Firewall, engine, Skybolts [37.5 hours]

Warning: count(): Parameter must be an array or an object that implements Countable in /home/tasrv100/public_html/wp-content/plugins/thumbel-slider/slider_versions/shortcodes_1.php on line 60

I’ve started work on the firewall penetrations. Since I’m using SDS/EFI there is a need for more wiring through the firewall than is the case for a conventional Lycoming installation. There is also a return fuel line associated with the duplex fuel system, and a fuel regulator. The location of all the parts and pass-throughs needs to be determined, so that I can drill the necessary holes. In accordance with current “best practice” for firewall insulation (in case of an engine compartment fire), I’m installing a Fiberfrax insulation layer on the engine side of the firewall. This requires a thin metal retaining layer on the front of the Fiberfrax. I bought some 0.002″ stainless steel foil for this purpose, but decided there was no way I could work with it without winding up with a crinkly mess, so I have switched to some 0.008″ Titanium.

I temporarily lifted the (Barrett) engine into position to ensure there’s plenty of room around the pass-through positions. I installed doublers for the AntiSplat air/oil separator and the A/C line pass-throughs, as well as various nutplates that are in addition to the plans.

Speaking of the engine, apart from desiccant plugs I’ve been using a home made “conditioner” to help preserve the engine until its first start. I placed about 2kg of (orange) silica gel in a sealed plastic cake container with an aquarium pump. The outlet of the aquarium pump goes through a respirator filter (to prevent any silica dust from entering the crank case), and then into the oil filler hole. A tube from the breather outlet goes into a glass bottle, and a separate tube from the bottle for return air into the plastic container. A humidity sensor in the bottle shows that the crankcase is being maintained at a humidity level of less than 10%, whereas the outside air humidity is often above 50%.

I also worked on the cowling mounts at this time. I’m going to use the Van’s hinge method for the firewall sides, and between the two cowl halves, but have decided to use Skybolts for the upper cowl firewall mount. To that end, I bought a kit of parts from Skybolt. Included in the kit are these nifty looking interlocking flanges, but when I started to work out how to lay them out, I found that I would have to adjust the spacing between skybolts in a fairly arbitrary way to have rivet holes in the flange overlaps occur in sensible places with adequate edge spacing. I really didn’t like the resulting uneven spacing, so I decided to make my own Skybolt flange mounts.

I used two sections of 0.032″ Alclad, overlapped in the middle (top). Marked out, drilled, cut, and then match drilled into place from the firewall edge with the top skin in place. I made a small overlap between the two halves, and dimpled all of the #40 holes to match the firewall. I also added a 0.02″ Alclad spacer to the assembly, same as would be used with the hinges, to allow for some filler on the cowl edge. With these cowl brackets complete, I put them aside since they can’t be riveted on until the upper front fuselage assembly is ready to rivet on. The hinges on each side, and spacers, were made in accordance with the standard plans and riveted in place. I haven’t done anything with the bottom hinges, since these are not used for the Showplanes cowl. I’ll most likely add a Skybolt or screw/nutplate on each bottom side but can’t really do that until I’ve fitted the cowl.

  • f33a
    Figuring out firewall pass-through locations
  • f33b
    Figuring out firewall pass-through locations
  • f33c
    Figuring out firewall pass-through locations
  • f33d
    Figuring out firewall pass-through locations
  • f33f
  • f33g
    Engine environment on right, outside environment on left
  • f34a
    Marking out second Skybolt bracket (first above)
  • f34b
    Cutting diagonals - carefully!
  • f34c
    Cutting diagonals - carefully!
  • f34d
    After initial drilling and cuts
  • f34e
    Overlap formed with hand seamer
  • f34f
    Match drilling to fuselage, with 0.02" spacer
  • f34g
    Finished skybolt brackets for upper cowl, side hinges riveted in place.
  • lola1
    RIP old friend

Clearcoat done [8.0 hours]

After a lot of preparation, I sprayed the clearcoat onto the overhead console. I’m not much of a painter so I’ve got one run to scrape out and the whole thing will need a cut and polish, but it turned out fairly well and I got the effect I was looking for in terms of the transition from the light grey overhead colour into the carbon fibre console. More importantly, the overhead console is now UV protected with the Durepox clearcoat product I used.

I’ve put the overhead aside for now, time for a break from fibreglass and painting so I’m going to get back to the firewall and work towards getting the fuselage up onto the gear.

  • f32a
    Set up on a nice day
  • f32b
    Spraying clearcoat
  • f32c
    Spraying clearcoat
  • f32d
    Clearcoat done, before cut and polish
  • f32e
    Clearcoat done, before cut and polish

Cabin Top Blues [45.0 hours]

My long battle with the Cabin Top is coming to a close, at least for the interior. Various choices made the task harder than it should be, but the end result is now in sight, and I’m happy with how it’s turning out.

Including A/C in this aircraft meant an overhead console would be needed, so I bought the Aerosport Products carbon fiber overhead console. I wanted wiring conduits up each front pillar to make wiring easier, so that brought about a whole load of work to glass these in and finish the pillars appropriately. I’m installing Visors so that requires mounting points, but Van’s have taken the position that holes should not be drilled in the front pillars because the holes have an unknown impact on the rollover protection, so I made up some metal inserts with Nyloc nutplates and glassed them in.

I bought the Aerosport cabin headliner kit, but the headliners were compromised during shipment. I could have repaired the problems, but overall I decided against using a headliner, which meant that I had to finish the entire cabin top interior. This amounts to a lot of work, because the Cabin Top as it comes from Van’s is quite rough, and there are a lot of complex shapes and curves involved, especially after glue-ing in the overhead console. The headliner would not have helped much either, the hardest parts are outside of the area that a headliner would encompass.

I decided to keep the Carbon Fiber look for the overhead console, this added yet another layer of complexity because it had to be masked off while the rest of the interior was painted, and masking off the line between the interior paint and the overhead console is difficult because of the afore-mentioned complex shapes.

Previous choices had an impact as well. Some time ago I chose to use the McMaster Carr door seal, so as described in previous posts I used a length of seal as a mould and built up the door frame to a constant 1/4″ width all the way around. When you do this, there is a “jag” on the inside of the seal which winds up making a nice groove on the interior side of the door edge. This is a good thing, because when you put the “real” seal in place, the seal jag slots into this groove and helps to hold it in place. While spraying fill primer, I didn’t want this groove to fill with with primer and disappear, so I had to tape off to just beyond the small groove around all the door frames. I removed this tape before spraying the primer and topcoat.

If you’re building an RV-10 and want to minimise the work on the cabin top interior, then don’t do any of the things I’ve done. Otherwise, here is the entire process I’ve used:

  1. Fit the cabin top etc., and create the door gap for the McMaster seal (see previous post)
  2. Fit the overhead console, and glue it in. I used Lord adhesive for this (see previous post).
  3. Fit the wiring conduits to the front struts, and glass them in with 3 layers of glass. See previous post for details on how I used a Nylon 3D printed part to transition into the overhead. I used straight epoxy, blackened with die, to seal the Nylon transition pieces into the overhead. The same technique was used to seal around the (black Aerosport) door strut brackets.
  4. Make metal inserts with Nyloc nutplates for the visors. Position these so that the visors just miss the overhead console when (unextended and) pushed to the front. Glass these inserts in at the same time as step (3) above. Triple check their position with the real visors before doing so!
  5. Apply filler (epoxy + micro-balloons + cabosil) to the front pillars to create the desired shape around the conduits, and transition into the visor mount points which must be kept flat.
  6. I didn’t like the amount of material left on the cabin top roof after the front seat belt mounting bolts were countersunk, so I added three layers of glass around these points to create a slight “bulge”. I added filler around this “bulge” to transition smoothly back into the cabin top.
  7. Apply filler everywhere else around the interior, and to transition from the overhead console back into the interior. This took several iterations to get right, so fill, sand, fill, sand, fill, sand ….
  8. I cleaned everything off with wax and grease remover, and scuffed the overhead console to prepare it for later clear coat, with a 600 grit wet sand. Then I cleaned everything off again and taped up the overhead, and fussed over the edge that I was going to tape to. This edge I made about 1/4″ back from the corner of the overhead – it’s really impossible to tape a straight line on the corner when you have to go around bends in three dimensions.
  9. After cleaning off again, I applied a thin glaze of straight epoxy over the “flat” parts of the cabin top, i.e. the parts where I could use a squeegee. This was to fill pin holes. Around the curves and steps, I didn’t do anything in particular for pin holes at this point.
  10. I rolled on a coat of Wattyl UC-230 primer-surfacer, let it dry, and sanded it back. I wouldn’t bother with this step again, I was still left with a zillion pin holes and it’s much more effective to just spray it on.
  11. Without bothering about pin holes, I sprayed on a second coat of UC-230, using a cheap gun I bought for priming that had a 1.8mm nozzle. Once this dried, I sanded it back. The result was a zillion pin holes, a lot of small low spots where further sand/surfacer operations were required to render the interior flat, and a few larger low spots that required a bit more filling.
  12. I filled the few “larger” low spots as required, and applied Ever-coat 440 express pinhole filler. Be careful if you use this product, it is highly toxic. After this, another coat of UC-230 and sand it back.
  13. I repeated the process – spray on UC-230, sand it back, fill pinholes – about another three times. Each time I probably sanded 80% of the material off – but never back to the previous layer – and got down to the point where (a) I couldn’t find any more pinholes, and (b) when I sanded it back, there were no “low spots” left. It’s easy to see the low spots, they remain shiny in reflected light while the material you’re sanding doesn’t.
  14. Once I was happy with all the surfacing, I sanded the surfacer down to the level of the tape along the transition line to the carbon fiber console, removed all the tape, and then carefully sanded the edge down further, leaving a very slight rise where the surfacer began (180 grit for this sanding). I used a sharp knife to scrape away surfacer in a few tiny places where it had crept under the tape. I removed the electrical tape around the door surrounds (to stop surfacer filling up the groove for the seal jag). Then I cleaned everything up with wax and grease remover, and finally with alcohol.
  15. I re-taped the centre console, putting down a careful line with 3M vinyl tape, just “inside” the edge of the UC-230. I fussed over this taping a lot, because the next step – polyurethane primer – is much thinner paint and will creep under a bad taping job. Moreover, it will stick to just about anything and be impossible to get off. I left the door surrounds un-taped.
  16. I sprayed PPG polyurethane primer, two coats. One in the evening, one the following morning. Drying time is 4 hours. It was a perfect day outside, so I decided to spray the top coat in the afternoon. For this I used my better quality spray gun, with a 1.2mm nozzle.
  17. I sprayed two coats of PPG polyurethane topcoat, 1.5 hours apart. The 1.5 hours is just to allow the first coat to flash off. After each of the PPG primer operations and each topcoat, I rushed the cabin top back inside the garage (which is closed up) and threw a drop cloth over the whole thing to try and reduce dust, since I don’t have booth that I could fit the cabin top in.
  18. Drying time for the topcoat is 12 hours, full cure in a week. Seven hours after the second coat, I carefully pulled up the vinyl tape. A sharp knife and tweezers helps, and it’s important to avoid dropping any “shards” of paint from the vinyl tape (which it doesn’t adhere to) back onto the topcoat – it’ll stick.
  19. After the topcoat had cured overnight, I removed the rest of the masking on the overhead.

Unavoidably, a few small bugs landed on the topcoat and suffered a cruel death. I waited two days so that the topcoat had hardened up, and used a very sharp/pointy knife to remove the remains. I think I did this in 4 places, fortunately they were all very tiny bugs. These, and a few dust spots, will clean up OK when I cut and polish the topcoat. I’m not going to do this until after the topcoat has fully cured.

The next job is to spray clear-coat on the carbon fiber. I’m going to tape off the grey paint along the transition line, protect the rest of the interior with plastic etc., and spray the clear. I’m using a marine product called Durepox for this. I haven’t done it yet, so I’ll update this post once it is done. I did a little test piece and the chemistry between this and the PPG polyurethane seemed OK. I do expect to run into some pinholes in the carbon fiber, I can see them under a magnifier, so I’ll probably have to use a small brush to fill them, sand back, and do several coats of the clear before it is good enough to finish with 2000 grit wet and a buffing polish.

Once the clear is done, I’ll cut and polish the grey topcoat, glue the windows in with Lord adhesive, and the cabin top will finally be ready to fit to the fuselage for good. To finish the interior window transition, I plan to cut a rubber seal in half and cement it in place, as described in a Van’s Air Force post some time ago.

Of course, once the cabin top is on, I get to do a lot of this all over again, filling in around the lower door surrounds, filling and joining everything up with the existing finished paintwork, in theory so that it is not possible to see where the join is.

  • f31b
    3 layers of glass over conduit
  • f31c
    3 layers of glass over conduit
  • f31a
    Slathering on micro to shape front right door pillar
  • f31d
    Front door pillar after sanding
  • f31e
    Both front door pillars shaped, rear section filler sanded
  • f31f
    Visor support, and lots of pinholes
  • f31g
  • f31h
    Cleaning up after final primer/surfacer coat
  • f31i
    Surfacing finished
  • f31j
    Taped up again, prepared for primer and topcoat
  • f31k
    Spraying on first layer of primer
  • f31l
    Spraying on second layer of primer
  • f31m
    First topcoat
  • f31n
    First topcoat
  • f31o
    Second topcoat
  • f31p
    After topcoat cured
  • f31q
    After topcoat cured
  • f31r
    After topcoat cured
  • f31s
    Visor mounting points, and a bit of MC seal in place
  • f31t
    The hardest part to deal with
  • f31u
    MC seal fits under aerosport door strut bracket
  • f31v
    MC seal fits under aerosport door strut bracket
  • f31w
    Detail of groove where MC seal jag sits.
  • f31x
    Conduit, glass, micro, surfacer, primer, topcoat
  • f31z
    Ready for attachment to fuselage

Winter design work [200.0 hours]

It’s been a few months since I posted to the build log. July and August have been very cold months at this latitude, and the days short. Terrible weather for fiberglass work. Rather than spend a lot of time being miserable in the workshop, I decided to do some desk work until spring.

I needed to get the Electrical design done. I used Eagle for the schematics and made a parts library from the ground up. It’s 90% done, the other 10% depend on some avionics choices and finishing details that I’d rather put off for a bit.

I also needed to figure out the sub-panel cutouts, sub-panel equipment mounting points, and firewall penetrations. That all interacts with the avionics and engine choices (using sdsefi has a lot of ramifications), and the electrically dependent engine interacts with the electrical design. Until I do the firewall penetrations, I can’t do the firewall fire insulation work, and until I do that work I can’t bolt on the engine mount, and until the engine mount is bolted on I can’t mount the nose wheel leg, and without that I can’t get the air-frame off the dolly and up on the gear. The whole thing is a vicious circle of dependencies, and the cold of winter was a good time to work it all out. Not quite there with all of this design work yet but I’m close.

Just this week we’ve had a good spell of warm days, so I’ve rolled the cabin top outside and have been filling and sanding. It’s good to get back to physical work on the air-frame, so I’ll have some meaningful posts for the build log in the coming weeks and months.¬† haven’t done a very good job of time keeping, but I’m estimating 200 hours of desk time across July and August for this work.

Lots of odd jobs and painting [135.5 hours]

I’ve continued with various cabin and door jobs over the past month. Mounted the door struts, and was quite surprised when the doors “worked” properly. Open the door, let it go, and the door goes up by itself. I filled over the door hinge gap covers with micro, and sanded it back to shape (no photo, but it worked out well). I also glued the Aerosport overhead on permanently, using Lord adhesive. Started doing some filling work, but then elected to put it aside for a while and catch up with fuselage work – that was the work I put aside a few months ago in order to use the last of the warmer weather before winter on the cabin top, doors and windows.

I drilled the fuselage for the engine mount, to get this out of the way before painting the interior. It is apparently quite normal for the holes in the firewall to NOT line up properly with the engine mount. I didn’t pay enough attention to this, and decided to simply follow the Van’s instructions and drill one of the top holes to size (3/8″). This turned out to be a mistake, because it established an arbitrary fixed point for that corner of the engine mount and what I SHOULD have done was to establish where the engine mount had to go with respect to ALL of the pilot holes in order to (a) take out all the pilot holes, and (b) keep the mount exactly centred. Why (b)? Because on the end of the engine on that mount there will be a hole in the cowling, and a spinner that is supposed to line up with that hole.

I should have “worked” that top hole¬† to move the centre of the 3/8″ bolt hole about one mm toward the middle, but I didn’t. I had to stretch the mount a bit with a clamp (not much, just a bit) in order to fully cover the opposite hole. The bottom middle holes were right on the edge of the 3/8″ guide on the engine mount, so much so that I could drill a #30 hole in the centre point and not actually break into the pilot hole. I was a bit concerned about how to drill these holes without having the off-centre pilot holes “pull” the drill away from where it needed to go and start scraping material off the engine mount tubing.

To resolve this, I 3D printed a bunch of drill guides. They were just cylinders with a 3/8″ outside diameter, and a #30 hole through the middle. Using plenty of cutting fluid, I drilled #30 holes in the places where I could, and “pinned” the mount in place using the shank of long #30 drill bits. In the places that couldn’t be pinned because the pilot hole already overlapped the centre, I used “other” 3D printed drill guides of various size internal holes, and stepped up the drill size in successive operations before consuming the pilot hole. The final step was to use a 3/8″ reamer to final drill the hole, and then put the 3/8″ engine mount bolt, washer and nut in place.

By the time I got to the worst two holes – which had #30 drill bits holding the mount in location, the other bolts around the mount held it in place so well that I simply ran through the same series of drill guides, stepping up the drill sizes. The mount simply could not move, so the fact that I was breaking through the very off centre pilot hole didn’t matter at all. I was able to consume those (badly off centre) pilot holes without dragging the drills off centre, again finally finishing up with the 3/8″ reamer.

At the end I had a handful of wrecked 3D printed drill guides, but they had done the job!

I found that I had installed the incorrect sized nutplates on the Antenna inspection covers, #8 rather than #6. It was a nuisance but fairly easy to drill them all out and replace them with the correct part.

Then it was on to painting. I primed and painted all the detachable interior panels, primed assembled and painted the rear seat frames, plus the remaining control rods, rod ends and some other miscellaneous bits. After that I prepared the fuselage and painted the internal floors and baggage area. Most of these areas will be covered by an Aerosport carpet set.

Once that cures, I’ll be able to go back and mount the seat rails, rudder pedals, brake lines, fuel valve and lines, and control system so quite a lot of parts I have lying around here will go into the airframe for good.

  • f30a
    Drilling cabin top for the door strut bracket
  • f30b
    Drilling out incorrect #8 nutplates
  • f30c
    Incorrect #8 nutplates removed
  • f30d
    Drilling strut bracket into door
  • f30e
    Door strut brackets fitted
  • f30f
    Door strut brackets fitted
  • f30g
    Filling over the hinge gaps and strut bracket rivets
  • f30h
    Filling over the hinge gaps and strut bracket rivets
  • f30i
    Ready to glue on the overhead with Lord adhesive
  • f30j
    Lord adhesive applied, overhead clamped/weighted in place
  • f30k
    Starting to do a bit of filling
  • f30l
    3D printing disposable drill guides
  • f30m
    Using #30 drills in guides to locate engine mount
  • f30n
    Stretching engine mount just a little
  • f30o
    Engine mount drill guides did their job
  • f30p
    Engine mount in place
  • f30q
    Backing plate to fill in rudder cable hole
  • f30s
    Filling rudder cable hole with micro
  • f30t
    Priming various internal panels
  • f30u
    Topcoat on various internal panels
  • f30v
    Priming rear seat frames and some control bits
  • f30w
    Riveting seat frames
  • f30x
    Helping paint to dry during a cold Tassie winter
  • f30y
    Top coat on seat frames
  • f30za
    Ready to paint interior floors, sides
  • f30zb
    Painting interior baggage area
  • f30zc
    Interior after painting
  • f30zd
    Compressor outdoors, cheer squad in the background


Windows, Conduits, overhead etc. [63.0 hours]

I’ve continued to chip away at the never ending cabin top work. This is clearly the character building part of the RV-10 build. It’s hard to separate activities because the doors, cabin top, windows, etc. are all interdependent. There’s a notorious flat spot on the pillars between the doors and rear windows in the cabin moulding, for instance, which has to be built up to match the door. The front side of the window also has to be spaced out so there isn’t a sudden transition between the door, pillar and window. Can’t do this until the door is fully fitted though. The inside surface where the window is raised will need to be sanded back, so it isn’t different than the rest of the window interior. All of this could have been avoided if the cabin top moulding was fixed up, but Van’s won’t do this, hence the character building.

I trimmed all the windows down to size, using the last of the warm days so I didn’t risk cracking anything by trying to work it in cold weather. I used a dremel tool with a Permagrit wheel for all of the rough trimming, followed by a belt sander with 60, 80, 120 grit belts. When the windows were all trimmed to size I ran around the edge with a 240 grit belt just to be paranoid about removing any scratches. Cutting the windows causes sharp material to fly everywhere, like shrapnel, and it clings to everything. Glad I was able to do all this work outside.

I wanted to run electrical conduits up the front window pillars, but the problem is how to transition them into the Aerosport overhead. I didn’t want to build a front assembly that jutted out, restricting the view forward and upwards from the cockpit. So I 3D printed transition pieces in Nylon that go from the 16mm conduit into a fairly flat channel, and some corresponding channel pieces which when laid together continue the Nylon wiring channel back to the point of accessibility near the forward edge of the front overhead panel. Cut some slits in the front of the overhead, it’s all fairly unobtrusive. A few anchor points inside the roof for Adel clamps completes the wiring capability, so I now have a 16mm conduit from each side up into the overhead space.

With the conduits clamped, I tacked them in place with some dobs of epoxy. Once that cured, I sprayed fireproof expanding foam into each pillar, around the conduits. I then cut the cured foam down to form the shape I wanted on the inside of each pillar. Once the rest of the overhead and door work is done, I’ll do fiberglass layups over this shape. The foam is a bit rough in parts but it’ll be fine as a base to do the layups.

I have a set of Rosen front/side adjustable visors. At one point when the cabin and overhead were on I established the position I wanted the visor mounts, so that the visors just missed the front strut by about 1/4″ when pushed forward. I made an oval metal mount out of 0.025″ Alclad, with two 10-32 nyloc nutplates in the position of the visor mount screws. I oriented these so that they were minimally invasive in the front pillar, which forms part of the rollover protection and shouldn’t be drilled. These were tacked in place with flox, and the way they are positioned I’ll be able to make them merge in when I do the front pillar inside layups, with a bit of filling and sanding.

I drilled the overhead and installed nutplates for the metal inserts. Also did the cutouts for the vents. The carbon fiber is incredibly hard, I completely wrecked a 2″ consumer grade hole saw after just the four holes.

I 3D printed a drill guide for drilling the four holes in the front strut up through the cabin top. The guide worked great.

At one point when I was sick of fiberglass work, I modified the front seat rails, installing AN4 nutplates so that the seats can be removed easily by undoing two bolts, rather than having to take the rail lock off. This is a well known modification that means the seats can be removed in 2 minutes rather than 10-20 minutes of cussing.

Next step is to fit the door struts, and then take the cabin top out, install the overhead, and start on the layups and filling/sanding I need to do in order to finish the interior surfaces.

  • f26h
    Drilling for seat rack modification
  • f26i
    AN4 nutplates fitted for seat rail modification
  • f26j
    Bolts now go in from the top, making seats easy to remove
  • f26k
    3D printed drill guide for cabin top strut
  • f26e
    Trimming left side door window
  • f26f
    Trimming windshield
  • f26g
    Fitting windshield
  • f26l
    Windshield trimmed, retaining clips in place
  • f26c
    Drilling #6 screw holes in overhead
  • f26d
    Drilling out overhead vent holes, ruining a hole saw.
  • f27a
    3D printing overhead wiring conduit
  • complete
    Conduit to cabin overhead adapter
  • cutaway
    Cutaway view of overhead conduit adapter
  • f27c
    3D printed overhead conduit adapter
  • f27d
    Overhead conduits, visor mounts in place
  • f27j
    Wires will come out of channels and be secured with Adel clamps
  • f27f
    Overhead conduit channels
  • f27i
    Overhead conduit channels tacked in place, plus nutplates for wiring clamps
  • f27e
    Overhead conduits, visor mounts in place
  • f27b
    Tacking overhead conduits and 3D printed adapters in place
  • f27g
    Filled out with fireproof spray on foam
  • f27h
    Foam cut to shape, ready for fiberglass layups


Door pins, and moulding a seat for the McMaster door seals [47.0 hours]

I managed to get the door latches worked out, after a fair bit of head scratching. The combination of kits works together well, just takes some patience to sort through the 3+ sets of instructions. I did various other bits of work with the doors – set the hinge positions properly with a layer of flox (left the washers that I had already set the position with in place), and epoxy’d some scrap pieces in position to cover over where the seal needs to go across the hinge recesses. I’ll fill and finish this with micro later on.

One of my awesome sons generously made me a pair of door pins, with points instead of flats, to help locate the position of the fuselage holes. He made them slightly larger in diameter than the Planearound pins, so that they were a friction fit inside the pin blocks – helping to make the hole location mark even more accurate. They worked great! I simply screwed them into the door in place of the Planearound pins, closed the door, and while holding the exterior of the door in the correct position, applied closing pressure to the handle to create a punch mark in the fiberglass. I then drilled these punch marks #40, #30, and then used a step drill to expand the hole to almost the pin size. I used a tapered hand reamer to finish each hole to size, taking off small amounts until the (real) door pins just fitted through the hole. Now the doors can be closed and latched! This sets the final door position all around, so I then final sanded to get the gap properly set to 1/4″ for the McMaster door seal. In a couple of places I had to build up the cabin surface about 1mm to close gaps that were too large.

Using the McMaster seals requires a 1/4″ thick edge all the way around the cabin door frame, and a 1/4″ gap between the door frame and the door. This gap comes from 1/16″ for the seal body, and 50% compression for the 3/8″ seal bulb. The problem is, the bottom rail of the door frame has to be built out to get to the 1/4″ gap, and around the rest of the door frame there is more or less a knife edge left after sanding down to the correct gap. I did two fiberglass operations per door in order to get the correct shape.

For the horizontal edge at the bottom, I cut a strip of 1/4″ Masonite, and covered it with packing tape. I also covered the bottom of the door with masking tape followed by packing tape, so there was no chance I could glue the door, Masonite and cabin top together. I roughed up the lower exterior surface of the cabin (which is concave and has to be filled), and then layered in a mixture of epoxy, flox and cabosil to just fill to the surface established at the top of the lip to be 1/4″. I had to cut some out in the middle to allow the Planearound gearbox shaft to get through. I then closed the door, clamped it shut, and wiped away excess epoxy that oozed out. This operation would be best done after the door pins were in place so the door could simply be latched. In my case I hadn’t done that yet, so I simply clamped the doors closed and used a few shims to ensure the Masonite was pressed firmly up against the top surface of the cabin, the surface I had already established as being 1/4″ from the door. After the epoxy set, I opened everything up, and except for a few voids that would need to be filled, the inside surface was smooth and correct. I then used a dremel tool with a Permagrit cutting blade and routing frame, and cut the top off this horizontal section of the door frame, down to just below the level where the Planearound gearbox shaft protrudes. Finally, I sanded the inside surface, and in some cases a bit off the outside surface, so that this bottom section was exactly 1/4″ from the door when closed (did this after the pins were in and door latched), and 1/4″ thick.

For the second operation, I did the remainder of the door frame in one step. I cut a piece of McMaster seal to use as a throwaway mould (after doing both door frames), long enough to do the top, both sides, and run along the bottom about 4″ each side. After final checks on the 1/4″ gap all around the door (the pins were done by this point so I could latch the door closed), I laid the seal down along a table, and filled the inside with epoxy/flox/cabosil. The mixture was not as thick as peanut butter, but just thick enough so that it would not run out after I pushed the seal in place. It has to be thin enough to be extrude-able from a ziploc bag, as shown in one of the photos. This is best done as a two person operation, I mixed two pumps of epoxy and started applying that while my wife mixed up a second lot. Once the seal was full of epoxy mix, we both picked it up. I had already marked the “centre” position of the seal and the corresponding place on the door. Starting at the centre, I pushed it into place and worked it around and down one side, then switched to the other side. Used a rubber mallet to make sure it was pushed all the way in. Cleaned off excess epoxy with paper towel, followed by paper towel soaked slightly with acetone. Make sure you lay a drop sheet inside the cabin before doing this! Then we drew another drop sheet across between the door and the opening, to prevent any chance of epoxy oozing out onto the door, closed the door and pulled the latch shut. This compressed the bulb of the seal and held the seal in position. After this, I did a final clean up and left it for a day. With the epoxy set, I removed the seal – which releases from the epoxy just fine – and was left with a perfectly formed lip around the cabin, the correct thickness and in the correct place. A bit of sanding to clean up is all that is required.

With this work done, and the overhead work which I’ll put in a separate post, I’m now ready to take the cabin top back off and do the layups and finishing required on the inside of the cabin.

  • f28e
    Custom turned door pins for locating pin holes
  • f28f
    Regular Planearound door pin
  • f28g
    Locating pins installed to locate drill holes in fuselage
  • f26a
    Epoxy on door hinge surface to get correct hinge position
  • f25a
    Preparing bottom edge of doorway for filling
  • f25b
    Taped off door with packing tape
  • f25c
    Applied epoxy/flox/cabosil mix
  • f25d
    1/4" wooden spacer, wrapped in packing tape, in position
  • f25e
    Door closed onto wooden spacer
  • f25f
    Making sure the spacer sits firmly against the cabin skirt
  • f25g
    After curing, good smooth surface with a few voids to fill
  • f25h
    After curing, good smooth surface with a few voids to fill
  • f28a
    Filling in door hinge gaps for McMaster seal
  • f28b
    Filling in door hinge gaps for McMaster seal
  • f28c
    Filling in door hinge gaps for McMaster seal
  • f28d
    Filling in door hinge gaps for McMaster seal
  • f28h
    Trimming the Planearound centre latch block
  • f28i
    Trimming the Planearound centre latch block
  • f28j
    Throwaway McMaster seal ready for setting up doorway
  • f28k
    Filling the throwaway McMaster seal with thickened Epoxy
  • f28l
    Throwaway McMaster seal in place, door latched shut, epoxy ooz cleaned up
  • f28m
    Removing the throwaway McMaster seal to reveal built up door surround
  • f28n
    Built up door surround, prior to cleanup/sanding

Doors! [75.0 hours]

Fiberglass hell continues (apart from two weeks off while I did an awesome flying trip) in the form of the two doors – an infamous part of the RV10 build. Each door comes as an inner and outer shell. You epoxy the two halves together, using the cabin top as a mold. Then you trim and sand each door until it fits. Sounds simple enough, but after installing the door, checking the fit, and taking it off for the umpteenth time it all gets a bit tiring.

There have been many incidents of RV-10 doors coming off in flight. Van’s released a supplement to the design, a “safety catch”, after the first few incidents. An aftermarket design from Planearound is generally regarded as the best solution. It provides a central gearbox with a CAM that “pulls the door in”, and 180 degree handle travel that provides longer pin penetration than the standard kit’s 90 degree travel. In addition, I’m fitting an external handle made by Aerosport Products along with a lock which is keyed the same as the baggage door lock.

I reviewed the information available about doors coming off in flight. The exterior of the door is a low pressure area, which tends to suck the door outward. The bottom of the door can flex and bow outward. If the door is poorly constructed, or if one of the pins is inadvertently not engaged properly, the door can disengage from the bottom and once that happens in flight, it is guaranteed to tear off around the hinges. Many RV-10 pilots never allow passengers to close the doors, electing to always do it themselves to ensure the door is correctly closed. Here is a list of what I’m doing during construction to avoid future door problems:

  • Install the Planearound safety lock (180 degree), instead of the Van’s safety lock. It seems overall a better design, there is greater pin travel and the centre cam lock doubles as a means to draw the door in so that the front and back pins cannot go anywhere but into their respective pin blocks.
  • Install four door pin proximity switches, which will act in series to switch a panel mounted annunciation light from “red” to “green” indicating all four door pins are correctly seated.
  • Adding some supplemental “stiffening” to the bottom edge of the door. To the extent the bottom of the door resists any tendency to “bow outwards” in flight, it is less likely to put stress on the door locking mechanisms.
  • Adding backing plates to the hinge mounting points for both the door and cabin hinge mounts. These came as a kit from Air Ward a long time ago. Not sure if I’ll use the exterior cabin support parts (since the standard screw heads are directly applied to the steel hinges), but for the cabin interior, which winds up hidden by the overhead, and the door interior, the support plates distribute the load across a larger area of fiberglass than four individual washers and nuts. Same for the door exterior mounting plates.

In addition to the above, I’m using an aftermarket bulb seal sourced from McMaster Carr. This seal is applied to the cabin side, and provides a more professional automotive style finish than the standard kit seal, which is applied to the door itself. There’s a bit more work though in that the cabin needs to be sanded down to a 1/4″ gap all around the door, and in turn built up to a 1/4″ edge all the way around to support the McMaster seal.

For the “additional structure” along the bottom of the door, I simply laid some 12mm conduit along the bottom edges in the inner shell, and secured it in a few spots with epoxy. In addition, I cut a handful of conduit sections, just under 3/4″ in length, and epoxy’d one end in place distributed around the large open areas of the lower door, as shown in the picture. When it came time to glue the two door halves together, I put down a couple of layers of glass across the conduits that ran across the bottom of the door, and then filled around it as normal with extra epoxy/flox/cabosil. When the two door halves were brought together, this created a “box section” along most of the bottom of the door. For the other open areas, I simply deposited a clump of flox on top of each short conduit, filling up the inside of each small conduit, and then securing onto the outside surface of the door once the halves were brought together.

I also pre-installed the Planearound gearbox and supports before the doors were glued together. I haven’t seen this done before. It means you don’t have to cut the bottom of the door open after the fact, and fill it back in. I simply drilled a hole in the right place on the inner shell, until the shaft just fitted inside the hole, then used the gearbox itself to match drill the four mounting holes, in turn countersinking these from the inside surface of the door. I pre-lubricated the inside of the gearbox with some Boelube, and then wrapped up the gearbox in packing tape, except for (a) some plain plastic on the rear under the packing tape so that the back of the shaft wouldn’t bind with the packing tape, and (b) the two rectangular holes in the sides of the gearbox where the racks go.

One concern is epoxy from the Parabeam draining down onto the gearbox, into the rack holes, and in turn into the gearbox – that would be a disaster. To ensure this couldn’t happen, I made up a thin Alclad plate which ran along the top of the gearbox, with a flange on each end. I epoxy’d this along the bottom edge prior to bringing the two door halves together. This means any epoxy that ran down after the door halves were brought together would drip down well away from the rack holes. I applied some thick epoxy/flox/cabosil mix along the top of this flange prior to bringing the door halves together, so that any (runny) epoxy from the Parabeam would tend to run away to each side rather than down the outer door shell. This all might be overkill, but it worked out OK because after the doors were cured, neither gearbox was seized up!

Gluing the two door halves together is definitely a two person operation. I had “Rosie” mixing up epoxy/flox/cabosil while I applied it. We both carefully placed the two halves together, and then onto the cabin/mold. I previously drilled #40 cleco holes through the door halves into the cabin. These are easily filled with (structural) epoxy later, and they allow uniform pressure to be applied while the door cures. Clamps can easily be over tightened, causing flat spots around the natural curve of the doors. After the doors were epoxy’d and placed on the cabin, I left for a two week flying trip, so there was plenty of time for the epoxy to cure!

The cleco’s came out OK, some requiring a bit of twisting and force to remove them due to the epoxy running down into the cleco/holes. It took around eight hours per door to trim off the excess and sand down the edges to match the cabin top. I used duplicator straps to keep the door alignment left/right during this step, and one on each top corner. I highly recommend the latter – four straps in total – because the top edge of the door is where you start the final fitment, this is where the hinges get drilled. Once the door is on the hinges, it hangs slightly differently because there’s more weight towards the front of the door than the rear (the hinges are not in the door’s C of G), so you can’t trim the lower parts of the door until the hinges are fitted. It’s all a bit of a chicken and egg problem, but after having each door on and off about 20 times the doors are a good fit. I only took enough off the doors to have them fit into place, more sanding will be required after the cabin top is properly fitted to the fuselage in order to set up the gap all around each door.

After all that, I have each door now hinged to the cabin, and quite an accurate fit, the bottom edge of each door is less than 0.5 mm proud on the front and back edges, and perhaps 0.5 mm under in the middle. Ultimately the transition from the door to the fuselage will have to be filled and sanded all around, along with the transition into the cabin top, so +/- 0.5 mm is good enough for now.

The doors (and cabin top) are a long part of the project. I have quite a way yet to go with them, but the weather’s been good so almost all of the sanding I’ve been able to do outside.

Notes from the future for builders:

  1. Don’t drill the bolt holes around the lower doorway from the cabin top into the fuselage, just drill #30 holes on the side and #40 holes on the bottom. You’ll want to remove and reinstall the cabin top on multiple occasions while you work on the doors and overhead/interior, and it’s much easier to use cleco’s than bolts. Enlarge these to the required bolt holes once you’re ready to final install the cabin top.
  2. If you’re going to use the Planearound kit, definitely install the gearbox onto the inner door shell before gluing the two door halves together. For the 3/8″ access hole required – use their measurement for the distance forward from the gearbox, but for the up/down distance, don’t measure it – insert a rack into top side of the gearbox, roll it forward until it intersects the “vertical” line you just drew for the forward distance, and use the hole in the rack as a drill guide to drill a #40 hole through the door. That makes a pilot hole in exactly the correct position, now enlarge the hole to 3/8″. Later on, place some masking tape over this hole while sanding the door, you’ll be creating clouds of fiberglass dust and there’s no need to have that accumulate inside the door or gearbox.
  3. If you’re going to use the Aerosport exterior handle, pre-drill #40 the three holes using the stainless steel striker plate as a template, through both the inner and outer door shells, while the two door halves are all cleco’d together for initial preparation. This is the first step in the Aerosport instructions. Why? Because when you glue the doors together, you can install three #40 cleco’s (could use #30’s) through these holes and that brings together the two surfaces where the door latch mechanism goes. The instructions tell you to coat these surfaces with regular epoxy, but that is fairly runny and once the door is placed on the cabin for the epoxy to cure, the epoxy can run and a poor bond can occur in this area. This happened to one of my doors, I had to use a syringe to squirt more epoxy into it while installing the latch, and this shouldn’t happen. In fact I’d put a thin layer of the thickened flox on this area in the first place, slathered on top of the thin straight epoxy which will run.
  4. Use plenty of the thickened epoxy, and remember on the aft side and the bottom that the final trim is quite close to where the two shells part. Be generous with the amount applied, you need to work fast and there’s no point in trying to economize on the amount of epoxy (it weighs next to nothing) or focusing on making the application pretty.
  5. My formula for the thickened epoxy was 4 pumps of epoxy, 4 pumps of slow hardener, 2 scoops of flox, and 2-3 scoops of cabosil. Mix the epoxy and hardener thoroughly first, then mix in the flox, then mix in the cabosil, until the mixture does not slump if held vertically. We made up 3 of these per door (in addition to the regular epoxy used for the Parabeam). Might have been a 4th on one side, can’t quite remember.

Next job is to install the door latches, which means rationalizing three sets of instructions – from Van’s, Planearound and Aerosport. Fortunately, plenty have gone down the same path before me, so between these instructions and some online reading I should be able to work it out.

  • f24a
    Starting initial trim of door shell halves
  • f24b
    Drilling door shells for cleco's, using cabin top as a mold.
  • f24c
    Drilling door shells for cleco's, using cabin top as a mold.
  • f24d
    Installing internal support for gas strut
  • f24e
    Pre-installing the Planearound gearbox
  • f24f
    Wrapped in plastic + packing tape, with shroud
  • f24g
    Additional supporting material
  • f24h
    Epoxy applied, door halves cleco'd to fuselage "mold"
  • f24i
    Epoxy applied, door halves cleco'd to fuselage "mold"
  • f24j
    After initial trim, and addition of duplicator straps
  • f24k
    Initial clearance
  • f24l
    Sanding initial (upper) door fit
  • f24m
    Sanding initial (upper) door fit
  • f24n
    Sanding initial (upper) door fit
  • f24o
    Drilling door hinges into cabin top
  • f24p
    Fitting Airward cabin top hinge reinforcements
  • f24q
    Fitting Airward cabin top hinge reinforcements
  • f24r
    Airward cabin top hinge reinforcements
  • f24s
    Initial door fit, after a lot of sanding
  • f24t
    Initial door fit, after a lot of sanding
  • f24u
    Broomstick will eventually be replaced by a gas strut!
  • f24v
    RH door initial fit, after a lot of sanding
  • f24w
    Initial trim of 1/4" gap for McMaster seal
  • f24x
    Initial trim of 1/4" gap for McMaster seal
  • f24y
    Door trim debris
  • f24z
    Now to figure out the latch mechanism...


Starting fiberglass hell [115.0 hours]

In the context of the Fuselage build plans, I’m up to section 36 – brake lines. That’s followed by fuel lines, rudder pedals and brake system, control system, flap system, upper forward fuse installation, rear seats and then the cabin cover, doors and windows. I’ve decided to change things around a bit and work on sections based on the weather. I don’t want to be working on the cabin cover and doors over winter when it’s cold and miserable working outside, and fiberglass doesn’t set enough overnight to be sanded the following day. So, I’ve started on the cabin cover. I’m going to work on other sections during the times when either the weather’s a problem, or I just can’t stomach doing fiberglass work.

I also need to paint the interior. This made me think about eventually painting the exterior. I decided to buy some topcoat, in a dark grey tint, and use the interior as a means to learn how to shoot it. The first test pieces were a disaster. On a hot summer day, there was so much moisture in the compressor air, it totally overwhelmed the water trap in my filters. To fix this I built a heavy duty water trap, consisting of 18 metres of 1/2″ copper tubing, immersed in cold water. I built it in a wheelie bin, with incoming (hot) compressor air tapped in near the bottom of the helical tube. A drain valve at the very bottom of the tube can be manually opened to release accumulated water. The cold immersion water comes from my bore pump, 75 metres underground, so the water is always cold. I turn the water feed on whenever I spray, and the outflow just drains into open ground in a paddock. As the hot compressor air cools, water condenses and runs to the bottom of the helical copper pipe. The outlet at the top of the pipe feeds into the regular three stage air filters (which I replaced). I also installed a final water trap on the belt regulator. This resolved the water problems and greatly improved the spray job. I still had a bit of silicone contamination, from my priming habit of using syringes (which are lubricated with silicone oil) for de-greasing solvent. I’ve ordered some materials to do a better job of cleaning, so we’ll see how that goes.

I built the back seats, they were easy and are in the spray booth ready to prime.

I built and installed the flap system. The drop saw with the blade I use for aluminium cutting worked well for trimming the UHMW bushings. I was a bit concerned about drilling the safety wire hole in the flap motor, but that turned out to be a non event. I forgot to final ream the 1/4″ holes in the flap crank, and had to use an angle drill to get down into the tunnel and clean the holes out after the components were assembled. I could have disassembled everything but that would have taken longer.

To trim the cabin cover, I used several different tools. My ancient jigsaw with a perma-grit blade worked a treat on the open sections. For trimming the door openings, I used the van’s supplied cutting blades in a die grinder for some parts, and a perma-grit circular cutter in a dremel tool for others. Plus a belt sander with 80 grit paper, and several different perma-grit hand files. The Van’s scribe lines weren’t much use. I used a series of measurements, triple checked it all, and cut to the measurements. I was able to get the cabin cover to fit in place on the second attempt. It came off and went on a few more times to adjust the sides and rear deck, which turned out to be a bit long and had to be trimmed.

I decided to drill everything #40 to start with. This way, I didn’t have to open up as I went to remove shavings from between the canopy and the skins. It also makes it very easy to drill the rivet backup strips. I match drilled #40 from the cabin cover to the rear backup strip. For the side rivet backup strips, I simply placed them on top of the longerons and match drilled from the side skins – easy. Then I re-installed the cabin cover, and cleco’d everything up including the rivet backup strips with #40 clecos. With everything installed and cleco’d, I then went around and final drilled #30, again not needing to clear out shavings.

I’ve got a long way to go with the cabin cover and doors.

  • f23a
    Dryer for compressor air
  • f23b
    Dryer for compressor air
  • f23c
    Building rear seats
  • f23d
    Starting to trim the cabin top
  • f23e
    Starting to trim the cabin top
  • f23f
    Making a template for the rear angle
  • f23g
    Cutting flap UHMW bushings
  • f23h
    Flap UHMW bushings
  • f23r
    Flap system being assembled
  • f23s
    Right flap horn
  • f23u
    Flap motor safety wire hole drilled without incident
  • f23v
    Flap motor installed
  • f23i
    Cabin cover offcuts
  • f23j
    Final adjustments before drilling cover
  • f23k
    Rear skin overlaps a bit too much
  • f23l
    Pushing cabin cover up against skins prior to drilling
  • f23n
    Ready to start drilling cabin
  • f23t
    Cabin cover drilled and cleco'd
  • f23p
    Initial drilling done
  • f23q
    Initial drilling done
  • f23w
    Preparing to drill rivet strip
  • f23x
    Drilling rivet strip
  • f23za
    Match drilling #30 skin, cabin top and rivet backing strip
  • f23zb
    Match drilling #30 skin, cabin top and rivet backing strip
  • f23zc
    Setting the correct position for the Aerosport overhead duct/console
  • f23zd
    All skin, cabin top and rivet backing strip holes now drilled #30
  • f23ze
    Work for the future - Aerosport 310 panel


Baggage door [31.0 hours]

Today I finished the baggage door. There’s a problem with the RV-10 baggage door. The top surface of the door structure has a curve in it, which doesn’t match the fuselage. The plans instruct you to build the door on the bench. If you do this, it may not fit very well, and it’ll bulge out about 2mm due to the mismatched curve problem.

I got some advice on these issues, and built the door on the fuselage, forcing it into the fuselage shape while riveting it together. I built the door from the hinge out, and it worked out really well. In addition, like many builders I arranged the hinge pin as two halves, which can be inserted from the inside, making the baggage door easy to remove. This also allows the hinge to be set almost flush with the fuselage, rather than protruding out as the plans indicate (since in the standard plans, the hinge pin is inserted from the outside).

Riveting the outside skin of the baggage door is a bit awkward. I managed to slip once and put a very slight dent in the skin near the hinge line. This is easy to fill prior to painting, annoyed me a bit though because it’s the first rivet gun blemish on the entire fuselage.

I’ve now done enough to the fuselage to paint the interior. Before doing that I have to make a water trap to properly dry the compressor air before the filters.

  • f22a
    Baggage door hinge pins
  • f22b
    Baggage door hinge pins
  • f22c
    Baggage door hinge halves
  • f22d
    Ready to rivet in fuselage hinge half
  • f22e
    Ready to rivet baggage door skins
  • f22f
    Baggage door complete
  • f22g
    Completed baggage door
  • f22h
    Completed baggage door