Aft Flapper Seal Inflght Test Report #2
I've been communicating with Pat Hatch about adapting his flapper seal to the aft canopy-skirt-to-fuselage junction, and would like to relay a few relevant points from that discussion.
The best canopy seal is to fit it precisely in the first place. With due respect to the builder of mine, this is an area in which he could have done better. Getting help building a new canopy isn't an option at the moment, so I'm stuck with finding a solution to deal with what I've got.
Airflows around the canopy are a complex mixture, and the fit of each individual canopy presents a unique sealing task. That said, the differential pressure along the sides is substantially greater than along the aft skirt, and I've verified that inflight with a short piece of yarn on a dowel. Before I added any seal, when the tip of the yarn was held near any side gap from the windscreen all the way back to the end of the canopy sill, airflow out of the cockpit grabbed the yarn and streamlined it like a coon tail on car antenna
That's why I added a piece of angle aft of the UHMW blocks to the end of the sill. This provided a vertical surface for gluing the seal, so that mine is a continuous piece from the windscreen all the way back. I also cut the EPDM rubber strip from McMaster-Carr right down the middle to create two 1-inch-wide, 36-inch-long strips. The little extra height ensured that the seal doesn't come close to popping out and under the bottom edge of the skirt, and the rubber conforms nicely to any variations in the inside surface of the skirt like a rivet, for example.
Inflight testing with the yarn has confirmed that the aggressive outflow of air from the cockpit continues up and around the aft skirt of my canopy for about 10-12", where it tapers off to nothing. The boundary gradient from high flow to zero is very sharp, and I think this is due to a combination of two factors: 1) the differential pressure between the cockpit and the atmosphere is much less back there, and 2) this is where the skirt on my canopy is the tightest against the fuselage.
One flow discovery really surprised me, however. When comparing the airflow from the center rail at the doghouse (which on my airplane is effectively zero) to the left and right along the aft skirt, the left side outflow is also zero, but the right side has a point a few inches long at which the yarn flutters
forward into the cockpit. That is occurring at a point where the canopy fits the tightest, but there is a short section in which the curve of the skirt lifts very slightly from the fuselage.
My explanation of this is that airflow across the canopy creates a suction that deforms it just a bit and lifts the trailing edge of the skirt.
But here's the really surprising part: if you think of viewing the aft canopy skirt from inside the cockpit inflight, from the center rail to either side for about 10-12" or so, the flapper seal is laying in its static position just like on the ground, and the gap above the seal to the bottom of the canopy skirt is about an inch or so.
And yet with the exception of that short section in which air flows into the cockpit, the net flow is zero.
It's important to note that I may have failed as a test pilot in assessing the effectiveness of the aft seal, however, because I changed more than one thing from the previous flight by adding some sticky-back pile Velcro to the trailing edge of the canopy skirt to fill in the gaps.
My canopy sits tight against the fuselage from the doghouse to about 18" around the skirt on both sides. At that point, a very slight gap begins, increasing to maximum at the bottom rear corners of the skirt, then begins decreasing along the horizontal skirt for about 6-8" where it reduces to nothing.
To determine where to put the Velcro, I slipped a piece into the gap between the skirt and the fuselage and worked it toward the point where the skirt was tight against the fuselage. As soon as it began to bind, I marked that spot, repeated the process at the other end of the variable gap, then attached a single layer strip of Velcro to the underside of the skirt along that perimeter. I repeated the process to add additional strips until the gap was filled.
This has to have an effect on the flow of air through the cockpit, but I can't separate that from having extended the flapper seals farther beyond the edge of the fuselage skin overhang before this second test flight.
Those two changes in combination, however, resulted in the following summary conclusions:
1. From just aft of the UHMW blocks where I added length to Pat's rail seals and up to about the first 10" of the fuselage skin overhang, it is clearly apparent that the rearward airflow is shoving the seal tight against the canopy skirt. If I pull the seal away from the skirt and drop the end of the yarn telltale into the gap, the yarn is visibly pulled to the rear with the escaping air. It's important to note that with my relatively ill-fitting canopy, this is also the area that has the largest gaps to contend with.
2. From there up to the doghouse, the seal appears to be in its static position with a fairly large gap between the seal and the canopy skirt. The same is true from the doghouse down to about 10" from the canopy sill on the opposite side.
3. When the telltale is placed in this gap on the right side of the aircraft, it streams forward. When placed in the vicinity of the doghouse, it falls straight down. From the doghouse all the way around the aft skirt on the left side of the aircraft to about 10" from the canopy sill, the telltale is static.
4. The combination of a flapper seal and pile Velcro has greatly reduced the flow of air out of the cockpit all around the canopy.
My original goals for the canopy were 1) get the best fit I could without having someone who knows what they are doing build me another one, 2) install the tipper mod, and 3) seal it better. I've accomplished all three, and at this point I see no reason to try a different approach to using an aft flapper seal (like attaching it to the underside of the skirt rather than on the fuselage).
For me, the most useful lesson learned is that Pat's original idea of the flapper seal can be extended along the entire perimeter of the canopy-to-airplane junction. There may be areas along the junction in which the seal isn't necessary because the canopy fit doesn't need it, but that's one of the advantages of the flapper idea to begin with. It reacts rather than being a static solution.
Anyone's final results will vary based upon the fit of the canopy, but there's no question in my mind that wherever the air is flowing out of the cockpit, Pat's seal can be an effective method for dealing with it.
Tosh