« Archives in January, 2011

Wiring and wing conduit

4 hours.

I took a drive down to Earl’s on Hawthorne at the 405 freeway to pick up the fittings I need to plumb the purge line. I wanted to save the 11 bucks for shipping, which is ricockulous considering it’s all of 15 miles away from my house. I’m glad I went. The place is an absolute gold mine of high performance vehicle plumbing. They have everything, including the weird stuff, and best of all, they can custom-make hoses while you wait. Still a pricey run. The stainless steel bulkhead elbow for the firewall was not cheap by any stretch. After I got back, I pulled the wings out of the garage to finish up all the old business, which was fabbing conduit support angles, installing Adel clamps, fitting and plugging in the pitot tube, and wiring the autopilot roll servo. I also spent a good amount of time rubbing off old masking tape glue with MEK. Don’t leave masking tape on anything for too long, especially in a hot, dry place The tape turns brittle and the glue turns solid. On the right wing I got wise and heated up the old tape with a heat gun which let me get most of it off in one piece. The rest came off like the stickers Van’s still puts on every part, which never peel off cleanly. This process took a while and by then I was completely beat. I only got 4 hours of sleep last night for some reason, so I was a zombie all day. I figured I’d nap for an hour and go back too it, but I started reading the last chapter of William Gibson’s “Zero History” and I was asleep before the end of the first page. I woke up two hours later, still exhausted, but I went back out there and wired up the autopilot pitch servo.

Both autopilots got wired using the D-Sub method: Use barrel-crimp d-sub pins and sockets on the wires, then individually shrink-wrap them, then put a big shrink tube around all of them. I did this for a couple of reasons. First, the Molex connectors didn’t have a decent locking mechanism. I’m sure it’s fine, but I didn’t like it. Second, I somehow lost a couple of the pins. Embarrassing, but whatever, this should work fine.

I didn’t get around to plumbing my purge line, and common wisdom seems to be to wait until the engine is hung for this. But the longer I wait, the more stuff is going to be in my way when I do a triple-bypass on the fuel system, so I want to get the tubing roughed in as much as possible.

Heated Pitot done.

6 hours.

I finished up the pitot heat stuff, got the heat module mounted and the wires spliced in. When I get the Adel clamps from spruce, that process is done and done. It doesn’t sound like much, but it took some time. I had to use platenuts because I have to plan on an extremely limited amount of access once the skin is riveted on. And as an added bonus to my day, I found that some condensation or spray from the hose left some kind of deposits all over my left wing. This sucks. I might be able to buff it out, but it looks like the lime or whatever is etching the alclad. Flat Olive Drab paint it is.

I also got the MGL com extender mounted, and got a good start on mounting the IO extender, but the 6R8 screws I have are too short, so I need to get some longer screws from ACS. Always with ACS. I want to finish this plane so I can stop buying stuff from them for a while.

Firewall fail.

5 hours.

Yesterday, I should have stayed in bed. It was that bad. I started out the day by attempting to mount the RDAC (engine monitor module) on the firewall. I did this by drilling out one rivet in the F-601L stiffener along the top of the firewall, then match-drilling the rest of the holes. Would have been a great plan, except when I was drilling the second hole in the firewall and stiffener, something moved and I put the hole about 3/32 of an inch below where it was supposed to go. Boom, violated edge distance. I have an email in to Van’s tech support to find out whether I’m going to have to replace that stiffener, which would suck like nothing has ever sucked before. It would involve removing the engine mount and the landing gear. The other possibility, since I’m not really placing a load on the stiffener with the 5-ounce RDAC, is that they’ll tell me to “build on.” If that happens, I just have a slightly crooked RDAC, but otherwise it’s fine. So I’m kind of freaking out about that, and really hoping I haven’t bought myself a ton of work with this one stupid mistake.

Then it started to get better: The LA public library emailed and said my copy of William Gibson’s ‘Zero History’ was available for pickup, so Shelley and I rode our bikes down there to get it. But on the way home I almost got run over by an idiot backing out of the car wash. What does this have to do with airplane building? Not much, other than it’s difficult to work on critical things in a state of elevated stress. But I needed a win of some kind, so I decided it was time to mount the heated pitot tube in the wing. Of course, that ballooned into wing wiring.

This pic doesn’t show the tubes, but I originally had the AoA and pitot lines running through the two grommets in the rib in the bottom of the frame.

But it’s a couple of years later, and I now know a lot more about EM interference than I did back then, as well as what equipment I’ll have on board. I pretty much rewired the whole thing. First thing I did was pull out all the wires and cut a section in the PVC conduit to allow the pitot heat wires and AoA tube an exit to mid-wing. I rounded off the edges of the PVC to mitigate chafing, then ran AoA tube, NAV antenna wire, pitot heat, landing light, and position light wires down the conduit, breaking out the pitot heat wires and the AoA tube, while sending the rest down to the wingtip. The strobe cable I ran through one of the grommets from wing root to wingtip, which will help isolate the strobe pulses from the NAV antenna. Supposedly RG400 cable keeps this from happening, but I’m going to be making a lot of connections at the wing root (building in the guest house, remember?), so I want to keep the strobe cable and the antenna as separated as possible. Unfortunately, I can’t close the deal, because I need two more 1 1/16″ Adel clamps for the cut ends of the conduit, otherwise they’ll vibrate against the ribs. Those should show up from ACS in a couple of days, but it really irks me to leave things unfinished because I don’t have supplies. It makes more to double check later.

The other wing got the same treatment, but I’d already cut a break in that conduit to allow for autopilot wiring. I still ran the strobe cable through the grommets near the spar, because I’m not terribly interested in having the strobes make the autopilot twitch. Also, depending on the performance of the Archer NAV antenna, I might put another one in the right wing for a second NAV radio at some point.

After all that, I didn’t get the pitot heat module actually mounted. That’s next.

Panel v2.0

1 hour.

Redesign of the panel based on some thinking and some comments and suggestions from VAF. Notable changes: Master and alt bus feed are now to the right of the EFIS, putting all the DC power stuff there, safely out of the way of casual bumping. The second is the addition of a start button, eschewing the spring-loaded dual-pole ignition switches in favor of dual-pole on-on switches per the Aeroelectric Connection. The start button will actually be on the Infinity stick grip, but if that wire snaps somehow, I want a way to start the plane. Another significant feature is the addition of the Pax Enable switch, which disables the buttons on the passenger side stick grip and also safeties the connector when the stick is disconnected. There’s no starter enable switch because there’s very little chance of me bumping the starter button in flight. It takes an actual effort to depress the Infinity switches, so I don’t think the risk of accidental starter engagement is enough to merit the complexity of another switch. I guess I could wire an off-on-(momentary on) toggle in place of the start button that would serve the same function, but it puts an extra step in the checklist (disable start) that I don’t think is necessary. Also, in case of an engine out, I want as few things between me and the starter as possible. Phones and Mic have been moved to the lowest point possible on the panel, but i’m not going to put them in the armrest. Eventually I’d like to get the Classic Aero interior, which would cover up any accessible jacks in the armrest. There will also be a wet compass mounted on the glareshield or in the panel, I haven’t decided. Ideally, I’d keep that million dollar view free of obstruction , but I’m concerned about interference from wiring. Also, the buttons have been moved to the top, so that nothing gets shut off or bumped during turbulence, and the transponder has been moved to the bottom of the stack.

Click to open full size.

More grip stuff

1 hour.

I finished changing out the on-off switches and toggle on the other Infinity stick grip. Now both grips will have the same configuration.

stick grip soldering

And this is what the finished product should look like:
grip on stick

grip on stick
Tested for ergonomics. This works just fine.

But there’s a problem. Remember that big fat blue cable in the first pic? There’s no good way to get it out the bottom of the stick, and on the right side, it’s pretty much impossible. I may have bitten off a fair bit of work for myself here, but I need to find a solution. One thing is for certain, and even though a lot of guys have done it, Van’s says “Thou shalt not drill holes in thy control stick, especially near the holy pivot.” The reason for this is that with a hole drilled in the stick, the yield strength of the tubing has just gone down the crapper. You can argue all you want about the likelihood of someone putting enough force on that stick to break it off at the hole, but I’d hate to be doing a high -g (for me anyway) maneuver and have the stick fold up.

So what to do about that fat blue cable? I have two issues, one is the exit out the bottom of the left stick. The cable chafes on the rod bearings connected to the ailerons, so that’s no good. The other is that on the right side, the stick is designed to be detachable, and it’s going to be, because the missus likes to knit and read books, and she’s not going to have a good time doing that if there’s a control stick waggling where she’s holding a book or a big-ass pair of needles.

In that blue cable are sixteen 20AWG wires, six of which are ground wires (earths, for my Commonwealth readers). I have read of several builders who have stripped off the blue outer cover and the shield, then connected all the grounds together in the upper cavity of the grip, which thins out the wire bundle considerably. This will allow the wires to exit the bottom of the control stick on the pilot’s side. On the passenger’s side, there’s no getting around it, there’s going to be a hole drilled in the stick for wires to exit, but not the honking 5/8″ monster required for the full cable, blue covering and all. That will also require some kind of multi-pin connector, such as a CPC circular connector or a DB-15, so I can detach the stick easily. It’s already sounding like a lot of work, but fortunately it’s not a messy job, and I can start/stop it at leisure.

Get a grip.

1 hour.

I felt fat, stupid, and guilty for watching a hulu’d Sons of Anarchy episode with a big bowl of mac and cheese, so I went out to the shop to get something, anything done. I wound up replacing the microswitches in one of the Infinity Aerospace stick grips. The military style grip is really cool, it’s like an old F4 Phantom grip, with a trigger PTT, a four way toggle (‘china hat’ is not the preferred nomenclature, Dude), and 3 pushbuttons, one momentary, the rest on-off. I had to replace the toggle, because it sticks on in the up position and is momentary down. I want that to be my flap controller, so it needs to be momentary in both directions. I replaced it with one I bought from ACS a few weeks ago. I also got momentary pushbuttons to replace the on-off ones. I imagine those will become AP disengage, transponder ident, frequency flip-flop on the 430, and something else TBD. Maybe flip flop on my spare comm, whatever that winds up being. I might change one back to on-off and run the fuel boost pump with it, we’ll see. I can probably do the other grip tomorrow morning before work if I don’t dawdle on the internet for too long.

Wiring and fiberglass.

12 hours.

This is for yesterday and today. I finished up the elevator and HS fiberglass tips. You’ve heard me piss and moan about fiberglass before, so I’ll spare you that this time, but yesterday’s experience was not great. I got the HS off the futon in the other room and set it up on the bench so I could mount and balance the elevators. When I last left the HS, I’d done the fiberglass tips, but I hadn’t done much else with them, and they needed a good amount of shaping before the elevators could swing freely. One side was fine after some filing and sandpaper, the other had the foam too close to the edge so I sanded all the way through the filler and started taking out foam.. This isn’t good. Raw foam isn’t helpful, so I had to mix up some flox, both to fill in the elevator tips after balancing and reinforce the left HS tip. I had this cool thing set up where a thin piece of sheet was curved round the elevator tip, making a form for the flox that would guarantee the free motion of the elevator through its travel.

Would have worked too if the tape hadn’t popped loose, leaving me with a distorted lump of flox on the front of one elevator. I had to file/sand that back into shape, then glop in a load of micro filler to get the final shape.

So there it is curing away.

I busied myself with antenna mounts while the epoxy was setting up. First was the transponder antenna. I’ve decided to put this on the floor behind the baggage bulkhead. All antennae need doublers, even the stubby transponder antenna. The skin of the aircraft isn’t thick enough to provide structure on its own, so I had to take a piece of .063 and make a plate that the antenna can grip onto.

Today I did the final shaping and priming of the HS and elevator tips. While I was waiting for paint to dry, I worked on the doubler for the Garmin GPS antenna.

Installing these doublers aft of the baggage bulkhead is interesting. First thing you do is drill the holes in the plate at four corners, then one in che center. Then you go inside the ship, down the Jeffries Tube, and position the plate where you want it for proper antenna location. Don’t forget to bring your drill. Use the plate as a guide, and drill the skin using the center hole in the plate as a guide. Then measure to square the plate up with surrounding stuff, then drill the corners of the plate. Now it gets fun. You have to dimple all four holes in the skin that match the corners of the plate. You need to use your 3/32″ pop-rivet dimpler for this, but you do it by crawling into the ship, placing the dimpler, getting back out, making the dimple, then repeating 3 more times. In the case of the GPS antenna, there are also screw holes to deal with, which means platenuts. Oh, don’t forget to countersink the skin side of the plate.

Since I was in the tailcone a lot today, I figured I’d do some nagging things I was going to have to do anyway, like the restraint cable clevises and removing the ELT antenna cable. Yeah, can’t run the ELT cable through bulkheads. If you crash, the airframe might fold and sever the ELT antenna cable where it passes through a piece of sharp metal. I decided to run the trim servo wire down that run in its place. Then I ran the transponder antenna wire alongside the two GPS wires on the left side and did the BNC connector.

I don’t want to spend a lot more time down there. It’s cramped, hot, and dusty in there. If I can get enough servo wire slack so I can work on the connector outside the ship, that will make me extremely happy.

MGL Avionics rocks!

3 hours.

Yesterday I took a long lunch and drove down to TOA (Zamperini Field in Torrance) and hung out with Matt at MGL Avionics for a demo of the Voyager EFIS. The Voyager is the Odyssey’s little brother, but it’s essentially the same EFIS. Matt showed me a lot of the features and a quick run-through of navigation techniques, as well as walked me through the setup screens for various aircraft parameters, and If I wasn’t sold before I walked in the door, I certainly was when I left. I was going to buy everything on the spot, but their stock got cleaned out over the holidays and they’re waiting for a fresh shipment from MGL South Africa.

If you go to the MGL Avionics website, you’ll see some video of the EFIS in action, but I have to tell you, the images on the website do not do justice to the actual unit. On the site, the screen looks really loud and playskool, but on the EFIS, it’s as high-quality and pleasing to the eye as any of the Garmin or Dynon full-synvis offerings. The unit itself is a very nice powdercoated black and the buttons are a nice metallic silver, with extremely good tactile response. My only gripe is a minor one: Terrain above you on the HSI flashes red. Not a primary, saturated red, but the flashing is annoying. I assume this is to let you know that you are potentially flying into a mountain, but a nice option would be to turn off the flashing unless you’re within a certain distance of the terrain.

I wrote a check and left with an RDAC X-D engine monitor module, while Matt packaged up the probes and senders to ship out later. All in all, a very good experience.

After work, I went home to find my shipment from B & C had arrived. Problem is, I ordered two extra E-bus diodes, so now I have to ship them back. Lesson: Always review your online order when the “review order” page comes up. I did manage to get one of them mounted, but I didn’t have my computer with me at the time, so I couldn’t wire it based on my schematic. I also got the 60 amp current limiter for the main bus mounted on the firewall. That was a 10 minute job that turned into an hour because I had to make a doubler and rivet it to the firewall, which sucked for mainly ergonomic reasons.

I’m also ordering the MGL current sensor to replace the 50mV shunt. I was originally going to go with this, then decided against it, now I’m going with it again. It simplifies wiring, and Matt has assured me that it shouldn’t be so messed up by the Earth’s magnetic field that it will be unreliable.

Tonight, when I get home, I’ll wire the E-bus feed and diode, then maybe clean up some wiring back in the Jeffries tube.

Fuse blocks mounted, panel laid out— for now.

5.5 hours.

4AWG and 6AWG wire suck to work with. Period. They don’t bend easily, and when they bend, they don’t stay bent until they’ve been bent that way for a while. But I got the main and e-bus fuse blocks mounted on the subpanel, facing forward, and I can actually get to the fuses in flight.

The main bus. The actual mounting is a bit higher, with the feed post at the bottom. It’s all about where the wires get run.

The e-bus fuse block. This feeds the avionics and EFIS. In case the alternator tanks, there’s an alternate feed to this that comes directly off the battery.

After this was the fun of getting the fat 6AWG wire to the main bus in such a way that it doesn’t chafe on anything. The last thing I want is a wire rubbed raw and arcing against the airframe. Along with this wire, I ran the starter contactor load wire and e-bus feed, as well as the master battery contactor load wire. At this point, I’m stuck. I can finish prettying up the wires in back and start sussing out where and how I’m going to run flap and trim wires from the sticks, but at this point, I’ve done all I can do before the next shipment of electrical bits shows up. I didn’t even know what bits I needed until today. I didn’t know it was OK to put more than one wire into the shank of a PIDG ring terminal. But today, I also finished my panel layout, for the most part. That means I know where the major components and switches are going to go. I really have no idea where to put music input and aux power, but I’ll get to that momentarily.

Click to open full size.

Stainless Steel Providers.

5 hours.

Oh yeah, it was a good day. Not only did it not require the use of an AK-47, I got a lot of momentum going on firewall-forward wiring. The overall schematic is still rather nebulous, but it’s based more or less on Bob Nuckolls’s Aeroelectric Connection, drawing Z11. In this scenario, there’s a main bus, an endurance bus, and a small always-hot bus. I’m toying with the idea of deleting the always-hot bus and just putting in a switch for the alternate e-bus feed because quite honestly, an always-hot bus is an excellent opportunity to drain every last molecular twitch out of an otherwise healthy battery.

Also, with the acquisition of a Garmin GNS430W, my avionics stack is now complete. EFIS, audio panel, transponder, nav/comm/gps, done. Good god, I’d love a cigarette right about now… But the upshot of this is that the the electrical picture is now complete. I have to provide power for these devices, plus the various other implements of flight, namely trim, strobes, lights, and autopilot servos. How does that work? Common wisdom is to start at the battery and work your way back. Instead, I ran loads for lighting and strobes, and today I ran starter and e-bus feed, and they’ll collide behind the panel in a Gotterdammerung of switchgear and fast-on tabs.

So the next step is, how to get the electrons from where they are to where they need to be? I had a few simple rules, gleaned from the Aeroelectric Connection and the mighty oracle of Van’s Air Force, to wit:

1. Thou shalt not run thy strobe cables alongside thy data cables.
2. Thou shalt not run thy data cables alongside thy power wires
3. Thou shalt provide ample room to service thy components after the holy top deck skin is on.
4. Thou shalt not run thy wires below tubes which carry fuel, for the drips from leaks onto that which arcs may beget the inferno.
5. Thou shalt not allow breath or light to pass between cabin and engine compartment.
6. Thou shalt not expose thy wires to sharp metal edges.

The first step was to figure out where to make holes in the firewall for pass-through of electrical cable and sensor data.

A little off from my original guesstimate, but this’ll do.  A 1-inch hole accommodates the SafeAir1 firewall Passthrough, a stainless steel gizmowith a rounded outlet to let wires exit in any direction without chafing on one side, and get sealed with fireproof tape and goop on the other.

After a little cleanup, it looks OK.

My original estimate for the size of the MGL RDAC engine monitor module was way off, and MGL doesn’t actually publish the dimensions of the unit in the installation guide, which is problematic. So I scoured the newly-minted mglavionicsusers.org forum and found the answer I needed, then made this ghetto-ass mockup from the battery box packaging and a roll of blue masking tape.  This made me reconsider the location for the data wire hole.   In theory, there should only be one data wire going aft, that of the RDAC itself.   All the engine probes and sensors should go from the engine to the RDAC.

How do you make a meal out of stainless steel?   Chew slowly.   Everybody gets all weird about stainless steel, and true, it’s a whore to work with, but remember, if you can scratch it, you can cut it.   This is a 1″ hole saw, about three bucks from B&B hardware.  The trick is to dunk the end of it in Boelube and go SLOWLY.   Make your pilot hole with a #40, then move up to 1/4″, which is the size of the hole saw’s pilot bit.   This one’s so dull it won’t go through warm cheese, but it serves as a good guide for the hole saw.   Then, if the teeth of the hole saw are sharp at all, you should be able to grind your way through the firewall fairly easily.   Keep it from heating up.  If it starts to smoke, put more Boelube on it.    The amazing torque of this Makita cordless drill is also helpful.  It’s relentless.   If you’re working above the battery, cover the battery with a sheet of plastic or something.    You don’t want stainless steel chips grinding away between your battery and the firewall.

For added fireproofing, I used a bead of my leftover Fire Barrier 2000 around the FPT (firewall pass through) flange, just to seal the deal.  Not that it’s very necessary; you get pretty much an airtight lock when the two halves are screwed together with the firewall between them.

And there you have it.   Starter load wire, main bus feed, and e-bus feed, all going through the firewall just like they’re supposed to.  What this photo doesn’t show is the firesleeve I forgot to put on the outside of the FPT before I ran the wires through it. It also doesn’t show the master contactor load wire I put in shortly after.  I put the fire sleeve on and clamped it down with one of the hose clamps provided in the kit.

Since I was feeling inordinately proud of myself, I figured I throw the engine mount isolators on there.   Still not sure how they go, I’ll have to check, but the red bolt protection nipples are a nice touch.

Up in the corner next to the VA-168 manifold, you can see the second FPT.   This will carry a data cable from the RDAC back to the EFIS, with room for future additions should I wish to take my life in my hands with some sort of electronic ignition.

And here I am, sitting in the focus of the Dynafocal brainprobe.   Maybe I can infuse it with some of my own sentience, such as it is.

I just got off the phone with my cousin, Navy SEAL and former SDV electronics tech, who assures me, despite my misgivings, that the switchgear on the panel and the power routing is not a problem.   Nor is anything else.  Where it gets tricky is the audio wiring, where impedances must be matched and other arcane spells must be cast.   I’m going to bring him out here from Yuma for the hard stuff, I think.