Chapter 18 Part-2 Canopy

Saturday, Feb 2, 2018

Time to build the canopy frame which means attaching the turtleback to the fuselage which also means first attaching the center spar to the fuselage which was done in Chapter 14 Center Spar. This also means attaching the plexiglass canopy to the turtle back and creating a canopy frame around the plexiglass canopy to provide structure for hinges and latches. Note: I'm adding this to the website in Feb, 2019, a year later, so I apologize if the descriptions leave out some detail.


The first step is to install the upper firewall which was made years ago. I don't have a picture but the firewall was just slightly too small. There was an 1/8" gap between the firewall and the turtle back no matter how much I tried to stretch or shape the turtleback. I could have filled the gap with flox but decided instead to make a new firewall. The firewall is made up of an expensive piece of plywood and a ply of BID on each side with two aluminum hard points installed for the engine mount. I made the hard points larger this time to allow for engine mount tolerances. Two wood uprights bolted to the lower firewall hold the upper firewall in place with bondo. As shown in Fig-27 there are several 2 ply BID layups to glass the firewall to the center spar. A plywood firewall doesn't sound like it would offer much protection but there will also be a layer of insulation called Fiberax and then a sheet of stainless steel over that.




The turtleback was glassed to the firewall and the longerons of the fuselage (only from the canopy cutline aft to the firewall) with flox and 2 plies of BID. A foam fillet was microed into the corner and 5 plies of BID was wrapped around the corner overlapping onto the engine mount insert.


With the turtleback glassed in place there was no way to avoid having to work with the plexiglass canopy. It amazed me how flexible it is, it is very noodley, it will widen and become shorter or narrower and taller. With some blocks screwed down on the bench I could control the width and then determine the height to see how it would match up with the turtleback. Also to determine at what location the canopy slope leveled out.



I did a lot of measuring, checking and fussing with the canopy to come to the conclusion that it was too big, even though I had increased the width and height of the turtleback. This is one of Tod's Texas Canopies so it is made larger than the standard canopy. The canopy needed to be trimmed at the aft edge so that template B.L.0 would position the nose of the canopy at the correct distance and angle from bulkhead F28. This first picture shows that the canopy is too far forward necessitating trimming the aft edge.


The group provided a lot of input regarding trimming the canopy and leaving a gap at the bottom edge, the discussion can be found here, Cozy Group Canopy Discussion.


With Tod's Texas canopy and a 1.75" wider and taller turtleback I trimmed 2.25 inches off the aft edge and .25 to .50 inches off the bottom edge. To mark an accurate line on the canopy I applied electrical tape to the turtleback where the aft edge of the canopy needed to be then with the canopy in place applied tape lining up the tape edge with the tape on the turtleback. I used the Fein sander to trim the canopy, making very light impressions going over the same line, 20/30 times until the cut was all the way through. It was very nerve racking not to slip and scratch the plexiglass.


I made full size side profile templates from hardboard so that the canopy frame would flow nicely into the top of the nose, this is why the nose section has been left unfinished for so long. The plans have you finish the nose section much earlier which many builders find it difficult to shape the canopy frame into the nose. Lynn and I will hot wire the top edge of the canopy frame forward of the plexiglass all the way down to the nose using these hardboard templates.


I pulled the fuselage outside so I could stand back and take a good look at how the plexiglass canopy fit with the turtle back and get an idea of the overall flow. It looks a bit stubby but I'm hoping that the canopy frame and nose section will help smooth things out, we'll see.


This shows the bottom edge of the plexiglass canopy trimmed with a minimum 1/4 inch gap to the fuselage. I taped a 3/8" piece of foam on top of the longerons to allow a gap for a seal between the canopy frame and fuselage. In the Cozy Group Canopy Discussion Marc Zeitlin highly recommended doing that. Also a notch is cut at the forward end to allow clearance for the instrument panel.


Before glassing the plexiglass canopy to the turtle back some temporary support structure is put in place to allow a surface upon which to build the canopy frame. This will provide a platform for the canopy frame foam.


The support structure and everything else possible is covered in duct tape. Also the plexiglass canopy is covered in 2" wide plastic plumber's tape, I opted for this instead of the Spraylat, both options were offered in the plans. Seemed to me that the tape provided better protection and some builders complained that if the Spraylat was left on for a long time it can be very difficult to get off. Others said that you just need to apply several coats to to make it thick and it's no problem. Considering how slow a builder I am I went for the tape.


Next is to flox the aft edge of the plexiglass canopy to the forward edge of the turtleback. The challenge here is to get the plexiglass canopy to conform to the shape of the turtle back, it wants to pop out at the sides and the top which the plans say will happen. Per the plans I bondo'd a board to each side so I could wedge the canopy to the turtleback and weighed down the top. With that it worked really well for a good bond.


After the flox bond of the plexiglass canopy to the turtleback cured a Ply of UNI is layed up at the bottom edge of the plexiglass canopy and overlapped onto the turtleback. Once that is cured 3 plies of BID is layed up over the plexiglass canopy to turtleback joint. (Sorry, don't seem to have pictures of that.)


I had some concern that the plans didn't show applying micro between the canopy frame foam and the bottom edge of the plexiglass canopy where the single ply of UNI is. The group said it would be fine to apply micro just not up to the top edge of the foam since that is dug out to create a groove where flox and BID will be layed up to create a hard edge between the canopy frame and the plexiglass.


The foam that comprises the canopy frame is made up of 2 inch wide pieces with a lip that fits under the plexiglas canopy. It was a bit challenging to make them but after a while you get the hang of it, using an angle finder you're able to mark the angles of the plexiglass since it's constantly changing. At the forward edge of the plexiglass canopy I couldn't shape a piece that would have a close enough fit so I made a template to hot wire the foam and micro'd a 1/4" of foam to the bottom to create a lip. It still took a couple templates to get the shape right.



Before rounding the corners I wanted to get the top profile established with the side templates. I microed some blocks of foam together to fill in the nose section.



The piece of foam that covers the canard was a little bit of a challenge to ensure that it sat at the correct height. So I created an inverse template from the original canard template to create a void for the foam to sit on top of the canard locating it accurately.




With all the foam blocks in place we could hot wire the top profile using the side templates. I hot glued some dowels to hold the templates out at the front otherwise the slope would not be correct.


I then shaped the corners trying to keep everything looking symetrical. I think the nose may look more blunt than other Cozy's or LongEZ's I've seen. But at some point you just have to call it good and move on.


It seems to be a pleasing shape which is the typical requirement called out in the plans.


Before glassing the nose the access doors are glassed over the foam to get the exact curvature of the nose. So I marked the foam where the access doors would be and then covered the area with packing tape and layed up 4 plies of BID over the marked areas. After the access doors cured they were removed (along with the packing tape) to be used later. Now the entire nose area could finally be glassed with 2 plies of BID.


Now the canopy frame could be glassed. I was a little concerned that the plans didn't mention allowing to overlap the plies of BID along the sides, I queried the group and the consensus was that it was fine as long as the overlaps were not in the same place causing a bump. It didn't seem possible to me to layout the pattern for one side of the canopy at a 45 degree angle on a roll of BID.


I hogged out the foam to create a groove at the base of the plexiglass, a 1/2" wide and deep. The first ply of BID is layed into the groove and then the groove is filled with flox and the second ply of BID is layed up over that to create a hard lip around the plexiglass canopy. 2 plies of UNI go over that for a total of 4 plies of glass on the outside of the canopy frame. Shaping the canopy frame foam was a bit challenging, again to maintain symmetry between both sides.



A curved recessed channel is sanded around the forward section of the canopy frame where it will be cut later. The recessed channel allows for a lip to cover the cut line.



The canopy frame is glassed with the two plies of BID, with a BID/flox sandwich at the base of the plexiglass canopy and 2 plies of UNI over that. It always feels good to finally have the fragile foam covered with glass to make it solid. The pictures make it look pretty smooth but when you run your hand over the glass it has hills and valleys. Hopefully I'll be good in the finishing stage to smooth this all out. When shaping the foam it's hard to smooth it out perfectly because you can only sand the foam away to smooth it out to the point where you've suddenly created a flat spot that doesn't look quite right or symmetrical with the other side. So it is what it is and you have to move on.



I had seen on Marc Zeitlin's site that he had made several improvement modifications to his plane, one of which was to have a removable canard cover. The reason for having a removal canard cover is to be able to have access to the screws that hold the "fuselage top" (the area forward of the canopy frame cut line) on. The fuselage top is removable to gain access to the area behind the instrument panel, generally where all the avionics equipment is. The plans have you bond (micro) the foam on to the canard and shape the foam and glass it, so the shaped area above the canard is permanently attached. Therefore you have to remove the canard to remove the "fuselage top" to access the avionics equipment. It seemed to me that it would be nice to have easy access to the wiring and avionics equipment, probably where most of your troubleshooting will be. So I decided to make the canard cover removable like Marc's. Here is how he described it on his website Marc Zeitlin's Improvement Mods scroll down to "New Removable Canard Cover". I was especially amazed how he made the complex curved flange on the canard that the cover attaches to. During the 2018 Columbia Flyin I talked to Marc about his removable canard cover and he took the time to explain how he did it and actually removed the cover to show how it was attached to the canard and fuselage. Builders are constantly seeking his advice and he is always generous with his time and knowledge, he is truly amazing! The pictures above show his removable canard cover, those are the actual hands of Marc Zeitlin!



So the secret to making that fancy flange is to sand an impression into the canard cover foam for the thickness of the flange and then put duct tape in the recess and layup BID into the recessed area and overlapping on to the canard 2 inches. I ended up with 4 plies of BID to fill the recessed area. This comprises the external part of the flange.



After the external part of the flange is cured, it is covered with duct tape and I cut the foam along the perimeter of the flange as deep as possible so the it would be possible to remove the canard cover after it was glassed. The entire canard cover was glassed with 2 plies of BID including the duct taped area covering the flange.



After the canard cover is cured you very carefully pry it from all the duct taped areas. It really does not want to let go, I find a narrow paint scraper works best to gently pry the glass from the duct tape and a lot of patience. You can see in the upper right photo the canard cover removed from the canard after glassing. The foam on the underside is shaped to a concave curvature leaving a thickness of 1 inch wherever possible.



I glassed the underside with 2 plies of Bid and added 2 ply strips of BID to the edges for extra thickness and strength for the attachment points. As you can see I also cut the opening for the access door. I used the Fein sander to cut the straight sections and a rotor zip bit for the rounded corners. I could then use a rounded Perma Grit sanding tool to finish the corners.



I love those shapely canard cover attach flanges. After the top of the canard cover was glassed and removed I could remove the foam on the interior side of the flange and added 4 plies of BID with a flox corner on the inside of the flange, so the flange has a total of 8 plies of BID.



More close ups of the flange after the 4 plies of BID were applied to the interior side.




Okay enough admiration of the canard cover and flanges. So now it was time to shape the underside of the nose section and glass that. Of course I couldn't follow the plans configuration for the access door, which is to have a removable door with screws to hold it on. At the 2018 Columbia Flyin I met Rick Hall who had a very elegant access door using "J" hinges and a latch with a cable release making opening and closing the access door quick and convenient. He explained to me the details of how he did it and I took several pictures but of course once you actually start trying to implement many more questions arise and Rick was very responsive to my emails. He even made me two different sized striker cylinders on his lathe. So in the picture the underside of the nose section foam is shaped and sanded down to the top side glass to allow flat areas for the J-hinge attach points. The square shaped foam is where it attaches to the forward part of the NG-30 side panels and nose gear box. At the bottom you can see a thick aluminum plate and nut which is where the nose gear pivots for extension/retraction. I layed up 2 plies of BID on a flat surface and after cure floxed it to the square opening at the top of the nose gear box. I made a channel of flox in the top of the panels that make up the square box and then put the cured 2 ply BID sheet on top. This wasn't in the plans but I thought this would give a solid surface for the squared surface on the underside of the nose section to bond to.



Here you can see that I've trimmed the nose door to fit the opening in the nose section and bondo'd in place to be flush with the upper surface. I don't have any pictures of glassing the underside but I wanted to have a foam rubber seal for the door so I bought 1/16 inch thick foam rubber that is 7/8 inch wide and applied that to the permitter of the door and then applied duct tape over that so that I could glass a lip to the underside of the nose section for the door to close against. So the entire underside of the nose section was glassed with 2 plies of BID then the permitter of the opening was glassed with an additional to plies of glass as well as the area where the J-hinges would attach. Here you can see some cardboard mock-ups of the J-hinges. I had first made a full scale side profile moveable model on paper to prove out the shape of the J hinge. Lynn's Dad, Erik also came up with some shapes that helped me determine the final shape. I though I could determine it with different radius circles but eventually with the full scale paper model I knew where the door attach points needed to be in the open and closed positions and then could arrive at a shape.




I cut the J-hinges from 1/8 inch thick 2024-T3 aluminum sheet. I used 5 minute epoxy to position the angle brackets and J-hinges to the door to establish the position and ensure the door opened and closed correctly. I marked the positions and used a heat gun to release the 5 minute epoxy and then floxed the angle brackets to the door and nose section.



I cut lightening holes in the J-hinges with a hole saw.



Then the Woolsey fire came through our backyard and singed the front living room window a bit. We were extremely lucky, when we evacuated the trees on our property were going up in flames and we were pretty sure that our house and everything would be gone, including the airplane. When the steel shed went up in flames it caught Firefighters attention that happened to be patrolling the area and they put out the fires and saved our house, garage (with airplane inside) and all our cars. After we evacuated I was convinced the airplane burnt up in the garage and couldn't even think of starting over, I'd probably buy a flying Cozy or partially built one, but luckily I still have mine. Fires come through our area about every 3 to 5 years (but seldom through our property), so I may need to come up with a plan to evacuate the airplane. When we finally decided to evacuate we first stopped in an Albertson's (grocery store) parking lot to be sure we wanted to leave the area, because once we left we would not be able to get back in, what was strange was that the parking lot was filled with cars but there was absolutely nobody around, it felt like the Twilight Zone. What we realized was, everybody who had evacuated left their extra cars in the parking lot since it was probably safer than at their home, another fire lesson learned.


Two days after evacuating Lynn and I managed to get back to our house by hiking back. The Sheriffs didn't seem to mind if you walked by road blocks so we hiked four miles to get back to our house. It was lucky that we did because spot fires kept flaring up that could have done more damage. We patrolled our property three times a day for a week putting out spot fires. Old stumps buried underground kept burning for a month, stuffing a hose down in the ground would result in a geyser of steam. The recovery from the fire interrupted the airplane build for a month or so. We had to redo the plumbing for our well, repair our chipper, cement mixer and other equipment. We still need to build a new shed to replace the steel Craftsman shed. We'll build another shed from a Yardline kit that is wood construction. There is no satisfaction in building or tearing down those steel sheds.


Here you can see all the brackets are floxed and glassed (2 plies BID) to the nose section and door. The J-hinge is riveted to the angle bracket attached to the door. The latch is also attached to an angle bracket with bolts threaded to a nut plate riveted to the angle bracket. The latch is SouthCo part number R4-10-10-501-20 purchased from Bisco Industries through Amazon. I initially ordered the wrong part number (right vs. left hand lever). Bisco provided the correct part number and let me keep the incorrect latch. SouthCo Part R4-10-10-501-20 Drawing


I used an 1/8 inch cable for a release and attached it to the latch with a thimble and crimp. I thought I would need another bracket to have the cable perfectly in line with the latch but I experimented with the cable pulling with an angle and it didn't take any more effort to pull than with it in line.



One problem is that the latch is in the way of removing the nose gear actuator cover. I've already had to cut the cover in half to make it fit through the access door. I may have to make it over.


The door was a bit flexible and with the latch pulling it closed in the center, the corners would warp up. So I added some stiffening ribs of 1/4 inch foam and 2 plies of Bid which eliminated flexing of the door.

Door Actuation Video

Return Home Links