Started Filtered Air Box – 4 hrs

Sunday April 14, 2013

Today I started working on the filtered air box, also known as FAB. I am hoping this does not turn out to be as big of a project as the engine baffles but it didn’t take long to encounter the first problem. The Airflow Performance fuel injection kit came with a nice gold anodized FAB adapter plate that is pre-drilled to fit the clamping collar that attaches it to the servo. They even made it thicker (.090) than the standard top plate – I assume because of the many reports of the adapter plate cracking. I clamped the adapter plate to the air box top plate and mounted it to the servo for a check fit to the cowling.

First I found that the air box top plate is too long and interferes with the cowl but this was to be expected. So I cut an inch off the front end and tried again. This time nothing interfered and I was able to see the impact of the offset of the fuel servo from the centerline of the cowl. I could rotate the air box top plate to point directly at the cowl air inlet but there was only about 3/8 inch clearance between the air box top plate and the inside surface of the cowl along the side of the air box. I was not comfortable with that little clearance because of the relative motion that occurs at engine start up and shut down. So I began investigating options to slide the adapter plate sideways a bit to increase clearance. I want at least an additional 1/4 inch or a total of about 5/8. Unfortunately when I tried to slide the AFP adapter plate that far to the side two of the six holes that will mount it to the air box top plate have less than acceptable edge distances. After looking for options for a while I finally concluded that I need to make a new adapter plate that has the center hole offset by at least 1/4 inch and is wider to maintain hole edge distances. I used the AFP part as a template. In this photo you can see the new plate underneath the gold AFP adapter. I roughed out the new adapter with extra material on the near side for the mounting holes.

I sanded about 1/8 inch off the side of the AFP plate so I could slide it a little further and transferred the center holes for the fuel servo to the new adapter plate. Here is the new adapter plate mounted to the air box top plate. I used two holes to connect them temporarily with clecos. You can’t see it in this photo but I also trimmed about 1/4 inch off the inside edge of the top plate center cut-out on that same side to allow the air filter to sit flush against the adapter plate all around. This means the shape of the air filter will be slightly elongated on the left side compared to the standard oval race track shape. I don’t see any way this can cause a problem.

When I put the new setup on the engine for another fit check I measured a clearance right at 3/4 inch which is the thickness of the middle knuckle on my right middle finger (precision gauging). This is much more comfortable to me than 3/8 inch. This is a lousy picture looking in from the exhaust exit of the cowl but you can get the idea of the clearance along the side of the silver top plate. The clearance will decrease about 1/16 inch when I install the fiberglass air box.

As far as the new adapter plate is concerned, I made it from .062 alclad 2024-T3 stock. I considered using .090 stock like the AFP part but there are reports of even those cracking. I suspect the problem is also related to the stiffness of the rubber coupling between the FAB and the cowl inlet which puts loads on the FAB every time the engine is started but that is really a hypothesis. Another theory is the vibrations of the engine are causing fatigue failure. So my approach is designed to address both theories.  I plan to make a doubler out of .032 thick 2024-T3 alclad and bond it to the new .062 thick adapter with a thin layer (~.010) of Proseal. That will give me a total aluminum thickness of .095 for strength and the Proseal will add some damping to the plate to attenuate vibration response. I will also make the rubber coupling between the air box and the cowl snout as compliant as I can. Of course I will inspect the new adapter part regularly to look for evidence of cracking but I think this design approach gives me a good chance of avoiding the problem. We will see.