadout lines will pass through these screws. I placed one ounce of tip weight into the outboard tip before gluing it in place, and then also put a tip weight box at the very tip of the outboard wing to allow for adjustment of the tip weight. The plane is now in one piece, and all that remains is detail work and finishing.
To cover the wing I enlisted the help of iron-on covering expert John Grigsby. When it comes to iron-on coverings, the most important thing I’ve learned from John over the years is to take your time. If you plan your covering procedure in advance, take your time doing it, and pay attention to the details; such as hinge slots and wing tips, you should be able to achieve a very nice covering job. For this project you must use a low heat covering such as: Ultracote, Econokote, or Black Baron. We used Ultracote with the iron set on about 225 – 250 degrees. Since the covering is being applied directly to the foam wing core special attention must be paid to not melt the foam. Keep the iron moving without too much pressure in any one spot. Since the wingtips are made of balsa you can turn the heat up on the iron to about 300+ degrees and stretch the covering and apply more pressure to cover the tips.
Once the wing is covered the fillets are applied. This is a very important step as it adds strength to the wing and seals the wing/fuse joint to prevent damage to the wing from fuel. On this type of plane I always use 5-minute epoxy mixed with Great Planes milled fiberglass as my fillet material. The milled fiberglass thickens up the epoxy to help prevent runs, and it adds a great deal of strength. I use it in many applications when building model airplanes. Mix a ratio of 1/3 epoxy, 1/3 hardener, and 1/3 milled fiberglass. Only mix up enough to do one or two fillets at a time, as the fiberglass will cause the epoxy to ‘kick off’ faster than normal. I simply put the epoxy in place with a craft stick, run my finger over it a few times to set it, and then run my finger over it again with alcohol to smooth it out.
At this time the flaps and elevator need to be installed. This is pretty straight forward, except that the flaps are actually in three pieces. Use a 1/8” flap horn to couple the inboard and center flap together, and a piece of 3/32” music wire to couple the center and outboard flap together. I used a piece of 1/8” carbon fiber sheet to make the elevator horn; but a piece of 1/8” ply strengthened with CyA glue will work just as well. I drilled three holes for mounting positions for the pushrod. The elevator horn is epoxied into the elevator. Simply cut a slot in the elevator with an X-Acto knife, being careful to not go all the through to the top, and epoxy the horn into place. For the pushrod, I always use carbon fiber pushrods with titanium ends from Central Hobbies. These are the best I have ever seen. When assembled properly with 24 hour J-B Weld they are practically indestructible.
The plane is ready for paint. The Hughes XF-11 had a very simple paint scheme. In the interest of speed any panel joints were filled and sanded smooth. It was then painted light grey and buffed to a smooth finish. I have used two part auto paint for years; either Dupont, PPG, Omni, or Matrix. All of these paint systems work well. I taped off the wing and applied two coats of light grey paint. I then unmasked the wing and applied the last bit of details. I used 3/16” strips cut from Ultracote to make the wing walk boundaries. I used a #3 Micro pen to ink the lines for the rudder and elevator hinge lines and Sig brand decals for the Stars and Bars. The lack of panel lines to be drawn made this a very simple model to detail. Once all this was completed, I remasked the wing and sprayed two coats of automotive clear on the painted areas. Move the tape about 1/8” back from the previous tape line to get a good seal with the clearcoat.
The final detail to add is the landing gear. The gear for this plane must be mounted in the fuselage pieces. To accomplish this I adapted a tailwheel mounting method for profile models by mounting the gear through a ¼” dowel and into a slot in the fuselage. Start by drilling a ¼” hole in a block of balsa. Next, place a ¼” dowel in the balsa block and drill a 1/8” hole in the dowel. It may be necessary to hold the dowel with a pair of pliers to keep it from spinning. Cut a 1/8” slot in one end of the dowel. Push a piece of 1/8” music wire through the dowel and bend. Cut off the excess music wire from the bend so it is flush with the dowel. Now you can bend the music wire to the desired shape for the nose wheel. The main gear are left straight and fitted with Dubro part #613 and a 1/8” music wire axle so that each main gear has two wheels. I also used Robart gear strut covers and Kavan wheels for a scale look.
My original intention had been to make a unique model to fly in the popular Profile-40 event. As the plane came together I quickly realized that two OS 20 FPs were going to have real trouble pulling it around. This plane just ended up being bigger than I had expected or intended. It just doesn’t look as big on the plans page as it does when it is built.
I began racking my brain to come up with a new motor combination. As I went through my stash of motors (wow, I have a lot more motors than I thought) I came up with a few different combos: 2 OS 25 FPs, 2 OS 32 SXs, 2 Saito 40s, etc. These were all motors that were readily available; i.e. in my garage. I decided to go with the Saito 40s. The sound of two 4 stroke motors running in unison was just too good to pass up. The Saitos also fit right into the space originally intended for the OS 20s, and are very fuel efficient.
To effectively run the Saito 40 in a CL Stunt mode I used the Ultra Hobby Products manifold/venturi kit. This kit consists of a manifold; which replaces the carburetor, a venturi and an OS needle valve assembly. Originally designed by Pat Johnston I have used these kits on several different engines; such as the Saito 56, Saito 72, and OS 52 FS.
To say that the 2 Saito 40 motors provide enough power for the Hughes XF-11 would be a huge understatement. They perform beautifully and provide more than adequate line tension. I am also very pleased with the ‘turning’ capabilities of this plane. It is very easy to make a hard flat corner. As a matter of fact, after the first couple of flights I slowed down the controls by moving the pushrod to the slowest setting on the elevator horn. Yet, the plane still turns plenty quick. The plane performs like a real stunter, yet looks very unique at the end of the lines.
I undertook this project intending to build a fun stunt ship that would also be capable of flying in P-40 contests. My main focus has always been to build a fun to fly plane, whether or not it was contest worthy. The use of larger engines precludes me from flying in P-40 contests, but I still consider the project a success because the XF-11 is an absolute blast to fly.
With that in mind I have already figured out many simple changes that could lead to better performance. My XF-11 is a bit overweight, which hampers performance. There are many ways to attack this problem: single wheels on the main landing gear, eliminate the gear strut covers, and less clearcoat (I put on a thick second coat, perhaps too thick) are just a few ways to eliminate weight and gain performance. Also, the use of lighter weight engines would reduce the overall weight; though this would require particular attention to keeping the rear portion of the plane very light. Mine balances perfectly on the spar with the Saito 40s in place. More lightening holes in the rear part of the fuselage booms and built up rudders would reduce weight in this area. I have really enjoyed designing, building and flying the Hughes XF-11. Should you decide to undertake this project, I hope you will enjoy it as much as I have.