FFW – Wiring and Fuel System Tasks

As I’ve mentioned before, the fuel system is just short of outrageous. It’s not bad. It’s just that there are plenty of bits and pieces to fit in awkward places, along with the ever present mandate of ensuring the robust integrity of every aspect of the system.

The Sling KAI provides guidance. Careful reading and understanding of the Rotax 912iS Installation Manual is a must. Let’s certainly not forget AC-43.13. All of the information lines up pretty well, yet there’s a lot left to the discretion of the builder.

I’m incorporating all of the fuel system features that appear in the KAI and the Rotax manual, with the exception of a fuel selector bypass check valve for the fuel return line. This valve appears in the Rotax fuel system schematic, but not in the KAI. The purpose of the check valve is to relieve pressure (of residual fuel in the system) to a fuel tank if one or both of the fuel pumps get turned on while the fuel selector is set to off. For now, I’m just going to be careful. But, after 20 years of maintenance experience with my Piper Warrior, I know that I’ve switched on the electric fuel pump [and started the engine] with the fuel turned off. Return pressure is not an issue for the Warrior, but the potential condition of fuel pump on and fuel selector off is common to both airplanes and I’ve already done it – more than once.

Also, as I’ve mentioned in other posts, I’m adding AN6 flare fittings in several places where only barbed fittings were provided with the kit. Key reasons for elimination of the barbs is to avoid awkward conditions for crimping the Oetiker ear clamps and to somewhat modularize sections of the fuel system. To my mind it seems advantageous to have separable pieces that can be unfastened and refastened with wrenches instead of having to cut and then crimp metal clamps. Adding bulkhead fittings for the supply and return lines pretty much demanded flare fittings. Being able to service/replace the fuel filter is another place where swivel fittings seem like they’d be somewhat more conventional and convenient.

The great majority of the fuel fittings are aluminum with just few brass NPT elbows at the fuel selector. Ahead of the firewall, steel fittings are often used, but for experimental applications anodized aluminum fitting are quite common, and as far a I know – not unacceptable. I’m not certain about the metal fuel lines themselves, but Piper put exclusively aluminum fuel fittings behind the firewall and in the wings of my ’84 Warrior II.

The kit provided a big fuel filter with barbs. The engine came with a Rotax part that pretty much everyone that’s building a Sling uses. Me to. The filter really needs to be secured in a suitable location. The sensible and popular location for the Sling 2 seems to be near the battery box. I designed and fabricated a bracket and clamp system.

Another consideration has to do with where to put the fuel pressure sender. Some builders have voiced concerns about making sure that the wet/pressure port faces down, to prevent pooling of fuel in the sender and orifice. I don’t know how big an issue that really is, but I found a place to put the sender and keep the port facing down. More AN6 fittings are involved. I designed and fabricated a bracket and attached the assembly to the engine mount with a cushioned clamp.

One of the things I’ve done is to upgrade the fuel hose. The kit provided SAE J30R7 hose for low pressure fuel supply and return lines. SAE J30R9 fuel injection hose was provided for the high pressure lines between the fuel pump an the engine. I’ve replaced all of the R7 hose with R9. The R9 hose has low permeability and resistance to methanol. There’s also R14 hose. It seems to be somewhat more available, but samples I’ve found are noticeably stiffer than R9 and I believe the pressure rating of the R14 is about half that of the R9. The R14 hose would probably be adequate, but I’ve gone exclusively with the R9.

I’ve built several of the short lines with connectors and firesleeve. I found push-lock fittings specifically for 5/16″ hose. Assembly with those fittings is easy. The ear clamps are solid.

The 5/16″ hose and and several of the fittings – especially the 1/8″ NPT fittings – are a bit on the narrow side. So franky are some of the barbed fittings supplied with the kit. I’m under the impression that the small variations I’ve made will not introduce inadequacies in the feeding of the miserly 912iS, even at full power. If it were a 914 or 915iS it would be a different story. From what I can see, I don’t think I’ve changed the game. I’ll test carefully, you can bet on that.

The fuel system also incorporates a tiny pressure relief channel between the supply and return lines that allow the high pressure side to bleed down into the return circuit when the [engine and] fuel pumps are turned off. Rotax shows this before the fuel filter, but the Sling 2 KAI shows it after. After lots of consideration, I think I’ll go with exactly how the KAI depicts it, using the T-fittings that were provided. I did however, purchase a clever manifold part from Lockwood Aviation that I’d seen on another builder’s blog. Since then I’ve seen the same part used on several Sling builds. The part incorporates the orifice for pressure relief, NPT threaded input and output fittings for pressure and return lines, and also a place to attach the fuel pressure sender, but all at the cost of more fittings that, in close proximity with the fuel filter and another check valve, presented an awkwardness that I don’t care for. I’ve decided not to use the Lockwood part.

Dealing with firewall penetrations is a subject with seemingly endless discussion. It’s a good topic, I’ll admit. The kit came with a big hole in the firewall and 2 halves of a cover plate. This is where avionics power and engine management cables go. The KAI makes a brief mention of the plate and setting some rivnuts, but not much more. I ended up making a sort of gland out of scrap silicone engine baffle material. My Warrior has the same thing where throttle and mixture cables penetrate the firewall. I replaced those cables some years ago. Piper used a paper-like material, but it had seen better days. The silicone material proved to be an excellent replacement for the Warrior. I’ve used the same material here. I’m not done with. It still needs to be sealed. But, it’s basically what I think I want.

While I’m working to get the FFW tasks finalized, my priority is getting the glare shield/dash, forward fuselage top-skin and windscreen done. I want the panel in final position and completely working before I do that. For confidence testing, I’m connecting up all of the power and grounds for the fusebox and engine.

Fuselage Tasks – Cables, Wire and Lines

For an E/A-B project there’s a seemingly endless list of small tasks. Some things are simply items in the KAI. Other tasks are things that come to mind as you think about how its all going to come together. Task creep just seems to happen. As long as it doesn’t get out of hand – it’s a good thing.

There is a daunting amount of wiring to support the avionics I’ve elected to squeeze into this airplane. Frankly, I’ve gone overboard. On the other hand, I’ve got all of the makings for a technologically advanced airplane (TAA), extremely well equipped for IFR training and cross-country adventuring. Finding good ways to dress and secure the wiring bundles is important. I’ve put a lot of thought into it and determined that an extra bracket will help me keep the harness up and out of the way. I designed and fabricated one.

Many things have to be squeezed into available space without undue conflict. Wiring, fuel lines, pneumatic lines, brake lines, valves, linkages and such. While there are plenty of best practices for how to do things in airplanes, finding ways to apply them to all of the situations that present themselves is what keeps this amateur builder scrambling for ideas. There’s always something new to learn. It’s fun. Challenging, but fun. Yes. It is. I think.

The fuel-injected Rotax engine has a somewhat complex fuel system. The fuel lines have a fuel supply circuit and a fuel return circuit. When you add filters and a duplex fuel selector and physical considerations, the plumbing details border on outrageous. How I do it is up to me. The KAI supplies only general information. I’m also introducing AN6 flare and push-lock swivel fittings in several places where barbs and ear clamps were originally called out and supplied. Figuring out where and how to mount and connect many of the fuel system components is ongoing.

Some tasks just need doing. Then they’re done. Cutting coax cables to length and adding connectors is straightforward enough. Done. The Rotax 30A Regulator B needs a heat sink that calls for the regulator to move from the fusebox to a nearby location on the engine mount. Done. There are a couple of GPS antennas for the PFD and G5 flight instrument that found homes on top of the GAD 29, above the center avionics stack.

Brake lines need to be routed. The time to do it was now. One line goes from the brake fluid reservoir, through the firewall, to just under where the throttle and brake quadrant is located. From there, another line originates at the pressure side of the master brake cylinder an runs to a compression T-fitting under the console compartment. Two other lines run from the T-fitting, down the main gear legs to each wheel. The combination of the 3 lines and the T-fitting were challenging to place. I should have done it earlier, when access was better. It’s done now.

I received some [evidently] old-style mounting hardware for the parachute activation handle, so I ended making a small bracket to secure the activation pull-handle to the back of the instrument panel. It’s nice. It’s done. My design turned out to be a slightly improved version of a bracket I eventually saw in a photo from the factory. Great minds think alike.

It was time for pneumatic lines to go in. I found a really nice way to route them from the left wing root, though the center console and up behind the instrument panel. Because I’ve seen that the Sling TSi documentation now shows 2 static ports, I added a second one in front of the LH fresh air NACA duct. Both of the equal length static port lines arrive at a push-to-connect T-fitting at the center console. A short line runs from the first tee to another T-fitting that provides two more equal length lines to the the G5 and to the GSU 25C ADAHRS unit. The pitot line has a single T-fitting that also splits between G5 and the GSU 25. The AOA line only goes to the GSU 25. All done.

WAAS GPS and COM 1 Antennas

Quite a few Sling 2 aircraft have been built with their GPS antennas mounted on a bracket under the engine cowling. That’s how the factory used to do it. Now guidance has been changed and a location on top of the rear fuselage is preferred. The new location reportedly offers consistently better reliability. I want reliability for IFR flying.

It was no simple task to get RG-400 coax cable for the GPS routed – and satisfactorily secured – on its way from the top-rear of the baggage area, down and then forward along the bottom of the rear fuselage to the center fuselage, and on to finally arrive behind the GTN 650Xi on the instrument panel. It’s done and it worked out very well.

Dealing with the COM 1 antenna coax cable was comparatively easier than for the GPS because the factory-supplied coax was already secured in place, routed from the mounting site on top of the rear fuselage, all the way to the instrument panel area. All I had to do was add the connectors.

The GPS antenna needed holes drilled in the fuselage skin. I used a doubler that came with my panel and harness and the screws provided with the Garmin GA 35 antenna. A TNC 90 degree solder/crimp connector by Amphenol made for a tidy connection at the top rear of the baggage compartment. A section of vinyl hose, covered with black heat shrink, provides good protection and satisfying aesthetics where the coax passes through the bulkhead, into the rear fuselage.

The COM 1 antenna needed 4 existing mounting holes to be expanded, in order to accept setting of rivnuts. The KAI called for M4 but the Rami antenna, supplied as part of my custom avionics package, came with 8-32 screws. I decided to set 8-32 rivnuts into the fuselage for this mounting. The way the coax was secured to the internal fuselage structure lent itself more to a 90 degree BNC connector. Again I selected an Amphenol solder/crimp part.

CF – Upholstery, Painting & Wiring

Upholstery work takes a bit of courage and determination – at least for me it does. A big part of the work involves trimming the vinyl leatherette-like covering to fit the center console and then bonding pieces of material to some rather sizable and awkwardly shaped panels – everything is frighteningly sticky with spray adhesive.

I got through it and it all seems to look rather nice. I can’t say I’d do it the same way again, but at the same time, I don’t exactly know what I’d have done differently. You’ve just got to do the best you can.

All of the center upholstery pieces came in a kit, along with fiberglass-backed side panels, seat and seat back cushions, and carpet panels for the floor and luggage compartment. The seats cushions are covered with genuine leather and stitched by professionals. I got to choose the colors. Methinks it’s going to be a rather sporty and plush setup for this little 2-seater. Oh well! What can I say?

Once the rear portion of the center console was covered, it seemed a good time to put it into position. The inside seatbelts pass through, so it was time to assemble those to their anchors. I haven’t riveted the rear console yet – just clecos for now.

The front console panels have 3 sections. The top panel of the LH side is removable for inspections, as are 4 other square panels – 2 of which are covered with the vinyl. The 2 forward most panels are painted, along with 2 panels for the baggage area.

Bonding the covering to the RH side panel was the biggest challenge. The section of material was large enough to be unwieldy. Working conditions inside the CF were awkward. I did, however, manage to lay in the material and get it satisfactorily positioned. I’m happy to have that task in my rearview mirror.

This seemed a good time to complete more of the wire connections to the rear fuselage (RF) section for the tail beacon and electric pitch trim servo. The PVC jacketed wires in the RF were routed and secured as part of the factory quick-build (QB). My custom avionics harness was provided with “pigtail” connectors. I connected the wires of the pigtail to the fuselage wires with Raychem D-436 series butt splice connectors. The connectors are a bit expensive and a special crimp tool is required. I believe the resulting connections are more than satisfactory. As usual, I insulate and protect the connections and secure them from movement.

Engine On

Bolting a factory new aircraft engine onto my own airplane is a dream come true. There’s so much going on it the world to be focused on this, but I am. I’m glad.

I’m also very glad to be amongst the extraordinary group of folks at EAA Chapter 430 – Olympic Peninsula. One of our chapter members so kindly shared his time and experience (and shop crane) to help me mount the new engine. Thanks Barry!!

Without further ado – a drum roll please…

Sling 2, N842RH (registration number reserved) — is today, powered by a Rotax 912iS Sport engine!

Wheels On

There’s not a whole lot to say except that the wheels are on!

The factory-supplied Matco wheels and brakes seem quite nice. I’d mounted the tubes and tires some time ago. I’ve packed the main wheel bearings with grease that came with the kit. The nose wheel has sealed bearings. I prepped the wheel fairing mounting brackets and treated them with alodine.

Jacking the airframe was a bit challenging. I’ll have to find a better way to do it for inspection and maintenance. Assembly went well.

It’s very good to see the thing up on its wheels.

FFW and CF Tasks

While the engine mount is off, I’m getting firewall forward and center fuselage tasks done that would be more challenging to do later.

I’m not keen about how the factory seems to expect the stiff-wire push-pull cable to go from the instrument panel, through the firewall and then to the heat box on the firewall. As with other firewall penetrations, I’m not content to just stuff the cable through the firewall with a grommet. I’m also not going to settle for having the cable penetrate the firewall at the absurd angle needed to even have a chance of getting to and working the heater box vane. Instead, I’ve designed and fabricated a jack-shaft bellcrank arrangement as an alternative. I’m still going to use the factory-supplied cable assembly – straight through the firewall, a bowden cable clamp and then attaching the wire to a nylon control horn. Another control horn, at the other end of a shaft, translates the push-pull control motion approximately 90 degrees – to be in line with the action needed to work the arm on the heater box vane. I made a couple of brackets out of aluminum angle and mounted the mechanism on the firewall.

Since my QB airframe was built and delivered, the factory has rethought how and where the ELT antenna goes – to just ahead of the vertical stabilizer. It’s too late for me. The structure was changed to accomodate a new mounting bracket and I’m not going to attempt a retrofit. The old location for the ELT antenna was inside the cabin. I’ve designed and fabricated a bracket to mount the antenna inside, just ahead of the rollover structure on the RH side of the fuselage.

Now that I’ve got my hands on the main battery – EarthX ETX 900, 16AH, LiFePo4 – I’ve been able to build and connect 4 AWG cables from the battery terminals to the 12V contactor and to the airframe ground lug. The high current cables are short and tidy.

I’ve had to acquire [standard AN] replacement hardware for re-mounting the engine mount, but this time, along with the front cables for the ballistic parachute. Longer bolts are needed to pass through the heavy cable-attachment tangs. Initially, I didn’t have the tangs. I eventually got those, along with a bunch of other factory parts that should have shipped with the main kit. The cables and the engine mount are on! Good deal.

With the engine mount in place, I’ve mounted the nose gear strut. Some months ago I accomplished fitting of the bushings, retainers and bolts. That made is super easy to just bolt it all together and connect the push-pull rods to the rudder pedals linkage.

I’m still waiting to put the wheels on because the fuselage is that much lower to the ground, making the inside of the center fuselage (CF) somewhat more accessible than it would be with it higher. I’m taking advantage of the easier access while I dress and secure the wiring and prepare parts of the control linkages and autopilot roll servo.

I’m pretty happy with my approach to securing wire harness bundles as they pass through various openings in the CF structure. I found a source for AN743-13 aluminum angle brackets. These brackets are just right for supporting insulated (Adel) clamps around the wire bundles. It was very challenging to drill holes and rivet the brackets at this late stage of the build. I didn’t have the luxury of doing it while the structure was open, sitting on the bench. Nevertheless, the brackets and clamps are in place and they’re pretty nice. I’ve also put some edge grommet in a few places, just for peace of mind.

I’ve previously tested the flap actuator with temporary connections, but now I’ve made the connections permanent with crimped butt-connectors and various layers of insulation and protective armoring. I’ve done checks to insure that the wiring will be clear of moving mechanisms. It all looks very promising and I’m feeling happy about the work.

Another thing I’m pretty happy about was my purchase of a simple jig for drilling nice cross-holes in the control tubes. Beautiful!

More Antennas and Fittings

After giving things a lot of thought, I finally summoned enough resolve to drill some holes in the aircraft skin, and most, where none had existed before. You’ve got to do what you’ve got to do. Antennas, static ports and bulkhead connectors need holes. Now I’ve got them.

Determining the location of antennas is where most of the thought went. I tried to get the transponder blade antenna pretty close to the fuselage centerline – right in the middle as it turned out. Airframe shading and interference between antennas and other equipment are key concerns. The COM 2 antenna is also on the bottom of the fuselage, just behind the rear wing spar carry-through. I tried to keep it as far away as I could from the the transponder antenna and GMU 11 magnetometer. I didn’t get all of the separation distance called for in various installation documents, but what could I do? This is a tiny little airplane — not a King Air. Time will tell if it’s enough. Build on.

All of the antennas are made by Rami and came along as part of the package with my panel, harness and avionics. Some builders have made rather large doublers. For the moment I’m using only the doublers that came with the antennas. I may eventually go beyond those, but for now, everything is in place.

Of the 14 holes that were already in the ELT mounting tray, none were where I wanted them to be. I made 4 more in the tray that allow 4.0 x 10mm rivets, through the fuselage skin and into the tray, without interfering with the plastic case of the ELT itself.

I made an airframe ground lug from a 5/16″ brass bolt and installed it at the location where the KAI called for a 12 AWG avionics ground bus wire to be passed through a grommet in the firewall on it’s way to the negative battery terminal. I’ve decided to have a 10 AWG cable for the avionics ground bus to the lug (inside) and a 4 AWG cable from the same lug, on other side of the firewall, to the battery. I plan to attach a 6 AWG engine [starter] ground cable to the lug as well.

Rubber fuel lines penetrating the firewall through ordinary grommets, as the kit assembly instructions (KAI) call for, didn’t float my boat. I’ve gone with AN6 bulkhead fittings for the fuel supply and return lines. These fittings need smaller holes than the ones pre-punched in the firewall by the factory. I designed and fabricated a doubler out of steel and riveted over the original holes. The fittings are mounted there.

A set of step-drills is absolutely essential for making holes in sheet metal. I’ve got a cheap set but they’ve worked well.

I’ve set M6 rivnuts (with some JB Weld epoxy for good measure) in the fuselage rib, behind the parachute compartment, where the Rotax ECU goes. Getting the 3 large engine wire bundles and connectors in place, given the minimal space remaining with the LRU rack mounted, is going to be super tight. Fingers crossed.

As long as I had rivnuts and JB Weld handy, I decided to rework the Andair Duplex Fuel Selector valve to receive M4 rivnuts and then set them in place. These steps are called out in the KAI.

My factory quick-build fuselage came with not one – but two – static ports in the rear fuselage. However, the factory now says not to use them. They’ve also said they haven’t officially decided where the new location(s) should be either. They think it’s going to be finalized “soon” and anticipate that it’s going to be just in front of the fresh air NACA duct near the front of the fuselage. They sent me a picture and I’ve gone with that location.

The outside air temperature probe found its home in the LH fresh air NACA duct.

Panel Power On Milestone

I’m happy to report that the panel powered up with no apparent issues. All good!

At this point, I’ve got good reason to believe that the avionics and and wiring harness are happy and healthy. There’s still a long way to go, but I’ve got something to hang my hat on.

The engine management and engine information systems aren’t connected yet, but the Garmin G3X Touch avionics are up and running, along with the Vertical Power, VP-X Pro electronic circuit breaker system. Of course this was done during setup and testing at Midwest Panel Builders, but it’s very nice to see it happening – here and now.

I’ve confidence-checked about as much of the harness as I can at this point. I’ll be able to do more when my Rotax 912iS Sport engine is mounted and its Engine Management System (EMS) connected. Stay tuned for that.

Wiring Harness Placement

After careful study of the wiring harness, I think I generally understand how it should be positioned. All of the connectors and each of the individual wires have labels that match the names of the LRU connectors and appear in the schematic wiring diagrams that came with it all. Sweet!

The harness has a central “gooseneck” from which various connector bundles extend. The several sub-harness bundles have been made to reach corresponding LRUs where they are mounted behind the panel. Other portions of the harness extend to places in the center and rear fuselage to connect with lights and autopilot servos.

There are probably several ways that some of the harness extremities could be routed within the center fuselage (CF). Some routes are undoubtedly better than others. I’ve tried to find what I consider to be the most natural – and if you will – most elegant way to dress and eventually secure the wiring, while always considering guidance in FAA AC-43.13-2B.

So far, so good.

In the interest of keeping things both lightweight and tidy, I’ve opted to replace a heavier and bulkier firewall-mounted circuit breaker enclosure with a simple bracket. I also decided to locate the battery and pitot heat power C/Bs and the ammeter shunt on the LRU rack instead of inside the engine compartment.