Bill's Aircraft Factory

First Flight

Trips

Horizontal Stabilizer

Vertical Stabilizer

Elevator

Trim Tab

Rudder

Wings & Fuel Tank

Ailerons

Flaps

Fuselage

  Page 1

  Page 2

  Page 3

  Page 4

Engine & FWF

  Page 1

  Page 2

  Page 3

Instrument Panel

Electrical System

Building in the Basement and Moving to the Airport

Dynon Autopilot Installation

Things to Consider

Tools

Products

Tell me again

Helpful Links

Plane Pictures

FAQ's

E-mail:

bill (at) repucci (dot) com

Engine Rebuild and Installation

aka Firewall Forward

Caution - Man running (with) power tools

E Previous Page

 

E Band clamps are used throughout the engine installation.  This is a simple way to safety wire them.  (11/4/07)

 

F A significant number of Adel clamps were used in my installation.  The purpose was to keep the wires, hoses, etc. from rubbing against anything.  Not only where the clamps used but so was lacing cord and zip ties as well.  Left to right, the pictures are: wiring behind the right cylinders, looking down at the right side of the engine mount, looking down on the center of the engine mount, and looking down on left side of the engine mount.

The thick black wire that runs across the top of the engine mount is not a wire but the vacuum advance line for the P-Mags.  Also note, due to the lighter weight of the O-290, the engine mount is 2" longer than those used for the O-320.  This provides a full 12" between the engine and the firewall.  I've also written the oil filter torque spec and Hobbs time on top of the filter for easy reference.  (11/4/07)

 

E This is a shot of the lower right firewall.  Note the Adel clamps holding the brake line as it crosses over to the right gear leg.  The second picture illustrates how I ran the brake line to the caliper.  Note how little wear there is on the tire after 65 hours.  That is one big advantage to the low takeoff and landing speeds that the -9 has.  (11/4/07)

 

F Side shots of the cylinder heads.  Note in the first picture how zip ties and some tubing were used to separate the EGT wires from ignition wires.  (11/4/07)

 

Cooling your engine properly is a challenge for all aircraft builders.  More so for those of us with tightly cowled engine compartments such as those found on the RV's.

After 147 hours and trying everything I could think of bring the temps of my #3 cylinder in line with the others I read an article in the RViator detailing how Dick VanGrunsven was struggling with the same issue on his new RV-10.  One of the things he tried was to smooth inside of the top cowl, where the airseal fabric mad contact.  The idea being that by smoothing the surface of the cowling, you will get a tighter seal and less air will escape between the cowl and the fabric.

E That made sense to me and when I pulled my top cowl off you could see where the air was blowing by the airseal.  Leaking air means less cooling air to go down between the cylinders.

F To solve this problem I sanded down the inside of the cowling where the airseal fabric made contact.  I then applied a layer of epoxy and micro-balloons.  When that hardened it was sanded down and a layer of pure epoxy was applied over the micro-balloons to seal it.  These pictures were taken before sanding down the micro-ballons.  (5/16/08)

 

Still struggling with cooling the engine but the good news is, the end looks like it might be in sight.

I'm making these changes one at a time.  Otherwise I will never know what had the most impact.

After the change above, I did see a difference in my CHT's but not enough to make a difference.  When I started this quest to balance my CHT's I was seeing the following temps in cruise:

#1: 363

#2: 358

#3: 408

#4: 364

After working on the top cowing I played around with adding air dams made of aluminum tape to the front of cylinders #1 and #2.  This worked to balance out cylinders 1, 2, & 4 but #3 was still running about 45 degrees higher than the other three.

Next up was to dded a second washer behind the #3 cylinder.  From the time I first installed the baffles, there was one washer there and I had heard that adding a 2nd one could only help.  This change did help some and the temps on 1, 2, & 4 came down to within 2 degrees of one another; however, #3 was still higher by around 35 degrees.  (Every change seemed to reduce the temps by five degrees, I can only hope I don't run out of ideas before getting the temperature of that cylinder down where I want it.)

Better but not good enough.  The next trick was to make a new corner piece out of airseal fabric and rivet it in the corner of cylinder #3.  That was an easy change, other than having to order more of those big headed rivets from Van's.  Subsequent test flights didn't indicate any noticeable change in the CHT's.  I suppose the original corner seal was fine.

Somewhere in here I thought maybe I was having an intake leak, so I changed out the intake gasket and rubber tube down at the sump.  No change, other than spending a few dollars with Aircraft Spruce for the parts.

The next test was to find out if my CHT probe was any good.  This was simple enough, just a simple swap of the #1 and #3 probes and go fly.  No joy.  :(

E What next?  After looking over the engine a friend commented that the O-320 baffles didn't wrap up tight against the cylinder base of #3.  Sure enough, after getting up on a chair and taking a closer look myself, you could see there was a one inch gap where the baffle did not wrap tightly against the cylinder base.  I emphasize "base" because the baffles fit the head just fine.  That must be where one of the major difference in the O-290 vs. O-320 cylinders is.  I quickly fabricated a small piece of aluminum to fill that hole, pop-riveted it in place, and RTV'ed every gap I could find anywhere on the engine.  Two subsequent test flights found my CHT"S running:

#1: 339

#2: 341

#3: 365

#4: 339

Get this, those temps were with OAT's of 70 and when the temps above were recorded, the OAT's were in the 40's and 50's so I'm definitely making progress but I still have that one outlier.  Unfortunately I didn't take a picture while running at cruise power on this test day and don't want to post erroneous numbers.  The good thing is that the CHT"s for 1, 2, & 4 where within two degrees of each other and #3 was the same 20 to 25 degrees out and better yet, it was below 400.  Progress has most definitely been made!

 

F The last change I wanted to make was to trim the air outlet back a little bit.  Remember, I live in the Southeast and from June to September, OAT's in the mid 90's is common.  When I fit my cowl, I realized there would be a little tail of the air outlet tunnel that would stick back behind the firewall.  Thinking this would look cool, I left it in place.  In researching my cooling issues, I found that it may look cool but this "tail" doesn't help cool the engine.  About five minutes with a body saw and belt sander and the air outlet on the lower cowling was even with the rest of the cowling.

 

Stand by for the results of more test flights.  (This sure makes me happy that the plane isn't painted yet!)

  (7/2/08)

 

E Previous Page