Firewall Forward

On this page I plan on documenting the construction of anything and everything that is related to the engine.  This will include things that are physically located inside the plane, but are not installed until you are working on the firewall forward components - think throttle cable, mixture, etc.

Powerplant
    I was able to find a 280 smoh Barret Precision Aircraft IO-320X with no damage history, 9.0:1 compression, AFP injector, dual mags, B&C starter, B&C alternator, oil cooler, exhaust, and a few other things for $12.5k. Soon after taking delivery I borrowed the Boroscope from work to inspect the internals. Everything appears to be in very good shape with no visible corrosion. The dyno tests show this thing making just shy of 180 hp, which means the RV should climb like a monkey.

Hanging the Engine
    If the back of your engine looks lke a birdsnest of hoses and wires then you will want to give yourself about three or four hours to install the engine.  After installing the increadibly expensive LORD mounts, the first step was to reposition the motor so that it would naturally hang on the hoist with the rear facing the airplane.  Next, my dad and I had to remove most of the inverted oil lines before we could wrestle it into position. Starting with one bolt at a time we were able to eventually slide all four bolts into place and torque them down. With the engine installed its time to cut up the One Design engine baffling so that it fits the RV-3 cowl!

Fitting Cowl
Dipstick Tube Trouble
    The engine I purchased came with all of the necessary baffling, unfortunately it was made to fit a One Design.  This meant I would have to trim the baffle while simultaneously fitting the cowl.  Half way through trimming the baffling I noticed that the dipstick cap was hitting the cowl.  Apparently this is fairly common on RV-3's and 4's because the cowl closely hugs the top of the engine.  I replaced the original 15.5" tube with a 6.5" tube and associated dipstick.  If you decide to replace your dipstick then you might want to consider replacing the lycoming paper gasket with a nice elastic rubber gasket - Part # RG-7205 from www.realgaskets.com

Front Cowl Symetry
   Before fitting the cowl to the firewall I had to trim the front cowl flanges so that the spinner back plate matched the front cowl area and that the cooling inlets were symetrical.  This was done with the cowl off of the airplane.  I recommend searching VAF for tips on how to do this properly.

Top Cowl
   With the dipstick tube out of the way and the baffle rough cut to size it was time to fit the upper cowl.  The first step was to mount the spinner back plate to the prop flange with 2.25" pvc spacers and hardware store bolts.  The 2.25" spacers are used to simulate the Sabre prop spacer that I will eventually install.

   Next, I set the upper cowl on top of the engine with the front face roughly 3/8" behind the spinner back plate. The height of the front face was adusted by unraveling a paper towel roll that was placed between the engine and cowl.  I set the height of the front face so that the top surface of the cowl was inline with the top surface of the spinner.  

Cowl Cheeks
   The cowl cheeks can be fitted after you have the cowl locked into position.  The kit comes with aluminum bulkheads that are installed on the front of the cheeks and aluminum sheets that are to be used to form the cheeks.  Per usual, none of these parts fit as they were intended to.  The aluminum bulkheads did not perfectly match the rear profile of my colwing and the cowl cheeks looked like a huge pain to work with.  As such, I decided fabricate the cheeks out of fiberglass.  
  The first step was to fabricate foam molds that matched the contour of the cowling. These were constructed of pink R-11 foam sheets that I purchased at the Home Depot.   Next, I used string and tape to help visualize the profile of the cheeks, followed by about 3 hours of sanding per side.  This would be a 15 minute job with a hot wire cutting machine.
  Next, I drilled and clecoed angle pieces on the top and bottom of the foam mold.  These will later serve as clamps, but are removed until the glass is layed in place.  I then wetted 3 layers of 8 oz bidirectional (bid for people in the know) glass, and layed up the mold.  The angles are then repositioned over the glass, resulting in a tight wrinkle and bubble free layup.

   

FAB
   I am using an Airflow Performance FM150 injection system that came with a 90 degree forward facing adapter elbow.  Unfortunately, this setup will not fit the stock FAB (filter air box).  I could have removed the injector and sent it back to AFP to be reconfigured for use with the stock FAB, but instead I decided to make what I had work.  
   The first step was to select an air filter that had at least as much media area as the stock filter (50 square inches).  I chose a fairly large cone filter with about 63 square inches of media area.  Next, I had to calculate the inlet area that would provide adequate flow rate for an IO-320 @ 2600 rpm @ stall speed - worste case.  You can find the formula for this online someplace.  I came up with an inlet area that corresponded to a diameter of a little less than 3 inches.  I then used this area to calculate the diameter of the FAB at the surface plane of the filter and the rear plane of the filter.  
   Next, I started on the foam plug mold.  I cut out 3 circles of aluminum equal to these 3 diameters and drilled holes in their centers.  The block of foam was then cut into two pieces at the axial location of the front face of the filter. The cirles were then placed in their correct axial locations and held in place with a threaded rod that was later used as a mandrel.  The entire assebly was mounted in my drill press and turned down to the shape of the FAB.  This plug mold was then covered in glass, yielding the final FAB shape.  A mounting plate that would be used to connect the FAB to the AFP servo was then attached to the bottom of the FAB with nutplates. Wow that was a lot of work, but now I know how to make a plug mold!

 

Lower Cowl Modifications
   My fancy FAB is too far forward to fit in the stock air scoop and will require a new larger scoop.  In other words I get to log a few dozen more hours of fiberglass work.  

Oil Door
   The oil door was a little trickier than I thought it would be. The first step was to stiffen it up a bit with two layers of 8oz bid glass.  I then cut a hole for the new door, installed backer plates, locking mechanisms, and the hinge.  In order for the stock locking mechanisms to work properly, you will have to slot them with a dremmel tool as shown below.  The door latches come down and sort of an angle instead of straight down, rendering them useless unless they are slotted.  


Exhaust
   The Veterman 4 pipe exhaust system is truly a piece of art.  I hope it sounds as good as it looks.

Firewall
   All firewall penetrations will be through steel eyeball fittings.  They are expensive and time consuming to install, but should prevent a fire from making its way through the firewall.  Seven total.  I also decided to mount the battery, starter and master solenoids on the firewall.


AFP Purge Valve
  The AFP injection system has a purge valve located between the injector servo and the spider valve.  This valve allows you to purge the stagnant hot fuel that is located in the engine compartment before attempting a hot start.  The hot fuel returns to the left fuel tank via a separate line.  This valve requires a vernier control cable for operation as well as a rod end bearing link to attach the cable to the valve.  The purge valve must be oriented sideways in order for it to fit under the cowl.  I sent my entire assembly back to airflow performance to have them reposition the valve so it would fit under my cowl.  The horizontal orientation forced me to run the fuel supply up through the right inlet baffle ramp and over the #1 cylinder.

Inverted Oil
   A Very tight fit...more to come

Cabin Heat
   I had to find a new spot for the cabin heat valve after the inverted oil three way ball valve took its spot.  One solution, and the one I chose, is to locate the valve inside the right cowl cheek.  The control cable can then be ported through the sideskin just aft of the valve.