Matt's RV-7 Project

I ran wires for the autopilot autopilot and got it powered on and talking. Lucky for me it fits perfectly in the panel, so I guess my CAD drawing of the cutout was accurate.

Here's a closer view. Notice that my panel has an on/off switch for the autopilot, so if it tries to kill me I can return the favor.

The autopilot is too deep to use a standard D-sub connector (it hits the subpanel) but I was able to make it work by using an expensive right-angle D-sub backshell. It was pretty dicey to get the two halves of the backshell clamped together and the thumbscrews inserted without pinching any wires.

All the fuselage wiring for the autopilot is in, but I haven't installed the connector for the pitch servo or the wing root disconnect for the roll servo yet. I'll do that later.

I've been thinking for a while that I'd eventually like to add rudder trim to this airplane, similar to what other builders have done. I may not actually get around to putting a servo in the rudder any time soon, but if I at least install a switch in the panel and run wires back to the tail it will be a lot easier to retrofit later. If I decide to fly the airplane before installing rudder trim, I'll just placard the switch "inop".

I had a Ray Allen rocker switch left over from a previous project, so I located it and dusted it off. There's a spot just below the radio stack, next to the flap switch, that's just about perfect for it.

I measured and marked cut lines, then used a unibit to hog out most of the material:

Then I used a hacksaw blade (held with leather gloves) to make a roughly square hole:

Then I filed away at the edges until the switch fit through the hole. It's not the kind of perfect cutout that I'd have had if I'd thought to include it when the panel was being CNC cut, but the switch bezel hides all that.

Test fitting in the panel:

To allow me to adjust the sensitivity of the rudder trim, I attached a Ray Allen servo speed control to the lower flange of the center subpanel rib.

Here's another view. I'm sure this thing is nothing but an LM317 and a trimpot, but it's easier to buy one of these than to roll my own.

Anyway, at least I can run all the wiring now and make the airplane ready for a future upgrade. Maybe some winter when it's too cold to fly I'll build a new rudder with a fancy trim tab… it would be nice to have a choice between building and flying! The airplane that I'm sort-of partners in has been down for maintenance since March, and of course the RV-7 is still in pieces in my garage.

I found the screws and lockwashers for the Garmin connector backshells, so I was able to get the audio panel connectors installed:

I'm not really happy with the way the wire bundles are pressed up against the reinforcing angle along the top of the radio stack. The wires are wrapped in silicone tape, and they're not free to move around very much, but then again that corner is sharp. I may drill out those rivets and replace the extruded angle with something that has a rounded edge instead. Once the top skin is riveted on, it's going to be very tough to get to these wires, so I want to eliminate any potential sources of future trouble.

Meanwhile, I installed an antenna splitter (aka quadraplexer) on the subpanel rib close to the back of the radio stack. This is an expensive little doodad that divides the output from the single antenna into separate nav and glideslope inputs for each 430.

I ran coax for both comm antennas, the nav antenna, and the marker beacon antenna, plus four short lengths from the radios to the nav antenna splitter. These wire bundles get big in a hurry when you start adding multiple parallel strands of RG-400.

Aerial view of the panel wiring so far:

Comm 1, nav, and marker antenna coax go down the left side of the floor, and comm 2 is the right. Not installed yet are the coax runs for the GPS (x3), transponder, and XM antennas. I'm waiting to run those until I get all the antennas installed in the aft fuselage, and until I have a solid plan for wire routing to avoid using up all the space in my spar bushings before I'm done running the small-gauge wires.

At the wing roots, the comm antenna coax is tied to the rib to keep it out of the way of the aileron pushrods, and the nav and marker antenna cables exit the fuselage through a snap bushing.

Eventually there'll be connectors for the wing wiring here – meanwhile I'm just leaving the fuselage wires extra long.

House chores on Sunday, no airplane work…

At long last, I am finished wiring the radio stack. I removed the backplates from the radio trays and bolted on the connectors and their backshells. Here's an inside view of one of the 430 backplates with connectors attached:

Here it is from the other side… the wire bundles are wrapped with silicone fusion tape to cushion them against the strain relief clamps.

The shield grounds all go to a card-edge connector that mates to a rib sticking out from the backplate casting. Seems to work okay, although it causes all the separate wire bundles to become inextricably tied together. A few product generations later, they switched to a different method of terminating shield grounds that I like better.

One of the audio panel connectors contains a "config module" (the small green thing in this photo) which is a little serial EEPROM that stores airframe-specific configuration data. The idea is that any changes you make to your audio panel configuration stay with the airplane, not the audio panel, so you don't have to re-configure things if you have to replace the unit. Since my overly fancy microprocessor-controlled audio panel requires you to connect a laptop to it to adjust certain settings, this may come in handy in the future.

The config module sits in a little pocket in the cast-aluminum connector housing:

I got the connectors and backplates for both 430s and the transponder installed, but I am stuck on the audio panel connectors for want of a handful of screws. I bet they're on my desk at work. Here you see the new style of connector backshell – it has threaded holes to which you attach the shield ground wires via ring terminals. This is a nicer way to do it, since multiple connectors on the same device can remain separate without their shield grounds getting all balled up.