Wooden It Be Lovely? Part Four: Fashioning the Fuselage
After a brief hiatus, AVweb's Matt Paxton is back to work on his Pietenpol homebuilding project. In earlier segments, Matt discussed kicking the project off, building the wings, and searching for new ideas at a type fly-in. This time around, he's assembling the fuselage. At the end of the day, he finally has a place to sit.
"Forgive me, oh spirit of Tony Bengilis, for I have been slothful. I have found excuses not to work on my airplane, and I have allowed myself to become distracted with other projects and with self-doubt. Please help me get off the dime and get back to work, like I know I should." This should have been my mantra last fall, when I let weeks go by with no work on the Pietenpol. At that time, the main structure of the wings and fuselage were done, but I was hung up on the steel work for the controls. But, I'm getting ahead of myself.
In previous installments to this series, I talked about getting started with an experimental aircraft project and progressed to building the wings. The next phase of my Pietenpol was logically the fuselage.
The fuselage and tail required another order from Jean Peters at Western Aircraft in Canada who, unfortunately for builders, closed his business right after I ordered my spruce from him. The wood arrived as before in a large cardboard tube. Like the spruce for the wings, all the longerons and braces were cut to the proper dimensions, the leading and trailing edges of the tail surfaces were cut and dadoed and cap strips for the tail ribs made. An inventory confirmed that the order was complete. Peters even bundled like pieces together with masking tape, with part descriptions written on the tape.
The fuselage is essentially square. The sides are made up in a jig, and then crosspieces and braces inserted between the two sides. The aft ends of the two sides are brought together and the end pieces become the tailpost, with some tricky sawing involved. The first order of business was to lay out the jig for the sides.
A Jig To Hold The Pieces
Bernie Pietenpol was very practical in his approach to the airplane. The plank wing only requires one rib jig. The fuselage side jig is a series of nails driven into the shop floor, or in my case, the assembly table. I used some wood blocks instead of nails in some places to minimize denting the relatively soft spruce. Basically, the top and bottom longerons are cut to length and inserted between the nails. At the front, the bottom longerons bend up and I ended up steaming the first foot and a half or so to get the longerons to form easily in the jig. The top longeron has no bend in it at all in the jig.
Once the longerons are in place, the braces can be cut to fit and the gussets glued and nailed in place. The plans call for a piece of eighth-inch ply on the outside of the fuselage from the firewall aft to just behind the rear seat. I decided to leave that off until I had the controls installed, so I built the fuselage sides with the inside of the fuselage to the outside of the jig. This allowed the inside gussets to be glued to all the joints. Once I had them all glued and tacked, I pulled the assembly out of the jig and gussetted the outside from the tailpost forward to where the plywood skin starts. One thing to watch for is that you don't end up building two left sides or two right sides. That's easy to do if you use the same jig for both sides, which I did. Some temporary gussets to hold everything in place until the permanent inside gussets could go on did the trick.
How to hold the fuselage halves upright in place while the cross pieces went in necessitated some thought. I ultimately used some hold down straps with ratchets, looping the straps under the table and up and over the fuselage sides. The straps held the sides reasonably in alignment until the first few cross members up forward and the front seat back could be inserted, then a final check with a square and a level insured that the fuselage was square. Next, the rest of the top cross braces back to the rear seat were put in and gussetted. Two two-inch by three-quarter-inch ash braces across the bottom of the fuselage which meet the bottom longerons where the fittings attaching the lift struts and the landing gear go went in next. All of this was given several days rest for the glue to fully cure.
Assembling The Sides
Now it was time to pull the two side together at the rear, giving the fuselage its taper from aft of the rear seat to the tailpost. The two fuselage sides get built with one-inch members forming the after end. However, the finished tailpost, formed by gluing these aft-most frame members together, has a width of one inch. This means that some careful vertical sawing, sanding and filing needs to be done to the aft ends of both sides to get the two halves to match up with no voids, and to be the proper width. That took the better part of a morning. When the tailpost halves matched up they were glued and clamped.
Another key benchmark was to ensure that the fuselage is straight after the sides are pinched together. I did that by snapping chalk lines on the table the fuselage sat on and by running a wire overhead parallel with the chalk lines. Small weights on fishing line suspended from the wire dangled down and provided reference points for the exact centerline of the fuselage. After it was all said and done, the fuselage appeared to be straight and square. The crosspieces and braces in the aft fuselage went in and it was rechecked. Again, everything looked good.
The fuselage floor, cut from quarter-inch birch ply, was glued and tacked in place to the bottom longerons and the ash cross members. The rear seat back and the front and rear seat bottom supports were cut and fitted next. I couldn't resist cutting out a temporary rear seat from some scrap plywood and climbing up onto the table to sit in the cockpit. It was tight, a bit tighter than I'd anticipated, but should be Ok.
The turtledeck, or rounded top of the aft fuselage, is made up with quarter by one-inch spruce strips set on edge in plywood bows with slots in them. A solid head board finishes the turtle deck behind the rear seat. I'd read that the turtledeck was a difficult part of the plane to get right, but for some reason, mine came together fairly easily. I made patterns out of cardboard for the plywood bows and sighted the cuts for the spruce strips for alignment before cutting the ply. The strips have very little bend to them running fore and aft. The finished turtledeck should cover nicely.
We Need A Tailwheel
Now that the basic fuselage structure was done, I turned my attention to the tailwheel. I decided early on that I was putting a wheel on the back end rather than a skid because I plan on operating off pavement occasionally. This will mean some kind of brakes for the mains. The plans don't give any guidance on installing a tailwheel or brakes, but the network of Piet builders proved very useful here. I was able to get drawings of a light, strong (I hope) mounting for a tailwheel and for installing toe brakes for the rear seat from other builders.
I ordered the smallest, lightest tailwheel Aircraft Spruce sells along with a single-leaf tailwheel spring. Mike Cuy's installation uses a wedge-shaped plug of one-inch wood between the lower longerons from where they meet at the tailpost forward about 8 inches. This is glued in place, and an eighth-inch ply gusset glued over and under the plug and lower longerons on the top and bottom. On the top, I cut a piece of .063 steel to cover the plug, overlapping the tops of the lower longerons. I drilled the appropriate holes for AN-5 bolts in the steel plate and through the wood plug and gusset. Bolts were run through the holes in the steel plate and the heads tack welded to the plate. The bolts went down through the wood and the steel plate sits on top of the longerons and the plug. The tailwheel spring then simply bolts up to the protruding AN-5 bolts. A piece of rubber between the tailwheel and the fuselage gives it some padding. Mike Cuy says this has worked on his Piet very well.
From Wood to Welding
Next, I turned to the steel tubing that makes up the control sticks and the torque tube they attach to. There are fittings for pulleys, a control horn that the aileron cables attach to and bushings that hold the torque tube and anchor it to the floor. All of these components have to be cut out of flat or tube steel and appropriately welded. This was the part of the project that most concerned me.
I'd attended a seminar on gas welding put on by the EAA, and learned the basics, but still felt pretty unsure of my abilities. I didn't have torches or tanks in my shop, although I had several offers to use equipment in other people's shops. One Saturday, I just jumped in and started cutting and drilling tubing. I used a friend's band saw to cut pieces for the control horn. Together, we tried welding the control stick bushings that run through the torque tube. My friend has a descriptive term for the final result, and it's not repeatable in polite company. Suffice to say, the first attempt at a torque tube is now decorative sculpture in my shop. I was pretty frustrated at this, and put the project aside for several weeks.
This was, in retrospect, the wrong thing to do. My failure at welding the torque tube began to magnify itself in my brain, to the point where I really dreaded going into the shop to start on it again. A few weeks later, I did cut another torque tube, drill for the bushings and cut fittings for the pulleys. This time, I build a jig out of a plank, several wood blocks and some clamps. I took the jigged assembly down to a local metal shop, and the shop owner agreed to help me weld it up. He was busy, though, and couldn't do anything for several weeks. The project sat in his shop gathering dust.
Meanwhile, Brodhead and Oshkosh were coming up, and I planned to fly our Bellanca to both events. I wrote about Brodhead in a previous installment in this series. What I didn't mention was that flying between Brodhead and Oshkosh, the Bellanca took a bird strike. A buzzard punched a six-inch hole in the leading edge plywood, the carcass going all the way back to the main spar. The plane didn't fly any differently, and I was able to make a temporary patch and fly home after the show. But, the wing needed fixing. That repair involved scarffing a patch into the wing skin and building a nose rib to anchor one side of the patch. Needless to say, while this was going on, nothing was happening on the Pietenpol project. By mid-September, though, the Bellanca was repaired, and I could get back to obsessing over the welding for the Piet.
Attending To Tail Surfaces
Mid-November, the welding continued to loom like Mount Everest over Mallory. I was going nowhere. I decided to postpone the welding and concentrate on another area where I could accomplish something, and get the project moving again. So, I started on the tail surfaces.
Each tail surface, the horizontal stabilizer, the two elevators, the vertical fin and the rudder all share the same basic design. A leading edge, dadoed to form a T-shaped section, a rectangular main spar and a trailing edge, also dadoed to a T-section. The T-shape allows half-inch by three-eighths-inch rib strips to attach to the leading and trailing edges. The tail surface side structure is also made up with this T-stock. The complication is that each of these wood components is a different thickness, requiring that some shaving and sanding be done to make the parts mate smoothly. Provision of eighth-inch ply gussets in all corners requires more shaving so that the gussets fair smoothly into the rest of the structure.
By snapping chalk lines and nailing wood blocks to the assembly table, a jig was made up for the stabilizer. The various pieces of spruce were placed in position, measured and cut, and the shaping and sanding to make everything fit up done. I opted to make patterns for the gussets, which are all different shapes, and cut all of them out of birch ply with a skil saw running a metal cutting blade. When all the pieces were cut and shaped to fit, and placed in the jig, I glued the side pieces to the leading and trailing edges, and the spar to the side pieces. The gussets were ready, and they were glued and tacked. The ribs were easy ó they just needed to be cut to the proper length and glued and tacked in place. Once the glue cured, an hour with the shaper and the surform tool, followed by medium grit sandpaper had the leading edges curved, the corners rounded, and all joints smooth and even. This procedure works the same for the left and right elevators, the vertical fin and the rudder.
Getting the hinges right for the movable surfaces wasn't necessarily difficult but required a lot of patience to get right. The first order of business was to mortice out pockets in the trailing edges of the stabilizer and fin and in the leading edges of the elevators and rudder. Each surface gets three hinges. I bought a set of cast aluminum hinges with mounting hardware from Vi Kapler at Brodhead last year. Kapler has been around Pietenpols most of his life. He drew the plans for the three-piece wing in the 1930s and worked with Bernie Pietenpol for decades. The hinges are robust and light, but do require a bit of filing of the ears to get full travel. After mounting the hinges in the elevator according to the spacing in the plans, I moved on to mounting the mates to them into the stabilizer. I measured the distances but also dry fit the hinge halves together and scribed lines on the stabilizer marking exactly where they fell. I was only off a little on one hinge, and with a minor adjustment the elevator moved freely. The other elevator and the rudder mounted up the same way.
Welding ... Second Try
With the tail surfaces finished, the spruce work on the plane is finished. I couldn't dodge the welding any longer. Fortunately, while I was working on the tail, a friend and co-worker offered me the use of his oxy-acetylene torches, hoses and regulators. I bought some tanks and set up a welding area in my shop. Having the equipment, with a good selection of tip sizes, allowed me to practice and experiment with welding various thicknesses of flat and tube steel, and I found that practice really does improve your skills and confidence. I cut steel for the elevator and rudder control horns, and edge welded them. The results were pretty good and with a little filing and grinding, I had three horns ready to go.
The fittings for the tail brace wires require bends to the proper angle. Bending had been another problem area since I only had one small vice. At Oshkosh last year, I saw a nice 18-inch benchtop metal bending brake, which I bought on the spot. That made bending work a snap and I could turn out pieces smoothly and accurately bent to the proper angle.
This spring, I plan to work from the tail forward, completing the bracing wires and fittings, making the elevator bellcrank and finally tackling those control sticks and torque tube. The third time should be the charm on that! I would like to have the fuselage standing on the landing gear by early summer.
I've learned a few lessons over the past year. Persistence pays off in the face of discouragement. The right equipment can overcome a lot of problems. When one route appears difficult, sometimes a detour isn't such a bad thing as long as you continue to make progress. And, though I violate this above, never make any predictions on when something is going to be finished.
So, one of these years, you should see my Pietenpol at Oshkosh. I'm just not saying which year that will be.