Matt's RV-7 Project

Since I am feeling professorial tonight, you get an illustrated tutorial showing my techniques for dealing with shielded aircraft wire. I am by no means an expert, but I won't let that stop me from showing you what works for me.

I buy most of my wire from Aircraft Spruce, since they have a volume discount and they'll put your wire on a spool for you without charging an extra fee. I always write the size of the wire right on the spool in big letters, so I can identify the right spool from across the garage. For example, the this photo shows some two-conductor, 24 AWG shielded wire. Trust me, you'll wish you did this once you accumulate a dozen spools of technically distinct but otherwise similar-looking white wire.

We're going to connect the #2 nav receiver audio output to the audio panel. The shield on this wire will be left floating at the radio end, and at the audio panel end we'll attach the shield to ground via a flying lead we'll be soldering to the shield. For the end where I want to ground the shield, I make a mark about four fingers' worth back from the end:

Gently score the outer insulation all the way around, then flex the wire back and forth until the insulation breaks. Don't try to cut all the way through – you only need to create a weak spot and the bending of the wire will finish the job. My favorite tool for this is the nail scissors on my pocket knife – I find I can control the depth of the cut they make better than with a regular wire stripper or pair of dykes.

Pull the insulation off, exposing the braided shield underneath. At this point it doesn't really matter if you nicked the shield or not, although it's a good habit to practice stripping the insulation off without touching the shield.

Scrunch up the shield so it expands out from the inner wires. You're going to snip away the shield, being careful not to cut the wires inside.

Trim off any stray pieces of the shield that are left sticking out. If you're obsessive like me about labeling all your wires, now is also a good time to slip on your un-shrunk label tube. By the way, the Wikipedia entry on heatshrink tubing is pretty interesting.

Now strip off an additional quarter-inch of insulation. Be very careful not to damage the shield, since we are going to solder another wire to it.

There are two ways to attach a flying lead to the shield. One is to solder it by hand, which works fine and is dirt cheap. The other way is to use solder sleeves, which is my preferred method despite being expensive. I like the kind that comes with a piece of pre-tinned 22 AWG wire already sticking out.

I like to first put the solder sleeve onto the wire and slide it up past where I'll be attaching it to the shield… this lets me smooth out the exposed shield and then slide the sleeve back down over it to keep it laying flat.

I blast the solder sleeve with a heat gun until the solder is melted all the way around, the red indicator dye is mostly boiled off, and the glue at the ends is smooshed out. You have to be somewhat quick about things so you don't scorch any insulation, but with a little practice the solder sleeve is pretty much foolproof.

The shielded wires have pins crimped on them in the normal way, and the flying ground lead gets attached either to a ground block or to the connector backshell. In this photo I haven't yet terminated the shield ground wire, since I will be adding more shielded wires to this connector later and I am planning to crimp multiple ground leads into each ring terminal.

The other end of the wire is much simpler, since we don't need to connect the shield to anything. In these cases I like to strip off three fingers' worth of shield and outer insulation instead of four.

A piece of heatshrink labels the wire and cleans up the cut end of the shield as well.

Done! Now repeat a couple dozen times throughout the whole airplane.

Today I started wiring the radio stack. For starters, here's a reference photo of how the pieces of the D-sub connector backshells for the GNS 430 and GTX 330 fit together; it took me a little while to figure it out from the drawing. And at the risk of biting the hand that feeds, I sure do wish Garmin would have used the same two-piece backshells here that they include with their audio panel. These one-piece jobs make you pass each wire through the backshell before you plug it into the connector, which means the backshell is floating around on the wiring harness the whole time you're trying to build it. It's a pain.

I found myself wishing for an easy way to temporarily pin together a bundle of wires, and in a rare burst of insight I realized that an ordinary clothespin with some self-adhesive weatherstrip on the jaws makes a pretty good little cushioned clamp. Total cost: zero dollars, since I had all the materials laying around the house.

I used the clothespin-clamps today when I was initially stringing wire bundles, then came back later and replaced them with still-temporary-but-not-quite-so-much twist ties. In this photo I'm starting on the wiring for the transponder:

Each radio has numerous power and ground pins, and in most cases you can run multiple wires into a single PIDG terminal on the other end. Here I've combined four 22AWG ground wires for one of the 430s into a single blue (14-16AWG) terminal.

The comm power input on the 430 runs off a 10-amp breaker and takes a doubled-up pair of 18-gauge power and ground wires… wow, that's a lot of current! Since 18AWG wire is way too big to insert in a standard D-sub pin, the 430 install kit includes a small handful of these special crimp sockets with heavy-gauge barrels. Since they stick out from the connector, you have to insulate them – no big deal for me, since I'm using my label printer to make labelled heat-shrink tubes for every one of these wires anyway. By the way, I crimped these suckers with my Daniels crimper – no positioner necessary, I just eyeballed it.

Following an afternoon of wiring, I now have the power and ground wires run for the audio panel, transponder, and one of the 430s. There is still a lot of wiring to be done for audio, data, and antenna wires, but this is enough to turn on three of the boxes. In the photo below you can see that I've left myself generous service loops in case I ever need to work on these connections later on down the road. I also have not bolted the connectors to the radio trays, since I don't want to have to worry about potentially bending one of the pins if I don't have to.

Apply power, and… it's alive! Of course I tested each component individually first, but this is the coolest picture.

According to the meter on my bench power supply, these three boxes running – without transmitting or doing anything strenuous – draw about three amps. That's a bit less than what I had in my power budget.

I would have wired the other 430 as well, but I ran out of both 18 and 22 gauge wire. Initially I thought that the 100-foot spools I bought would be more than enough, but now I see that there is way more wire in this project than I would have guessed. I bet I'll go through a quarter-mile of wire before all is said and done. Time for another visit from the UPS man…

I spent a big chunk of today messing with two little adel clamps that I installed in the center tunnel just forward of the F-705 bulkhead:

What you see above is the finished result, but it took a long time to get there – mostly due to my own stupidity. At first I had the adel clamps attached to the seat ribs with screws and nuts that went through some convenient tooling holes, but then I realized that was dumb: once the floor panels were riveted in place, I wouldn't be able to get a wrench on the nuts. Duh.

So I took them back out, but somewhere during this process I managed to accidentally squash the insulation on a couple of the wires in the right-side bundle. That didn't sit too well with me, and I didn't feel like splicing in new sections of wire, so I ended up removing and replacing about twenty feet of wire. I also had to replace one of the replacement wires after I managed to cut it about three inches too short. So that was pretty frustrating.

Then I riveted in some nutplates for the adel clamps to screw into, which was relatively straightforward except for the usual difficulties that come from trying to work in tight spaces in the bottom of the airplane.

Oh, and the reason for having these two clamps in the tunnel in the first place is that they'll provide a way for wires to cross from one side of the tunnel to the other without interfering with the elevator pushrod. All will be made clear once I start wiring the strobes.

Now that the holiday season is officially over, I have no excuse for not working on the airplane. Tonight after work I managed to find the time and energy to finish the wiring for the electric fuel pump. Here you can see how the power for the fuel pump comes off of the bus 1 fuse block, runs along the throttle support bracket to the fuel pump switch, then goes back up the subpanel and disappears into a snap bushing. The metal plate hanging on the throttle cable below is supposed to be there – ignore that.

On the reverse side of the subpanel, the wire heads upward, snakes through some adel clamps, and passes through the center subpanel rib via another snap bushing. By the way, all the adel clamps in these photos are currently way oversized for ease of running wires – when I'm finished wiring and I know the size of each wire bundle, I'll install correctly-sized clamps.

On the starboard side of the center rib, it continues through some more adel clamps, then takes a dive down through snap bushing in the avionics shelf. I know it seems like a circuitous route, but it's actually more direct than it appears.

From there the wire goes down the firewall, across the floor, and connects to the fuel pump via a Molex connector. A corresponding ground wire then goes back up the firewall to the ground block.

I powered up the bench supply, turned on the master, and briefly flipped on the fuel pump switch just to check the wiring – it ran! I turned it off immediately to keep it from sucking up any fod. I had previously blown out and capped the fuel lines, but given how dirty the floor of my airplane is right now, you can't be too careful.

One fact worth noting is that my fuel pump is connected to bus 1 – the "main bus" in Bob Nuckolls parlance – instead of bus 2, the "endurance bus". I would have liked to put it on bus 2, but since the fuel pump can draw up to 7 amps, I calculated that I just don't have the power budget. This means that if the main alternator fails and I turn off bus 1 to shed load, I won't have an electric fuel pump. However, the electric pump is just a backup to the engine-driven pump – the odds of seeing simultaneous failures of the main alternator and the mechanical fuel pump seem so remote as to not be worth worrying about. Besides, I will still have the option of leaving bus 1 powered up if I'm not worried about maximizing alternator-out run time… i.e. when I'm about to land anyway.

I spent my two days of christmas vacation stringing more wires. Here you can see that most of the panel switches now have their power leads connected:

The black object attached to the center subpanel rib is the flasher for the landing lights. I got it all hooked up save for the wires that go out to the wings.

I got the two voltage regulators mostly connected, and ran a bunch of wires to the annunciator controller:

Pretty soon I'm going to need to run a whole bunch more wires back through the center tunnel, so I decided to remove the sticks and associated control system parts to make it easier to get my hands in there. I ended up cutting the two outboard ribs as you see here, so I could lift the whole stick assembly out as one piece – much easier that way. I'll have to fabricate some more F-716B splice plates, but that should be no big deal. I'm not sure why the plans don't have you cut all four of those center section ribs from the very beginning anyway.