Saturday, December 31, 2011

The relay race

On the plus side, I have a borrowed engine hoist in the shop, delivered courtesy of a fine gentlemen who lives in the next county over. Brought it right to the hangar door, by golly, and unloaded it too. You meet the nicest class of people in the flying game... usually. More on that later.

The hoist is on loan for an indefinite period, but as it is a vital tool in the donor's work, I don't want to have it around any longer than I need to. Unfortunately, I'm still not sure how long I'll need it given the two steps forward, two steps back nature of the jobs I'm slogging through.

You may remember that I was in need of an adapter plate to adjust for a change in the spacing in the firewall holes that the starter relay mounts to, and after (not) convincing the doubtful Mr. Ken S. at Van's that I had not, in fact, already received said adapter in the past, one was sent.

I waited more than a week for its arrival.

In the intervening days I heard from a number of other builders that told me that they too had to run the Gauntlet of Cynical Doubt in order to acquire the part, and that the part in question presented other difficulties in the area of fit. In other words, "I have the new version firewall also and had the same issue. The Vans transfer plate pushed the starter contactor into the rudder support bracket, about a 1/4 inch lower then the position in your photo."

Well, the part arrived via US Turtle Express this afternoon and, with the clock ticking on the hoist rental, I wasted no time in getting out to the hangar to get it installed. And sure enough, the positioning of it put the relay solidly in metal-to-metal contact with the brace beneath it:


A situation like this could lead to no end of future problems such as electrical noise being generated as the relay rubs against the brace or damage to the relay as it wears its protective case away against the more rigid brace. The solution to this problem was the same as the solution to nearly all of life's little problems: if duct tape can't fix it, a Dremel tool can.


Now there's a nice little gap where there used to be inappropriate contact.


With the relay mounted, I was able to move onto the next step: remove the panel cover. Not easy, that, since it was still covered with obdurate duct tape that insisted on leaving a nasty residue when removed, right up until the moment when Cadillac Pete, expert in all manner of adhesive recalcitrance problems, suggested heating the tape with a heat gun. That worked like a charm! Once removed, I was unsure where to store the screws so that (brace yourselves!) I wouldn't lose them. Pete had another fine suggestion: take a picture of where they're stowed.


The reason that the cover had to be removed is that Van's assumes that the avionics are installed by this time. However, as we're still firmly stuck in Skyview purgatory, no avionics have been installed in my, or anyone else's for that matter, RV-12 since spring of last year. Still, there's nothing stopping me from getting the wiring harnesses in place.


Then it was time to install the fuel pressure sensor. See if you can find the problem here:


That's right: I have the plans and hardware for the new Skyview sensor, but when it comes to the sensor itself, I have the older model. That will require another call to Van's to rectify. I swear, if that guy accuses me of losing the new one that they already sent me....

I decided that I'd move onto the next step which was to draw another set of wires through the firewall. These wires were supposed to go through a pair of cushion clamps mounted to the engine mount standoff. The only problem was that I did not have those clamps installed, nor could I find the step in the plans where I was supposed to install them.


As I was puzzling over that, Pete stepped to the hangar door to see what type of aircraft was taxiing by. It turned out to be the big Piper Aztec that hangars next door to me. That was bad news as my hangar door was hanging wide open and the pilot of that particular airplane uses a lot of throttle on the left engine to get the plane to turn in front of his hangar. This causes a hurricane of sand, gravel, dead worms, and anything else laying on the taxiway to blow into my hangar. It also causes everything in my hangar to blow around. I ran to the door to bring it down. As it was just five seconds from being closed, I thought I had it made.

Wrong.

Too impatient to wait those last five seconds, Hurricane Jack(ass) hit us with nearly full throttle.

I was not happy.

I figured that was as good of a time as any to call it a day.

Not that it will do any good, but on my way out I reminded Windy that the way he makes his turn "makes a hell of a mess in my hangar, and if he could just wait a few seconds for me to finish getting the door down..."

I didn't bother mentioning how little I enjoy it when he sandblasts my car; I suspect he knows and simply doesn't care.

Usually you meet the nicest class of people in the flying game...

Friday, December 30, 2011

I HATE banjos...

I didn't used to, though. Dueling Banjos? Loved it. That was then, this is now - I'll never be able to listen to it again.

Why? Well, it's all because someone, possibly at Rotax, possibly at Van's, decided that the bolts that act as some kind of fuel line crossing guard look like banjos and should be referred to as Banjo bolts. That in and of itself would not be sufficient to induce my ire; no, I find them rebarbative solely because of the nature of the work on them that I did today.

I'm getting ahead of the narrative, though, so let's start at the beginning. I've spent the last couple of days making repeated trips to the hangar and to Harbor Freight (The Home of Un-Pawnable Tools*) in order to gather up the tools required for the installation of the carburetor drip trays. These seemed easy enough but I underestimated the dependency that I have on the tools that are normally available to me at the hangar, and just how many different metric sized tools that would be required.

Finally ready to go, I started on them this morning after one final trip to the hangar for a #40 drill bit (the pre-drilled rivet holes on the trays didn't align very well at all) and a handful of clecos. As it turns out, I grabbed the wrong size clecos, but fortunately I didn't really need them. The trays get four AN4263-3 rivets each, but none of that size are supplied with the engine kit. AN4263-3.5 worked fine. The trays also require a little filling in the corners to prevent any fuel that may drip onto them from leaking out - that was easily accomplished.

That was the easy part. The next steps involve removing the carburetors (one at a time; I started with the left side) and the rubber flanges that hold them onto the engine. This is all started by removing a spring that apparently keeps the carbs from moving around too much given they they are housed in flexible rubber flanges. It was somewhat hard to get the spring off of the little clamp it was attached to, but I consoled myself with the hard-earned knowledge that it would be even harder to put back on. That's just the way these things work, I've learned.


The next step is easy (no, really!) - it's just a phillips head screw that needs to be loosened. You can (and should) loosen it up quite a bit - the screw is captive in the clamp so you don't have to worry about it coming all the way out.


Once the screw/clamp is loose, it's just a matter of rotating and rocking the carb around to get it loose enough to pull out of the flange.


Next, Van's suggests removing the retaining nut behind the cushion clamp that holds the fuel line. If this looks impossible, that would be because it is impossible.


I found it easier to just wiggle a wrench (11mm) back in there to get as much grip on it as possible and use a 14mm socket to remove the bolt. The bolt comes out in the next step anyway, so you might as well just do it now.


When the flange is off, check to make sure that the O-rings are still in place on the engine and in the flange.



The drip trays fit right in. Replacing the clamp was easy; I just used the indentations in the sides where the bolts pass through to align the clamp appropriately.


It all went back together pretty much in way it came apart: the nut was hard to get at and the spring was a real stretch (heh!) to get back on. And, believe it or not, the other side was actually harder. The big ignition module black box gets in the way. Still, I think it was an hour or so to finish both sides.

Then it was on to the banjo bolts. The goal here is to remove the thingy on the top with the black rubber cover over it, remove some kind of restrictor thingy inside the top bolt, and turn the whole remaining mess upside down. Of the three tasks, it is the third that set me against banjos forever.


The assembly is mounted on an aluminum tube, so any torque applied to the banjo bolts needs a countervailing torque applied too in order to keep from bending/crimping/cracking the aluminum tube. So, armed with two metric wrenches working in opposition, I loosened the first bolt.


I removed the black-covered nipple thingy and set it aside. I finger tightened the bolt back in just to keep track of it and the two washers that had sandwiched the nipple thingy.


The bottom bolt holds two hoses. Here's one of them:


The other hose supported by the bottom bolt doesn't move very much - it's still staying in position under the banjo assembly. There is an allen head bolt on top that gets loosened in order to rotate the assembly upside-down. I loosened it and turned everything over, trying to keep it in the same relative position on the aluminum tube.


Before putting it all back together, I had to take the top bolt back out and remove the little restrictor device.



It was in there pretty tight, so I had to hold the bolt with a wrench while I used a small screwdriver to remove the brass restrictor.


When I tried to put it all back together, I followed Van's order of business, which was to tighten the allen head bolt first. That didn't work out. Van's kind of subtly warns that the fuel lines might not line up very well anymore; this, as it turns out, isn't a possibility; it is a certainty. As such, I found it worked better to leave the allen head bolt loose while trying to get the fuel lines back in place. Because...

Remember how I said that one of the hoses doesn't move very far away from where it was when the bottom bolt was removed? Well, that's because it doesn't want to. It very much doesn't want to. Getting the two hoses lined up with the newly located bolt hole while not dropping any of the washers or cross-threading the bolt was enormously difficult, and it would have been more so had I not been able to move the mounting block easily along the aluminum tube. My advice: get the bolt started, then go back and tighten the allen head bolt. Even done in that order, you'll find it hard to get good tool access to both the allen head and the other bolt. How you're expected to use a torque wrench is beyond me.


The only way I could get the allen head tightened up was to use the allen head socket and a wrench:


It took quite awhile to get it all back together, but it was a perfect opportunity to re-learn a couple of valuable lessons:

- coffee will help you think more clearly, but any benefit from that is lost to nervous, twitchy hands.
- it is better that I write about this stuff than record it on video; the language used to encourage hardware to comply with my wishes would result in an R rating.



* We stopped in at a local pawn shop yesterday on impulse, what with our being somewhat curious as to what we'd find after watching numerous episodes of Pawn Stars on the History Channel (oddly enough, unless you consider that a lot of the items they get are of historical significance). While we were there, a guy came in wanting to pawn a bunch of Harbor Freight tools and the pawn guy didn't even want to look at them.

"We don't buy those," he said.

Thursday, December 29, 2011

Two steps forward, one step back

It's a truism in building an airplane that the harder something is to fit/fabricate/install, the sooner it will have to be removed. Normally that's quite a downer, but in the case of the cooling shroud it wasn't so bad since I needed to remove it and patch it. Because of that, "Now remove the cooling shroud and set it aside" was just fine by me, albeit while being fully aware that there would be no immediate setting aside.

The patching job was straightforward and, other than the typical hassles that come with messing around with smelly, sticky resin and fiberglass cloth that's ever eager to shed threads, it went well. It required a half dozen more round trips on and off of the engine to get it trimmed properly, but the result looks okay.


The reasons that the shroud needed to be temporarily removed were primarily to mark its location on the top of the engine to show where the RTV (high temp tolerent goop to seal the edge of the shroud to the engine surface) needed to go, and to make plenty of room for the installation of the engine mount that will mate the engine to the firewall.

For that mount mounting to happen, a few other things had to be moved out of the way, starting with the four coolant hoses. The clown at Van's that was responsible for such comedy gold as "metric Crescent wrench" and some of the meaner little practical jokes (the longerons in general come to mind, metric Crescent wrench included) must have been on vacation the day that these particular instructions were written because the plans clearly warn us that there might just be some coolant left in the hoses from when the factory did the test run of the engine. It must have killed him to let an opportunity like this slide past....

Forewarned, I put a catch tray under the hoses before starting the removal of the clamps.


It's good that I did!


As I worked on these clamps, I soon realized that The Van's Joker hadn't taken a vacation after all. Rather, knowing what a royal pain these clamps are to deal with, he came up with an even more diabolical joke: "Remove the clamps and push them forward off of the hoses to allow the hoses to be removed.". Or something like that; I'm paraphrasing from memory. In any event, those clamps are very difficult to work with, at least without the special tool that surely must exist. Even on the rare occasions when I could get a decent enough grip on one of them to loosen it enough to move it, the area between the end of the hose and an inconvenient bend in the pipe was too small for the clamp to move completely into, leaving a corner of the hose still clamped. In one case, I had to remove the far end of the hose too in order to get enough slack in the hose to remove it from the pump end.

It was later, when I was trying to put the hoses back on, that I realized for sure (I already strongly suspected) that the whole thing would have been much easier if I had slid the clamps back onto the hoses rather than forward onto the pipes.

Ha ha, Van's, you got me on this one!

Literally got me, that is: it is possible (likely, even) that at least once when a clamp snaps loose from the inferior tool in use, the ostensible fancy purpose-built tool probably not available to us mere mortals, it will catch a small fold if your fingertip, resulting in a nice, livid blood blister on your finger.

Here at home, we now call that "the 'OUCH' heard around the block!"

It was the upper left hose (looking at it from behind the engine) that needed both ends removed:


With the hoses finally off, it was back to safer work like removing a big, long, seriously important looking bolt. Van's, perhaps realizing that we wouldn't be in the greatest mood at this point, uncharacteristically tells us why this bolt is being removed: it will be used to hold the engine mount.


Here's a hint: you know something is going to be a frustratingly tight fit when an entire step is devoted to applying a protective layer of tape to the part first:


Further foreshadowing of the difficulty ahead: another entire step that has us removing the powder coat from the areas that will need to fit in a tight space. Although, I suppose, this could also be required to provide a firmer fit or a better electrical ground. Either way, it turns out that my fancy new sander was up to the task.


And, for any remaining optimists, step-by-step instructions for getting the part to fit, complete with a "it will fit if you do it like we tell you to" admonishment.

Like so:

Start at the lower left, fitting the mount around the lower coolant hoses:


Fit the upper right around the ignition module:


And it will now align with the bolt holes.

Unless....

...like me, you found the packing braces to be a good way to support the engine while its sitting on the work bench, so didn't remove them:


Which will now be hard to do without an awkwardly positioned allen head wrench slipping as the bolt breaks loose, leading to a nasty cut on your inside wrist. Trust me on this one.

Once the braces are gone (and the bleeding stopped), the mount does, in fact, slide into place nicely.


There are four bolts that mount the mount to the engine. I remember reading about these bolts a couple of years ago; there was a big problem with people finding that they had worked loose, in many cases with all four having backed significantly out of the holes. That justifiably scared the beans out of those involved; an engine separating in flight is no different than the airframe falling apart in flight: 100% fatal. So, extra caution required here, over and above the routine high degree of care one uses when building a contrivance such as an airplane that one intends to fly in his very own self.

The plans don't include the all-important torque value, instead having you go try to find them yourself in the Rotax documentation. I'll save you the prolonged search: 26 - 30 foot-pounds. The upshot of the lengthy discussion around how to keep these bolts in place came down to two camps: use some blue LocTite, or buy some fancy wedged lock washers. I used LocTite.


I also marked each bolt with torque lock; this will show at a glance whether or not a bolt has moved.


Once the mount is bolted in, the RTV can be squeezed along the marked lines and the cooling shroud can be re-installed for the final time.


Once the RTV has set up, the coolant hoses can be replaced on the pump. Remember, it is easier to put the clamp on the hose, get a good tight grip on it to open it as much as possible, then slide the hose and clamp over the pipe together. Also note that, as always, 'easier' does not mean 'easy.'

Sunday, December 25, 2011

A Fitting Result

It's just after 8:30 am Christmas morning and, very unlike the years when Co-pilot Egg couldn't sleep because she was so wound up over the thought of a passel of gifts under the tree, I am the only one up. Even the dogs who, in the vernacular of the household have "been and had" (been outside, had their morning Milkbones), are curled up and sleeping.

There are no presents under the tree for me, what with my having started utilizing my new K-cup coffee maker more than a month ago, but I still eagerly await the gift opening. I'm not sure why; it's not as if young Egg will be surprised by anything. The only thing we got for her that we thought would be completely unexpected is a new set of beads to hang in front of her bedroom door (her only nod to any form of impending teenage hippy-tude) which she correctly guessed within two seconds of picking up and shaking the wrapped package. Why she can't be that deeply intuitive about something important like, say, statistics is beyond me.

The CEO of Schmetterling Aviation will be visiting today, along with the corporate office's CFO and it is always nice to have progress to show on The Project. I thought having the Shroud of Culin fitted and mounted would be, well, a fitting tribute.

But first.... I had to excise the superfluous grounding lug from the starter motor. I had spent a little time wondering if the removal of the lug was one of those silly weight saving things Van's does now and then, but those are typically called out as being optional. This job was not. I finally realized that the lug has to go because it might interfere with the fit of the motor mount. I'll find out for sure soon enough, but for now it doesn't really matter. Van's says "amputate," I say "Gimme a saw."

Or, as in this case, "Where's that Dremel?"

The cutting disk on the Dremel wasn't beefy enough to do it in one cut; I ended up making multiple passes at it, each pass cutting slightly deeper than the one before. When the trench got too deep for the ever-decreasing diameter of the apparently ablative cutting disk, I resorted to finishing it off with a hacksaw.


The starting step of fitting the Shroud of Culin (just in case you still don't get the pun, read here) is to mark it in a way that will show that the fitting is done.


This process involves two disjointed hands. Well, it requires three, actually, but two is what I have.


The horizontal line on the top is the measurement line. When the shroud is correctly fitted, it will be flush with the top fins of the cylinders. The line inside is traced along the almost invisible scribe lines molded into the shroud. No mention is made of them in the plans, but I thought that it might be helpful to see where they are, just in case they might end up being useful.


I took a swag at where the first cuts should be.


After the first cut, it was apparent that I'd have to do a lot more cutting. In fact, it looked like the scribe lines might be a useful place to start.


I didn't go straight to the scribe lines, but after a few tentative approaches I found that the scribe lines were pretty accurate. There were a few spots where the lines didn't show - for those I used trial and error. Quite a few times, as it turns out. Test positioning the shroud on the engine before I got the trimming done resulted in the shroud getting well and truly stuck in place more than a few times. Eventually, though, I got it to fit.


There was one corner of the raw, unmolested shroud that was just resin with no underlying fiberglass. I managed to cut too deeply into that area (without the cloth, it looked like an area that should be cut out) and it will need to be repaired. The tolerance provided in the plans only allows for a 3/16" gap and I created one that's more like a half inch. I need to mix up another batch of epoxy for another smoothing coat on the canopy, so I'll just add a little piece of fiberglass cloth back over the spot that I cut too deeply. That's the thing about fiberglass: you can always add some back!

Friday, December 23, 2011

A short day

I was skipping ahead when I started into the deconstruction and (ostensible) reconstruction of the engine - there's actually some on-airplane work to do before that. This on-airplane work involves the installation of the electrical and electronic components that support the engine. The complexity of the components runs a century-wide gamut, from the stone-age starter relay all the way up to the solid state manifold pressure sensor for the Dynon computerized engine management display.

Today, though, it matters not how complex the innards of the components are; they all get bolted to the firewall the same way. How hard could that be? [Ominous foreshadowing organ music here...]

The weather was in the low 40's, which is cold in my current pre-acclimated state, but will soon be considered balmy. Once we spend a few months in the 20's, the 40's are going to feel great! I figured I'd have Pete come down and we'd get all of these pieces/parts installed before the weather realizes that it's officially winter now.

The electrical cycle starts at the power source which in this case is an Odyssey gel cell. Put it in the box and bolt it down:


Then the master relay gets bolted in. Two bolts, two washers, done. Nothing instills what will ultimately be proven to be a false sense of confidence like a part that fits right in, no pushing, shoving, trimming, tweaking, or swearing required.

The starter relay looked at first glance to be just as easy, albeit with slightly smaller bolts. Such was not the case: the holes don't align. The hole spacing on the relay is 1 11/16", the hole spacing on the firewall (mine, anyway) firewall is 1 7/16":


This was a real poser, but fortunately Van's tech support was open - I made my first ever call to them to get this resolved. This turns out to be a well-known problem. So well-known, in fact, that the tech person at Van's was shocked, shocked that I was just now tripping over it. Solved years ago, it was, through the expedient of mailing out an adapter plate and a sheet of instructions for everyone that they determined needed it.

I don't think I ever received one.  I told him that, but I get the feeling that he didn't believe me.  He could be right, of course, given my propensity for losing things that I sock away in the interest of "not losing them."  Either way, there's one in the mail now.  One of the things Van's is really good about is sending replacement stuff quickly (and at their cost!) when things like this come up.

Not that there's any rush; the lack of the adapter plate didn't really require that we stop work, but I was already getting cold, having failed to dress appropriately to the season, and was quite happy to call it a day after only an hour. I did go ahead and install the engine ground wires. I was more than a little surprised to see a wire this critical to the operation of, well, just about everything attached with a tiny little bolt and an even tinier nut, hidden well behind what will soon be a large oil tank. I was very surprised that a nutplate wasn't used, given the inaccessibility of the location. I hope I never have to remove this thing! It's possible (or likely, even) that a nutplate wouldn't provide a good enough ground.


Here is the instruction page for the adapter: