Wednesday, July 21, 2010

I knew this would happen

Remember how I needed to order the fuel nipple that would pass fuel through the firewall and into the back of the gascolator because I had failed to do a complete enough inventory? Ans remember how I was missing two of the 3/8" sleeves to hold nuts onto flared 3/8" tube? Well, I knew it would turn out that these parts were, in fact, included in the kit and I had done all of that ordering of replacement bits and pieces for naught.  Well, I was right, and I was wrong. I found the sleeves.

The nipple has not yet been found, but its replacement finally arrived from Van's. You may also remember that I had attempted to order the nipple from Aircraft Spruce, but ended up getting the wrong size. Why, then, you ask, did you not just order it from Van's in the first place? Well, consider this: shipping from Aircraft Spruce was $1.28. Shipping from Van's was $8.50. They annoy me this way. Van's typically provides top-notch customer service, but they have yet to provide customer-friendly tools to allow the purchaser to make an informed decision regarding shipping method. In my opinion, that is something they could and should fix.

Do I seem a little crabby tonight? Eh, the paying job has a way of wearing me down towards the middle of the week. Some days go smoothly, others are pockmarked with unwanted interruptions, meetings that inevitably dump ever more work onto my crowded platter, and a crappy commute both ways. Two out of three, today. One bright point though was my sharing of an original, piquant, and pithy insight in one of the meetings: "The fact that I have a hidden agenda may make me duplicitous, but it does not follow that it makes me wrong." I don't think it worked, though. I ended up with the additional work anyway.

Nothing more relaxing than twisting a nipple, I figure, so I treated myself to a little shop time. You may recall that the hole that I drilled for the gascolator looked like it might have been a little off-center. Well, it was. It only took a little filing to get it rounded out to where it needed to be. Two bolts later, the gascolator was attached.

That was the final step in the outgoing fuel system. Now for the return line! The return line starts (well, the fabrication of it starts) at the fuel tank, follows the supply line around a little bit, and branches away to feed down to the bulkhead fitting that I enjoyed installing last Saturday. Because it is only 1/4" diameter tubing, it's a little tricky to get it to follow the 3/8" diameter around turns because the turn radius is a bit tighter. Or so it seemed to me; it's just as likely that I screwed up the math again. Either way, I found that the best thing to do was to be very careful to get the separation distance correct where the two tubes make their vertical climb up to where the fuel tank will be and just eyeball the rest. The two tubes will be clamped together for part of their run and I found it quite helpful to clamp them together early in the process to help maintain the required 1 5/8" separation.

The clamps are quite simple: they're just hose clamp and rubber tube.

I used the scrap of paper with the two Sharpie(tm) marks to set the spacing between the vertical tubes and used the first clamp to hold them in place. Then I just made the bends needed to shape the return tube to the path of the supply line. The hardest part was figuring out the bend required to angle the return line down to attach to the bulkhead fitting. I gave myself a few extra inches of tube to work with, and through a process of trial and error that saw the paired tubes in and out of the fuselage a dozen times, I finally got the length right, the tube flared (with the nut and sleeve on it!!), and the assembly installed for what will hopefully be the last time.

And now it's time for dinner!

Saturday, July 17, 2010


It's beginning to dawn on me that there is a reason that people buy airplanes already built. It's just for the fuel systems, I think.

Having finished the rudder pedals, it was back to the fuel system. But a little housekeeping would be needed first. It's been awhile since I've straightened up and it was getting hard to find things that I had just set down moments before.

It's a cluttery process, this airplane building. Every tool has its place, but it seldom visits it. After a little rearranging, I was ready to start.

I still don't have the nipple for the gascolator, but I did somehow manage to order the correct sleeves so I could finish the fuel supply line that runs from the fuel tank to the fuel pump. I had already made the first 90 degree bend in that line, but progress came to a screeching halt when I went to get the nut sleeve from the parts bag. As it turns out, even if the sleeves hadn't been missing, I was in for a rude awakening. You see, in my fervor to get the first 90 degree bend up tight to the end of the tube (as shown in the apparently fictional rendition in the plans) I hadn't put the sleeve that I didn't have on the tube. Had I done so, I would have realized that the flaring tool wouldn't work with the short length of tube left after making the bend. With the sleeve pushed back as far as it would go, the end of the tube fell 1/8" short of where it needed to be for the flaring tool to work. BAM! Two feet of aluminum tube tossed aside as ruined.

Fortunately for me (but not so much for the fellas I was going to donate my extra fuel line to - better order some, Kyle) I had a lot left over from the spare length that I had bought. I rolled out another two feet (plus two inches, just in case), put the nut and sleeve on and made the flare, then did the first 90 degree bend. Which, oddly enough, is the exact order written in the directions. Imagine! Still, my tube bender can't get the bend where Van's wants it, but the extra two inches of tube I had to work with would cover the extra distance the tube would have to run. And the best news is that there would still be enough tube left to give at least a couple of feet to Kyle.

For the second 90 degree bend, I did the measuring to find the location of the 'L' line as specified in the instructions that came with the tube bender. That bend went perfectly. Then there is a slight fifteen degree bend that I had to do by eyeballing - the tube bender instructions are notably mute on the topic of bends less than 45 degrees. That left only one more bend: another 90 degree bend that would turn the line up from the plane of the first three bends to attach to the bottom of the fuel tank. I measured an 'L' line, very, VERY carefully made sure that I was bending in the right direction (it's surprisingly easy to go the wrong way), and made the bend. I set the finished line down on the plans and found the the last bend was about 1/2" out of position. Argh! That's way more than can be waved away, besides which using the extra length of tube to get the bend to the wrong location had eaten up enough tube that the vertical part would be too short to reach the tank.

Remember the old adage "measure twice, cut once?" I've amended it for fuel line fabrication: "Measure twice, bend once, start over."

I grabbed the last of my tubing, made the flare with the nut and sleeve installed, made the first 90 degree bend, measured the 'L' line for the second degree bend, and confidently made the second 90 degree bend in the wrong place. Yep. The wrong place. I got something wrong in either marking the 'L' line or getting it positioned in the tube bender and the distance between the first and second 90 degree bends ended up being too far. Desperate, I decided to try to save the tube by making the fifteen degree bend a 30 degree bend, my thinking being that the vertical 90 degree bend only had to be in the right point in space, no matter how convoluted its path in getting there.

That seemed to work. If I could only make the third 90 degree bend in the right place this time, all would be well. I didn't bother with an 'L' line this time, that method having failed me twice already. I eyeballed the last bend and very, VERY carefully ensured that I was bending in the right direction, drove to KFC for a bucket of chicken to sacrifice at the altar of all deities affiliated with tube bending just to increase my chances of getting it right, and made the bend.


Well, you'll have to take my word for it. All of the stuff described above is hidden under the baggage floor.

Flush with victory, I went on to the next step. With the fuel delivery line completed, I now have to build the fuel return lines. It would seem that not all of the fuel that leaves the tank ends up getting burned, or at least not on that particular trip. Some apparently gets returned to the tank. Not as much as leaves the tank, though; these are 1/4" lines, not the 3/8" of the delivery lines.

Step 1: put a fitting through a bulkhead. A bulkhead that is now accessible only through an inspection port in the belly skin. The belly skin that hasn't always been there. It's the belly skin that I put there. I put it there when I was told to. You know what would have been nice? If I had also been told to put this fitting in before putting the belly skin on. It would have taken two minutes. Instead it took hours.

Putting it through the bulkhead wasn't so bad.

Getting a nut and washer tightened onto the other side was. It ended up taking hours and giving me a painful case of conniptions. Whatever those are. I think it's something my grandparents had, back in the day. We get vaccinated for it now, but even modern medicine has its limits.

The first problem that I ran into was that I needed a way to hold the fitting in place while I wrenched on the nut. I tried having Co-pilot Egg hold it, but she couldn't get enough of a grip on it. Also, when she held it flush the corners of the flats on the nut would hit the belly skin and I wouldn't be able to turn it. I was at a loss. It was time for the last resort, which is to search the forums on the Vans Air Force web site to see if anyone else had run into this problem. Someone had. The rest had been smart enough to put the thing on before riveting on the belly skin.


The trick is to hold the fitting in place with a pair of needle-nosed vice grips clamped to the side of the bulkhead. Then you go to Harbor Freight and buy a set of Crow Foot (I think they're Native American technology, going by the name) wrenches.

With the vice-grips holding the fitting, I was able to reach up into the inspection port with the Crow Foot wrench and tighten the nut. It still wasn't easy, but it was manageable.

All I have to do now is see if those clever Crow Foot indians ever came up with a cure for the conniptions.

Thursday, July 15, 2010

So, where ya been?

If it seems like it's been awhile since I've worked on the plane, well, it has been. It's been one of those weeks. Sick one day, then a day when I was too burned out from the paying job where my already limited patience had been worn down to a paper-thin veneer and the miserable accident-strewn ordeal getting home in a deluge, one evening spent in front of the computer ruminating (yes, ruminating! Can ya believe it??) about my idea for next year's personal vacation (in lieu of a third consecutive Oshkosh), and the day that I decided to brave the shop and put the replacement nipple fitting from Aircraft Spruce on the gascolator? Wrong size. Got fed up, ordered another, this time from Van's, and called it a day. If the one from Van's is also the wrong size, then they printed the wrong part number of the plans. Cross your fingers, because I'm gonna be some kind of irate if that happens!

Tonight, though, all conditions were favorable. I haven't received the part from Van's, but I did still have a little more work to do on the brake lines. There were two sets left to do: the pair that run from the left side master cylinders down to the wheel brakes, and the pair that run from the right side cylinders up to the fluid reservoir. I chose the long pair to do first because they would be the bigger challenge. Better to get them while fresh, I figured. These lines would be a little trickier since two of the nuts would have to be installed with the lines already installed in the plane. The order of business would be measure and cut the lines, install a nut on each, feed the lines through three bulkheads from the back of the fuselage to the front, install the nuts on the front of the lines, restrain the lines with an Adel clamp, and screw the nuts onto the fittings on the master cylinders.

The measuring and cutting were easy, as was the installation of the nuts. The feeding of the lines through the bulkhead was easy once I had decoded the correct holes to feed them through from the "looks backwards to me" drawing in the plans. That left the nuts and the Adel clamp to be installed. The nuts were only tricky because they had to be done in the more limited confines of the fuselage.

Each of the lines needed to be measured to the point where they would be clamped in place. 

I've always hated Adel clamps, but this one at least had the benefit of being easy to get at. It will not always be thus.

Screwing the nuts onto the master cylinders was trivially easy. So was the fabrication and installation of the other two lines.

Friday, July 9, 2010

A Fitting End

Well, no, not really "a fitting end." More accurately, "fitting the end fittings."

The idea behind having a brake master cylinder and brake pedal on all four of the rudder pedals is to allow the brakes to be used from either seat. This is useful for instructing new pilots (don't you wish you had had a full set of controls when you were teaching your kids to drive? Same idea, higher stakes in the airplane.) The thing is, though, that even with two full sets of brake controls we are left with the inconvenient truth that there is only one set of brakes. To allow two sets of pedals to work with one pair of brakes, the master cylinders need to be tied together somehow. That 'somehow' is done by running lines between the master cylinders to share not only brake fluid but pressure applied thereto. I fabricated those lines tonight.

There are two types of brass fittings in use: 90 degree elbows and 45 degree elbows. The plans are clear about which goes where, but you do have to pay attention because there is an asymmetry here. The left (pilot's side) pair of cylinders use three 90's and a 45, while the right (screaming passenger's side) uses two 90's and two 45's. I did the pilot's side fittings first because really, it is all about me, isn't it?

I have to confess that I don't know the official answer as to how far these things are to be screwed in. The method that I have been using is to screw them in until I feel significant resistance, then go one more full turn.

UPDATE - Nov. 21, 2010: I suggest getting yourself some of this stuff (also available at car parts stores):

I re-did all of my fittings using this stuff and was able to easily get at least one more full turn on each fitting.

Here's the rider's side:

The two 45 degree elbows at the top will eventually pass through the firewall and attach to the brake fluid reservoir. The bottom 90's are the fittings for the crossovers. On both the pilot's and petrified rider's side, the 90's are deliberately canted inwards to allow the stiff plastic lines to pass between the other lines at the top of the cylinders.

So, now for the hard part. The lines need to be cut and have fittings put on. They're very much like the fuel lines in that way. The only real difference is that they are flexible and don't need to have bends put in them. The kit comes with two 15' lengths of line, and each 15' gets cut into three sections. I wrote the measurements of the six lengths into two columns and added them up to make sure I didn't end up running short of line in one of the lengths. You never know, right? There are a 180" in each 15' length, so if I cut them just like I wrote them down, everything should come out just fine. In theory, anyway. Time will be the final arbiter.

Just as with the fuel lines, the order of doing things is critically important. The nuts go on first, then the sleeves, then what passes for the flare. The flare in the fuel lines was created with a special tool. In these lines, it is done by shoving a little brass tube into the end of the line. For something that sounds so simple, it's surprising how difficult it can be. For the first one, anyway. Once I learned the tricks, the remaining three were simple.

So, the first order of business is putting on the sleeves and nuts.

The sleeve is supposed to end up 1/8" from the end of the line. I measured and marked the 1/8" point, slid the sleeve down the tube to the mark, and held it in place while getting the brass flare started. It takes quite a bit of a push to get the brass flare into the tube far enough to hold the sleeve into place, and once you do that's about all the further you can get it to go in no matter how hard you push.

This is where the plans suggest soaking the thing in boiling water for a minute, then pressing the flare the rest of the way in.

It sounded oh so easy. You have to consider, though, that the brass sleeve is going to be hot as blazes and the tube is going to be wet and hard to get a good grip on. After numerous (painful) tries, this was as close as I could get:

That's nowhere near good enough. Brakes are just one of those things that you want to work each and every time, preferably without spraying leaking brake fluid all over the place. I was going to have to figure out a way to get this to work. I boiled it down (heh!!) to two distinct possibilities:

  - if I could get the water hotter, it might loosen things up a bit,
  - and if I could find a way to exert pressure without having to try to get a grip on a wet, slippery plastic tube,

I just might have a chance.

First thing first: making water hotter than normal boiling. Two ways to do that: put it under pressure, or add impurities to the water. First one, hard. Second one, easy. I added a tablespoon of salt to the water. I doubt if it makes a whole lot of difference, but the placebo effect alone made it worth the effort. It worked for Dumbo, after all.

As far as finding a better way to apply the downward pressure, I realize that I could use the nut for that. As long as I didn't let it soak in the super-heated salt water, anyway. I don't know which of those (if not both) did the trick, but the trick was done.

The 45 degree elbow was a little trickier. The sleeve is permanently attached to the nut, so the entire process had to be modified. I had to be a little more careful about getting the sleeve's 1/8" gap from the end of the hose correct, and I had to wear a glove to press down on the nut since I had no choice but to allow the nut get heat soaked too. The concepts still worked, though.

With the fittings on, there's no difficulty whatsoever in getting the lines installed.

I sure wish there was a way to leak check them! They are never going to be this easy to get at again. If I ever have to tighten or replace one of these fittings, it's going to be done by crawling down in there. As I've learned with the RV-6, that's not a fitting end to any story.

Wednesday, July 7, 2010

Go "pedal" your sob story somewhere else, fella

It's kind of hard for me to believe, but I'm actually a little choked up. Tonight I installed the rudder/brake pedals. That doesn't seem overly significant, I suppose, but as I was tightening up the bolts to hold the pedal assembly in place (temporarily - it will all come back out once the brake lines are done to aid accessibility for other stuff yet to go in) I realized that I had just finished installing my first flight controls. It's really starting to become an airplane! And what's more, it's really starting to become my airplane.

On the occasions when I have shared the idea that I'm not sure which airplane I will sell (the RV-6 or this RV-12) when the time comes, I've been told that the decision is already made. I will be too emotionally involved with the RV-12 to sell it, they said. I didn't believe it, thinking for some reason that I'd be able to keep a business-like detachment about the whole thing, but I now know that to not be even remotely possible. If the simple act of bolting a collection of tubes to a frame of aluminum nearly brought me to tears, I don't see how I could ever sell this thing after we have shared a moment as deeply profound as its first flight as an airplane and my first flight in a machine that I assembled by myself. Those, my friends, are emotional bonds that won't be easily broken.

And, ignominiously enough, it all started with grease. The brake torque tubes need to pivot nicely inside the rudder pedals, and for that we need lubricant. Nothing fancy; just plane plain old grease.

Here's a free hint: you don't need to grease the whole length. Fully half of the thing is going to hang out the other side and have a brake pedal bolted to it. Just grease the half closest to the flange.

The pedal part of the rudder pedals then get bolted on. They're just little rubber grippy things to give some traction to your feet. Notice the arrow pointing at what appears to be a missing flange? Well, I didn't. Not until later, anyway.

To be fair, it was enough trouble holding tools and parts in place with grease all over my fingers - it's no wonder that I missed that!

The brake master cylinders go on next. There's a washer that sits between the clevis bolt and the cotter pin that holds it in. It's not a particularly easy washer to get into place.

I got those tweezers at the dental tools section of Harbor Freight. One really has to wonder about a dentist that buys his tools at Harbor Freight. [shudder] I got as far as the second cylinder before I realized I had a little problem:

Over the months that I've been working on the plane, I've developed a highly tuned sensitivity to things that don't look right, no matter how subtle the hint may be. In this case, I don't think an inexperienced builder would have seen anything wrong, but my finely tuned senses immediately detected the whiff of a problem here. Something just didn't look right. But what could it be???

Quick fix, although it might have been easier if my hands weren't so greasy and I hadn't gotten that bolt so tight.

Then there was more dental work required to bend the cotter pins.

Finally, time to install it all into the fuselage! Which, I'm here to tell you, is a royal pain. You see, the plastic blocks holding the whole thing together were only in place temporarily. They have to be removed to put the pedals in the plane. Note the relationship between "holding the whole thing together" and "removed." What happens when you remove the blocks should be pretty obvious: everything falls apart. You then get to try to put it all back together under the shelf of the firewall. Tricky!

They sure do look good, though!

Taking a "brake" from the fuel system

Well, a "break," too. A break to start on the brakes, if you will. It's 90+ degrees outside with a humidity level to match, so I need to keep myself busy down in the shop. The fuel system fittings haven't arrived yet, so I thought I'd try to get started on the brake/rudder pedals. For those unfamiliar with the relationship between the two, here's a quick synopsis.

An aircraft is normally steered on the ground by use of the rudder pedals. The rudder pedals are attached via some linking mechanism to the nose wheel. Application of left rudder pedal will cause the aircraft to turn left, and vice-versa for the right. Braking of the aircraft is accomplished with differential braking. This means that the left or right brake can be applied individually, which enables the aircraft to turn about on its axis. In the case of the RV-12, there is no linkage between the rudder pedals and the nose wheel. All ground steering is therefore accomplished with differential braking until such time as sufficient airspeed is attained (typically during the takeoff roll) for the rudder to become effective. The differential braking is accomplished with "toe brakes." The rudder pedals either swivel at their base or have a separate pedal at the top that is activated by the top of the pilot's foot. In the case of the RV-12, each rudder pedal was a separate brake pedal.

The rudder pedals are a push type of control. You push on the right pedal for right rudder, and the left pedal for left rudder. Pushing on a pedal causes a tube to rotate. Flanges on the tube are connected to the rudder with steel cables. The upshot of all of this is that the tube needs something to support it yet allow it to rotate. This "something" is a set of plastic blocks. The first step is to cut them in half. The band saw had no trouble at all getting though them.

Because neither the band saw nor its operator is genetically capable of cutting a perfectly straight line, I marked each block half so that I'd be able to match them up again. Any deformities resulting from my poor directional skills with the saw would net out to a perfectly matched edge. Or so I like to think, anyway.

The two rudder pedal tubes are identical. Each has what the plans call a "short end" and a "long end." I set the two parts on top of each other with their short and long ends together.

Then I simply flipped one over to get the pair of them properly oriented to each other.

The two parts then get temporarily assembled into the plastic blocks. The plans show the need for washers to get the nuts to fit correctly on the bolts, but don't specify what size. I had a bag of the AN960 washers that eventually get used during the final installation lying around, so I just used those.

As it sits on the bench, the rudder pedal assembly is actually upside down. When it eventually gets installed into the airplane, it will get turned over and the blocks will be on the top.

To install the brake pedals, the tubes that support them are slipped through the rudder pedals. It is important to note here  that the "pedals" are actually just horizontal steel tubes. The picture below shows two of the rudder pedals with the brake pedal supports inserted.

The brake pedals (which are also tubes) will pivot on the bare steel parts of the rudder pedal tubes. Trust me, it's all going to make sense soon. The brake pedals are going to be slid onto these tubes and match drilled for a bolt to hold them on. When a pedal is pushed, it will rotate the tube it is bolted to. That will move the flanges  that you can see on the ends of the tubes, between the rudder pedals. The flanges will be attached to brake master cylinders. It's really quite clever.

To lock the pedals into place to aid in the drilling, one of the brake support tubes is inserted through the rudder pedals on the other side

It was a bit of a tight fit. The pedals didn't line up perfectly up & down, so it took a bit of twisting and turning to the brake support through both rudder pedals.

Back to the other side, two brake master cylinders need to be temporarily installed. I marked the 'L' and 'R', but these are parts that I don't think will fit in the wrong way if you lose track of which is which.

Here they are installed:

Now you can start to see how they're going to work. Once the actual brake pedals are bolted onto the bare steel part of the rotating tube, you can see how the rotation of the tube will activate the brake master cylinder.

The idea is for the brake pedals to be perpendicular to the rudder pedal assembly, so I used a square to get them as well lined up as I could.

It's hard to see what's what in that picture (the replacement photographer is much better about making sure my bald(ing) spot is out of the picture, but his overall composition skills are weak)  so I added arrows to show the two parts of the brake pedal. Those are the two points that are used to ensure a close approximation of perpendicularity. I say "approximation" because 1) the rudder pedal tubes block good access for the square to align the inner brake pedal, and 2) because I had a great deal of trouble in keeping the pedals from moving as I clamped them in place. Oh, and 3) I also had trouble keeping them from moving as I was drilling them because my clamps suck. Hopefully the alignment isn't super critical.

The holes get drilled #30, then #12, then 1/4". There is not a single one of those sizes that wasn't an absolute bugger to drill. My little 12 volt lithium drill is wonderful for almost every situation, but drilling through hardened steel is not one of its strong suits. And quite painfully, there are eight holes to be drilled. With three separate sizes for each, it was a good two dozen holes to drill. I was mighty glad when it was done. At one point I got a 1/4" bit well and truly stuck, and I think I came very close to stripping the gears in the drill.

What with these being match drilled holes, the odds are enormous against them ever actually matching. That said, it's prudent to mark each part in order to aid in getting them back into position after removing them for deburring.