So after trying unsuccessfully to connect with my EAA Technical Counselor to come look at my elevator fiasco I decided to jump in the deep end and attempt a fix myself. After much thought and a few opinions from Van’s Air Force (including one that amounted to, “just give it a big ol’ yank”) I was pretty convinced the fix I had in mind would work. Specifically, my plan was:
Remove the elevator counterweights
Drill out all the rivets in the counterweight skins
Using fluting pliers to impart a corrective bend in the end ribs
Reattach the skins and counterweights
I started with the right elevator because it had the most interference, figuring if I could get that fixed the left one would be a piece of cake. (Yeah, right.) Step one, unbolt the counterweights. Step two, turn several dozen of the previous builder’s carefully driven rivets into metal confetti:
With the skin removed I had unfettered access to the end ribs and proceeded with bending them (to my will). I created a slightly exaggerated correction figuring that when reattaching the skin (using the same holes) would have a tendency to pull in back towards the h-stab. It did as expected after I clecoed the skin but did leave enough of a gap (3/32″) to resolve the problem.
From here all there was to do was to re-rivet the skins. This went pretty quickly thanks to the pneumatic squeezer I got from Cleveland Aircraft Tool. (I also got a t-shirt but this didn’t materially affect the process.)
Happy with my work on the right elevator I took the night off and tackled the left elevator the next morning. Oddly, even though I now understood the process it still took me about the same time to finish. Here’s a gratuitous time lapse of me drilling out the rivets:
In the end, I spent almost 10 hours correcting an issue that should have been done right in the first place. Happy days!
With the elevators themselves in limbo, I decided to move on to the elevator pushrod. In theory this is a straightforward step. You trim a piece of aluminum tube to length and then rivet in threaded inserts on each end. The aluminum tube is deceptively light and you wonder how it handles the stresses of control inputs. Of course, like an aluminum can, as long as it isn’t deformed in any way it has tremendous longitudinal strength. That means don’t dent it!
Second, you want to make sure to get the finished length correct. This is particular important with a piece like this one since, at 6+ feet in length, if you cut it short you’ll be paying as much for shipping as for a replacement piece. To complicate matters I had just broken my bandsaw blade and so needed to go “old school” with a hacksaw.
As is good practice whenever working with aluminum I measured the correct distance but then cut the tube slightly longer. I then set up some blocks on an adjacent workbench level with my sander. This would allow me to sand off the excess aluminum (being careful not to overheat it) until I had the specified length.
This ended up working very nicely. I measured several times as I was getting close to my mark and sanded a bit more until I had a perfect length (or at least as perfect as a my tape measure). You can see this in action here:
After sanding the pushrod to length the next step is to drill six equidistant holes for blind rivets. The plans provide an easy way to do this–cut a strip of paper that matches the circumference of the tube, mark the hole locations, then wrap it around the pushrod. It worked a treat! Now just measure the edge offset and drill #30 holes through the pushrod and insert.
As you may have noticed, the previous builder dispensed with primer on the interior surfaces. Again, I won’t get into the endless debate here. Van’s is typically indifferent but does recommend priming the inside of the pushrod tube since, once it’s rivetted, you have no way to inspect it. I took their advice and sprayed some self-etching primer in both ends. It was a bit difficult to confirm coverage but spray was coming out the opposite end and there was enough accumulation to swirl around the inside.
Additionally I thought the outside would just look nicer with a smooth coat of primer so I scuffed it with Scotchbrite (probably not needed with self-etching primer but did anyway) and cleaned with acetone. To get a consistent coat I fashioned some simple hangers from 1×2 redwood and nails clamped to a shelf protected with plastic. I got some drips at first but quickly realized they were being caused by ill-fitting nitrile gloves hanging over the nozzle. A quick adjustment and the priming was done.
After a few days of curing there was nothing left to do but make use of my new Ace Hardware rivet pulled and attach the threaded inserts, followed by some rod end bearings.
Many builders look forward to the moment when, after months of diligent and careful work assembling numerous subcomponents, you get to Section 11. The title of the this section is brief and to the point: Empennage Attach. Like all sections, it begins with an overview diagram that conveys what you will accomplish by the end. The image associated with Section 11 tells the story. You will now assemble those numerous components into something resembling a real airplane (at least if you stand at the F-1006 bulkhead looking aft and squint your eyes). I feel kind of bad that this is one of my first tasks as the new kit owner and that the previous one won’t get to witness it (but don’t worry, I’ll get over it 😁).
One of the first steps is to attach the elevators to the horizontal stabilizer. Each is first attached by two rod end bearings secured with AN6 bolts, then you drill through the (very important) control horns, at which point both elevators are tied in to a common attach point. I was sweating this step because once you drill through the steel control horns you are, as they say, committed (in terms of geometry). Anyone who has worked with door hinges know that with two attach points you can get away with a little imprecision but with three everything needs to be lined up pretty closely. Such as it is with an RV-10 elevator. You are committing almost 11 feet of painstakingly and lovingly rivetted aluminum to two 1/4″ holes drilled using nothing but the central bearing as a guide. Screw this up and you will forever feel whatever wonkiness you’ve introduced in the control stick. Eek!
Fortunately I time to contemplate this most significant hole drilling exercise as the E-drill bushing (which is used to protect the central bearing from the nastiness of the drill) could not be located in the purchased kit inventory. Another order to Van’s and waiting for USPS was in order.
After the bushing arrived I decided it was time to forge ahead, the first step being to secure the horizontal stabilizer to a workbench “just so”. You position it so that the elevators hang over the edge of the workbench–all the better to test their range of motion.
The next step involves setting the offset distance of the rod end bearings to a specific measurement (7/8″), attaching each elevator, checking the edge clearances, and adjusting the bearing distances as needed. No problem!
Before I go on I want to point out that whomever decided to put rod end bearings in such a tight location (recessed inside the elevator skins) and expected someone to be able to adequately manipulate them must have been a masochist. After several clumsy attempts to adjust them using “off the shelf” tools (and mucking up the bearings in the process) I decided to search for a better option. I did find one–a custom tool for sale on line–but not wanting to (a) lay out more cash for a finished product and (b) endure the inevitable shipping delay. I decided to fabricate my own based on what I saw. So it was back in the truck and off to Lowe’s for raw materials.
One thing that has always appealed to me about the building process is that it will inevitably involve solving problems. (Keep than in mind–it becomes relevant again later in this post.) I do truly enjoy the process of being presented with a problem and having to stare, contemplate, theorize, diagram, experiment and (eventually) solve the riddle. For this particular conundrum I decided to fall back to my woodworking and plumbing experience and fabricate a tool that would do the job. It consisted of a plug made from an oak dowel (shaped using my drill press and small milling vise) and some PVC pipe.
With a handy new tool in, um, hand, I set about to satisfy the plans instructions regarding attaching the elevators. That’s when I had my first (only?) really big “WTF” moment thus far.
I was feeling pretty clever after figuring out how to support the port elevator so I could insert the AN6 bolts into each bracket, securing the rod end bearings into place, without assistance. Then, given the precision of modern, CNC-punched parts that Van’s produces, I expected to quickly confirm that the gap between the elevator end rib and horizontal stabilizer was a consistent 1/8″ as the plans specified. Well, not only was the gap not consistent, the forward tip of the end rib was actually binding on the h-stab. (As I said, WTF?!)
I won’t relay all of the thoughts that raced through my head at that moment (as some of them are not appropriate for a family audience) but suffice it to say I did pause and stare at the situation in front of me for some time as the stages of grief started to set in. The first phase, of course, is denial, as in, “wait, that can’t possible be happening.” I checked the plans and my work and indeed confirmed that the tip was binding. This quickly gave way to anger and several more curse words were uttered. Then the bargaining began. If only I adjusted this a little here and that a little there the maybe, just maybe…no. Actually I did get the tip to stop binding by backing the outboard rod end bearing out of its mount until it was hanging by just a couple threads but that didn’t seem like a structurally sound solution. After several iterations taking the elevator off, making adjustments, and putting it back on to no avail, depression set in.
Not ready for acceptance just yet I decided to try fitting the starboard elevator. Surely this was an isolated problem–or maybe I’d just misread the plans–and the second elevator would reveal the error of my ways. You can probably guess the result. It was a mirror image of the first elevator, if not worse. What was going on here?
I took a carpenter’s square and lined it up with the centers of rivets on the elevator’s longitudinal axis and attempted to also align it with the rivets on the end rib/counterweight skin. If you look closely at the images above you can see that, while the longitudinal rivets are perfectly aligned, the end rib rivets are askew, progressively getting worse as you go forward.
I shot off an email to Van’s and their response was, paraphrasing, “unpossible”. Well, yes, I agreed. If you follow the plans and align the pre-punched holes this should not be possible. We discussed potential ways to affect a fix, including longer rod end bearings and grinding down the edges. Neither seemed workable for various reasons.
Going back and reviewing the plans related to fabrication of the end ribs I noticed that it is up to the builder to straighten the ribs using fluting pliers. The skin is then clecoed to the ribs and the holes are match drilled. I speculated that the ribs might not have been completely straight and, after clecoing, might have put some sort of tension on the skins such that they developed a small deformation. Match drilling and riveting the skins locked this in. If you follow the plan’s assembly sequence you build up the entire elevator before fitting it to the h-stab so it’s possible the previous builder didn’t catch the problem.
So, what to do? I think the best solution is to drill out all the rivets that attach the counterbalance skin, remove the skin, and attempt to straighten the ribs, re-attaching the skins after that. This will likely involve replacing the skins since the existing holes would not align completely. Before committing to this course of action I decided to reach out to by build instructor Troy (who is also an EAA Technical Counselor) to get his opinion. I’m still waiting for him to take a look so, until then, this part of the build is on hold. Problem solving indeed!
With much of the tail cone already done by the builder I purchased it from I only a have a few options for things to work on until the next kit arrives (in a few months 😒). In preparation for attaching the tail feathers I decided to work on the elevator trim servo. In particular I wanted to (a) complete a bench test to ensure it works and (b) sort out how I was going to connect the wiring to the eventual fuselage and instrument panel.
The Ray Allen servo setup is pretty straightforward. The kit comes with the servo, a three position switch, and a position indicator display. The wiring diagram is also straightforward–connect the colors as shown.
Since I have previously dabbled in microelectronics I have no shortage of stuff for connecting components like this. One breadboard, a power supply and some hookup wire later…let there be light (emitting diodes)!
I ran the servo through its complete range and after confirming the servo electronics I moved on to the hardware. The first step was to assemble the servo linkage and it was then I ran into my first real aircraft builder decision. The plans call for the three pieces of the linkage hardware to be rivetted together using two AN470AD4-7 rivets but for whatever reason the remaining hardware that came with the kit contained but a single rivet of this specification. Now, being new to this whole every-decision-you-make-could-be-life-or-death thing I sought out some advice from the “hive mind” of RV aircraft building: Van’s Air Force. One respondent suggested I just take a longer rivet and cut it down using a rivet cutter. While I admit it’s a great suggestion I don’t (yet) own a rivet cutter. Not wanted to delay the process by ordering a single rivet to be delivered via USPS during the Christmas crush I decided to use a AN470AD4-8 (just a mere 1/16″ longer than specified). I figured as long I managed to create a good shop head I would be fine. The Cleveland Aircraft Tools “Main Squeeze” rivet squeezer made easy work of that.
The next step was to deburr the trim bellcrank and assemble everything. That’s when I ran into conundrum #2. The bellcrank mounts were too close together and I could not fit all the called-for hardware in the gap between the two angle pieces. With the holes in the F-1095A servo bracket already drilled I decided to drill out the rivets in one of the brackets and fabricate a new one. Fortunately the kit came with some leftover 3/4×3/4 aluminum angle stock and this went pretty quick thanks to band saw, sander and grinder with Scotch-Brite wheel.
Because I had all the hardware already out I sorted out the positioning by first drilling the 1/4″ hole for the bellcrank hardware, connected the bellcrank, and then match-drilled the three mounting holes using the holes in the servo mount as a guide. In the end it probably shifted the remade mount outboard about 1/8″ but that was enough for the bellcrank to move freely. Four rivets later (I screwed up one and had to drill it out) and the bellcrank was in business.
At this point I couldn’t help but wire up the whole assembly and watch the bell, er, crank!
I am curious to see how this all goes together since the forward bellcrank attach point dips quite a bit below the bracket and seems like it would interfere but, hey, after all these years I assume someone would have fixed it if it were a problem.
The next step was to create an attach point for the wiring harness I planned to build. I wanted to be able to (somewhat) easily remove the servo assembly as needed and so there needed to be a way to disconnect the wiring running up front. The servo lead wires are pretty short and so you need to route them to a point quite close to the servo. Based on prior art from Mouser I decided to fashion a bracket that would hold a CPC (see below) from .020 aluminum sheet.
I started by fabricating a cardboard template that would provide enough space for the connector and a tab that will be used to rivet the harness mount to the servo mount. I also added flanges to provide stiffness to the part.
I transferred the cardboard dimensions to aluminum and drilled out holes for the connector using a step drill and the mounting holes (not quite getting them aligned straight but, heck, it’s buried in the back of the empennage after all). Then I cut, sanded and deburred the final piece (being careful to drill stress-relief holes at the bend points) and bent the flanges in a vise.
I then had to decide where on the servo bracket to mount it. It needed to be close to the servo but could not interfere with the mounting screws. I offset it enough to provide clearance and then test fit the entire assembly in the empennage.
With the hardware sorted and connected I moved to how I was going to mount and route the wiring. As I mentioned earlier, I got the ideal to use CPC (Circular Plastic Connectors) from Mouser, and so did some research on them. I liked the fact that they are pretty foolproof and easy to manipulate in tight spaces and so, after shopping around, I ordered a bunch of components from Arrow.com:
Now, all of those components are relatively inexpensive. What’s not inexpensive is the tool needed to crimp these particular connectors. At $200+ it seems like an extravagance at this stage but I see myself using these connectors a lot so I’m sure I’ll recoup my investment over time. (And, by the way, I also got the pin removal tool because I know I will screw something up at some point.) In addition to the specialty bits I also got some wire loom and silicone tape from Amazon and was ready to fabricate a wire hardness.
The first step was to cut the wire loom to length and thread the bare wires through. Of course I failed to recognize this as the first step and went ahead and attached the female sockets. This required carefully unrolling/re-rolling the loom around the wires. Argh…
Before I crimped the wires on the servo I tried practicing with some scrap wire, sacrificing a few connectors in the process. I found it a bit cumbersome to position the wire in the connector in the right way and also hold the two at the right spot in the tool. Also, the tool uses a ratcheting mechanism that doesn’t release until the connector is fully crimped so it’s nigh impossible to correct the alignment if you misalign something. Eventually I discovered that if I closed the tool until the first ratchet position I could maneuver a pin into the correct position and it would stay there. Then I could feed the wire into the open end and finish the crimp. In the end I got all five servo wires crimped without any screw-ups. Time to declare victory and go home!
Actually I forged ahead and finished the servo-side wiring by attaching the wire loom and silicone tape, inserting the pins, and attaching the shell clamp.
To finish out the wiring tasks I created a temporary wire harness to use during the install/adjustment process and tested everything on the bench. I only had one brief moment of panic when the servo didn’t work before realizing I’d inserted one of wires into the wrong row on the breadboard!
The final elevator trim task was to finish the E-616-PP trim cable cover plates, to which are rivetted a set of cable brackets. The previous builder had purchased a set of milled cable brackets from iflyrv10.com, but only after match drilling the holes using the stock Van’s brackets so these ended up as scrap. (Actually I ended up using them as #30 countersink guides.) Van’s brackets are nothing more than a nut welded to a piece of steel and if I had built the tail from scratch I would have upgraded as well.
For a reason never fully explained, he had ordered one replacement plate so one of my first orders placed to Van’s was for a second. Since, at that point, I had not transferred the project over to my name I’m sure the folks processing the order were perplexed why a “non-builder” was ordering this very specific part. A few days later I had my part and was ready to complete the step…until I realized I had somehow neglected to include a rivet puller with my tool order. Fortunately this is a non-aircraft specific tool so a trip to the local Ace Hardware store rectified the problem.
The assembly process went pretty quickly. It would have gone faster but it was the first time I got to use my dimpling table so I spent considerable time making sure I had the right dies, adjusting the table height and ensuring I was dimpling on the correct side of the skin.