Roee Kalinsky's RV-7A Project

Canopy Frame and Plexiglass
Home Up News Specifications

 

Back Up Next

Canopy Frame and Plexiglass (Acrylic)

Running Total Hours: 0.0

 

2008.06.26: (0.0) The canopy frame starts with WD-716A, a fairly complicated aluminum assembly that comes welded from Van's.  Nice idea, but nothing fits.  Not sure how they jig these for welding, but there's significant twist and mis-shapen bends in the individual pieces, innacuracies in the relative positions of the individual pieces, pre-punched holes in individual pieces that are therefore misaligned relative to others, you name it.  This assembly needs to mate with the C-702 skin as well as other components, and ultimately match the contours of the fuselage very closely at the front and sides to form a smooth continuation of those surfaces.  I've been working with this thing for over a week now, using every trick in the arsenal to persuade it to fit.  Getting closer, but man, what a pain in the ass!  I've heard other builders complain about this, and now I understand.

2008.07.29: (0.0) Progress report: Still working with this damn thing, but getting closer.  The WD-716A is now primed and riveted with the center splice plate, and also the hinge pieces are riveted to the forward channels (with various shims inbetween to set the right distance to the aft tube).  I'm still fine-tuning the fit of the canopy frame skin to the forward fuselage skin by filing the WD-716A and adding shims as needed.  It's close enough now that I went ahead and drilled the hinges.  As others have experienced, the holes didn't end up quite centered on the hinge material.  So for edge distance considerations, I decided not to use the big brass bushing inserts from Van's, and make my own thinner ones.  Whereas Van's bushings are 3/8" outer diameter, I made mine from thin walled brass tube with 9/32" outer diameter.  That way I only had to enlarge the holes in the frame to 9/32" (actually, I used a slightly smaller "K" drill bit), thereby maintaining comfortable edge distance.  I easily press fit the bushings using the hand squeezer.

To break up the monotony, I also started working on the stiffeners for this assembly.  I think this was an afterthought by Van's, first sold as an extra sub-kit, and now part of the main kit.  Anyway, these three pieces form a truss between the bottom flanges of the WD-716A and the skin, which should indeed stiffen up this floppy structure considerably.  The stiffeners have a total of nine 2" diameter lightening holes already cut in them, but the builder (me) has to bend a flange along the outline of the hole for stiffness.  Van's suggests going progressively around the hole with a bending tool, but I doubt one could get nice results that way.  I decided to make what effectively amounts to a really big dimple dye.  I made it out of 3/4" plywood, and with a 1/4" bolt in the center to provide the pressure.  A picture is worth 1000 words, so without further ado:


I was reasonably happy with the result.  The flanges looked more-or-less uniform, and they actually did stiffen up the part considerably.  For the outboard parts, the lighening holes also had a groove on one end.  For them, this "dimpling" operation also put a little bit of curvature into the part in the direction that it will need to curve to conform to the WD-716A.  Cool!

2008.08.19: (0.0) Progress report: Still working with this damn thing, but getting closer.  The forward canopy frame weldment and skin are coming together nicely, and now fit the coutours of the fuselage pretty well.  It took five rather complicated shims, and a fair amount of filing on the weldment's flanges.  The first shim is along the forward top rivet line.  It is 0.062" thick in the center, and tapers down and bends down on the outboard ends (2 rivets outboard of the hinges).  The outboard sections of the forward rivet line are the opposite, where material had to be removed from the weldment's flanges so that the canopy skin will not sit higher than the forward fuselage skin (the infamous 2 and 10 o'clock "air scoops").  There are also shims along the upper of the two longitudinal rivet lines on each side.  Rather than force the skin to pull in to the weldment, I decided to measure the gaps through the rivet holes, and make shims to fit.  This allows the skin to follow a more natural curvature, giving a nicer appearance and also providing better overall stiffnes to the structure.  These shims vary in thickness along their length, and also have some twist, most noticabley at the aft end.  The are also shims along the two longitudinal rivet lines at the top (the weldment's members that run back from the hinge arms).  These compensate for the added height due to the first (forward) shim, and also some undesired curvature that these weldment members took on as a result of welding.  In any event, the varying thicknesses of these shims along their lengths were also determined by measuring depth through the rivet holes in the skin.

Now with the aft canopy decks installed again in the fuselage, I'm starting to fit the canopy frame sides.  The bottom flanges on these parts have already been shrunk by Van's to provide some of the necessary curvature to match the fuselage sides, but not quite enough.  I'm trying to get a hold of a shrinker to complete the job.  Hoping to find one I can borrow, and not have to buy one just for this one job...

2008.09.23: (0.0) Progress report: I borrowed a shrinker from Gene and it made quick work of the side rails.  Got them fitted and riveted to the forward frame weldment.


Fitted the stiffeners to the frame.  This also required some complex shims for a good fit.  I added one shim between the center stiffener and the skin, which tapers down at the outboard ends to match fit the flat stiffener to the curved skin.  I also added shims to fill the gaps between the weldment and the skin where the outboard 3 rivet holes are located by the stiffeners.

Fabricated and fitted the mounting blocks for the canopy strut attachement on the frame side.  See section Canopy Support Struts.

The outboard 3 rivet locations (both sides) between the skin and the aft tube of the weldment have considerable gaps.  The most outboard rivet has a gap of approximately 0.300", the next one approximately 0.250", and the next approximately 0.030".  Furthermore, the gaps are significantly tapered, so even a simple spacer wouldn't be a very good solution.  I decided to shim the 0.030" gap and build a little pedastal for the 0.250" gap.  The outboard most location (0.300" gap) is not conducive to any good solution I could think of, so I'll just leave that one alone and put a dummy rivet in the prepunched hole.

(photo was taken at a later date, but shows the various shims discussed here)

WIth all this stuff fitted, I'm contemplating the next step.  Van's would have you start fitting the plexi now, with the frame still held together with clecos.  But I'm thinking that it would be better to rivet the frame together now so it's stiff while fitting the plexi, and also to reduce the risk of scratching the plexi with a cleco.  This does present a couple of chicken and egg problems having to do with the "ears" in the skin and the "tabs" between the forward weldment and the sides.  A search on "canopy, ears" on the forum yields much discussion...  I'm currently thinking that I will remove then ears completely, partly because they are problematic in terms of fit and fastening, but also this will eliminate the chicken and egg problems with riveting the skin now.  With the ears eliminated, there is adequate access to the tabs, so they can be adjusted and rivet at a later time even with the skin already riveted on.

2008.10.04: (0.0) Progress report: I decided after all to proceed as instructed and at least do the initial trimming of the canopy before riveting the frame.  So the frame is still held together with clecos, for now.

Before even doing any trimming on the plexi bubble, the builder's manual says to mark a centerline on it.  Sounds simple, but...  I found that it's not quite symmetric, and there are no good reference points.  Neither the clamp marks around the edges, nor the extra material outside the clamp marks, are uniform.  Knowing that none of these are perfect references, I located a series of center points based on each set of possible references, which varied by as much as a half inch.  I used a taught string from front to back to make the line more or less straight, and ran it down the "average" (eyeballed) of the center points I marked.  As best as I can tell, the accuracy of this centerline isn't actually important.  The only way I see it being used in the manual is to keep the forward and aft canopy pieces in proper alignment after they've been separated.  I suppose this marked centerline could also help in aligning the canopy on the fuse, but really I think that fit rather than this sloppily marked centerline will be determine final alignment.

On to cutting.  The manual instructs to initially just cut off the clamp areas on the sides and aft, and the extra flange in the front.  Sounds reasonable, but again there are no really good references, so it takes some good eyeball work.  Not knowing how much extra material there is in the molded canopy, I was a little nervous about cutting off too much.  [But fast forward a couple of days, all turned out well.]  I did the sides first, making the cuts about 1/4" in from the clamp marks, where the plexi is again "clean" -- free of distortions and texture.  I then did the back, same idea.  Lastly I did the front, which is different.  As it comes from the mold, the front of the canopy has a substantial "flange" area, which actually closely mimics the shape of the forward canopy frame skin.  The transition between this flange and the usable part of the canopy is well defined around the center with about a 1/2" radius bend, gradually becoming smoother and less defined outboard.  I made my cut such as to completely eliminate the transition area, which was very clear in the center, and again used some artistic eyeball to conservatively extend the curved cut outboard, where the transition was no longer distinct.

Various tools and techniques for cutting and dressing the plexi have been widely discussed on the forums.  I also paid a visit to Ridout Plastics, a local plastics shop (with online store as well), to buy some tools and ask for any advice they might have to offer.  On the advice side, they were definitely very knowledgeable about working with acrylic (a.k.a. "plexi"), and were happy to chat about it.  Their advice about cleaning up edges after cutting was similar to that used by many builders -- a scraper tool, and sand paper.  On the cut itself, they suggested a hand-held jig saw.  To me that seems like asking for trouble because of the reciprocating action and lack of support to the oddly shaped canopy.  I decided to first try Van's tried and true method using a cutting disc in a die grinder.  In any case, I also picked up some scrap pieces of acrylic at Ridout so I could experiment and practice on before doing the real thing.  The die grinder worked ok, but not great.  Its low torque would tend to get it stuck, and also necessitated running it at higher RPMs than I would like, making it hard to control.  Using Van's cutting disc in the pneumatic drill worked out much better.  Its high low end torque made for much better controllability, especially for the first pass.  To summarize, either the die grinder or the drill can do the job, but the drill is more controllable and ultimately leaves a cleaner cut.  BTW, whether I used the die grinder or the drill, I didn't make the cut in one pass.  I took about 5 or 6 passes to cut all the way through the material.  My technique was to first mark the cut line by putting down blue masking tape.  I put 3 layers of masking tape to give it enough thickness that it would actually guide the tool back somewhat if I strayed too close to the line.  I then did a very slow and careful initial pass, using low RPMs, to cut a shallow groove that will henceforth be the guide.  I then made additional passes with increasing RPM and pressure to progressively deepen the groove, until the final pass cutting all the way through.  After the cut is made, the edge has to be smoothed out.  I found that the best way to do it is with a sanding block.  First with 60 or 80 grit, moving to 150 grit, and finally 400 grit.  Then, chamfer the corners with the scraper tool.  Note that the side of the scraper tool without the grooves can also be used to work down the edge, but I found that sandpaper works better.  The cutting operation can leave little burrs, and the scraper would have a tendency to catch on these, whereas sand paper does not.  Note that all work on acrylic should be done at warm temperatures (75° to 80° F is good).  We've had some hot days recently, which worked out perfectly.  And otherwise I use a 1500W space heater to warm up the garage when necessary.

So after the initial trim, today I laid the plexi on the airplane for the first time.  Looks cool!  To my surprise, it actually fits pretty well (that's a first with anything related to the canopy).  Around the front there are some gaps of no more than about 1/8" where it doesn't fit the contour of the canopy frame quite right, but that should be easily fixed with another couple of minor trims to the bubble.  Above the roll bar there's only about a 1/4" gap, which is better than most from what I've heard.  At the aft end there's about 2" of extra material, and about 0.5" along the sides.  So I have to say, so far so good.

2008.10.17: (0.0) Progress report: Several rounds of incremental trimming around the front of the canopy, and it is now fitting the contour of the skin reasonably well.  It's not perfect yet, but I it will also change somewhat after the big cut, with the canopy acutally resting on the cabin frame, and also with the sides pulled in all the way to the side rails, so good enough for now.

2008.10.19: (0.0) Progress report: I made "the big cut", splitting the canopy into two at the center of the cabin frame.  It was a delicate job, but by taking it slow and careful (over an hour just to cut an initial groove), the results were about as good as one could hope for.  Nice clean straight perpendicular cuts with no major material removal required to finish the edge.  A couple of hours of sanding later, and this milestone is now behind me.  Now, to continue fitting the forward canopy...

2008.12.02: (0.0) Progress report: I was away on vacation for 3 weeks, but still since the last log entry there's plenty to report.  Progress is being made on several fronts.

"Tabs": I've finished trimming the bottom (i.e. front and sides) of the forward half of the canopy, and have adjusted and riveted the "tabs" which connect the forward frame to the side rails.  This was a very iterative process, and the results are not perfect but pretty good.  The fit in the area with the "tabs" and the "ears" still leaves some to be desired, but at this point, the tabs and the frame's inner structure are no longer the limiting factor.  It's the contour of the canopy vs. the contour of the outer skin of the canopy frame around the "ears" is now the issue.  It's not the ears themselves that are the problem, but rather the fact the the glareshield skin just forward of the ears does not provide any room for the canopy bubble to transition through.  A bit hard to describe, but I'll add a picture that should clarify the situation.  Back to the problem, I could see two possible solutions: 1) notch out more of the glareshield skin forward of the ears, as some builders have done, or 2) bend a notch into the skin without removing any material.  It seemed to me that enlarging the notch forward would significantly weaken the structure, or at least make it lose some rigidity, so I opted for option 2.  I used a 1/4" steel rod to push in the skin and create a channel for the bottom of the canopy to ride in.  Again, this took some trial and error, but ultimately the results were satisfactory.

"Ears": I've also been deferring the decision on what to do with the "ears", and now as I'm getting ready to rivet the canopy frame skin on, it's decision time.  The goals are to ensure that the canopy glass is robustly held in position (it wants to bow out), and to achieve a nice fit and finish.  The deficiencies in Van's design are that the ears don't naturally conform very well to the canopy bubble, and that there are no fasteners or any other means to secure the ears to the canopy glass.  Forward of the ears a fiberglass fairing will be formed in place, which will also blend over the ears, but I want to make sure that the fiberglass and the aluminum ears work together and not against each other... Again, hard to describe, a picture should help.  Anyway, my solution was to add a screw through the aft end of the ear, sandwiching the canopy glass between the ear and the frame just like along the side rails.  So this secures the aft end of the ear.  And then forward of that, I trimmed the ear down to about only 1/4" above the bottom of the canopy.  This will provide support for the canopy glass and keep it from bowing out, but will ultimately be fully covered by fiberglass.  Since the forward part of the ear is cut that low, the fiberglass fairing will be extended back over this area, and will reach higher up than the aluminum on the canopy bubble, with the fiberglass forming the final lines.


Lights: I decided to install an LED light strip under the glareshield to provide panel lighting (and possibly cabin lighting, if I find it sufficient).  See Cockpit Lighting.

2008.12.08: (0.0) Progress report: The canopy frame is primed, partially painted, and I've riveted on the forward skin with Jimmy's help.  So far its holding its shape quite nicely.  I've been doing everything as much as possible with the canopy frame all clecoed together and clamped down to the fuselage.  I think that helped.  I'm now riveting on the stiffener kit.

2008.12.11: (0.0) Riveted the stiffeners to the forward canopy frame.  I was able to do the right side on my own, with gun in right hand and bucking bar in left hand.  But the opposite feels too awkward, so Greg Larson came over and helped me do the left side.

2008.12.15: (0.0) Progress report: Final-fitted and riveted the rear bow to the canopy frame.


Also, contrary to Van's recommended sequence, I went ahead and fitted the side skins now, before the plexi is drilled.  My reasoning is that for drilling the plexi, the side skins can be clecoed in place to provide the ideal "clamp" for holding the plexi in place, and also serving as a drill guide.  There is, as Van's notes, quite a bit of flex in the side rails, so they will probably move around some when pulled tight to the plexi.  But that's an even better reason to use them to hold everything in its final position when drilling the plexi.  And worst case if it does significantly change the fit of the side skins, I can then always fabricate a new set if necessary.

2008.12.18: (0.0) With everything clamped in place, I marked the hole pattern for the rear canopy along the rollbar, and the hole pattern for the forward canopy along the aft bow of the canopy frame.  I ended up using approximatly 2.25" spacing (Van's recommends 2.5") to avoid edge distance problems with the rivets in the canopy frame.  I also started experimenting with drilling on some scrap pieces of plexi.  Doesn't seem too bad.  So I'm just about ready to start drilling the real thing, but I have a temperature problem.  Yes, it gets cold in winter even in San Diego.  Yes, cold is relative, but this 60° F weather we've been having is too cold for my liking, and too cold for safely working with plexi.  I'd feel safer with 75° F.  Well, I ran my little 1500 W space heater for over an hour, and it only got the garage up to about 65° F.  I'll have to find a heater with a little more UMPH!  I don't think I want to wait until summer to drill the canopy.

2009.01.05: (0.0) Progress report:

The weather isn't getting warmer any time soon, and I'm just not that patient, so I went out and bought a second 1500 W heater.  With a total of 3 KW going, I was able to heat the garage to 75° F, and hold it pretty accurately with the thermostats.  I'm not looking forward to the electric bill, but oh well.  To keep things precise, I'll will try to do all the plexi final fitting and drilling as close as possible to 75° F.

On to the drilling.  I clamped the canopy frame to the fuselage and positioned the plexi on it.  I clecoed on the canopy frame side skins, which nicely sandwiched the sides of the plexi in place.  I used duct tape to pull the plexi down into position all around the sides, front, and back.


I then match-drilled the sides using the holes in the side skins as a guide.  Note that I did this using a 1/8" plexi drill bit, even though I was drilling through a sandwich of aluminum - plexi - aluminum.  Since both pieces of aluminum were already pilot drilled to 3/32", the plexi bit only had to enlarge those holes to 1/8", which it did without too much difficulty.  After drilling each hole to 1/8", I then also ran a #30 reamer through it, which produced immaculately clean holes, and also prepared the canopy to eventually be countersunk using a bit with a #30 pilot.  (I also deburred the holes using a permagrit countersink bit once the plexi was removed from the frame.)


With the sides drilled and clecoed, I moved on to the aft edge (i.e. the arch that matches the rollbar).  These, I did not drill in place.  I transferred the hole pattern that I marked on the canopy frame arch onto the canopy, just in sharpie on the protective plastic, so that I could double check the straightness and evenness of the hole pattern.  I then only pilot drilled less than half way through the plexi, just to set the hole locations.  [Note: In the photo it looks like the holes are messy with big burrs, but that's just the protective plastic.  The holes in the plexi are actually very clean with no discernable burr.]


I then removed the plexi and drilled the holes through it on the bench (1/8" plexi drill bit, #30 reamer, permagrit countersink for deburr).  When drilling the plexi I clamped another piece of scrap plexi with matching curvature to the inside as backing material, to prevent the bit from punching through or chipping the back of the hole.


I then put the plexi back in position on the canopy frame, of course clecoing all around the sides, and match-drilled the aft arch using the holes I just drilled in the plexi as guides.  I drilled the frame using a standard (i.e. non-plexi) 1/8" drill bit, and then ran the #30 reamer through the plexi and frame again.  Great results.

I have only two more holes to drill through the plexi of the forward canopy (there's always something...).  Recall that the forward-most hole on each side is one that I added that goes through the forward skin, not the side skins.  And I had to already enlarge and dimple the hole in the skin for the #6 screw before I riveted on the skin.  I wisely left the opposite hole through the inner side rail at 3/32".  But this does mean that I have to match-drill this one from the inside rather than from the outside.  So I'll do this at the next convenient opportunity.

I also started fitting the rear window.  I can see this is going to be a real PITA, because the top of the rear window plexi is substantially concave along the fore-aft axis (what Van's calls a "duck tail"?), which causes it to press against the structure between the roll bar at the front and the bulkhead at the back.  The plexi will be under stress, and will not interface with the skin nicely in the same plane.  Damn it...  I anticipate this being another one of these "massaging" situations.  In the mean time though, I prepared the top skin, chamfered its edges, and marked and pilot-drilled a hole pattern for the screws that will connect the plexi.  I also took this opportunity to correct the angles of the flanges in the bulkheads that mate up with this skin.  The QB factory over-bent these by a good 30 degrees.

2009.01.14: (0.0) Progress report:

Working on the rear window.  Despite the concavity issue, the fit has actually turned out pretty descent, with the plexi pulled down to the roll bar with only a moderate amount of force -- not enough to cause any problems.  

I drilled the rear window starting in the aft center, working outwards left and right while the forward edge of the window is clamped down to the roll bar, and Stacey pushing the window from the inside tightly against the fuselage skin.  About the photo: after 5 attempts I gave up on getting a photo with her eyes open, oh well.


I then drilled the holes along the front edge of the rear window in a similar fashion as I did the aft edge of the forward canopy.  I first transferred the hole locations to the plexi (covered with transparent celotape) with sharpie pen, and then I drilled.  But unlike the aft edge of the forward canopy, this time I went ahead and drilled the plexi while still in assembly on the fuselage, all the way until the drill bit hit the roll bar.  The holes in the roll bar will ultimately be drilled and tapped for 6-32 screws, so that's a smaller hole than 1/8".  Letting the 1/8" plexi bit just hit the roll bar will later help as a center guide for drilling through the roll bar.  I removed the plexi from the fuselage and finished drilling the holes all the way through the plexi, and then reaming them to #30.  I then clecoed and clamped the rear window back to the fuselage, and drilled through the roll bar to 1/16" (which will serve as pilot holes for the #36 drill prior to tapping).  I chose to have the plexi on the fuselage for this operation only in order to help drill the holes at the right angle, perpendicular to the surface of the plexi, which is not exactly the same as the roll bar.  Anyway, it is also much easier to sight the angle with the plexi in place, because I can clearly see the reflection of the drill bit.  Anyway, that's done.  But then I discovered that I don't have a #36 drill bit, so that'll have to wait.

The forward edge of the rear window matches nicely to the roll bar center line, with only a couple of high spots that still need a bit of sanding.  I marked these with masking tape and will take care of them at the next convenient opportunity.

My next major task for the rear window will be to make the "optional" backing strips shown on the plans.  I think this is a good idea, as it will help spread the load from the fasteners more evenly across the edge of the plexi.  It will also make for a nicer looking finish on the inside.

Today I also drilled the two remaining holes through the forward canopy.  Just clecoed the plexi back to the frame, and drilled through from the inside.  I still need to trim the aft edge, and will probably do so in the next few days.  I also marked with masking tape the line where the canopy meets the glare shield.  I plan to paint the glare shield (the portion inside the canopy) flat black.

2009.01.24: (0.0) Progress report:

Fabricated 0.025" aluminum backing strips for the rear window.  Came out nice.



Did a couple of last iterations of final fitting (sanding only, no cutting) to get the gap between the forward canopy and the rear window to an even 1/32".  Now with everything clecoed in place, the canopy opens and closes with no interference.  Looks sweet!



The fit between the forward canopy and the rear window is almost perfect along the top of the roll bar, but not so great on the sides.  Basically, the rear window sticks out further than the canopy, presenting a "step" to the airstream.  A careful examination of the airframe's geometry makes it clear why this is the case.  At the roll bar, the fuselage is tapering back, becoming narrower as you go aft.  The plexiglass canopy also has a similar contour.  The roll bar however is simply square, it doesn't match this taper.  As a result, the plexiglass and the outside of the roll bar are not exactly in the same plane.  Now, the forward canopy has been fitted to mate neatly against the centerline of the roll bar all the way around.  So only the very aft edge of the forward canopy plexi actually contacts the roll bar near its centerline, and the plexi sits a little higher (i.e. not making contact) at the very forward edge of the roll bar. In the case of the rear window, the mismatch is in the opposite direction, such that the window plexi only contacts the very aft edge of the roll bar, and the very forward edge of the window sits a little bit higher (i.e. not making contact) at the centerline of the roll bar.  Hence the mismatch at the roll bar centerline.  Again, along the top this is alost indiscernible.  But on the sides the mismatch peaks at about 0.040", quite noticeable to me and to the airstream (drag!!!).  We'll I've been debating anyway for some time whether or not to make a targa strip.  Now it looks like I pretty much have to.


Started experimenting with various tools and techniques for countersinking the plexi (using scrap for now).  I haven't gotten a result I've been really happy with yet, so this effort will continue.

2009.02.16: (0.0) Progress report:

The plexi (canopy and rear window) is countersunk and drilled to final size.

I experimented with several methods (on scrap) for countersinking suggested by other builders, and some worked better than others.  First attempt: The permagrit countersink bit gave surprisingly rough results.  The grit was just too coarse.  Second attempt: My standard 3-flute machine countersink bit in a cage first gave mixed results.  It was difficult to start, required a scary amount of pressure, and left a mostly smooth cut except for some subtle rings.  Third attempt: I had a hunch that the problems with the 3-flute cutter may simply have to do with it being well used and not so sharp anymore.  I bought a new cutter, which worked much better.  It took less pressure, although still quite a bit, but cut consistently and left a totally smooth countersink.  I did the real thing using this technique.

After countersinking, I still had to enlarge the hole to its final size.  This was relatively easy, with several good options.  Just running a 5/32" plexi bit through it does the job pretty well.  Being extra cautious, I thought I would also follow that up with a #22 reamer, to leave an even cleaner hole.  That also worked well.  But it turned out that I could skip the 5/32" drill, and use the #22 reamer directly on the existing #30 holes.  That's how did it.

After doing this I did a final fit check, and ended up trimming just a tad more from the left side bottom edge of the canopy to make sure it doesn't interfere with the joggle in the side rail.  Hopefully, that's the last plexi trimming I will ever have to do! ;-)

With everything trimmed and fitted, it was time to button up the canopy frame.  I final drilled/reamed and countersunk the side rails and skins, and riveted on the skins.  Also made an external lifting handle on the pilot side.  Another fit check after riveting and everything looked pretty good.  I had to trim the bottom edges of the side skins a touch more, and everything was perfect again.  And I was pleasantly surprised at how easy it was to insert and remove the plexi despite the dimpled screw holes in the side skins.

I decided to attach a pair of spacers below the aft ends of the canopy side rails that would sit against the fuselage canopy deck when the canopy is closed.  That way, the vertical position of the canopy is limited by the metal frame rather than the aft edge of the plexi.  I carefully adjusted the thickness of the spacers (1/8" main spacer, plus thin adjustment shims), trimmed the canopy latch cams ("fingers") a smidge, and in doing so got the thing to sit solidly when closed, and let the latch handle operate smoothly without excessive stress/strain during the over-center transition.

Gave a coat of gray paint to the remaining interior surface of the canopy frame, except for the top surface of the glare shield, which will get a coat of black paint.

Another little refinement done on many tip-ups, I create a funnel-like guide along the roll bar that lines up the canopy laterally as it closes.  The canopy structure is still pretty flexible, and this mod helps prevent careless bumps and snags that could damage the canopy if closed while out of alignment.  Note that it is still possible with extreme twisting of the canopy frame to miss the funnel guide completely.  And in any case, the canopy should always be closed carefully and gently, so it doesn't just slam down.  It is not nearly as foolproof as, say, a car door.

2009.02.22: (0.0) Progress report:

I bought a "deluxe glare shield trim" piece from Classic Aero Designs.  This provides a padded covering for the glare shield edge, which is a good safety feature.  It also hangs down about 3/8" below the glare shield edge, which will help hide the light strip from direct line of sight.  Hiding the light strip is not for aesthetics, but rather to prevent looking directly at the light source and ruining night vision.  On aesthetics though, this trim piece is nicely upholstered in a black synthetic leather and looks very nice.

I fitted the trim piece to the glareshield, which merely involves locating some screw holes.  I then painted the top of the glare shield flat black (Krylon Fusion 2519 Flat Black).  And once that dried, I installed the LED strip on the bottom of the glare shield.

2009.02.25: (0.0) A big day (or actually, a long night).  I attached the plexi bubble to the canopy frame for good.  Difficulty-wise, this would have been a non-event, except that I decided to use fuel tank sealant (a.k.a. "pro-seal") to seal the interface between the plexi and the side skins.  That stuff is a nasty, sticky, goopy, stinky compound with a limited working time, and is very tenacious once cured.  So I had one chance to get it right.  And I did.  Whew.

Some other details to note.  Rather than the nylock nuts called out on the plans, I decided to use acorn nuts.  This is motivated by my goal to not have any sharp protrusions in the cockpit (the sharp protrusions being the end of the screw coming through a nylock nut).  Since the acorn nuts don't have locking nylon inserts, I used a dab of thread locking compound (Loctite ???).  I did this VERY CAREFULLY so it never comes in contact with the plexi.  To reduce glare I also painted the acorn nuts and washers with my standard interior paint.  I did this about a week prior to the installation.  With the hardware pre-painted before assembly, I don't have to worry about painting anywhere near the plexi.

Man, it's nice to have this step behind me.  I'll take a moment to admire it before getting back to work :-).

2009.03.06: (0.0) Quick update: I'm getting ready to do the composite fairing around the forward edge of thre canopy.  Before I do though, I'm going to get my hands dirty with some less ambitious composite work on the wing tips and empennage tips.  Just to gain experience and confidence working with composites, and lessen the probability of making a high-stakes mistake on the canopy.  Stay tuned.

2009.03.09: (0.0) Did a bit of composite work on the wing tips, and that went pretty well.  So back to the canopy.  I started by creating a structural (flox) fillet all the way around the front, trying to basically continue the contour of the outer surface of the plexi all the way down to the aluminum skin.  Came out pretty good.  But I think I'll do one more pass to fill in some low spots before doing the layups.  My goal here is for the layups to have a continuous surface with a uniform contour to lay on.  I don't expect this to be perfect and it will not be the final finished surface.  Non-structural fairing filler (i.e. "micro") over the layups will serve that purpose.  But having the layups conforming to even gentle contours should provide maximum strength, as well as a good foundation for the subsequent fairing filler.

Note in the photo that the flox fillet is gray, not white.  This is due to black pigment that I added to the mix.  I will do the same with the layups.  The idea is to keep the cockpit interior color-neutral so it's not visually distracting, and to minimize glare.

Note also the red electrical tape in the photo.  This masks off my planned boundaries for the fiberglass fairing.  The top boundary is approx 5/8" from the edge of the plexi all the way around, and approx 1" from the aluminum skin in the front and a bit more around the sides due to the fillet angle.  Van's drawing shows 1" overlap of the plexi proper (i.e. excluding the fillet), but I believe that 5/8" should be sufficient structurally in this application.  The reduced overlap makes the fairing match the side skins better aesthetically, and also is less obtrusive to the pilot's forward visibility from the cockpit.  I have a STRONG preference for cockpits that give a good high view over the cowl.  The RV-7A is actually not stellar in that regard, but I'll do what I can to that end.

P.S. one more note about surface prep.  1. Clean with VM&P naphtha (what is "VM&P"???).  This is one of the few solvents that is reportedly safe to use on plexi.  2. Abrade with red scotch brite pads.  3. Another pass with naphtha.

2009.03.16: (0.0) Did several more rounds of fill & sand on the flox fillet, and it is finally a beautifully contoured work of art!  Around the front it's basically a continuation of the plexi's contour, plus a radiused transition to the aluminum.  The areas around the sides were a bit more tricky, where the continuation of the plexi's contour would actually end up never intersecting the aluminum skin.  So here I just used an artistic eye to blend the contours.  I also blended a bit of flox around the aluminum "ears", so that the fiberglass will have a continuous surface to conform to.

2009.03.19: (0.0) Laid up fiberglass around the front of the canopy.  Following the advice of Dan Horton (composites guru on VAF forums), I ended up doing this using the pre-preg sandwich method, three plies of RA7781 all at once, a full-length strip reaching all the way around.  John Nichols came over to help me with this, which was a good thing.  With the two of us, this was a difficult, messy, aweful task.  By myself this would have been impossible.  The results were decent.  Inevitable there are some small voids that need feeling, and lots of blending (fill, sand, repeat) that I have to look forward to.  But that's how this goes.

I also added black pigment to the epoxy, which had mixed results.  I added enough to make the epoxy appear pitch black in the mixing cup.  But when I used it to saturate the cloth, it became apparent that much more would have been needed to really do the job.  The wet cloth was a bit darker in appearance than normal, but still quite transparent.  Now that it's cured, it looks dark gray from the outside, but not very dark at all from inside the canopy.  The fiberglass is still quite transparent when back-lit.  We'll see how it does once its exterior is primed and painted.  Anyway, however it is, I'll have to live with it at this point.

As a side note, I am now soooooo happy that I built a (mostly) metal airplane!  I couldn't imagine doing nothing but this nasty composites work for several years!


2009.04.16: (0.0) Sanding...

 

Back Up Next

 

 

Send mail to roee@kalinskyconsulting.com with questions or comments about this web site.
Copyright © 2003 Roee Kalinsky
Last modified: April 18, 2009

Visit my consulting web site at www.kalinskyconsulting.com for your engineering needs.