Main Menu

Peterson Javelin J-4

The Javelin dates back to the early 70’s in America when one person tried to produce a low-cost glider in an effort to stem the rising importation of the popular German gliders.
Early sketches were for the best performing aircraft in the world, but it was only a couple of weeks until that philosophy had changed. What was really needed was a Certified Sailplane that the public could rent and afford to buy. What was needed was a sailplane that was not perfect, just good. The new priorities were: low cost. 30: 1 glide and easy maintenance. The project became known as “Easy-Do” at that stage. Six years, later it seemed a good idea to drop that title and just call it the J-4 Javelin.
The ship was literally designed from the inside out. Cost was paramount, so design of the wing structure was the first step. A spar without rivets, pieces, or bolts was what was needed, and a tube spar was chosen as the best solution It uses a six-inch tube and tapered its wall thickness by chem-milling. The wall of the tube tapers in thickness from .188 in. at the root to .050 in the first ten feet and remains constant after that. Full tube diameter rings under each rib were left, so that rib cut-outs would be the same. Also an uncut strip spanwise on top and bottom made the tube tangent to each skin surface. Studies showed that twelve spars could be chem-milled at once for a cost of $8.00 each for the tank time. Masking the tube for cutting was only $18.00 each-an obvious saving compared to driving 1500 to 2000 rivets and root fittings.
With the basic method of fabrication chosen, the wing size was selected: a six-inch diameter tube with an 18% airfoil gave a 33.33-inch chord, and a forty-nine foot span equalled an area and aspect ratio that were acceptable. Wing skins were sized to take advantage of full or half sheets to minimise wastage and keep cost down. The trailing edge is glued on (no rivets) and the rear spar and aft ten inches of the wing are bent from one piece of metal. (Thet saved a few rivets) A straight wing was designed first. It seemed the obvious way to keep cost down. Flight tests with the straight wing showed what any aerodynamicist knows — it has a higher drag than a pure elliptical planform. Since an ellipse is expensive and impractical, and a straight taper from root to tip is just about as bad cost wise, the Javelin wing tapered starting two-thirds of the way out front the root. Incidentally, this approaches the elliptical plan form better than the complete taper.
Before building the first set of wings, another important decision was made. Most aerodynamic studies of surfaces moving through air at 60-70 mph seemed to indicate that exposed rivets didn’t damage performance very much, and flush riveting is expensive. Also, since blind rivets were easier to install than solid two-man bucked rivets, the design would be tailored for blind rivets. The FAA-approved structural blind rivets are the Huck and Cherrylock rivet types. These sold for fifteen cents each, even in the large quantities. That would have been $1500 just for the rivets if it had the usual 10,000 per airplane!
A blind rivet produced by a local firm had one length that would serve all needs. The head was custom moulded when installed to a low .030-high domed rivet, and best of all it swelled sideways to fill out-of round or oversized holes. It also sealed itself so it was pressure tight. Cost per aircraft would be $60.
The FAA sent its “Dear John”- a letter saying they had never heard of the rivet, and that it wasn’t on the approved list. The FAA were asked, “If it holds the static test loads, we’re going to use it.”— They said, “How about vibration?’ Question - “In a sailplane?” They said, “Our rules (which don’t cover sailplanes) say eight million cycles while under shear loads.”
After $1000 worth of vibration testing by one of those special labs that shake, bend and twist things, the FAA sent a nice letter approving the rivet.
Cost studies suggested that an aileron didn’t earn its keep, and if a wing could he built without one it would be a lot less expensive. That’s when it was decided to use spoilers for roll control.
They were less expensive and the yawing drag was in the right direction (which isn’t true with the aileron). As a test a linkage to the right spoiler in a single-place was disconnected, leaving only the left spoiler operative. It soared quite well; making left turns with the spoiler handle. The first Javelin wing had spoilers laying flat upon the top skin surface; later recessed them into the wing to improve performance.
The tail assembly has all three surfaces identical (i.e. one aerodynamic surface can serve interchangeably as a rudder or either elevator). The main spar is a two and one-half inch steel tube cold-swaged in a straight taper down to a one and one-half inch diameter.
It is simple with no built-up parts, provides an ideal pivot for a full-flying surface, is easy to attach and again the cost to swage the heart of the structure was only $9.00. Since flying slab surface was aerodynamically balanced, a small anti-servo tab was added to give control feel. The leading edge has a full length weight inside to partially static balance the surface. Originally piece of concrete reinforcement rod bonded in place with a rubber mould compound was used. They couldn’t prove to FAA that it wouldn’t fall out, so instead of arguing, the weight was changed to a thick-wall tube riveted in place. (Incidentally, those old leading edge tail surfaces still have the bonded concrete reinforcement rods in them and they can’t get them out)
The fuselage is three major basic assemblies. A square steel frame (easy to cut and fit joints) goes from nose to tail and offers good pilot protection plus holding the wheel and all the controls right out in the open where assembly is easy. Fibreglass lower shell 24-ft. long acts as a structure and fairing for the fuselage. The third piece, an aluminium sheet metal deck behind the wing and extending to the tail, completes the basic fuselage. All three pieces are riveted together with one row of rivets through the aluminium sheet, fibreglass hull, and steel frame.
The wing-mount is a piece of tubing that has been bent 4 degrees (2 degree dihedral) and attaches to routed plates–the inside has been chem-milled to increase the diameter so that the wing spar will slide into it. Since the tube will take bending and shear loads in all direction, the rear spar serves only as a reaction for wing torque; hence simpler structure and lower cost. I looked for three weeks to find out how to bend a six and one-half inch diameter, 1/4 in.wall tube four degrees. An old gentleman who owns an oil well equipment manufacturing company in Tulsa, Oklahoma, and he gave a note to his foreman and told Peterson to send the parts back there. Both airfreighted and within a week 5 bent tubes were done. He never did send a bill.
The canopy is 26 inches wide five feet long, free blown and optically perfect. Eye level is at the top of the wing so that you can watch the action of the spoilers and spoilerons and have almost unlimited visibility. The canopy is en-trapped within a special extrusion and is not drilled for bolts.




On Nov. 4, 1969, Peterson flew the ship for the first time on El Mirage dry lake using auto-tow. Five times up and five times down. The spoilerons did work.
At this point a couple of USAF test pilots and they took over the test flying. The first air-tow was at Tehachapi and Major Mike Love was at the controls. On auto-tow, at 60 mph, the spoilers stayed closed: at 65 mph on air-tow they came open different amounts This caused yaw and roll problems. At release (5000 ft. AGL), Mike put the nose down and at 70 mph the ship flew straight again because all spoilers were open the same amount. It had achieved an L/D of about 4 to 1. Mike added 20 mph and flew his downwind and base at 90 mph with touchdown at 80. The end result of research was to recess the spoilers. With spoilers cut into the wing surfaces and a new tapered wing built, things started to brighten up. They had made parts for five or six ships, so a new wing or fuselage really wasn’t too big of a problem.
We our flight test work moved from Tehachapi to Rosamond Airport for convenience, since it was closer to home and pilots, Major Love and Major Staten. An exact ground distance between 2 points of 3 miles + 85.6 ft. checked airspeed indicator accurately.
Flight test work for FAA information was expensive and time-consuming. Between pilot schedules, weather, and schedule, the weeks did drag on. Most spectacular were the spins. Three turns were required in each direction with controls in every position possible.
The FAA pilot, Norm Moentmann, had landed and Peterson pulled the Javelin to the sidelines and started to tie it down. A NASA DC-3, sitting on the line, started its engines and the open canopy caught the blast. With short hinge pins in it for the flight test work the air loads knocked it off the ship. A few small scratches and a broken hinge on the fuselage side were the only damage. A quick trip back to the shop in Newhall, picked up some parts, and spent part of the night fixing it so that testing could continue the next day,
In January 1973 the FAA couldn’t have issued the Javelin an ATC if they had wanted to because the environmental laws on noise abatement arbitrarily included gliders! The law seems to have been written without specific values-just anything new coming out had to be made quieter. Finally reason prevailed somewhere, and they got a letter saying it wouldn’t apply to the J-4.
On Thursday, February lst, 1973 FAA pilot, Norm Moentmann landed the ship. “I’ve flown it enough,” he said, “it’s okay.”
So the J-4 is an ATC’d sailplane that goes together in ten minutes with all the controls automatically hooked up, is low in cost to produce, looks reasonable even with exposed rivets, has a 30-to-1 glide, and flies well. It has been flying up to 150 mph and have done a complete flutter test program up to 140mph.
Eight Javelin were built.
Wing span: 14.94 m / 49 ft
Wing area: 11.71 sq.m / 126 sq.ft
Chord: 848 mm
Aspect ratio: 19.2
Airfoil: NACA 64(3)-618
Empty Weight: 260 kg / 574 lb
Payload: 105 kg / 230 lb
Gross Weight: 365 kg / 804 lb
Wing Load: 31.17 kg/sq.m / 6.38 lb/sq.ft
MinSink: 0.70 m/s / 2.3 fps / 1.36 kt
L/DMax: 32 89 kph / 48 kt / 55 mph
Seats: 1



Copyright © 2020 all-aero. All Rights Reserved.
Joomla! is Free Software released under the GNU General Public License.