Joshua's RV-9A Project

Today I received a reply to my query to Van’s regarding the twisted elevator problem. From Ken Scott:

This is a significant twist, but we have no test data that would enable us to know how much twist is too much. We haven’t built elevators with different twists and flown them, in other words.

We know there are RVs flying with twisted elevators and nothing bad has happened, but again, how much is too much?

You have two options:

1. rebuild the elevator to the intended configuration. This doesn’t necessarily mean building an all-new elevator — just drilling out rivets until you can straighten the one you have would probably work. No flight risks here — we know straight elevators work.

2. fly it the way it is and let the airplane tell you if it needs to be changed. Problems would most likely show up as rigging/trim problems rather than structural ones. Some risk here, as we really can’t say what the flight characteristics might be

Pretty much what I expected to hear, though perhaps not what I was hoping to hear. So, in reading that and thinking about it, I decided to take a look and see how difficult it would be to fix the thing, rather than hoping it would work and (possibly) having to fix it later. I’m a semi-reformed perfectionist, and every once in awhile I have these tendencies. (If there’s one place I don’t mind having perfectionist leanings, it’s in the building of the plane…)

Out to the shop I went, and did some checking at various places on the table (which I’ve come to believe isn’t perfectly flat, and is probably the cause of at least some of the twist), to determine how and where the twist was built in. I drilled off the skin from both the root and counterbalance ribs first. Not enough. The trailing edge had to come undone too, and I believe this is what happened: the skin closure at the trailing edge wasn’t perfectly even when drilled originally, rather the top sheet was offset to one side, which caused the whole thing to twist.

In case there’s a question, applying heat to Proseal does not cause it to become more fluid; it just gets hot. I tried this on some scrap angle which had the stuff on it from when I sealed the TE’s, with no luck. There wasn’t enough “give” in the Prosealed trailing edges to let it move enough, so I ended up using a thin flat ‘blade’ of sorts to split the bottom skin from the AEX wedge. I didn’t spread it enough to disturb the ends of the stiffeners which had been sealed together, nor did I disturb the top skin.

Next I clamped the whole thing to the table quite well, used a 4′ level and spare angle to prove that it was flat, and proceeded to re-squeeze rivets into the trailing edge, using the same approach as used the first time. Also squeezed a few rivets in each end rib. Initial testing after the clamps were removed looked promising, so I squeezed the rest of the rivets, and pulled the blind ones (which I had to steal from the wing hardware, so I suspect I’ll need to order some extras). Yes, my elevator has one more blind rivet than it did before, because I got carried away and drilled out an inaccessible rivet (one of the counterbalance skin to elevator skin rivets on the inside of the end rib)…the other four are at the hard-to-buck ends of the ribs.

Test time…I expect the problem to be improved but not perfect. Cleaned up the table and brought the HS back up, the proceeded to hang both elevators. Clamped the counterbalances in trail, clamped the long angle across both elevators. In my hand, it looked mighty close… Rested the center of the angle (behind the control horns) on the shop chair, and took the ruler in hand to compare the two sides (the chair doesn’t go high enough to get the counterbalances flush, so I measured the height of the counterbalance nose above the HS skin. Left…6/32″. Right (rebuilt)…6/32″.

I’d call that fixed. Almost three hours, but well-spent both in solving the problem, and in regaining my confidence.

(despite how it looks in the photo, the right one did measure around 6/32″; I was supporting the camera on the HS with one hand, which caused it to tip down and increase the gap.)

Finishing up random elevator tasks today…mounted, drilled, primed, and riveted the brackets for the trim tab motor, then set it aside for now. Mounted the trim tab and marvelled at the moving part! Pulled all the plastic off the empennage components for storage, and took a crack at hanging the elevators on the horizontal stabilizer.

I adjusted the rod-end bearings to their proper length and reamed out the holes in the brackets for the bolts (the powdercoat clogs the holes somewhat). It was here that I discovered an issue…

It seems that when both elevator counterbalance assemblies are clamped “in trail” (aligned with the horizontal stabilizer), the right trailing edge is about a half inch higher than the left. They’re both straight, and the gap is consistent along the whole length. Playing around with a 4′ level, I checked both elevators for flatness and eventually came to the belief that the right elevator is twisted. The result is that either the TE’s match, and the counterbalances are split, or the counterbalances match, and the trailing edges are split.

I did some searching on the various online resources, including VAF and builder logs, and it seems that this problem isn’t unheard of. The opinions on its solution vary widely, however-some state that it’s no problem at all, others say it will cause a rolling moment which will always need opposite aileron to counteract it. Some claim it can be fixed by drilling out some rivets and “massaging” the end rib, others report they built a whole new elevator. RV-9A builder Mike Hoover had a similar problem which he wrote about on his build log (related thread at VAF), and included advice he received from Van’s:

Talked with Scott at Vans about the misaligned TE on the elevators as they are currently mounted on the HS. He suggested three alternatives: 1) leave as is, if any undesirable flight characterists, then use the second alternative. He said this happens frequently to varying degrees and may not be that noticeable in flight. 2) Align the TEs, reposition elevator horn bolt holes as required and accept the downward position of the left elevator counterbalance or the upward position of the right one. 3) Remove the elevators, find where any twist is that is causing the misalignment, [and] work the twist out by using more rivets, which will result in the elevator having pre-loaded stresses.

I believe he had already drilled his elevator control horns when this problem was noticed, which put his counterbalance arms in trail, but left his TE’s split. I would be able to go directly to option #2, which would align the TE’s and split the counterbalances. In any event, I have sent an email to Van’s builder support, and await their response. I suspect it will be some variant of the above, and the problem is not as big as I imagine it to be. From the above linked VAF thread:

I ended up with about 3/8 vans said, you won’t know it during flight, many rv’s have this problem, (we all start on the emp) also he said only you will know it so at the fly in’s or when you go somewhere other might see it, just tie the stick back with the seat belt so they are not aligned and no one will ever know. build on..

OTOH, if I need a new elevator…well, that’s life I suppose.

As planned, I was able to rivet, epoxy, and clamp the trim tab this morning. It’s weighted down to the flat table while the epoxy sets. The package claimed 30 minute set, 7 hours to handle, and full cure in 24 hours–which will be about when I next have time to play with it. I also riveted the hinge to the elevator, bent the pin, and drilled the safety wire hole in the rib. Just a few things left to finish the tail…attaching the elevator weights, installing the trim motor, and fiberglass.

I did find a bit of light corrosion starting at the edges of the trim tab when the plastic was removed for riveting, so that will have to be polished off. This is a “known issue” with leaving the plastic on; one of my next steps will be to pull the plastic off all the finished empennage parts, since they’re pretty much done being moved about, and won’t need the scratch protection.

Primed the trim tab components. The rib locations were masked off so that the adhesive (I will use some epoxy) has a better surface to stick to. From here, it should be a quick job to finish out the tab, and thus, the empennage! (Well, except for the fiberglass…)

The AFS primer/sealer works really nice in weather like this; you can keep moving at a good pace since the light coats dry quickly. (Today: scattered clouds, 86*F, 43% humidity, and a light breeze.) The last priming session wasn’t so much fun, since the humidity was high (90%+) for several weeks, and things were taking forever to dry. I had to touch up a couple spots where fingertips had goofed up the coating. Today, by the time I had sprayed the last pieces, the first pieces were ready for their second coat. I’m quite happy with this stuff.

Finished drilling, dimpling, and etching the trim tab components. Ended up dimpling the hinges rather than countersinking–I goofed on the elevator and dimpled the rear spar, instead of countersinking. I tested the dimples with the excess length of the hinge, and didn’t see any problems at all; no deformation of the eyes, and it nested nicely with the other dimples. Only “problem” is that it may require a longer length rivet than the plans call out, but that’s not really a problem since I have a whole case full of rivets.

Also fabricated the foam ribs, a task made easy by taping tracings of the pattern onto the foam block, rough cutting with a hacksaw, then going to the lines with the disk sander. Masked the rib locations inside the skin so everything’s ready for primer. Will try to pick up some epoxy today, so if I can get things primed tonight, tomorrow’s session should make quick work of the assembly.

Pulled all the clecoes from the trailing edges & angles…the Proseal has been drying for quite awhile, and all appeared to be straight and solid. Squeezed in rivets partway, then used a cut-off bit of the AEX wedge to shim the squeezer to the proper angle to set the rivets fully. Worked well, and they turned out nicely; there may be a teeny variation from straight here or there, but well within the allowable range, and nothing to worry about. I found that sighting the edges was made a bit frustrating by the blobs of Proseal hanging off–is that a wave in the edge, or is it just a lumpy blob of sealant?

Following dinner, I jumped on the trim tab, which went together pretty quickly. Got the hinge drilled to the tab & trimmed to length, the ends bent up, match drilled the holes, and shaped the control horn. Should be able to prep & get some primer on it tomorrow. Need to pick up some epoxy to bond in the foam riblets.

Bought a couple 8′ pieces of aluminum angle from Menards the other day, so I got the trailing edges sealed.  Using the severely expired Proseal (exp. 11/07) might not have been the best idea…I had kept the tube in the fridge since receiving it, and it warmed up just fine, but the mixing head broke off the “dasher rod” when I tried to mix it.  I squeezed it all out onto a piece of cardboard for manual mixing, but it took a very lumpy/ropy consistency when mixed.  It was still quite sticky, so I used it by scraping off the lumps…hopefully it will work.  This validates my decision to NOT order the Proseal for the fuel tanks with the wing kit.  I’m not worried, since Van’s sells expired Proseal at a discount, with a note that it’ll be fine for this sort of application, but not recommended for fuel tanks.

In any event…the empennage trailing edges are clecoed to angle, awaiting the cure of the Proseal before starting to rivet.  I plan to use an angled set in the squeezer to do these, and will experiment on the practice kit first.

Spent the morning & early afternoon in the shop and got quite a bit done. After only one goof-up (I had te re-fabricate the gusset I made yesterday when it slipped while drilling and the hole was too close to the edge of the piece), I was able to finish up the riveting on the left elevator skin, mount the rod ends, and call it good.

I decided that since there was time left in the afternoon, I would start on the trailing edges, so I drilled, dimpled, and countersunk the appropriate pieces for both elevators. (The rudder was done previously.) I need to pick up a couple pieces of angle to use when pro-sealing the edges to keep them straight…I looked into using the tabletop, but it has just a bit of a wave in it. It would probably be okay (less than the 0.1″ noted in the manual), but why not try for perfect?

Began the riveting of the left elevator tonight, with the stiffeners, spar skeleton, counterbalance assembly, and the trim access plate. For some reason, that trim access hole just seems very impressive. Maybe it’s the nutplates, or maybe it’s knowing that a moving part will go in there. Also fabricated the small gusset which goes between the end rib and the rear spar; somehow I missed that earlier. Brushed on a light coat of primer so it’s ready to go for tomorrow’s session. Left to do are the tricky rivets around the inner part of the counterbalance assembly, and then closing it up.

I did take a moment yesterday to clean (acetone) and re-prime a small section on the skin where the primer was blotchy and came off very easily. When getting ready to prime, I noticed what looked like stains from our funky water, so I used some of the etch to clean them off, followed by acetone…I suspect that it didn’t get all the etch cleaned off. in any event, I wiped off the loose primer with acetone, scrubbed with some scotchbrite, cleaned again with acetone, blew dry with the compressor, and re-primed, and it’s in good shape now. Lesson learned…be more careful rinsing the parts, and be sure to get the etch cleaned off well.

Primed the left elevator components (minus trim tab) before the rain started.  It’s taking forever to dry in the near-90% humidity, but I suspect an overnight spell will have it ready for some riveting tomorrow or the next day.  Also pulled the right elevator out of storage and cleaned the cobwebs off.  Will probably remove the blue plastic from all the empennage parts so they don’t develop corrosion, as some have reported can occur.