Restoring the A-10D Mitchell Wing
SN# 275
By
Episode VI
The Instrument Panel, Electrical System, some Mylar, Paint And Cranky Wingtip Shrouds
So, after clicking my heels together three times and flashing my credit card, I had returned from Kansas with a new outlook. I was convinced that if this thing ever flew it could be a lot of fun, something the restoration process didn’t have a lot of. Nevertheless, I went back to the garage/hanger with new determination. The next move would be to continue the installation of the electrical cabling that ran from the engine to the cockpit. I had decided to group the wires into three major cables. The one on the left side of the airframe would house the ground and kill switch wires as well as two green magneto wires to drive the tachometer and Hobbs meter and two yellow magneto wires that would provide electrical power to the cockpit through a Key West power regulator. In addition, another pair of wires would be brought into the cable at a point just behind the seat from the Facet electric fuel pump. All the wires would be long enough to reach the front of the cockpit in the general area of the battery (wherever that might end up) plus some extra for last minute changes. The second cable would run down the right side of the airframe. This one would house the EGT, CHT and the fuel depth sensor wires. It was important to keep these separated from the magneto wires because they had to transfer very small DC voltages to the gauges up front. I didn’t want to risk them picking up any induced AC from the magneto that might bias the readings. The starter button wires would be brought into the cable from the side panel next to the seat. The third cable would actually be only the two heavy, #10 copper, wires that went from the starter solenoid / battery back to the starter. I then set off to the local auto parts store and came back with two ten foot lengths of flexible, half inch, plastic conduit as well as some rolls of 18 gauge and 20 gauge wire in a rainbow of colors. The bag also contained a variety of crimp-on terminals and connectors. Next, I searched through the piles of parts boxes in the garage and succeeded in locating the CHT and EGT sensors, along with two eight foot extensions purchased sometime in the dim past. After uncoiling the sensors it was obvious that the extensions were too long by a good four feet. A call to Aircraft Spruce had two, expensive, four footers on the way by return mail (Unfortunately, as it always happens, I learned too late that I could have simply extended the existing sensor wires by splicing in some 18-gauge wire. Another entry for the ever growing “stupid is expensive” notebook.).
Assembling the cables was fairly straightforward, cutting twelve-foot lengths of wire and then stuffing them into the conduit through the side split. These were then temporarily held in place on the engine diagonal tubes by tie wraps to allow the usual design changes and “better ideas” as things progressed. And, although the wires did indeed fit, were I to do it again, I’d probably go with ¾” conduit in order to make wire positioning a bit easier. Twenty gauge wire served for everything except the fuel pump, engine ground and starter button which were #18.
Mounting the battery was the next problem, along with the power regulator and starter solenoid. I had the notion that the battery should be movable fore and aft as it would act as part of the ballast to change the weight on the nose wheel for proper CG positioning. After toying with a number of schemes, it occurred to me that the percentage change might be so small as to be moot. A few minutes with some cardboard and the small lead brick that is an 18 Amp-Hour gel cell confirmed my suspicions. After some reality brainstorming I decided to mount the battery along with the power regulator and the solenoid on a small aluminum plate cut to fit the forward part of the main frame tubes. I secured the battery to the plate using some short lengths of padded aluminum angle and a nylon utility strap that I’d picked up at a local “we sell everything cheap” store. The regulator, two terminal strips and the solenoid took up the remaining space on the plate. It was here that I decided to implement an idea that I’d gotten from a Phantom owner at an EAA fly-in. Noticing that he had very little shielding on his electrical system I asked how his handheld transceiver reception was. “No problems”, was the reply. And the reason was that he’d gone to a single point ground rather than using the entire airframe. No ground loops, no electrical noise. Or at least that was his theory. It made some sense so I used a longer than necessary bolt for mounting the solenoid, to which all the ground wires would be attached. Then, with everything in place, I Cleco’ed the plate to the frame tubes to check the fit. It looked pretty good and hopefully would get me close to the magic 19%-20% nose wheel weight for docile flight characteristics.
I was now ready to tackle the instrument panel, a definite milestone in the restoration process. The first step in the process was to retrieve the cardboard mockup panel that I’d made eons ago, before the pod had been removed. It still had the moveable, sticky-backed cardboard “instruments”, made by my kids, to help me work out the positions of the various dials and gauges. Climbing into the pilot’s seat I plopped the faux panel in front of me and immediately found out that, without the pod, I had no idea where it would actually attach to the airframe. So, out came the pod. The problem was that I’d have to remove the nose wheel to put it back in place. Harking back to the factory construction video I realized that they had attached the pilot’s cage to the supported center wing section before jockeying the pod into place on the frame. Obviously I wasn’t going to make that happen since we were well into another miserable Rhode Island winter and the wing was outside in the “hanger”. Hmmm… More brainstorming. Rifling through my bicycle stuff I came up with two hooks that could hold a bike in the rafters. A short time later the hooks were installed overhead and some polypro line, used for the kayaks, pressed into service to lift the airframe. Unfortunately, the pod wouldn’t fit. The spacing between the two forward wing attachment points was too close. Damn. A search of the scrape woodpile produced a piece of 1x3 furring that I notched and forced between the lift lines. The pod finally went into position, more or less, and I replaced the nose wheel to take the strain off the lift lines. It was immediately a new world, a smaller, tighter world at that. Whoa, and it looked like, dare I say it, an airplane.
Grabbing the cardboard panel I squirmed into the cockpit and slid down into the seat. “Now for some “imagineering”. I shifted the mushroom shaped panel into position beneath the pod’s cowling and then mentally “took off” to do some ergonomics checks. At first it felt a bit silly, moving the non-functional stick, pulling levers connected to nothing and pushing buttons that didn’t do much but click. But, after about half an hour of this make-believe I was convinced that the panel, as it stood, wasn’t going to work. The mushroom design was good in that it bought additional real estate, but the stick obscured most of the vertical “stem”. There was also a problem in the fact that the altimeter and the VSI would project some six inches behind the panel. As it was, the forward diagonals of the airframe would interfere with the mounting of the larger instruments. A major reorganization / redesign was required. Rats…
Alright, maybe I’ve got too much stuff in too small a bag. What did I really need to fly this thing? An airspeed indicator? Airspeed is life to a pilot, so yes, the 3 1/8” ASI would stay. But my time in Kansas told me that a Hall wind speed indicator clamped to a diagonal outside the cockpit would work just as well. So, if need be, the dial unit could go back in the box, if not back to Spruce. Altimeter? If airspeed is life, altitude is life insurance. This was especially true in the confines of Southeastern New England, where I’d have to thread my way in and around the controlled airspaces at various airports in Rhode Island and The People’s Republic of Massachusetts. The 3 1/8” altimeter would stay. I wanted both the ASI and altimeter in the largest size to be as readable as possible. What about the Vertical Speed Indicator? The time in Kansas said it was a good thing to have. But was it critical? I did have the antique pellet variometer from the old panel. On the other hand, I had no idea if it still worked, and if it did, the tiny display head would be at quite a distance from these aging, tri-focaled eyes. A Ball Electric would work, but those were heavy money. I could dump the vario altogether, but then I’d be screwed for soaring, and that was one of the main reasons I bought the A-10 in the first place. Command decision, the 2 ¼” VSI stayed and the vario was off the panel (anguish…), for now. Tachometer? It stayed. The combined CHT/EGT gauge? A keeper. The Hobbs meter? Nice to have but expendable if necessary. The Fuel Gauge? Unlike the old tank between the knees on the standard A-10s, there was no way I could see the fuel level in a tank behind the seat. I had considered a tube gauge like the Falcon used, but that would mean additional plumbing that would bring the fuel into the cockpit. Nope. I’d also be out a hundred bucks and would have a large hole in the JAZ jug tank. Hmmm, I suppose I could move the gauge somewhere else. Maybe a bullet housing overhead? Whatever, it stayed. The combined Volt-Ammeter? A great tool for troubleshooting electrical problems and evaluating battery condition. But was it critical? I guess not. What’s left? Some fuses. Nah. There’s always room for fuses. How about the keyed main switch? That stayed as a safety item. The last thing I wanted was some kid, or adult wannbe, climbing into the cockpit at a fly-in and spinning up the engine. I’d also need space for some 12-volt DC adapter sockets to run things like the radio and GPS, not to mention the electronic compass. Still, those didn’t necessarily have to be on the panel. And if all else failed I could backtrack and go digital with one of the expensive, electronic, do-everything-show-everything-in-a-tiny-space monitoring systems. I rather dreaded this, as I was not a big fan of digital readouts. Even digital watches annoyed me. Humans, after all, are analog creatures. Okay, now I had some options. So, it was back to ground zero with a new cardboard mushroom and the sticky cutout instruments.
Once again I climbed into the cockpit with Mushroom MOD 2 and began shifting the cardboard cutouts in sort of an instrument shell game. It still wasn’t working. Then a miracle happened. I discovered that if the panel was tilted forward, not only could I read them more easily, but also the instruments actually cleared the airframe tubes. On a hunch I lengthened the vertical stem, kinked it four inches above the bottom and bent it down to support the bottom of the panel. Viola! MOD 3, the bent mushroom. I even got some extra real estate, enough, in fact, to mount the fuses, the key switch and all three toggle switches that had been relegated to the small (and inconvenient with the pod in place) left side panel. Whipping out my Sharpie pen and ruler I then transferred the mushroom outline from cardboard to aluminum sheet. The next stop was Roger’s Machine Shop to have him cut out the complex shape. While that was in progress, I turned my attention to getting my variometer back. But first I had to know a little more about it. I hadn’t seen a pellet vario since I’d climbed into a glider at age 17. Did they still make these things? A search of the Internet turned up nothing and calls to instrument companies brought only muffled laughter. Too bad, it was really kind of a cute little system. Closer examination revealed that it was small block of Lucite that had been precision drilled in multiple dimensions with two pith pellets that moved up and down in parallel tubes as air moved in an out of a small reservoir (in my case, what appeared to be an insulated peanut can). A piece of clear tape with graduations in ft/min covered the front, labeling one tube with a green “UP” arrow indicating climb (thus the term “green air” for lift, used among older soaring pilots) and the other with a red “DOWN” arrow, meaning sink. There was also a tiny, but worn label on the top with what I took to be the manufacturer’s name: something, something, KIKI Electronics. While the name of the manufacturer was almost unreadable, the phone number on the bottom was fairly clear, except for the area code which ended in “08”. It was a mystery worthy of Sherlock Holmes, or a determined soaring nut like me. Since Holmes and Watson weren’t about, I picked up the magnifying glass and began searching for clues. The glass revealed what appeared to be the remnants of an address still embossed in the label. It ended with a postal zip code that the USPS website quickly identified. It was a little town in… Hawaii? Area Code 808? Could it be? Holy hula, this thing was made on Oahu? Having been there some thirty-two times one would assume I’d have heard of these guys. But such was not the case, perhaps the result of a laid-back Hawaiian marketing strategy. Whatever, the next day found me on the phone to the manufacturer located on the northeast short of the concrete island.
The “aloha” from the other end of the line sounded a bit mechanical, probably from the stress of trying hard to filter out the local dialect. “You gotta what? No kiddin’? Wait one…okay?” The phone thumped down, steps were heard and all pretense disappeared as the same voice shifted to pidgin. “Hey ‘Bra, some lolo got one your vario tings. Sound like some kine mainside houle. You wan talk to ‘em?”
“Mainside, huh?” answered an older voice.
“Yeah, Rhode Island.”
“Wow, far east, huh? Not no @#%! Lawyer?”
“Nah, sound okay. Talk funny though.”
“Well….okay.” After a short a short pause, slower steps approached the phone and then, “Good afternoon, may I help you?”
“Howzit,” said I in my best pidgin. “Hey, you’ll never guess what I’ve got in my instrument panel.” There followed a very interesting conversation with the designer of the variometer. No, they’re not made anymore. Stopped making them back in the late seventies. Sold out to an Aussie who made them for a few years down in Oz. Parts? Who needs parts? Nothing moves, bra, except the pellets. Long as the can is good and the tube don’t leak you good to go. Test? Sure, just take the can off an’ blow real gentle into the tube. If the green side pellet move up, that side okay. Move the tube to the other outlet on the back of the head an’ do da same. Red side pellet move up, hey, pau hanna, it’s Miller time. An if sometime they get stuck, it’s probably static, just rub the Lucite real good with a dryer sheet. Okay?
“Okay,” I said, now very relaxed. Waves were lapping softly on the sands in the back of my mind. I liked this guy. And I was suddenly consumed by the urge to “check the sets”, grab my “stick” and go surfing. But then the revere broke and I was back in the “far east”.
“So, what’s it like there in Rhode Island?” came the voice.
I looked out at the freezing rain. “Sucks, here, how ‘bout you?”
“Ah, you know, just another shitty day in paradise.”
“Thought so,” I ventured. “Big kine mahalo, bra. Aloha nui.”
“Anytime.”
And so, I decided to keep the pellet variometer. I’d make it fit. The question was how? It took a few days of trial fitting, and rethinking, but I did it. The head now lives on the top of the instrument panel in a tiny Radio Shack plastic project box mounted on an extension to the compass base.
When the blank instrument panel came back from Roger I covered it with masking tape, pulled out my handy-dandy instrument template from Aircraft Spruce and started drawing in the instruments. Three more imaginary flights with the panel in place brought about three more design changes. Then, a strange thing happened. Everything fit. Not only that, I was able to move the starter button from the side panel to the mushroom stem. There was even some blank space for future additions, if such should ever occur. I toyed with putting the pellet vario on the panel but realized that the slope would prevent the pellets from reliably returning to zero. The head had to be near vertical. Thus, it stayed on top. A better place, anyway, I thought. It was closer to the line of sight as I’d be constantly looking outside the cockpit while circling, trying to center a thermal. And it was only a short distance, visually, from the ASI, another important instrument to monitor while trying to balance the plane on the edge of minimum sink speed, clawing my way upward to cloudbase.
And so, finally satisfied with the redesigned redesign, I dropped the marked blank back with Roger, who was now on a first name basis with the “airplane guy”. While that was underway I was in a quandary as to what to attack next. I decided to make an attempt on the fiberglass wingtip shrouds that had been epoxied into oblivion by the well-meaning repair / painting person. Ordinarily, the shrouds are unusual clamshell affairs that open to the rear in order to slip over the rudder hinges, streamlining and protecting them. Once fitted and pop-riveted in place, their rear sections are put together, semi-permanently, using Bondo, which actually doesn’t adhere all that well to fiberglass. This is the secret that allows them to be reopened for removal at a later date. My problem was to somehow sever the epoxy bond between the paper-thin rear shell halves of the shroud. The initial attempt brought out my trusty Dremel tool with an abrasive cutting wheel on the business end. It was only marginally successful as the wheel was unable to cut deep enough to get through the inch and a half of epoxy. It did, however, cover me, and most of the interior of the garage, with fine, white, itchy dust. The wheel also had a habit of wandering off the straight and narrow, which didn’t help matters in the least. Next came a hacksaw blade, which had the depth of cut required, but was nearly impossible to control. It was obvious that I had been defeated. Thus it was that I e-mailed Larry Smith at Ameriplanes to order a new set of shrouds. A phone call and $200.00 later, they were pulled from the mould, assembled, and on the way east.
It was Spring of 2003 now. The weather was grudgingly turning from cold, wet and miserable to just miserable with isolated days of sun and a hint of warmth. While the panel was being cut and the wing tip shrouds were on the way, I decided to wander into the hanger and see what could be done on the main wing section. A close inspection of the Mylar film on the wing’s underside showed it to be in very good condition except for the small diamond shaped section at the trailing edge. This showed evidence of tearing and, oddly enough, burn holes that had been repaired with what looked like clear packaging tape. The section would have to go. Fortunately it was not part of the large center piece. So, with a bit of patience I managed to carefully pull off the packaging tape and then the aluminized Mylar tape from the edges. Pulling the Mylar film from the double-sided foam tape turned out to be easier than expected because of the tape’s age. Getting the foam tape off the aluminum, however, was going to be a real pain, just as it was when I removed the Mylar film from the wing end sections earlier in the restoration. Dipping into my reservoir of experience in this area I pulled out a variety of solvents and began attacking the tape. A citrus product called “Goof-Off” appeared to be a definite winner as a non-toxic answer to tape adhesive remover. The secret seemed to be in letting the stuff soak into the tape adhesive for about fifteen minutes and then carefully pulling and coaxing it off the aluminum. It was messy and time consuming but it worked.
Looking at the old aluminized Mylar film I noted that the underside was dirty, covered with the red mud common to the southern reaches of the country. Peering into the hole confirmed my suspicion that another colony of mud wasps had left a ghost town in this part of the wing. Quick work with a screwdriver and the shop vac cleaned up the mess and removed a pound or two of dried mud as well as numerous wasp remains. It was now time to recover the hole. The Mylar sheet purchased from Aircraft Spruce came out of the tube and was cut to the rough dimensions of the hole with overlap for the double sided foam tape. Outside cutting was done with some very sharp scissors while the openings for the pilot cage attachment points were done with a hot knife attachment on my soldering gun. The sheet then went into place a bit a time in order to reduce the chance of a wrinkle in the Mylar. This was accomplished by pulling back the tape’s paper covering just ahead of the Mylar as it went down. Once satisfied that the new film was in place and pressed down firmly, I pulled out the roll of 2” aluminized Mylar tape that would hold the edges firmly to the aluminum wing. It didn’t look too bad. Still, it needed to be tightened. Left too loose it would vibrate in the wind and eventually fail. The heat gun was therefore pressed into service to shrink the film. Carefully applied, the Mylar quickly tightened like a drumhead and I crossed another job off the “to do” list. The only problem was that it looked new and crystal clear compared to the older film of the center section. Loath to take off the center section I decided I could easily live with it. The next item was the wing’s fiberglass nosepiece, which was still a faded red.
And so, I switched into paint mode. After a quick conference with my paint / fiberglass guy, I pulled out a green abrasive pad and began roughing up the fiberglass gel-coat. Masking with some old plastic sheet was next and then the paint. I rummaged through the cardboard box that held a variety of spray paint cans until I came up with some DOT Hi Viz Warning Yellow. The cap promised a reasonable match to the automotive Mayflower van yellow on the pod, so I decided to try it. The spraying went fairly quickly and produced a very nice finish. The color, however, was nowhere near that on the pod. Harking back to my days as a color matcher and paint tech at Ford Motor Co., it occurred to me that the small panel I’d sprayed out in the garage came up close as a match only because of the miserable partial spectrum of the fluorescent lighting. Daylight made a big difference. Nuts. Another conference with Al the painter and a rag soaked with acetone quickly removed the offending pale yellow. Mayflower it would be. The local hardware store provided a disposable aerosol paint system and the auto parts store the thinner and hardener for the paint. Another spray session and the nosepiece was a glowing healthy yellow. I thought about painting the newly arrived wing tip shrouds, but decided to do some fitting on them first. It was a wise move.
Making the shrouds fit the wing turned out to be another descent into purgatory. I had taken the more expensive option and ordered the shrouds assembled in order to save some time. When they came out of the box, however, I noted that the profile of the surfaces that mated to the wing tip was considerably different than the originals. The new units were thinner and the edges that mated to the wing were curled inward. The forward section was also narrower and did not match the wing’s leading edge. Damn, another puzzle. I decided that the best thing to do was to cut the slot for the wing support and find out just how bad it was going to be.
Using the old shroud as a guide I measured the slot position and then put my rotary tool to work with an abrasive disk to make the cut. Moving to the wing tip I bent and twisted the new shroud until it grudgingly slipped under the rudder only to discover that the rudder was now jammed. The slot was cut too low and the inside edges only vaguely matched the wingtip. On the positive side, the flat upper surface of the shroud that fits beneath the rudder was now at the same cant angle as the rudder base. At least this part of the new shroud was in synch with the new hinges installed earlier. Successive fittings and cuts with the Dremel resulted in the narrow slot being widened to an inch, and the thing still didn’t fit. Worse, the leading edge of the wing in no way matched the narrow profile of the shroud. What to do? I had to somehow widen the inside of the fiberglass shroud to match the wingtip and at the same time adjust the fit of the top to allow free motion of the rudder. Okay, one step at a time. After some thought I ended up tracing the profile of the wingtip from the leading edge back about eight inches. I then found some scrap aluminum plate and cut it to the profile, along with some relief to clear the rudder hinge. This, I jammed into the inside of the shroud to hold the fiberglass open and was pleased to see that there was now a reasonable match to the wingtip, except for the leading edge. I now began trimming the raw edges of the fiberglass to make it fit reasonably well against the wingtip. The lower edge ended up with more material being cut off. This allowed the top to cant outward to match the cant of the rudder. Small, shallow cuts on the top and bottom allowed clearance for the rudder fold hinges. After many hours of work, the rudder now moved freely, but there was a major league slot in the side of the shroud and the leading edges still didn’t match. Damn. It was nearly impossible to stretch the fiberglass nose of the shroud any farther to get a good match. There had to be another way. Then the obvious struck me. When in doubt fill it out, with fairing material, that is. I had to repair the overlarge slot, which meant a small fiberglass job that would look like crap from the outside unless faired with compound and sanded. Why not just fair in the whole nose while I was at it? I had to fill the numerous flaws and voids left from the molding of the shroud anyway. So, off I went to the local West Marine supplier and rounded up some epoxy and featherweight fairing compound. It was certainly not cheap, but the West system was proven and easy to work with.
The next move was to thoroughly clean the raw fiberglass with MEK to remove the mold release agent and to lightly sand the thin fiberglass to remove the blobs of epoxy left over from the joining of the two halves. This took some time and considerable care. During the process I noted that the fiberglass had begun to crack at the thinnest point on top, where the shroud had to be repeatedly bent while installing and removing it for fitting. A bit of epoxy would, I hoped, would cure that problem. The next move was to pop-rivet the aluminum profile plate in place to stabilize the structure while the fairing compound was added. This done, I pulled out the lightweight fiberglass cloth, purchased in the dim past from Aircraft Spruce, and cut a small strip to fill in the large slot. Epoxy was then mixed, using pumps to get the right ratio, and applied to the inside of the shroud around the slot. The cloth was then carefully laid inside the shroud, over the one-inch slot with more epoxy to hold it in place. Both sides of the cloth were then carefully wetted with the sticky goo. The entire thing was then set aside to cure overnight. The results weren’t too bad.
A few days later I was again mixing epoxy, this time adding the fairing powder until the whole gooey mess had the consistency of peanut butter. In order to get the right profile I had previously traced the wingtip’s leading edge on paper and then cut it out as a guide. Then, using some large, flat, tongue depressor like craft sticks I applied a thick coating of the fairing paste until the nose of the shroud was slightly larger than the paper profile. I also filled in the voids and imperfections in the raw fiberglass.
Sanding and shaping occupied most of the next work session. Latex gloves and a dust mask were primary safety items for this, as the fine dust flew everywhere. Doing it outside is a great idea, assuming you don’t have to deal with the Southeastern New England drizzle. What I ended up with was a pretty reasonable approximation of the original shroud profile but pockmarked with some annoying voids. A second, thinner coat of filler eliminated the major holes and allowed me to feather the additional material into the shroud with no abrupt transitions. The second sanding session was done with the care that Michelangelo might have employed on his David. I was definitely getting better at this. Still, close examination revealed a relatively porous surface with the occasional pinhole void. To correct this, I applied a very, very thin coating of Bondo Glazing putty right from the tube. Then some 400 grit wet / dry sand paper and presto! The surface was the next best thing to glass. With this done I noted the lessons learned to ease the work on the other shroud, assuming I lived long enough to get to it. I hoped that writing it all down would help me remember the complex gyrations required to wrestle the shrouds in place when the time came.
And so the restoration saga continues. Tune in again for Episode VII when our hero battles the paint monster, finishes the wingtip shrouds, attacks the pilot’s pod, again, and more.
Episode I Episode II Episode III Episode IV Episode V Episode VII Episode VIII Episode IX Episode X