Archive for the ‘Fuselage’ Category

Comm antennas

Monday, March 30th, 2009

My brother helped me install the belly-mounted comm antennas while he was in town this weekend. In the photo below we're lining up the antennas and drilling the mounting holes… much easier to do it with the antennas inside the fuselage than outside:

My camera was low on batteries so I didn't get a picture, but we made doublers for the antennas out of 0.050" alclad. You can kind of see one in this photo:

Since the antennas are grounded via their mounting screws, ensuring good electrical contact is essential for proper antenna performance. I used an alodine pen to corrosion-proof all the mating surfaces:

Here's a better shot of the fuselage doublers, with the alodine drying. It looks blotchy but that doesn't really matter.

I likewise cleaned, scuffed, and alodined the areas of the fuselage floor destined to be covered by the doublers:

Adam helped me rivet the doublers to the floor:

They're also attached to the seat ribs with blind rivets – plenty strong. Each antenna is attached with four screws that go into floating nutplates, and the BNC connector protrudes through the hole in the middle.

Voila, antennas. This arrangement should make for easy wiring, keep the antennas out of the plume of exhaust gunk, and satisfy the radios' requirement for minimum antenna separation distance.

Fuel flow sensor

Sunday, May 4th, 2008

If you choose to research the topic obsessively, as I have been doing lately, you'll find that there are about three locations where people have been mounting the fuel flow sensor, and everyone thinks the way they installed theirs is best. You can put it on the cabin floor downstream of the electric fuel pump, near the engine between the mechanical fuel pump and fuel injection servo, or in the line between the fuel servo and the fuel distribution spider on top of the crankcase. I accept the argument that your fuel flow readings will be more accurate the farther downstream your sensor is placed, but there are also vibratory and thermal factors to consider when mounting a somewhat delicate sensor in the harsh environment of an engine compartment. After a weekend of head-scratching, I couldn't settle on a location forward of the firewall that I found completely satisfactory, so I gave up and elected to mount my flow sensor on the cabin floor. I can live with the fact that I may get inaccurate readings when the electric fuel pump is turned on, but that will only occur for a few minutes per flight. There is theoretically an increased chance of vapor lock as a result of adding another restriction upstream of the mechanical fuel pump, but I don't know of any evidence of this actually happening in real life (and if it does, I can still turn on the electric pump). There's also the fact that I put the flow sensor on the cabin floor in my last airplane, and it seemed to work fine.

Anyway, end of rationalization section. I pulled all the fuel plumbing out of the fuselage – it's easier than it sounds: you only have to undo three flare nuts and four screws and the whole works lifts right out.

I cut the line that previously went from the electric fuel pump outlet up to the firewall, re-flared it, and installed the fuel flow sensor.

The sensor body is partially tucked under the mounting plate, so it won't hit the fuel pump housing. It's not attached to anything, just sandwiched between the mounting plate and the floor. I stuck some foam weatherstrip material to the bottom, to hopefully isolate it from exhaust vibrations coming up through the floor.

With the flow sensor spliced in, the fuel line runs up to the firewall as before. Notice that this location satisfies the manufacturer's recommendation to give the sensor 5" of straight line on both the inlet and outlet sides, which is much more difficult on the other side of the firewall.

The housing fits without hitting the sensor or any of the plumbing:

Throttle quadrant cable bracket

Sunday, March 16th, 2008

To attach the engine control cables underneath the panel, I made this little bracket out of 2"x2"x1/8" aluminum angle. The three small holes are for the cables, and the other holes are just to lighten it a bit.

Because the bracket will be somewhat visible hanging down under the panel (at least when you're standing on the ground behind the wing looking in) I decided to paint it black. The paint was taking forever to dry in the cold garage, so I laid the bracket on top of my little electric radiator, using a piece of scrap alclad as a tray. It cured fully after just a couple hours in the sauna.

Here is the cable bracket clecoed to the throttle quadrant supports. The green cable is from Van's and the clevis ends are from Aircraft Spruce (the clevises that Van's sells won't work).

Another view. I had to have at least one cable on hand so I could know where to mount the bracket, but without mounting the bracket I wouldn't know what length of cables to get. I chose to solve this dilemma by buying one cable in a standard 48" length, and using it both to locate the bracket and to determine how long the real cables actually need to be. Since it's a standard length, I can return it for 90% credit later, and I don't have to waste money trial-fitting with expensive custom cables that can't be returned if I guess the wrong length.

I hooked my test cable up to each of the three engine controls in turn. Here's a view of how the prop cable is routed from the eyeball in the upper-left corner of the firewall, down to the prop governor bracket. With this part adjusted correctly, the portion of the cable inside the fuselage needs to be a few inches longer.

The throttle cable comes through the firewall low and in the middle, and runs up to the bracket I previously bolted to the sump.

Inside the fuselage, the throttle cable is about an inch too short:

I forgot to take a picture of the mixture cable, but you get the idea of how this process went. In the end, I decided I needed cables that were 49", 51", and 52" long. I called Van's to order these and found that they also have standard-length cables in 49.5" and 52.5" lengths… the cable lengths can be varied somewhat, so I decided to order those two standard lengths and one custom 51" cable.

By the way, these cables come with a most interesting label on them:

Not for use on aircraft indeed.

Throttle quadrant

Sunday, March 2nd, 2008

I've been agonizing for quite a while about how to mount the throttle quadrant that Mary gave me for our anniversary last year. Specifically, I was concerned about the degree to which the throttle quadrant was going to impact the amount of available legroom – I'm somewhat tall and I want to be comfortable sitting in this airplane on long trips. Another perplexing issue was the inconvenient lack of substantial structure behind the panel to which a throttle quadrant could be mounted.

Well, recently I started playing with the standard Cessna-style, plunger-type control cables, and was surprised to find that the bracket required to mount them was going to come uncomfortably close to my right knee. You see, my panel is an inch or so taller than stock, and the bracket that the plunger controls attach to has to be pretty wide because the knobs are arranged side-by-side, with a couple inches of spacing in between adjacent knobs… the combination of those two factors made the bracket intrude on my personal space. So, I started taking a hard look at just how the quadrant might be mounted. I started by mocking up the mounting location using spring clamps and scrap aluminum, then installing the pilot's seat and climbing into the fuselage for an ergonomics check. You can see in this photo that there's at least a couple of inches between the throttle quadrant and my right knee under normal circumstances:

On long solo trips in calm air in my RV-9A, I always used to like slouching sort of sideways across the cabin, stretching my right leg over to the passenger side. That airplane had the short, stock panel, so it was easy to do. With the throttle quadrant here, I can't get my leg all the way across without my knee hitting it, but a decent amount of slouching-ability is still retained. I can deal.

I put the transponder into its rack, just to make sure that my hand won't bump it when the throttle is fully open. With the way I have the quadrant suspended below and aft of the bottom of the panel, there's plenty of clearance. And yes, I did make airplane noises.

This is how the quadrant mounting setup looked about halfway through the process. I sort of made up the design as I went along, using various bits of scrap angle. With the airplane up on the gear, access to the under-panel area is somewhat inconvenient, involving a lot of crawling in and out of the fuselage with the help of a stepladder, so it was tricky to get all these pieces measured and aligned.

To close out the top of the throttle quadrant and make it look nicer, I freehanded this little cosmetic trim piece out of some scrap aluminum. I don't fancy myself much of a metalworking artist, but I'm pleased with how it turned out.

I painted the trim piece flat black to match the quadrant and panel. I also painted the exterior faces of the side mounting rails, since they'll be visible after they're mounted.

Here's the finished product, bolted in place. It's very strong – if I had any strength or agility whatsoever I could probably do handstands on it. The prop lever (i.e. the middle one) is centered laterally on the panel, so the pilot and passenger have an equal amount of legroom. Since the travel-sized Mary will be my main passenger, I thought about biasing the quadrant to the right a bit to give myself more room, but in the end I decided to take a more egalitarian approach so giants like John can ride in comfort.

The fasteners at the aft (panel) end of the mounting structure are primarily loaded in tension, so I used #8 structural screws (AN525's) instead of rivets there.

At the forward end, I drilled out one of the subpanel reinforcement brackets and replaced it with a longer one that ties into the port-side quadrant mounting rail. A couple more little pieces of angle tie into the other mounting rail and give it some additional torsional stiffness. All this aluminum angle is starting to make the under-panel area look like a jungle gym, but it's strong and light and nobody will see it.

The painted trim piece looks pretty sharp:

When the throttle lever is in the idle position, the sticks can get tangled up with it if you use full forward-stick and full aileron deflection. I'll be trimming the sticks to eliminate this potential safety issue – they're overly long as received, so I'd planned to shorten them anyway – but I'll wait until I have the elevators and ailerons hooked up before I do it. I have a feeling that the stick has a wider range of motion when it's not hooked to the rest of the control system, compared to how it will actually move in the finished airplane.

Now that I have the throttle quadrant installed – and I must say, it turned out great – I'm left with the problem of determining what length of cables I'll need to connect it to the engine. Dave Parsons was kind enough to share the cable lengths he used for his quadrant setup, so I'll use those as one data point. To avoid wasting money having custom-length cables made only to find I picked the wrong size, I decided to order from Van's a single 48" cable, which is one of their standard sizes. If that works for one of the three controls then great, I will keep it. If not, I can send it back unused for a credit, and having it on hand should get me close enough to be able to measure how long I need my custom cables to be.

Gearleg nuts

Tuesday, February 5th, 2008

In this week's Aircraft Spruce shipment I received the proper nuts to go on the bolts that secure the gearlegs to the engine mount, so on they went:

The plans call for regular AN365 nyloc nuts here, but I've already decided to use only all-metal locknuts forward of the firewall, no exceptions. AC43.13 only forbids nylon nuts where temperatures exceed 250°F, but since I'm not in a position to measure peak temperatures in various locations forward of the firewall, I'd rather just outlaw nylocs entirely.