Top spark plugs

December 14th, 2013

Today I installed the top spark plugs on all four cylinders. For posterity, here's a photo of the box of plugs I used… NGK BR8EIX iridium plugs, as recommended in the P-Mag install manual. The 6747 part number has solid terminals, not the two-piece screw-on variety, which is preferred for this application.

The back of the box has a "not for airplanes" pictogram. That's encouraging.

The engine came with a set of cheap Denso copper plugs. I'm sure these would work fine, but I thought I'd go with iridium plugs based on past experience with an airplane that had a car engine in it. Also I've been really impressed with how long the OEM iridium plugs in my car have been going, with no appreciable wear.

View of the old and new plugs showing the difference between the two types of electrode.

You're not supposed to gap iridium plugs by hand, but I did check the plug gaps just to make sure they were manufactured correctly. The manual recommends a gap of between 0.030" and 0.035"… all four of my plugs were 0.028", which I'll consider close enough.

To use auto plugs in an aircraft cylinder, you need adapters to make the 14mm auto plugs for the 18mm aircraft plug bosses. When I sold my Lightspeed system, I kept the Lightspeed steel adapters rather than buying new ones from the P-mag people. That just means that I had to refer to the LSE documentation to install the plugs.

Using the LSE instructions, first the adapters get installed and torqued to 35-40 ft-lbs, using anti-seize and a new copper gasket:

Then the plugs are installed in the adapters, again with anti-seize, and torqued to 18-20 ft-lbs. This is different from the instructions for the P-mag adapters, which have you thread the plugs into the adapters and then torque them into the heads all as a single unit (to 18 ft-lbs). Anyway, since I have LSE adapters I decided to stick with the LSE instructions.

I put a light coat of silicone grease on the outside of the plugs to help the boots seat properly, and pressed the boots onto the plugs until they clicked:

I also re-did the ignition wire support arrangement a bit… I had previously piggybacked the support brackets onto the existing adel clamps that secure the fuel injection lines, but a VAF thread made me consider that such an arrangement could lead to increased vibration and potentially even cracking of the stainless lines. Probably a remote possibility, but just to make myself feel better I put in an extra set of adel clamps for the ignition wire supports, and restored the fuel injection line clamps to their previous configuration. Now there's no connection between the fuel lines and the plug wire supports.

Meanwhile, I haven't final-installed and torqued the bottom plugs yet, so used some red tape to remind myself they they are only temporarily installed:

Air duct bolts

December 7th, 2013

With the alternate air door complete, it's time to bolt on the air duct for good. But first I had to take it off and remove the duct tape I put on there five years ago to keep stuff out… how time flies:

Photo for posterity showing what the business end of the fuel injection servo looks like – an inlet throat with a butterfly valve and a set of little air metering tubes:

The plans want you to use plain bolts plus lockwashers to keep them from backing out. I decided to use proper drilled-head bolts instead, just to be contrary. The part number for these is MS20074-04-06, in case you're wondering.

Safety wired the pair of bolts on the right side:

And here's the left side. Not my best safety-wiring job ever, but it's tricky to get the tools back in there.

The engine air duct is now permanently attached – bolts at the bottom and screws on top. That's a good feeling after all this time.

I installed the air filter to keep objects from falling into the engine, then taped a plastic bag over the top to keep schmutz from getting all over the air filter:

Alternate air cable improvement

December 7th, 2013

I wasn't happy with the way the alternate air door could be opened with the cable, but not closed again the same way. After a couple days, though, I had an idea for a simple way to make it better. First I bought a package of spring washers from McMaster. These are just little pieces of round spring steel with a bend in the middle. You can get them in many different sizes and spring rates; these are the ones I used:

I made a little 1/4" OD brass bushing on my lathe, and enlarged the pivot hole in the door up to 9/32" so it would be a loose fit over the bushing. A quick trip to the hardware store for some brass washers and I was all set…

Here is the hardware stackup. Around a #8 AN525 screw is the brass bushing, surrounded by the two spring washers which are sandwiched between a pair of brass washers. A thin AN washer goes between the lower brass washer and the screw head.

Some blue loctite on the screw threads for extra security:

Here's what it looks like when it's torqued down. The screw captures the bushing between the outer washer and the steel door aperture; the bushing remains stationary while the door rotates around it. The outer washer compresses the springs, which in turn bear on the door to keep it from flopping outward when it's open. The brass washers are just there to provide a bearing surface for the steel spring washers as they rotate.

Here's what it looks like all assembled. In theory the moving washers will tend to try to loosen the screw, so this could still be improved by using a proper bolted joint with a castle nut and cotter pin. I already had the #8 nutplate installed on the inside of the air duct, though, so I'll keep an eye on this area and make sure it's lubed periodically. Between that and the loctite, I expect it should be fine.

The little thingy on the push-pull cable in the previous photo is one of these – a machined collar more typically found on model airplane landing gear:

I decided to put the collar on the wire to give it a positive stop. In retrospect I cut the cable sheath an inch too short, but the collar helps to make sure the cable doesn't pull too far out.

I thought about making a little video of the door being opened and closed with the cable, but it's too cold in the garage. Trust me, it now works in both directions! I'm happy with the way this turned out.

Alternate air cable

November 4th, 2013

Returning to the alternate air door, which needs a control so the pilot can open it if required. It all starts with a bracket, made of 1/8" aluminum and designed to fit in a very specific place:

That place is the side of the throttle quadrant, attached using two of the existing mounting bolts:

Locking push-pull cable housing installed in the bracket:

Here's what the cable knob looks like when it's in the "closed" (normal) position. I arranged things so the face is flush with the panel when it's not in use.

Marking and drilling the firewall for another cable passthrough fitting… this took some figuring in order to find the ideal routing:

Just enough clearance between the engine mount tubes to cut, drill, and deburr. Duct tape is there to protect the engine mount from tool-induced scratches:

Eyeball fitting installed and cable sleeve in place:

More custom-made brackets, this time from 4130 steel:

Primed and painted, with the paint baked on for durability (don't tell Mary I used her oven for airplane parts again).

These brackets attach to the engine using a couple of bolts on the intake tubes, and support a pair of clamps around the cable sleeve:

This routing is much simpler than the oddball idea shown in the plans, which I think would probably have conflicted with the mounting of the oil cooler anyway. Here the cable is just a straight shot to the front of the engine, with some slack at the aft end for when the engine moves around:

Here you can see that the cable runs well inside the loops formed in the ignition wire harness, and has plenty of clearance:

No problems at the aft end either, thanks to careful positioning of the firewall fitting:

Next, we need a way to attach the cable to the door itself, so it will open when you pull on it. The plans suggest heating the cable with a torch and bending it around a screw. That seemed pretty lame to me, so instead I fabricated this little aluminum block on my milling machine. You'll see in a minute how it works…

At the place on the door where the plans call for a simple screw and nut, with the cable hooked around it somehow, I instead fabricated a brass bushing that's captured securely between a pair of washers:

The bushing is just slightly longer than the width of the block, so the block can pivot freely on the bushing but is kept in place by the outer washer:

And now here's how the cable attaches to the block. The cable wire passes through the small hole, and is clamped between a pair of set screws which come in from either side. The principle is the same as the standard "cable B-nut", adapted to fit this application. Once I go to install everything for good, I'll use blue Loctite on the screws to keep them from backing out.

Alternate air door in the closed (normal) position:

And here it is in the open (emergency) position:

When the door is open and the knob is fully extended, it looks like it might interfere with the throttle lever, but it doesn't. Luckily I thought of that before I went to all this trouble.

You can see from the above photos how it's unlikely you could close the door again, once opened, just by using the cable. The door doesn't want to slip under the little retaining tab at the top, hence why this is an emergency-only control that has to be reset via cowl removal once used. I don't particularly like that, since you can't test it before takeoff. I guess I'll put a "emergency use only" placard on it and hopefully never have to use it except at annual inspection time.

Update: See the next post for how I figured out how to work around this shortcoming.

P-mag wiring

September 21st, 2013

Time to run wires through this mess of engine mount structure and over to the P-mag:

An adel clamp on a handy screw provides strain relief for the incoming wires. And here we are looking at that obnoxious non-aircraft connector again.

I bought a cheap wire ferrule crimping kit, as recommended by certain folks who've been down this road before me. The ferrules are basically just little tubes which get swaged onto the ends of the wires.

This is what a wire looks like after having a ferrule crimped on, and then adhesive heatshrink tube applied as a further strain relief. Now at least there aren't any bare wire strands to get broken by that terrible connector.

Wires go along the upper engine mount tube, then loop down to the P-mag:

Hard to get photos of this area, too much stuff in the way:

Oil pressure hose reinstalled:

I actually didn't get around to testing these connections yet – I will eventually put 12V power on the airplane and see if the LED on the P-mag lights up.