Thursday, April 29, 2010

The Tom Sawyer Technique

Of all of the books that I read under duress back in high school but appreciate having a dim remembrance of today, Tom Sawyer has to be in the top five. Not coincidentally, I had a part in the school play of the same name. It wasn't much of a part, mind you. I didn't have a single line (as I remember it) - my sole function as Dr. Robinson was to be murdered by Injun Joe. At least it was a nice, soft piece of Styrofoam that he hit me with; on a public school budget, there wasn't much money for good special effects. A real rock would have been cheaper.

The reason I bring up Tom Sawyer is that one of the things from the story that has stuck with me the most is the way he tricked the other kids into whitewashing the fence his Aunt Polly wanted painted. You know what's just like that?


As much as I'd like to include installing nutplates in my attempts to fool other people to do work for me, the person has not yet been born that is naive enough to believe that to be any fun at all.

Riveting is much easier, though, because the pneumatic rivet puller is such a cool gadget. PfwuhhhhPOOSH! PfwuhhhhPOOSH! And before you know it, the job is done.

It's sad, in a way, that the only job I can easily foist onto the unsuspecting is one I actually enjoy myself. Life's cruel ironies can be trying at times.

Tonight it was riveting the floors onto the underlying rib structures. As with most large riveting jobs, it's the payoff to a long series of mundane tasks. And, as with all large riveting jobs, it is worth the effort of hauling the parts out to the hangar to use the air riveter. Tonight was no different, and had the added bonus of a new helper: Co-pilot Egg's boyfriend Lucas. I took then out to the hangar, pulled a couple of instructional rivets myself, and turned them loose.

My only mistake was telling Co-pilot Egg that she was pulling the rivets right where my tender behind would someday be sitting. I caught her trying to leave one rivet un-pulled. Man, that would have hurt to sit one!

I had a little more time available to finish up the job, but...


Amazing how quickly my helpers disappeared.

Wednesday, April 28, 2010

My new addiction

I'm going to have to look up the location of the nearest DUA. That's Dremel Users Anonymous, for those of you fortunate enough to have not become addicted to the power one wields when using a cutting wheel spinning faster than a Formula 1 racing engine. I hear there are two types of program available for treatment of the addiction: the normal 12 step process, and for a few bucks more, a variable speed program.

I tried to quit cold turkey tonight. Really, I did! I even picked up the hacksaw and tried to use it, but...

I'm getting ahead of things. The job for tonight was to cut, fit, and install the hinges that will be used to hold the seat backs. These hinges are really nothing more than strips of aluminum with interlocking hoops through which a piece of piano wire gets threaded. Vans airplanes use them for all sorts of things, up to and including holding the fiberglass engine cowls in place. The number of places that they are used as attachments probably rivals the number of locations where they are used to actually, you know, hinge things in the more traditional sense.

The seat back attachments actually do both. Their primary purpose is to hold the seat back in place, but it is convenient that the seat backs will pivot forward a little bit when it comes time to reach behind them for the wing attachment pins. The hinge material comes in various lengths and is simply (well, not so simply as it turns out) cut to the length you need. I needed four 10" lengths. The reason for four rather than two (one for each seat back) is to allow for a little bit of adjustability. Not much, truth be told, but some.

I measured out the 10" lengths and marked a couple of the hoops that were to be removed:

A hole gets drilled in each piece 1/4" in from the horizontal and vertical edges of one side. You might be able to see the dot where I marked one of the locations.

With the measuring and marking done, the band saw would make easy work of cutting the 10" lengths. Or it would have, if I didn't own a 9" band saw.

An inch too short. Again. Story of my life. I'm hear to tell you, it has been my experience in life that size does matter. Not to worry, though, as I still have my trusty hacksaw. But boy oh boy was that taking a long time, and there was that luscious Dremel sitting right there, just calling out for me to pick it up and cut, cut, CUT!

Lord help me, I did.

And I liked it! Is that so wrong? I can quit any time. I swear it!

I drilled the holes and used them to cleco each of the hinges to the cockpit floors. The idea was to then turn the floors over and line the holes up with a line that had been drawn horizontally along the other side of the hinge. Then the existing holes in the floors would be used to match drill the hinges.

It was easy enough to see the line through the holes, but not so easy to figure out a way to secure the position of the hinge while I match drilled it. When you consider that I believe match drilling to be the best way on earth to create two holes that don't line up ever again, you can understand why I wanted to be sure that the hinge wouldn't move while I was drilling it. The best I could do was a big spring clamp.

Between the clamp and a block of scrap wood that I put under the floor to give me something to press against, the drilling went OK.

On the other floor, though...

The clamp wouldn't reach. All I could do was try to hold the hinge in place long enough to get a second hole drilled and clecoed.

I was going to keep going, but the next step was to dimple and install more nutplates.

Yeah. Not tonight. Maybe if you could do that with a Dremel...

I did trial fit the floors to the rib sections I've been building for the last few days, though. Just to see what all of the fuss has been about.

Tuesday, April 27, 2010

In which I get my fill of nutplates

Having had a couple of days to rest up from the four building sessions over the weekend, I was ready to replace the worn down Dremel cutting wheel and remove the piece of metal from the seat rib that had to come out to make an opening for the autopilot servo control arm to fit through. Or something related to the autopilot, anyway. I really have no idea. It's ongoing mysteries like this that keep it interesting.

The Dremel is an absolute marvel for this kind of thing. I'm not sure how I would have done this without it.

At this point, the only thing being installed for the autopilot is a doubler that (I assume) one of the servos will get bolted to. The standard rib wouldn't have been strong enough to support the torque of the big servo motor pushing against the aerodynamic forces acting upon whichever control surface this part of the autopilot ends up controlling. Hence, this relatively beefy doubler. It required a few match-drilled holes first.

There are three holes that the plans insisted be left open, so I stuffed them full of clecos to remind me not to rivet them. That is anything but a foolproof method, of course, because the act of remove-cleco-insert-rivet is fully ingrained into my psyche at this point. Still, I thought I could manage to leave a measly three holes unmolested.

Just to help with the math, the separate page of autopilot bracket instructions called for six holes to be riveted in the case of what they call a PRECONST installation. The other option is POSTCONST; that's for the pitifully burdened people that are trying to install these things into a completed airplane. Not only do they have to drill out some very hard to get at rivets, they get to put in nine rivets for this step. I was happier to only have to deal with the six, thank you very much.

There were another three rivets after those six, but trying to determine that from the plans was a bit of a chore. I think Van's must have had Obscuro the Obfuscating Ostrich come in on Mondays to write some of these directions. Foreshadowing: it got worse. I finally figured out that the three rivets went in the flange of the bracket and the "aft intercostal." What's an intercostal? I have no idea. Ask the darn ostrich.

Here's where Obscuro really got me. See, I'm in the PRECONST clan, so I skipped ahead to Step 3, as directed, completely skipping over Step 2. Well, no, that's not right. I did my Step 2, which was to ignore the POSTCONST Step 2. Which, as it turns out, was not the right thing to do at all. Well, it was the right thing in that it's what I was told to do, but not the right thing in that one of the steps of POSTCONST Step 2 is also necessary for us PRECONST folk. Confused? Here, let me show you.

The part that I was instructed to skip but needed to do had to do with removing a rivet (one that I had just put in two days ago; I think they want me to skip this step out of embarrassment for making me remove something that I wouldn't have put in at all if they had thought to edit the directions on the prior page) in order to swing a nutplate out of the way of a screw that needs to be put in. With the nutplate still in place, that screw ain't going in, PRECONST or POSTCONST.

Of course, what with my having skipped the meatier of the Step 2s, I had no idea why this was the case. I figured I'd just have to remove the nutplate. I went back and read the real Step 2 and realized I only needed to drill out one of the rivets, so there is that. Fifty percent less rivet removal using their method. That ain't nothing.

Unfortunately, the drill bit went through the hole and nutplate as the rivet was drilled out and bent the nutplate down. To get it back where it belonged so I could rivet it back in, I had to clamp it.

I thought it might be a good idea to get the don't-rivet-here clecos out onto the other side of the rib so that I wouldn't end up building so much airplane around them that I couldn't get them out. That was also beneficial in reminding me not to rivet into those holes as I riveted all of the rest of them. Obscuro's drawing showing where rivets needed to go was obscured (or course) on the far side of the ribs and there was no "LP4-3, 12 places" (or whatever - I don't remember exactly how many it was) to tell me how many to expect - just "LP4-3 TYP." That usually means fill every hole you can find with a rivet, but doesn't necessarily countermand a "LEAVE OPEN" directive. Man, this stuff can get confusing!

So what's my beef with nutplates? Well, I thought that I was at a pretty good stopping point, but the next step was simple: put eight nutplates in the brackets that will hold one of the seat belts and rivet those brackets into the three holes that I had been obediently keeping open. How hard could that be?

Normally? Pretty easy. Especially since four of them used flush blind rivets - the easiest kind! Except this time, one of the nails broke off in the rivet puller, leaving an obnoxiously long remainder in the rivet. Obviously too long to just leave in there, but too short for the rivet puller to get a grip on.

That's never happened before!

But good things? They come in three. It was only a couple of rivets later that it happened again!

The third time it happened, I decided to remove the broken nails and call it a night. The only way I could get them out of there was to remove the tip from the rivet puller so that it could get a grip on them and pull them out. Without the tip, though, the head of the rivets extruded themselves up into the rivet puller. When that happened, they were no longer flush and had to be drilled out. When I was putting the replacement rivets in, I found that using the next smaller size tip would pull them without breaking them as the larger tip had, but the nails would jam in the puller. I had to take it apart after each rivet to get the nail out.

And that is how I came to have my absolute fill of nutplates.

I did go ahead and cleco the parts together, though.

Sunday, April 25, 2010

In which I end with an emphatic NO!

Well, I started a new page which I shall call 21-05. There's something you need to know right up front about old 21-05: there's a lot of room for mistakes to be made on this page. All eight of the seat ribs will come into play, but they are divided into four groups of two, each group being subtly different in its mix of belt restraints and nutplates.

Here's some free advice: as each of the pairs get completed, use the Sharpie to mark which step you were on when you put them together. That's how they're going to be referred to when you have to select them in later steps.

The first strange thing I noticed was that those factory-made dimples I was so happy about? They missed a few.

Odd. As with things like this that I've noticed in the past, I'm sure there was a reason for it. I'm less than certain that I will ever know what that reason was. It's not a problem by any stretch - I'm so used to dimpling now that I do it almost robotically.

I struggled with the math again, though. Step 3 has four nutplates to be dimpled, but they then get riveted onto two seat ribs. That math flat doesn't work. I had to assume that I actually needed to do eight of these.

The riveting too has become second nature, although every now and then I run up against something like this that appears to be completely impossible:

And it appears to be impossible because, well, as much as I hate to admit it, it is. A blind rivet gets used here.

A little acronymal advice for you there on the picture. If you don't get it, google it. Although, it may be NSFW. (Google will know that one, too)

On the opposite end of the confusion spectrum, I saw this little tidbit:

Long time readers will remember how put out I was when a similar notice was NOT given when the same type of flush blind rivets were to be used on the horizontal spar box. I only found out about it before making an expensive mistake by seeing it mentioned on someone else's blog. I still wonder why (well, "if" - it's not like I've gone back and checked) Van's hasn't released a revision for that page in the empennage kit.

The parts that were to be countersunk are a pair of braces that will provide additional beefitude to the ribs that hold the mystery mixer box and the other little box that I built.

These braces were interesting because while it was the case that they would only match the holes in the rib in one direction, they would still match if you flipped them over. For a perfectly flat piece, that would make absolutely no difference whatsoever. In this case, though:

They aren't. Perfectly flat, that is. They're bent. Curved. Parabolic, as it were. Is this a problem? I don't know. It is not mentioned at all in the plans. I tossed a coin and decided that I'd install them bowed down into the ribs. It didn't seem to matter.

From the other side:

The plans specifically said to rivet on the brace; they said nothing about riveting the rib to the flanges of the mystery box. So I didn't. I marked them so I remember to keep an eye out for the step during which those rivets get installed. Just in case.

I have to apologize that I can't provide any help regarding how to keep track of the eight ribs and which goes where (or when or why, for that matter) - I puzzled my way through it and ended up with another pair of ribs holding a mystery box. In this case, the mystery box. Yes, it's the magical flaperon mixer!

I had been wondering when the autopilot servo brackets were to be installed. The autopilot kit came out after the RV-12 was in production, and I wasn't sure if the standard plans included instructions regarding when to install the brackets. The brackets and a single page of instructions came with the fuselage kit, though, so I knew it had to be done eventually. I'm keen on getting it done during the build because I've heard that while they can be installed later, it's a PITA. (You'll have to go to for that one. Hint: it's not bread.) In light of that, I was happy to see this:

I was less thrilled about this:

A single-word command: REMOVE.

I replied with a single word myself: NO!

More nutplates on the horizon

I started on the seat ribs today. These are the same things as the baggage ribs in that they will support a floor that in turn supports baggage, although in this case the baggage is me. I wasn't clear from the first step about just how many of these I was supposed to make since the counting scheme when it comes to parts that need to be separated is somewhat inconsistent. As I got further into it, I learned that this is what I needed to assemble:

By the end of the day, I'd also end up being thoroughly confused about 'left' and 'right', something I thought I had mastered years ago. The 'left' part goes to the left of the 'right' part, except when the 'right' part goes to the left of the 'left' part. And vice-versa. In this case, my advice is to try not to think about it and just put the parts where they're shown in the drawing. In fact, to reduce confusion you could just label the parts Ralph and Larry to avoid falling prey to preconceived notions as to positional location based on apparently arbitrary labellings like 'left' and 'right'.

Once you get your head around how many of these to assemble and the idea that you don't really know left from right, you get hit with another puzzler. Now I'm the first to admit that I get spoiled very easily when it comes to time-saving conveniences, and that is certainly the case with the band saw, but I was shocked to see that I was going to have to remove this relatively large chunk of material sans band saw:

Times six eight, mind you, although only four today. The reason the band saw wouldn't work is spelled out for all to see, right there in its name: band saw. Closed loop. No way in. This is only news to anyone that thought 'band' referred somehow to the sound the saw makes when it's running, an idea which would be patently ridiculous if there was no such thing as modern jazz.

I'd have to find an alternative. Quickly discounting any methods lacking the application of a power tool, I arrived at the Dremel tool. Second day in a row!

And there's that bald(ing) spot again! I need to fire that photographer.

The Dremel (well, more accurately, the operator of same) left a very rough edge, but the little ScotchBrite drum took care of it.

Someone at Van's must have realized there'd be some malcontent like me that would grouse about having to do a little more work, so they threw us a bone: factory-dimpled nutplate holes!

Given the easy location for the dimpling of these holes, I can't figure out any other reason for them to have been dimpled at the factory. Still, it's not a trade I would have made. Van's got the easier job out of that deal.

Carefully following the drawings and mentally replacing 'left' and 'right' with 'Larry' and 'Ralph', I clecoed it all together. These tabs are eventually going to support a bolt that will secure the seat belts.

Somewhat out of the ordinary, the holes have to be final drilled for LP4-3 rivets. Usually the rivets will drop right into the holes as-is.

I was a little surprised at the selection of LP4-3 rivets instead of the slightly longer LP4-4 rivets, given that they were going through relatively thick pieces of metal. That said, if there is ever serious stress on these rivets, it will be in shear. In that case, it's the body of the rivet that will take the load, not the manufactured head.

They don't look like much, but all in all it took a couple of hours to get them assembled.

That was the end of page 21-05, and the first step on page 21-05 was.... the installation of nutplates.

Maybe later.

Saturday, April 24, 2010

The what??

I had a little time to head back down to the shop after mowing the lawn (it didn't rain, but there were some light sprinkles) to build the "flaperon mixer arm."

"The what?" you say?

Beats me. I went all the way to the end of the plans that I have and while this thing does get installed, nothing gets attached to it.

Let's think about it, though. First, you may be wondering what a flaperon even is. Well, the word itself is a conglomeration of the words "flap" and "aileron." Most of my readers will know what a flap and an aileron are, but if nothing else, I'm inclusive. I'm reclusive too, but that's not the same thing at all. Anyway, here's a quick paragraph or two from Explaino the Wonder Donkey:

Ailerons are hinged control surfaces attached to the trailing edge of the wing of a fixed-wing aircraft. The ailerons are used to control the aircraft in roll. The two ailerons are typically interconnected so that one goes down when the other goes up: the downgoing aileron increases the lift on its wing while the upgoing aileron reduces the lift on its wing, producing a rolling moment about the aircraft's longitudinal axis. The word aileron is French for "little wing".
I didn't know that last part. And now that I do, I realize that I didn't know because I don't care. On to the flap:

Flaps are hinged surfaces on the trailing edge of the wings of a fixed-wing aircraft. As flaps are extended, the stalling speed of the aircraft is reduced, which means that the aircraft can fly safely at slower speeds (especially during take off and landing). Flaps are also used on the leading edge of the wings of some high-speed jet aircraft, where they may be called Krueger flaps.

Extending flaps increases the camber of the wing airfoil, thus raising the maximum lift coefficient. This increase in maximum lift coefficient allows the aircraft to generate a given amount of lift with a slower speed. Therefore, extending the flaps reduces the stalling speed of the aircraft.

Extending flaps also increases drag. This can be beneficial in the approach and landing phase because it helps to slow the aircraft. Another useful side-effect of flap deployment is a decrease in aircraft pitch angle. This provides the pilot with a greater view over the nose of the aircraft and allows a better view of the runway during approach and landing.
Wow, all those words and Explaino missed one of the most pertinent things: where ailerons move in opposite directions, flaps move concurrently in the same direction. If one flap goes down, so does the other. The same goes for up.

As long as the donkey is on a roll, here's his take on the flaperon:

A flaperon is a type of control surface that combines aspects of both flaps and ailerons. In addition to controlling the roll or bank of an aircraft like conventional ailerons, both flaperons can be lowered together to function much the same as a dedicated set of flaps would. Both ailerons could also be raised, which would give spoilerons.

The pilot has separate controls for ailerons and flaps. A mixer is used to combine the separate pilot input into this single set of control surfaces called flaperons. The use of flaperons instead of separate ailerons and flaps can reduce the weight of an aircraft. The complexity is transferred from having a double set of control surfaces (flaps and ailerons) to the mixer.
I added the emphasis - he's just a donkey, he doesn't know HTML for crying out loud.

My guess, then, is that I just assembled the mixer. I have no idea how it works - all it does is provide a pair of arms that swivel back and forth a few degrees.

But why does the RV-12 have flaperons in the first place? None of the other Van's airplanes do.

I've found that when I come across a "why did they do it this way" question with regards to the design of this airplane, it pays to start looking for an answer in a "follow the money" kind of way, but by replacing "money" with "removable wings."

For example, one would ask if the design element in question benefits the removable wing feature.

In this case, the answer is yes.

One of the most frightening things about removable wings is the fear that you will forget to re-attach the control surfaces when the wings are put back on. To counteract this (and most assuredly to reduce liability), most manufacturers of removable wing aircraft will make the connection of the control surfaces automatic. This is the case with the RV-12. The flaperon has a tab that fits into a slot on the side of the fuselage when the wing is installed. The control stick moves the slot, the slot moves the tab, and the flaperons do what flaperons do. With this type of control linkage, it would be very difficult to provide a separate slot/tab for the flaps and ailerons. With flaperons, only one slot/tab is needed.

That's one slot/tab per side, for you lovers of symmetry and/or pedantry.

With all of that complexity, then, this mixer seems woefully simple.

It's actually pretty easy to put together, too. It starts out just being a simple box.

I thought that this would be a nice break from installing nutplates, but it wasn't to be. There are two that get installed. Then there is a lengthy easter egg hunt for a collection of bolts, washers, nuts, and bushings. The search for the bushings would have been easier if the measurements hadn't been truncated on the inventory sheet.

I'm just being churlish. There are only three aluminum bushings in the entire kit, all three are in the same bag, and the measurements are plainly written on the drawings. It was easy to find the correct pair. I laid everything out prior to assembly, mostly because there are six washers that are the same diameter, but two are thinner than the other four and I didn't want to get those mixed up with each other.

I probably won't know how this thing actually works until next year when I'm doing the finishing kit. I have a couple of pictures of the slot and tab, though, from a nearly finished plane.

Below the flaperon control tab, you can also see the tab that goes into the receptacle I was working on a few days ago.