There is an old saying in the kit aircraft world that when you are 90% done, you only have 90% left to go. The fact is that experienced builders know that when you have something that looks like an airplane, you are rarely more than halfway complete—that last “50%” takes way more time than the first “50%” did. What are we talking about? Wiring, plumbing, hanging an engine, building baffles, and endless fiberglass fitting and finishing—even when you are building an “all-metal” airplane!
And so I have been at the point where our F1 Rocket project has looked much like an airplane for a long time now. I can sit in the cockpit, turn on the power, and play with the avionics and lights. The wings are “finished” in their racks, ready to mount (although we still have to do all the fit-and-finish work to make the fiberglass tips look pretty). If we bolted the prop onto the engine, we could roll it all outside and turn it over (although that is a bad idea without the wings mounted because the wing spar stubs add considerable torsional stiffness to the fuselage). But there is so much left to do that only the experienced builders really understand. So as an example, what have I been working on this week?
Endless fiberglass fitting and finishing to start with. All of the fiberglass empennage tips are riveted in place, but the seams aren’t pretty, and the pulled rivets need to be filled, so we mix up some thick micro, squeegee it on, wait twelve hours, then bring out the sander to take off much of what we put on. Lather, rinse, repeat. If I am working efficiently, the stuff goes on just before I quit for the day, so I can sand first thing in the morning, then get another batch started. I have a love-hate relationship with composites—love what you can do with it, hate the dusty mess.
Once I have put on some goop, I move to something else. We have a primary GPS antenna that feeds the IFR navigator, and all of us who—20 years ago—mounted our good GPS antennas under the fiberglass cowl learned when we upgraded to WAAS that signals kept dropping out because of blockage from the firewall and baffles. So it needs to go on top of the airplane with a clear view of the horizon. On my RV-8, it sits behind the passenger on the turtle deck, under the bubble canopy. I intended to do that on the F1, but when I started looking at the fastback canopy, I didn’t think that it was going to work there, so even though that cable had been run, I “abandoned it in place” and intended to mount the antenna up in front of the windshield (there is virtually no glare shield behind the Plexiglas). But then I finished the canopy skirt and realized that the amount the canopy lifts in back (due to the shape of the track) would allow me to put the antenna there after all. So, I mocked it up out of foam to make sure it would work, and now I am headed into the fabrication phase. Glad I left that cable going back there!
Oh—another antenna thing—the simple GPS antenna for the G3X screen (a backup if you will) is a little thing that isn’t WAAS, so it can go under the cowling, and that meant I needed another tiny cable penetration in the firewall and a mount on the back of the engine baffle to attach the antenna. An hour-long project—done!
One more example of “things that come up.” There has to be a little support fabricated that attaches the front end of the rear canopy track to the bulkhead behind the passenger. A bolt, spacer, and a washer to spread the load. I wanted something bigger than a large-diameter AN-3 washer, so that was another little fabrication exercise. It goes on and on—a zip tie here, an Adel clamp there, a fancy inspection cover on the tail. That last 10% really does take a LOT of time! Even if you have a modern kit with step-by-step instructions, you’re going to find that you retreat to the chair in the corner of the shop trying to figure out exactly how to execute a step—or you’re off on the computer shopping for a few extra bolts from Aircraft Spruce. As you get closer and closer to finishing, little things cause more delays because if you have to wait for a part to arrive, you don’t have much else you can work on.
Finishing a homebuilt aircraft often ends up in that category of you don’t know what you don’t know—even if you are an experienced builder (if you are building something new). In fact, it can get worse for the experienced builder as you try out new techniques and strive to make things better than on your previous build. I find that I more often make prototype parts before making the final part as I strive for “better.”
What do all of these little tasks have in common? They all need to be done before the airplane is finished, for one thing. Another is that predicting them in advance of knowing that you need to do them is difficult, if not impossible. Yes, if you’re building a cookie-cutter plane, you can have a list of tasks done in advance (by people who have built many of them), but if you’re doing anything special—at all—you’ll not know you have to do some things until you come upon them.
One common recommendation—which I endorse—is when you get to this point, start making a list of things that still need to be done, and tape it on the fuselage. Leave lots of empty room at the bottom, because once you get started crossing things off the list, you’ll see more things to add. The list will initially grow as you add more things than you cross off. When the total number of lines that are not yet crossed off begins to shrink, you can begin to believe that you are seeing the light at the end of the tunnel.
And that’s the point of this little essay. As a DAR, I get calls: “Hello, Mr. Dye, I am about to move my RV-7 project to the airport, and I’d like an inspection.” OK, do I tell him that he is at least a year away, or not?
“Hey, not a problem, send me an email with your N-Number and I’ll start a file for you, then keep in touch and let me know how you’re coming. No need to try and set a date right now, we’ll have plenty of time for that in the future. Keep working, and you’ll get it done!” I’m just not going to even give them a guess as to when.
Excellent piece Paul
I am entering this stage now with my RV-14. People are telling me I’ll be flying within a year. I’m not so sure!
Right here at the stage of my 4 build.great article Paul. And I thought I was the only one who built prototype parts.
I usually start by making a cardboard mockup of a part – that’s my “intentional” prototype. If it fits, I move on to making the actual part out of aluminum. Often the first one or two of those become “unintentional” prototypes, before I finally make one that works. Got a whole collection of ’em.
One thing I found helpful with my “things left to do” list (which I kept as a Google doc) was to try my best to put everything in the correct order. That is, if Item A needed to be done before Item B, it was listed first. Having an electronic list was useful because I could insert steps as I found those “wait, I need to do this first!” things. I remember when “run engine” first got added to the list, and when I crossed off the last item before flying.
That document is still in use though — it’s now the keeper for things like little bugs that need to be fixed, additions or changes planned for the next CI, etc.
I kept another list of “things checked”, which was all of those little things I had to look up or measure at some point (What’s the diameter of the engine mount tubes? What size is the nut for the alternator pulley?)
Mike: thanks for this!
I’ve run out of space on the garage whiteboard, and sequencing of tasks makes it look more like spaghetti than serious planning.