Re-Envisioning a 30-Year-Old Classic Cross-Country Design
Taking an iconic design from the 1980s and recreating it with 21st century materials and techniques.
I was flying cross-country soaring and a friend came by to watch. After a while he mentioned that he and his brother used to fly Constellations from early morning till late afternoon on a single launch while sitting in their lawn chairs. Sometimes, they would get in their Country Cadillac (pick up truck) and run down the road chasing the Constellation when it was flying fast. It sounded like so much fun. I decided to seek out a 30-year-old kit to begin my journey with the Constellation. But no kits were available.
I asked around until I found someone who knew how to get in contact with Bob Sealy to see if he was still selling plans for his kits including the Constellation. A couple of calls later I reached Bob and was surprised when he told me he could make up a new Constellation fuselage set for me. A new fuselage and a set of plans were on their way shortly after that. The plans arrived and they were in line with the best plans of that day, 30 years ago and were for a builder’s kit. Forget about what you see for plans for today’s kits, if they have them, these plans had the basics in a scaled down plan set. After a few emails back and forth with Bob, I had all I needed to start acquiring the parts to put this beauty together.
Even back in the day, nobody built ‘exactly’ to someone else’s plans. We all have our pet ideas on what works best for us and what is familiar. That is when the idea struck me to build this sailplane in the spirit of the original design re-envisioned using today’s materials and techniques that were just not available back then.
The first change was to replace using white beaded foam as the cores for the wings, stabilator, and rudder. I went with Hiload 60 blue foam from Owens Corning. I have used this foam for all of my vacuum bagged wings for the last decade, or more. Next up was that the original Constellation, Connie for short, had a polyhedral wing. With a span of 167” those inner and outer panels were 42” long and they had a 12-degree dihedral in them in the middle. Hmmm, they were not going to fit in my car that way.
This led to the second change: each wing side would be made of two panels with a 12-degree joiner to ease storage and transport by packing flat. Although washout was not called for on the plans, I had 2-degrees of washout built into the inner panels for better flight characteristics. This would force the wing inner panels to stall before the outer panels and would still allow for some aileron effectiveness. By building the washout into the inner panels instead of the outer panels I could have more washout effect at the tips for the same number of degrees of washout. It is more efficient this way.
The third change was to replace the Connie’s main spar, which was built up from spruce and shear webs, with a carbon fiber tube. The aft spar was relocated to be on the CG to make it easier to balance the Connie. This spar ran the full length of the inner panel and continued halfway through the outer panel. This allowed for a joiner to be fashioned from a steel tube bent to 12-degrees then trimmed to fit snugly in the carbon fiber tube. But what size carbon fiber tube would I use? I selected a strong carbon fiber tube with an inner diameter (ID) of 0.625" (5/8”) so that I could use 5/8” outside diameter (OD) steel tubing for the joiner between the inner and outer panel. That worked out great!
Turning my focus back to the root of the wing where it met the fuselage, I ran into a problem. Since these wings plug into a 3.5” wide fuselage at a 4-degree angle per side. Bending a joiner to make the 8-degree dihedral angle would be easy. But there was that fuselage in the way. I would need two 4-degree bends in the same plane and 3.5” apart. This was beyond my tube bending skills. I did try though.
To solve this problem, I went back to Bob’s plans for the Connie and saw he used a straight joiner through the fuselage into 4-degree angled tubes buried inside the spar trapped by spruce triangles. That was a great two-dimensional solution. But how could I do that with the carbon fiber tubes I was using as a spar in three dimensions? I needed a tubular adapter that slid into the carbon fiber tube and had a bore hole through the adapter at 4-degrees. I tried to manufacture these from wood, but ultimately failed because of needing a longer drill bit than I had available and more depth than my drill press could muster.
Then the answer came to me: 3D-print the part. A quick drawing in 3D CAD and off went the Standard Tessellation Language (STL) file to two friends who each had 3D printers. The idea was to 3D-print the adapter with a 3/8” OD bore hole through it at a 4-degree angle.
I planned to insert a 3/8” OD brass tube into the adapter and then the 11/32” hardened steel ejector pin would be the joiner. After the prototype adapters came in, we selected a different 3D-printing material and then printed the four adapters needed. Instead of using the brass tube sleeves, I went with a 3/8” diameter hardened steel ejector pin as the joiner. I felt the thicker joiner would be more beneficial going through the fuselage than the brass sleeve would give me in the carbon fiber tube. The main spar joiner assembly was now solved. That left the second spar to deal with.
Here I opted to move the spar forward towards the thicker part of the airfoil up front. Instead of using a full-length carbon fiber tube for this spar, I only used a length of tube a little longer than the 3D-printed adapter. I added three laser-cut 3mm plywood ribs to support the tubes and spread the stress to ensure both of these carbon fiber tubes did not push through the bottom of the inner panels when the wings lifted up. However, I was not finished. A year before I ever knew about the Connie, I built a 48” long carbon fiber spar press.
This press allows me to seriously compress carbon fiber while the epoxy is curing and to squeeze out as much epoxy as possible. The entire jig fits into a long heavy carboard mailing tube which served as a hot box or hot tube. Using MGS epoxy, 24 hours in the hot box, and I could get six layers of C68 unidirectional high modulus carbon fiber and one middle layer of 1.7-ounce Kevlar at a 45-degree bias compressed down to 2mm. I decided the Connie would benefit from a blade spar like this. I cut the groove for the blade spar using my table saw with a blade kerf to match the thickness of the blade spar.
The spar was 1.00” wide, and I located the kerf cut exactly at 1.00” thickness of the airfoil and scraped up against the back of the forward carbon fiber tube spar (the short one). This gave me 6” of overlap contact between the short tube spar and the blade spar. The tube spars were glued in place with MGS epoxy along with the three plywood ribs and the blade spar and the assembly was allowed to cure overnight.
The next change on the agenda was that the original white foam cores were sheeted with 1/16” balsa. I chose to use a 5.9oz per square yard carbon fiber cloth in a twill 2x2 weave with an interlaced thread of purple nylon (adds sparkle) for the inner panels. For the outer panels I used the same 2x2 twill pattern but used C68 carbon fiber running lengthwise and orange-dyed Kevlar running widthwise for three-quarters of the outer panel with the last quarter being just 1-ounce fiberglass cloth, no bias.
The bottoms of both inner and outer panels were laminated with C40 unidirectional carbon fiber. The blade spar contacts the top and bottom of the foam cores so that the carbon cloth of the top and bottom of the inner panels contacts it and forms an I-beam and the wrap around carbon fiber around the nose of the airfoil completes the D-tube structure.
Having designed, built and flown, several cross-country sailplanes, I know there is a tremendous amount of stress on the inner panels when winch launching. The hollow carbon fiber tube might compress and bend under that stress, compromising the panel. To combat this, I obtained several very light hardwood dowels, one for each side cut to length. I made score marks along the length and a few divots here and there, and glued them into the carbon fiber spar tubes leaving the ends open for the adapter on the root side and the 12-degree joiner on the tip side. I used slow MGS epoxy so that I had plenty of time to get these dowels in the right place in the carbon fiber spar tubes.
When specifying the foam to be cut for the wings, I failed to recognize that while the foam inner panel I ordered would be a perfect rectangle and the fuselage intended to have the wing meet at a slight angle. Therefore, the root of the inner panel was intended to be a trapezoid. I discovered this after the inner panels were vacuum bagged while I making my first trial fitting of the wing panel to the fuselage. Why is there a gap at the trailing edge? Sigh. I did not want to cut away the root edge of the inner panel because of all the work I had done there to support the two spars and joiners. The only option left was to extend the fuselage wing fairing so that it mated to a rectangular inner panel. The process was simple. I used clear packing tape to cover the wing’s root edge and up onto the wing itself both top and bottom. A little Maguire’s paste wax was added as a mold release to the tape’s smooth exterior. I attached the inner panels to the fuselage leaving the gap at the rear. Using 3mm clear packing tape, I sealed the gap on the bottom and rear of the joint. I mixed up a slurry of MGS epoxy with slow hardener, chopped fiberglass, and colloidal silica to act as a thickening agent, and proceeded to trowel it into the cavity. I used a toothpick to poke at the slurry to ensure all the air bubbles were released and to ensure the slurry made its way into the corners of the gap. A quick pass with a MonoKote heat gun (to remove small bubbles from the slurry) and then it was left to set overnight. The next day, I removed the tape to find a close to perfect fairing between the wing and the fuselage. Now, the moment of truth came — would the wing inner panel separate from the new fairing to the fuselage? I lost sleep wondering if I had done enough to ensure a clean separation. With a slight tug, the wing’s inner panel separated from the now cured slurry mixture. Success! There were a couple of minor imperfections that I fixed with the next batch of slurry and a finish sanding and all was good to go. There was also the added benefit of a much thicker, and stronger, fuselage side wall to support the ejector pin joiners going through the fuselage.
The Connie was designed as a rudder, elevator (t-tail), flap, and spoiler cross-country sailplane. The combination of spoilers and flaps intrigued me and I had to try it. But having flown several large rudder, elevator, spoiler (RES) sailplanes, I know they can wallow on landing unless you have a nice straight approach. On a cross-country course, you do not always have that luxury. I wanted more control. My solution was to keep the 12-degree dihedral between the inner and outer panels and to add a tilleron (small aileron) to the outside edge of the outer panels.
The spoilers on the plans were drawn in the traditional way with a pull string to a servo in the fuselage and a spring to pull them closed again. Great old-school design. I tossed that notion aside and purchased a set of electronic blade spoilers. After the wing was bagged, I marked the top skin and cut away the carbon fiber skin and then, using a hand-held hot wire jig, cut the trench for the blade spoiler and its little integrated servo. I located the spoiler right in between the two spars to avoid compromising the strength of the wing. I used shoe glue to secure the spoilers into their trenches.
All that was left to complete on the wings were the flaps. Instead of cutting out the 2.25” wide by 20” flaps, I used my table saw to cut through the top skin of carbon fiber and through most of the blue foam leaving the bottom skin intact. I enlarged the flap chord to 3.00”. I like flap authority. I calculated, like any engineer on a napkin, the amount of drag and down-force the 18” spoilers were going to generate and balanced that against the up-force the flaps were going to generate. I used a hobby knife to cut through the remaining blue foam to expose the 45-degree bias 1.7-ounce Kevlar hinge I had built in before vacuum bagging the wing skins. The flaps were born! The kerf of the table saw allowed for the flaps to have reflex capability for speed mode. As a side note, I go over every hinge with a very thin smear of silicon sealer to form a living hinge and on the other side I use a 1” strip of book tape to ensure my hinges don’t fail in flight, from a high-speed run, or hard landing.
With all these changes, someone might think I was done. Not so. We have yet to speak of the elevator and rudder servo installation. I don’t like pushrods that flex. Cross-country sailplanes are heavy and have a lot of stress on them. I decided to mount the servos in the vertical fin. What? Servos in the tail! That is a huge weight penalty! Ah, but think about it. Servos back 30-years ago were analog and weak compared to today’s modern F3J-capable mini servos or flat wing servos! For the weight of half of one older servo, I could get two modern wing servos in the tail with three times the torque and direct solid linkages to the rudder and elevator. One custom 4-wire cable from the RX though the fuselage to the tail completed the servo installation. Oh, and yes, I used modern lightweight but very strong servos for the flaps and lightweight servos for the tillerons. My target weight for the Connie was the 5kg maximum weight for an F3H cross-country sailplane.
All the fuselage needed now was the flight pack battery, the 10-channel receiver, and nose weight for balance. I used the biggest battery I could fit in the fuselage, enough for a full day of cross-country flying, and used lead bird shot for the balance secured in a Ziploc bag and stuffed into the nose of the fuselage. All up weight, ready to fly was 5kg. I did consider making the fuselage in two pieces. In fact, it comes from Bob as a forward and aft section that is glued together with a 5” overlap. However, I was running short on time and decided to glue the fuselage halves together.
Of course, no model is finished without graphics! I used my laser cutter to make up custom graphics for the Connie from self-stick reflective material. I cut graphics that included the name of the model, my AMA number, phone number, and my FAA number.
I had some friends give me a hand doing hand tosses to ensure I had the elevator set at neutral and the flaps + spoilers were balanced. Yes, the flaps, and spoilers were both activated by the flap stick and were tied together through my computer radio. The first toss, made by Wally ‘By-Golly’ Adasczik LSF President extraordinaire, went perfectly. No adjustment to the elevator was needed. Not even one click.
The second toss confirmed this and that the I hit the correct balance of flap to spoiler so that no elevator compensation was required! I’d like to say that the painstaking calculations I did during the pre-build, build, and selection of throws and spoilers etc. was responsible for hitting that mark. However, it was dumb luck. Even a blind squirrel finds a nut once in a while.
After two more hand tosses, Wally was tired and wanted to go home. In fairness, he had just finished being the Contest Director (CD) for a two day ALES contest. Now, it was time for the winch launch. Rise off Ground (ROG) was selected. Going back a moment, there is nothing on the plans that describes how much throw to put on any surface. I guessed that I would use as much throw as the surface could give me. That is how the radio was set up for the first winch launch. Get ready for a wild ride!
One helper held the wing tip level, another was recording the launch, and I stepped on the winch pedal and the launch commenced. If you have ever seen a graceful straight like an arrow launch from a winch for a new sailplane and the joy that brings…
You would have been in shock at the sailplane veering hard to the left. But I kept it on the line and corrected its flight path, but overcorrected and the sailplane veered hard to the right. I corrected less this time and got her centered and up like an arrow she went.
We were going for an initial short line launch with 120m of line from the winch to the turnaround pully. At the top of the line as the Connie came off the line, she stalled, hard just like a Sailaire. I could see her going into the stall and pushed all the down elevator I had. After the stall, the Connie headed for the ground and I pulled up with everything she had and she pulled herself back up and into the second stall but this time dropping the right wing and turning as she flew and now heading downwind. I corrected, or overcorrected, and she came nose up again and stalled a third time to the left. I got her back out of the stall without inducing a fourth stall and now she was heading towards us. I can hear the photographer, Mike Bergerson, saying “John’s just getting the feel for her and he’s going to smooth it out any moment now”. Another bystander, who was running for the safety of the pits, called out “the balance is off, she’s tail-heavy.” I love it how everyone is willing to give you advice when you are fully concentrating on flying your sailplane as if you have time to listen to them. But just as Mike said, and as if on cue, the Constellation calmed down and flew gracefully around the sky even gaining a little altitude when turned into the wind. A part of me wishes Mike had said that two stalls earlier.
The plane was fine, I just needed to recalibrate my fingers on the sticks and stopping the pilot induced oscillations. I flew the Connie for a few more minutes and at around 50 feet high decided it was time to land. A 270-degree turn to head into the wind and set up for the landing, I engaged the flaps and spoilers and the Connie slowed down to a crawl but did not pitch up or down. The Connie made a beautiful landing.
I succeeded in resurrecting a 30-year-old design, building it with modern materials and construction techniques. I came away with a cross-country sailplane at 5kg that can take a full-pedal winch launch from a powerful winch without flexing the wings or having the sailplane breakup on launch.
I was very happy with my creation and I sent a note to Bob through email describing the changes I made in materials while keeping true to his design. Bob replied with a very heart-felt letter, an excerpt is reproduced here:
“I can’t thank you enough for the feedback and pictures. I read your email several times. Your detailed descriptions of your construction techniques were awesome. It makes me feel good, make that great, seeing some of the old designs from 30+ years ago, flying these days…Your pictures and email bring back many fond memories of years, make that decades, ago. I can’t thank you enough for sharing your experiences with me…May you have many, many, great flights with the Constellation and Catalina.” — Bob Sealy (provided with Bob’s permission).
I encourage all of you to try your hand at resurrecting an older design, re-envision it with today’s knowledge and materials, build memories, get out there, bring a buddy, and enjoy soaring!
The Connie, and my Catalina (see Resources, immediately below), are both vying for my attention at the flying field. The Rabbit is happy it gets to carry either of them into the air!
©2021 John Marien
Resources
- Cross-Country Soaring with a Rabbit My adventures with my Catalina sailplane as recounted in the October, 2021 edition of the New RC Soaring Digest.
Images by the author unless otherwise noted. Read the next article in this issue, return to the previous article in this issue or go to the table of contents. A PDF version of this article, or the entire issue, is available upon request.
