Static Port Installation Sequence

After reading (and before flying!) the "squirrely nature and high speeds required" to land this airplane on our 2000' grass strip, I wanted to upgrade N133RM to add Angle Of Attack.
(Note: I later found out on the very first flight N133RM is NOT an immense handful to fly; rather it is quite nice!)
AOA is part of the Dynon D10A EFIS along with AutoPilot control and a bunch of other things. So you have to use the Dynon Pitot/ AOA probe. That means adding a static source (if I would eventually remove the original pitot/ static setup originally installed in 1993.)

The parts kit came from Vans Aircraft.

Here is the location; 10 1/2" aft of the firewall & 4" below the top longeron. Be careful you don't drill through any structural framework!

RTV weather-proofs the installation. Safety wire gives something to insert the lock nut onto when feeding through the inner fuselage skin. See below on this little detail.

Stacking the nuts provided depth so the thin lock nut didn't recess back into the nut driver...

Taping the safety wire in place locked it into position to switch positions and work from the inside...

[Note that Roy built this airframe with an inner mahogany skin on top of the 5/8" structural frame. This acts as a inner sheet gusset.]
The safety wire provided a guide to slide the nuts onto the obscured static port hose end & threads.
Then use the nutdriver to lock the assembly into place. Finally slide the static hose onto the installed port end. Done. Or so I thought...

It turns out this location gave me too much pressure on the static port. My readings, compared to the legacy steam gauge were way off.

I temporarily placed a blocking plate behind the port to adjust the static pressure like the RV guys do but it didn't work. 2 plates was better at cruise but the low end was off by 20 MPH IAS. I tried these diverter plates in many configurations and locations. Some were better than others and none were close enough to accept.

As a test, I plumbed the Dynon D10A into the steam gauge legacy static system Roy built and it indicated air speeds matching within 2-3 MPH. That means finding a new location for the static ports. Rats...

At the 2021 KLXT KRGathering, I got photos of Roger Baalman & Robert Pesak's static installations. Both those guys installed their static ports toward the rear on the outside of the skin, like some Cessna and Piper products. Roger said since it works for the factories, that was good enough for him. Who am I to argue detail points with a Bronze Lindy award winner? Ok, so much for the new location...

-->February 2022 Update
After more flights still fighting this issue, I gave up trying the forward mounted static port and resigned myself to drilling more holes in the airplane. I ordered a Dynon Pitot/AOA/Static install kit. The installation instructions say "Dynonís static port has been designed to be mounted proud of the aircraft skin, in order to give more accurate readings." It is designed to use on metal skinned airplanes so the finished installation depth doesn't match up for our 3/32" plywood fuselage skins.

I called and spoke with Jeff at Dynon Tech Support. He said to just install the whole flange outside the skin. That's not what I want. So I'll machine the flange thickness for the dome face to just protrude through the plywood, bond it inside and use this skin doubler inside to further support the new 1/2" hole.

Sorry, I didn't get a photo of the lathe "turned-thinner" flange.

The ports installed on my first try at the top of this page were just aft of the firewall. The locations Robert and Roger used are behind the cabin. I measured 45" back from the canopy's rear edge and halfway between the floor and the top longeron. I hated drilling into Roy's pretty paint job and you gotta do what you gotta do. Finish with a step drill bit.

Inside, the two ports are T-88 epoxyed in place, doubler rings installed and a clamp stick forced the ports into firm contact with the fuselage skins while the epoxy cured.

 I'm really satified with the end result. I ran the flexible tubing and tested the installation; it held a 6" water column for 12 hours with just a 3/4" bleed down. As Mark says, "Close enough for a KR."

Now for a test flight to make sure it provides an accurate static pressure source.