What have I gotten myself into!: Fuel Injection Conversion

Engine Removal

Time to get started with one of the least glamorous, but as you'll see not less dramatic part of this fuel injection conversion.

Removing JT's engine turned out to be quite an ordeal, but not because I had never done it before, so I thought you might find my experience to be a bit… refreshing.

The short of it is that the engine attaches to the engine-mount with four huge bolts, and the engine-mount in turn attaches to four aluminum extrusions using eight smaller bolts, but this fact alone doesn’t do justice to the complexity of the endeavor.

One O-320 ready for travel

When the Rutan Aircraft Factory closed its doors in the early 2000s, it left everyone to their own ingenuity and resourcefulness in dealing with the challenges that inevitably crop up during a build as involved as this, and over the years standardization morphed into a four letter word whose original positive connotation was lost to time.

That “no two Long-EZs are ever the same” is such an understatement to anyone who has seen more than one example, that you would be excused for thinking that factory options must have been numerous, when in fact there was only one set of plans ever published (albeit with many corrections). In spite of that, anytime you get a chance to encounter one of these incredible machines, the marks of individualism and personalization are always in full display, and customization is a point of great pride achieved at very high cost in both time and effort.

These planes could hardly be any more different up close

Knowing that every other Long-EZ will be somewhat different, sometimes significantly, I will concentrate on JT’s hardware only, and what it took to get her stripped… of its power-plant.

Right off the bat one has to decide whether he wants to remove just the engine, or the entire engine and engine-mount assembly. The difference is not trivial, and it highly depends on the type of maintenance you are looking to perform.

If you had to remove the engine to do a complete overhaul, you would probably elect to remove only the engine, and leave as many of the subsystems untouched as possible, like fuel and oil lines, starter relays, electronic ignition computers, engine monitor computers, etc.

On the other hand, if you needed to do a more thorough job, like replacing the firewall, upgrading the fuel system, replacing the engine monitoring system, etc. you might want to remove both, possibly together, to facilitate doing the work.

In JT’s case, the changes would be massive, so all of it had to come off.

But, why?

You see, one of the issues I had been living with since the beginning, was that the aluminum firewall’s protective coating had been slowly flaking off for the past two decades. The exposed bare aluminum was ill equipped to withstand the heat of a gasoline flame, leaving the prospect of an engine fire even more dreadful than it already was.

Heat resistant coating has seen better days

To remedy this drawback, I wanted to manufacture a new firewall out of stainless steel, or perhaps titanium, but replacing the firewall meant that everything had to go, and I mean EVERYTHING.

All of the engine temperature and pressure sensors interfacing with the VM-1000 engine monitor, including the Engine Computer, and the Instrument Panel Display would be removed and sold, and a new modern GRT-Mini Engine Monitor system would be installed in its place. This will allow for software updates, audio alerts, recordings, easier leaning, PFD backup, better readability in direct sunlight, customization, multiple pages, technical support, repairs, etc. No, it wouldn’t be cheap to buy, nor easy to install, but in a now signature move, I’d kick this issue down the road for later-solving like so many others.

The VM-1000 engine monitor is a quarter of a century old

There's nothing like the smell of new avionics in the morning 👃🏼

Removing the engine began with disconnecting everything that went from the engine to the firewall, such as throttle/mixture/carb-heat cables, alternator and starter cables, numerous fuel lines, quite a few oil lines, primer lines, EGT probes, CHT probes, electronic ignition leads, Manifold Pressure hose, magneto leads, and a few more I cannot think of right now.

Demolition day

CHT and EGT probes

Fuel primer lines

Chris and I supported the engine with a crane while removing the four bolts connecting it to the engine-mount, then we slowly pulled it back while checking for other tethers. Pretty soon we reached the point of no return, and the first time we saw the engine dangling from the hoist was pretty dramatic.

No turning back now

The scary side of commitment

With the engine out of the way, I was impressed at the amount of hardware still connected to the firewall, and made a mental note to relocate them and simplify the firewall as much as possible during a future reassembly.

Effective, but way too busy for my liking.

"What's all of this stuff anyway?!"

Took me a while to map it all out

Until I could remove the engine-mount off the plane (more on that in a moment), and fabricate a custom engine-mount to automotive engine-stand interface, I rested the power-plant on a thrown together temporary structure made out of 2x4s I had in the hangar, and called it done for the time being, kicking another can into my distant future.

Improvised engine stand

It would be a little while before I can use a real engine stand

Removing the engine-mount turned out to be very troublesome. While seven of the eight nuts and bolts were easily removed, number eight (there’s always one) proved to be a major pain in my ass. Let’s take a look at the bolts so you can understand.

Lower left engine-mount to extrusion interface

Top left engine-mount to extrusion interface

Top right engine-mount to extrusion interface

Lower right engine-mount to extrusion interface

In the last photo you can see that the aileron control-rod pivot-plate (aka CS124) has a special shape that allows for left and right movement near the bottom right engine-mount. Unfortunately the proximity to the aluminum extrusion prevents the horizontal bolt from being removed, even when turning the control stick fully left, or even disconnecting the control linkages altogether. No amount of counterclockwise rotation will allow for enough clearance to remove that bolt.

Highlighting the lack of space to remove the horizontal bolt

Blue color highlights the lack of available space

Even with the plate rotated 90º the bolt cannot be removed

The antidote to this dilemma is swapping the horizontal and vertical bolt positions, but this can only be done during the initial build, the way Mike Toomey did on Chris’ plane… Smart!

The horizontal and vertical bolt hole locations were swapped here

My idea was to pull the pivoting plate backward about 1" from the firewall, then remove the bolt by sliding it between the firewall and the newly repositioned plate.

Looking in the plans this seemed possible after first removing the bolt that fastened the swiveling plate to the control rod. Unfortunately this bolt lived on the other side of the firewall, in one of the least accessible locations on the plane, up high in the aptly named Hell Hole.

Even finding this bolt proved to be challenging

The bolt from Hell

Not only is this bolt impossible to reach, but it is also impossible to see. The photo above was taken by sticking an iPhone through the hole in the passenger’s seat-back.

Hell Hole is behind this top hinged Engine Monitor computer (location of the bolt highlighted)

What I really needed was an access panel on the bottom of the plane, below the Hell Hole, that would allow for direct access. The access panel would end up right in the middle of the engine NACA air intake, a super thin molded carbon fiber piece epoxied to JT's bottom.

To make a long story short, NACA ducts are notoriously difficult to make work as any irregularities in the surface smoothness would trip the air into avoiding the duct altogether, compromising JT’s cooling airflow. So, cutting a hole in the middle of one, especially a very thin one, is not to be taken lightly, and great care must be taken to prevent warpage of the carbon fiber, and restore surface continuity and smoothness of the finished cover plate.

But the cover plate is a problem for future Marco.

Let’s cut some Carbon!

A template for cutting will allow for precision should I need to machine the new cover

Testing the location of the new access panel

I wanted the hole to be in the flat surface of the NACA to make the cover easier to produce

Just another frightening moment

Main drilling is over

Now, buzz saw all pencil lines.

The plan was easy to execute. The harder part was choosing where to put the hole.

A little sanding later

There it is

I wish I could say that this alone did the trick, but there was one last wrinkle in this plan. Once I unscrewed the nut, and pushed the bolt up to remove it from the control rod, the bolt hit the fiberglass ceiling (bottom of the Center Section Spar), and couldn’t be removed.

"Well, F@#% me!"

Ended up building myself an improvised bushing puller, and disconnected the rear Joystick from the control torque tube, rotated the tube sideways, and was finally able to remove the blasted bolt.

Using a long bolt and a few washers to remove a pressed bushing

This bushing will have to be pressed into the torque tube by reversing the puller action

Success at last!

This allowed me to push the aileron pivot plate backward about an inch, which in turn allowed me to remove the last bolt from the engine-mount, and remove the bloody mount.

Pivot plate pushed back leaving enough room to remove the last engine-mount bolt

Engine-mount removed

After that it was just a matter of removing the rudder pulleys, electrical fuel pump, oil and fuel lines, aileron servo, and a few more odds and ends.

Removing an oil pressure manifold


Some wires had to be harmed in the process

Rudder cable pulley removal

Andair gasculator, and old gyro vacuum pipe (capped) were discarded.

This pipe fed vacuum all the way to the original instrument panel #1

I eventually repurposed the vacuum pipe hole for the new throttle push pull cable

The aluminum firewall came off with little effort, but harbored a terrifying surprise.

Aluminum firewall removed and flipped over for inspection

Looking at the picture above, you can see a whitish cloth-like material attached to the back of the firewall. That is a ceramic heat rejecting mat called FiberFrax, and it’s supposed to be completely white.

Note how the further down you look the darker it gets?

What happened throughout 20 years of flying was that engine oil, spilled at every oil filter replacement, dripped to the top of the NACA diffuser (the part of the NACA duct entering the engine compartment) and collected there. Because of the many systems attached to the firewall, detection and cleanup were impossible (I tried many times), and the spilled oil slowly wicked its way up the porous ceramic mat by capillary action.

The big question in my mind at this point was… is this FiberFrax still able to reject heat, or would it go up in flames while in contact with the structural fiberglass spar?

Only one way to find out…