Avionics

So, today’s main focus was on preparing for putting this thing into the airplane. That’s not necessarily to imply that I’m really close to that point, but more that I wanted to try and identify any potential issues. First and foremost was the question of whether I could feed this thing through the spar center section, and that was what most of the early-day stuff revolved around.

To start with, I just began pulling the harness up from the table. This entailed temporarily bundling up some of the branches, more than I had before; I used a bunch of small zip ties for this. I also made sure that each branch was labeled so I’d know what it was. In some cases I put this info on individual wire labels, but I wasn’t consistent about it.

The real fun came with the harness portion behind the spar. My general plan of action was to remove from the harness anything that was behind that pass-through, and didn’t go through did pass-through. Basically I wanted to slim down that section as much as possible. Next, I took the branches from behind the spar, and worked them in alongside the main trunk. I used masking tape to wrap each junction so I could restore this stuff down the line, and also extensively wrapped where these wires laid alongside the trunk – especially where the ends of the wires pointed aft. I wanted to provide as smooth of a profile as possible.

Once that was done, I grabbed a snap bushing of the same size used in the spar passthroughs, and started trying to slide it along the bundle, working from the tail forward – basically simulating feeding this thing through the fuselage. Unfortunately, it didn’t go too well – once I hit the ends of some of those runs under the seats, the bushing just wouldn’t move any more. I actually had to cut it off the bundle to get it off.

So, what to do from here? I considered emailing Van’s to see if it was permissible to enlarge the passthrough hole a bit. I figured going up from 3/4” to 7/8” would give me plenty of room. But when I checked Spruce to see if they had the larger snap bushings…nope, the biggest they have are what I already have.

Fortunately I’d already been considering a plan B. There are actually two passthrough holes in close proximity here; my plan had been to use one for the wiring bundle and the other for one of the com antenna coax cables. What I’m going to do instead is split the harness just forward of the spar. There’s a convenient branch for this; all the stuff that goes across the fuselage to the right wing root should work well, since 1) it already has the length to reach over there and 2) all of this stuff has slack built in, and the connecting harnesses can be lengthened if needed to fit (when I make them).

So after redoing my wrapping for this idea, I decided it was time to try putting the harness into the airplane. I decided not to deal with maneuvering it through the spar for this temporary installation; the main thing I wanted to do was verify that the runs behind the panel were long enough. With that confirmed, the plan would be to go ahead and terminate all these on the table.

And so, after a whole lot of wrangling, muttering, and other questionable activities, I had the thing in place:

The harness doesn’t look too great here since it’s not laced up at all, and in general it’s kinda messy. But the branches reach their devices without issue, so this should be good to go. So…I pulled the thing right back out.

One thing I learned while doing this is that my idea if the upper shelf bolts also holding adel clamps to secure the bundle is…not a lot of fun. For one thing, I don’t have adel clamps big enough to wrap the root of this harness. Second, it’s a huge pain trying to start those bolts through a clamp that’s wrapped around a wire (I did this for the center bolt). Pretty sure if I stick with this, I’ll have massive regret the first time I have to pull the shelf. So I think I’m going replace the adel clamps with simple plastic wire guides; since these aren’t intended to hold the wire tight, they should be easier to work with.

Anyway, with the harness back on the bench, it was time to get back to lacing. This is a little bit fun; I really shouldn’t lace some of these junctions, since some runs from the engine compartment won’t exist until much later (like the CPI harnesses or the thermocouples for the EMS). But the lacing needs to proceeds in an orderly fashion to make the bundles nice and neat…so I’m lacing everything, even though I know I’ll have to cut some of these off down the road.

That’s where I finished the night, after lacing up the trunk across the upper shelf. Tomorrow I figure I’ll keep lacing, getting to the service loop and trunk to the panel, and then…I guess it’ll be time to start building up some connectors…

Yeah, I got tired of trying to keep track of what part of harness stuff I was on. Besides, some better variety in these post titles will make it easier if I ever need to go back and refer to anything here.

I picked up this morning with the com radio connector; not super complicated, but it did call for a few window splices (doubled power/ground inputs, plus a combined phone/mic/PTT ground). All in all, though, a pretty uncrowded connector. But that just left me with the granddaddy of them all: the EMS sensor connector.

Even the pile of heat shrink labels I printed for this thing looks impressive:

It was really fun trying to go through all 30 or so wires in this bundle and making really sure I was getting the labels right. The one nice thing is that these were almost entirely one wire-one pin; the only exceptions were the power and ground lines for the trim servo indicators. For the former, I window spliced them into the +5V out wire (which will go FWF for sensors up there), while the latter I was able to just twist the striped ends together and combine into a single pin, since the trim wires are pretty small.

Then it was time to pin everything up, again being really careful and triple-checking each pin before inserting it into the connector. Yeah, that’s a lot of stuff:

After that, I just did two more fairly simple connectors. One was the Skyview network cable that will connect the forward and aft network hubs; the other was the partial Skyview network cable for the autopilot roll servo. The autopilot servos use the same data pairs as the regular Skyview network cables, but with a single higher-current power/ground pair and a disconnect line. The latter three wires are run totally independently, but the network data lines I want to tap into the hub with everything else.

Finally, I got out the second EMS connector – which contains nothing but thermocouple lines for FWF stuff – and laid that out alongside the connector I’d already done, with the actual EMS unit. That allowed me to lay the thermocouple lines along the rest of the harness and get the relevant sections zip-tied together, so they’re secured temporarily. Then, by taking the existing FWF bundle and adding the CPI2 harnesses (which will have to be inserted from the FWF side later on), I verified that a 3/4” hole in the gear tower and firewall would be sufficient to pass this bundle.

That last part was important because I still needed a hole in the forward side of the right gear tower, for this bundle to go through, and I needed it done before I stuffed the harness into the fuselage. This was where I transitioned to preparing the fuselage to receive the harness. In addition to drilling that 3/4” hole in the gear tower and adding a snap bushing, I also added protection on the larger holes that wires needed to pass through.

At the tops of the gear towers, I took some of my surplus nylon pitot tubing, split it down one side, and slipped it in place around the edge of the hole. It doesn’t look amazing or anything, but it gets the job done:

And finally, there was nothing left but to start trying to put the harness into the fuselage. This is a pretty awkward operation; the main thrust of it is feeding the entire harness down through the right gear tower and out the lower exit to the rear. Since there are a couple harness branches to deal with here, it’s just sort of a cumbersome case of feeding in a little, fiddling with the far end, pulling slack, feeding a bit more, and so on.

I got the harness fed in about as far as I think it needs to go, which allowed me to get both the switch console and FWF branches through their respective holes. The next step is to get this thing fed through the spar center section, but I decided that was a job for another day. I’ve had enough wrangling this thing for one night, I think.

Booo work. It was fun having nice long productive days on the plane, but today was back to the normal grind. So now I’ve got to work on keeping the daily habit going.

I started out tonight trying to continue pulling the harness through the fuselage. I fed the long tail through the spar center section, and got most of it through there before realizing that this would be a little more complicated than I thought. Since I had to bundle up all the stuff that branches off this trunk aft of the spar, I’ve got a few different taped-up sections. Most notably one of those sections is to the forward side, and if I just pull the trunk through, that part will be right inside the center section. That’s not going to work too well for removing the tape…

Presumably this section is some wires from aft of the center section pulled forward, so hopefully what I can do is unwrap this section forward of the center section, then finish pulling it through. Might be a bit of a struggle though. That prospect was enough for me to decide to go work on something else and let this stew a little bit.

So I picked something simple instead. I’ve been considering for a while how to label the fuse blocks. There’s nowhere near enough room on the things to put the actual name of things by each slot, so I decided here to just number the slots and then have a key somewhere else that will reference slots to actual names – not too far off from what you find in a typical automotive installation.

I spent a bit of time trying to manually work up labels on the printer that would line up with the slots, then discovered in the manual that it has a setting for printing a series of equally-spaced “module” labels. Perfect for this setup, except that there wasn’t a spacing option that was perfect, so I had to just cope with some misalignment here. But it worked out pretty well once I figured things out and trimmed the printed labels down a bit:

Next, I got to looking at the wiring run for the fuse shelf. I’m a little concerned that even giving what I thought was some extra length on the wire runs, they may not work to go all the way through the routing on the shelf to the various fuse positions. If so, it won’t really be a huge disaster – I can solve that problem just by adding a disconnect plug on the shelf. Honestly that’s something I might want anyway, if I ever want to remove this shelf for service. Without a disconnect I’d have to manually disconnect twenty-some tab terminals plus some cable securing stuff, and then be sure I reconnected it all right later on. I was willing to deal with that, but I’d also be willing to work with a disconnect plug as well.

OK well, before I even consider whether that’ll reach, I really need to start with the stuff I’ve already terminated, which has no real room for adjustment. So I decided to get the upper shelf installed, route the harness where it needs to go, and see that all the connectors I added already will actually work out. And it looks like I’m in pretty good shape; some of the routings are a bit longer or shorter than I’d like in a perfect world, but they’ll work. I’m probably going to want to add some chafing protection in some spots, particularly where the main network trunk cable is going to be touching the edge of the CPI2 controller.

Also, I think I’ve decided that I don’t like my fancy ground bus clamp thing I put together a few days ago for the intercom. I think I’m just going to use a short screw and nut to tie the ring terminals together, and cover the whole thing with heat shrink.

Anyway, here’s a look at some stuff actually hooked up:

I guess tomorrow I’ll see about that bundle that bamboozled me earlier. I figure my next general move should be to verify that the harness stuff at the aft avionics shelf also reaches without issue. Everything else should be relatively straightforward outside of that.

Also, I’m going to rework the conduit runs back under the baggage compartment. I attached the conduit ends with these little zip-tie pads, using 3M VHB tape. For reasons I don’t understand, these pads have stuck just fine in the forward spots, but at the aft ends they’ve come off multiple times. So I think I’m going to see about pop-riveting the zip tie pads to the baggage ribs instead; they have little holes that ought to work for this. That’ll be something I want to fix before pulling the harness through the conduit.

Hard to believe I spent an hour and a half just finishing pulling wire bundles through the spar center section, but such is life. I worked on very carefully unwrapping the taped-up section that hung me up on Monday. There were a couple reasons for this – for one, some of the bundled-up wires have tape flags to identify them, and losing those and mixing things up would make life kinda interesting. For another, some of the spots I tapped were to indicate where branches were supposed to go off. Losing that reference wouldn’t be such a huge deal, but still something I wanted to avoid.

It turned out hat taped-up portion was only harboring three wires, the power/ground/disconnect leads for the autopilot roll servo. As I figured, it was easy to free their loose ends and then pull them through the spar separately. With that part settled, I finally finished pulling the main bundle through the spar, and got the snap bushings (which protect the edges of the passthrough holes) worked into place and installed. I also reinstalled the silicon-tubing protection on the gear tower penetration, which had come off when I was fishing stuff through the first time.

Next up was the secondary bundle that needed to go through the spar. This was the stuff I had to split off previously because everything wrapped together was just too bulky to go through one hole. I’d figured this would be the easy part, since it was a relatively small wire bundle, but…nope, not at all. Working from outside the fuselage, I just couldn’t get into a position to feed them through, so I resigned myself to having to work inside. A bit of a hassle since there are no floors in the fuselage right now, but nothing that can’t be “solved” by the strategic use of scrap lumber:

It worked, but wasn’t exactly comfortable. It did, however, provide an opportunity to take a photo illustrating the glamor of homebuilding:

That photo was taken after I realized that even wiring inside, I couldn’t fish the loose wire bundle through. The wires want to curve since they came off a spool, making it impossible to blindly fish them through two small holes about 6” apart. So I got Josie to bring me a roll of masking tape and, while lying on my wonderful lumber mattress, I carefully bound up the whole bundle. With that done, I was finally able to get it fished through:

I also spent a bit of time untagging the rest of the main bundle here and essentially figuring out all over again where things were supposed to split and branch. Next up, I get to take all the wires I removed for this routing expedition and add them back to the bundle, before moving on to the next big fun of pulling the bundle through the conduit to the tail. Except first, I still have to do my new riveted anchors to the baggage ribs.

Realistically, though, I think before I do that I’ll finish at least the rough routing under the seat floors. I need to make a couple holes in the center seat ribs for a small run over to the left side of the aircraft; that branch will serve few different things: the rear stick wiring, the roll trim servo, and the left wing disconnects. I figure if I can get some organization done in this area, it’ll maybe simplify and clarify what needs to be pulled through the conduit.

I’ve also been thinking through where to go once I get the main thrust of this wiring stuff done. My engine is still very delayed due to supply chain issues that Titan is having, but after talking with some other builders it sounds like it makes sense to work on some of the upper fuselage stuff ahead of hanging the engine anyway. So maybe from here, I can get the top skins at least clecoed in place, and then start working on the canopy. I figure I’ll also reattach the empennage along the way, and eventually transition into a life of fiberglass work. I guess I also need to decide at what point I should get this thing standing on the actual gear legs.

Not a ton to report tonight. I took care of securing the aft ends of the conduit runs to the baggage ribs, so those should be squared away now. Then I looked at allowing for routing wiring across the fuselage under  the seat floors; this is needed to get my left wing connections in place. I just drilled 3/4” hols through each of the indoor seat ribs and added snap bushings.

At that point, I figured it was time to go ahead and get the wire bundle routed over to the left side. So I started lacing it up starting at the spar, got it routed through the first snap bushing, continued lacing the part between the ribs, right up to where the rear stick run needs to split off. Then I realized my mistake – this run isn’t complete. These were a couple of the wire runs I removed to allow for routing through the spar…so I needed to add the left nav/strobe and flap control wiring. Which means I’ll have to remove and redo all this lacing. Not a huge deal, but still annoying.

I think what I’ll do for this next bit is go ahead and terminate the shielded wire runs from the nav/strobe controller board. All three of those were among the stuff I temporarily removed, so I can do this work on the table instead of inside the fuselage. Then, I figure with those bundles already terminated, working out the rest of their routing will be a bit easier.

But hey, here’s a look at the bundle I routed across the fuselage before realizing I’d jumped the gun:

As mentioned last time, after realizing I’d forgotten to re-add some local wire runs, I decided to work on the nav/strobe controller plug outside the airplane, since the three output lines were all local. These are shielded wire runs to the wingtip and tail lights, so I got to work with the solder sleeves and shield terminations and such. Then I just had to terminate the ground and two switched power wires (for activating the nav and strobe functions) inside the fuselage, and finish up assembling the connector.

I was a little worried about whether this short branch might interfere with the right passenger footwell, but after a bit of thinking, it should be positioned well by the main trunk once that gets routed into the conduit:

Next up, I worked on redoing the branch that crosses over through the seat ribs, the one I laced up last time out before realizing I forgot some stuff. I got to remove the old lacing and rework this one, but once that was done I got the pertinent part routed through the second seat rib:

The harness splits here into two branches: the part that doesn’t go through the second rib is the branch to the rear control stick. The rest is stuff in the left bay: the roll trim servo, flap control relay deck, and the left wing connections.

One thing I’m realizing is that I probably need to clean out the fuselage and get it back on the rotisserie. Leaning over the side and doing this work isn’t really a whole lot of fun, and there’s plenty more termination and lacing to be done. I think from here I’m going to work in lacing up the right wing connection branch, and once that’s started, I should be ready to bundle up the remainder of the main trunk and get it pulled through the conduit.

Picked up tonight with some more cable lacing. This time I completed adding some of the previously orphaned wire runs to other bundles, and put a few laces on the stub for the right wing. At that point, I decided it was time to address the big fun of running the trunk through the conduit.

I spent some time looking at the length of the trunk, ensuring that every wire ends was wrapped with tape so there’d be nothing to hang up inside the conduit. In a few places some of the wrapping I’d put on previously needed to be removed and tweaked a bit. The one part I was a bit concerned about was where the trunk splits at the rear avionics shelf; I had those two branches taped off individually for organizational purposes, but that meant that when combined, I’d have a fatter bundle than I could otherwise get away with. But I liked this better than having to carefully re-separate those two bundles after pulling.

Just for more fun, as I was attaching my pull string to the end of the harness at the forward end, I pulled the aft end of the string into the conduit. I still haven’t tied those off at the aft end, a decision that finally came back to bite me. Still, it wasn’t a huge deal to fix – just like fishing the string through the first time, I pulled it out, tied the end to a cotton ball, and blasted it back down the tube with compressed air.

Pulling the harness itself was surprisingly easy. Things got a bit stubborn when the fattest part of the trunk was going through the conduit, and it took a decent bit of tugging, but overall it worked well. The only real casualty was that one of my retainer pads up front pulled loose of the 3M tape it was stuck to. That should be easy to fix, though, and honestly I could probably do without that particular pad if I wanted to.

In any case, everything’s looking a lot more tidy behind the spar now. Getting the trunk routed shifted the wingtip light controller branch over just like I thought it would, so it’ll be out of the way of the footwell. All I have left to do in this area now is to add the terminations for the wing connections: