Forward Belly Skin
Working on the forward belly skin, F-321. This one is the .040 piece that goes to the firewall.
The first bit of work I had to do was to correct the bottom edge of the forward lower longerons and the F-312 (mine aren't stock) ribs so that their aft edges fair to the seat-area belly skins. I'd made them too high. Fortunately I had enough material to do these things.
I put a description of this on the RV-3 Clarifications page.
After that, I had to bevel the forward edge of the forward belly skin to fit between the firewall bottom cross-piece and the firewall lower flange - wish that were wider, but it's not.
Then I could start clecoing the skin on. I worked from the going front aft and things went well enough until I got to the spar bulkhead. That has the wing dihedral angle in it so that the bottom of the skin isn't flat. Since the forward belly skin is one-piece, that meant that there's a compound curve going over the bulkhead flange.
I made a tooling piece and suppose I should call this a Howe fitting like the ones I used on the wing. Same sort of affair, held on and pulled tight with straps. Fortunately the .040 skin is relatively robust and it pulled down without trying to buckle.
Putting the carpet on the fitting, with a much-used roll of carpet on the right,
And then I added a post in the center area, where I wanted local pressure.
The skin overlap at the center of the spar bulkhead, where the seat belly skins overlap, left a bit of a gap under the F-321 forward belly skin. I glued on some shims and they're curing now. I can strap down the Howe fitting when they're set.
The photos are also at this place and this one too. Just in case.
I clecoed the forward belly skin back on and used straps to tighten the Howe Fitting in place over the spar bulkhead. That is, it's "over" because the fuselage is being built upside down, but as far as the airplane is concerned, it's under the belly.
The compound curve due to the overall geometry still produced a ripply in the skin. I could not find a way to position the Howe Fitting to avoid that, even using shims. This shows the ripple.
With no other alternative, I consulted my mentor, who said that he'd cut the skin to relive the ripple. I decided to try it. A colleague in Australia had sent me a photo showing that he'd done it too; plus the RV-4 plans show something similar in the same location.
The cut was made using a Dremel cut-off disk and took a very short time. It's about .050 to .060 wide and 1.72 long. The hole at the end is for stress relief and is reamed to #30.
Here it is in place. Bottom line, it works. The ripple disappeared.
Now I wish that I hadn't thrown out the nicely drilled exoskeleton for the spar bulkhead after gluing on the seat belly skins. But it's a small workshop and alas, I did that.
The pictures might also be found here, here, here and here.
Wow, what a great documentary on building the kit. I have picked up a kit ( SN 10006) with a lot of the work already done. But reading about your building helps me understand what all goes into one of these. I am a couple of years away from having a place to finish building the kit, (if I do) and will continue to review your building to learn more. I am also looking at possibly selling the kit but want to pass on as much information as I can. My wife(as do I) filp/flops on sit on it versus sell it. Storage may become a problem.
I will be researching the RV-3 more anyway. I would be a first time builder and need to move to an area with a good building chapter for help. Right now I am working towards the CFI rating as I enjoy sharing the sky?s and teaching/coaching people.
Thanks for sharing all your work on this with us.
Garwin, I tend to be slower than a lot of builders, so don't get the impression that an RV-3 is a huge amount of work, as it might not be for someone else. It's entirely doable and it's been an enjoyable process. I started this kit more for the project than for a flyable airplane and that might tend to draw the project out a bit, too.
You're right about documenting the construction of it. I hope to do that just to show how I've approached certain tasks and some of the RV-3 hidden gotchas.
Lately I've been working on the tunnel and that's been one of the more fun little jobs. I'm still figuring out how to add some Fiberfrax and titanium to the belly for insulation and as a firewall. I just need to settle on a construction sequence but there are no real stumbling blocks with it. I'll discuss that when I have some more time as the April - June period tends to be my busy season.
With the tunnel shells and bulkheads drilled and fit, it was time to consider the belly insulation and fire protection. The insulation is 1/16" Fiberfrax and the fire protection is .020 Titanium. I chose that because the FAA recognizes .016 Ti as acceptable for firewalls and .020 was available from McMaster at a reasonable price.
Most people who do this use stainless steel foil. KatieB used that foil, for example.
But I got to thinking that the protection is more than from heat; it's to guard against actual fire under the belly behind the engine hot air outlet. The belly is aluminum, .040 thick to be sure, but still aluminum. I decided that primary firewall protection should be added. Hence the .020 Titanium.
At one point, it looked as if the Fiberfrax and Titanium sandwich would both go between the tunnel shell inner flange and the belly skin, and made this test piece to see how much the Fiberfrax compresses after being riveted with LP4-3 rivets. The side view shows minimal compression.
When looked at in the right light at the right angle, though, some dimpling is visible in the simulated belly skin.
Later, I riveted an AN470AD4 rivet in one end and that seriously did compress the Fiberfrax.
Therefore: the Fiberfrax should not be part of the structural connection between the skin and the tunnel. I decided to stop the Fiberfrax by the inboard edge of the tunnel shell. This led to some rivets on the belly going through the Titanium and others not. It appeared easy to screw up (still does) so I made this guide.
One issue is that with the Fiberfrax stopping at the inboard side of the tunnel shell, and the rudder pedals above the tunnel shells, what about insulation there? But I think that the tunnel shells themselves will give me some insulation, and if I add a bit of the Titanium to the inner wall, it'll be somewhat protected, for a short while anyway, against flame. With no Fiberfrax on the inner walls, heat on the Titanium can work through to the aluminum tunnel shell. No problem for normal flying but in a real fire situation, they could melt. Still, that'll give me some protection for a while.
The photos can also be found -
After making the titanium belly overlay, I clecoed it in place. Forward is to the left. That overhanging part is to allow for a rounded entry for the air exit.
Then I glued the belly piece, the aluminum one, on to the frame. We'll get to that photo shortly.
With the belly on it was time to rough-fit the forward side skin. I started on the right-hand side. I had assumed that the cut to fit the overhanging seat belly pan would be difficult and it wasn't too hard. It was just a matter of marking the edge of the belly skin, fitting, repeat. Another part that I'd thought would be hard was locating the rear spar stub, which pokes through the side skin.It wasn't so bad, either. I drilled holes at either end of the cut-out and then used my Dremel to cut the slot out.
Did you know that I've already worn out my first Dremel-like tool? Now I'm on a name-brand one, and frankly preferred the older one, a "Master Mechanic."
Here's the rough-cut of the right-hand forward side skin in place. You can see that I went slightly too far on the bottom of the rear spar slot.
And here's a photo of that forward belly skin I promised you, plus the untrimmed bottom edge of the side skin. The rectangles on the belly are there for the double-flush middle rivets to the titanium overlay. These are also stand-offs to ensure that the Fiberfrax doesn't get compressed by these rivets. The rivets are double-flush so that they don't interfere with my feet in case I leave the floor uncovered.
I'd taken the opportunity of the above photo to mark the side skin for trimming. Here it is trimmed and unfinished. You can also see the titanium overlay, now dimpled, resting on the fuselage out of the way.
When people say that the RV-3B isn't prepunched, this is what they mean. This side started as a rectangular piece of aluminum, cut slightly oversize to allow for trimming.
You might note that several skins haven't been riveted yet: the right-hand tailcone side skin and the belly skins forward of the rear spar. These need to be riveted before these forward side skins get glued on.
The photos are also hosted here, here, here, and here.
Here's the titanium belly overlay in place. Nothing unusual about this except that the .063 spacers in the middle area under it are forcing it to lay in a mild compound curve, since the perimeter lays flat on the aluminum belly. I just clecoed it on so that I could verify that indeed, this works.
Then it was time to start shimming the area at the seat bulkhead where the bottom longeron and the angle that goes by the seat area all intersect. The design is somewhat crude there and I ended up with a bit of space to shim on the angle at the seat. This is the area at the red ellipse below.
If I ever build another RV-3, I'll pay a lot more attention to this area during the construction. And no, I don't have a solution that you can implement on yours. Please understand that since I'm not a great builder, that I could have simply screwed up -- but I don't think so; this is not the only one of these to need something here.
I'd bent the angles, somewhat, to allow for this, but there was still some space to fill. So I clamped an angle on the outside of this area and measured the gaps. From that I selected some shims from my scrap pile, cut them, abraded them and cleaned them, and then glued them on. Here they are with the glue still wet and the clamps still on. First the right side.
And the left side,
I anticipate that these will be sort of a starting point, with more shims necessary. But I won't know that until I've cleaned these up.
The photos are also here, here, here and here.
"The design is somewhat crude there and I ended up with a bit of space to shim on the angle at the seat."
Yes, a good description! I shimmed mine here too but not to the same extent, same idea just not quite as far.
The Titanium layer on the forward belly has a portion that needed to be rounded to be part of a more efficient cooling air exit than the stock design. In this view, it?s already partly bent. I just grabbed one end of the pipe with Channel Lock pliers and rolled it up. The pipe, of course, was taped to the Titanium.
Then I rolled the initial curve into it the same way as we roll the empennage surface leading edges.
Getting a pipe with a fatter diameter, I was able to give a more gradual aft radius.
This is the final shape, except that I expect I?ll need to do some trimming when I put the engine mount on.
The front bottom of the belly has some nuts and bolts that attach the engine mount fittings. I?ve got these nuts torqued. The reason they?re being shown here is that the tunnel shell will eventually get riveted over them and they?ll not only be inaccessible, they?ll be hidden from sight. However, if absolutely necessary, an access hole could be put into the tunnel bulkheads, which are now pilot-dried for that.
The alternate hosting has them here, here, here, here, and here.
This isn't the main stuff I've been working on lately but it's part of the project.
I’d gotten one of Gil Alexander's Skyview network hub printed wiring boards and the rest of the components, all connectors, made it a kit. I don’t need it yet but I thought that I ought to build it up now anyway. A friend of mine builds electronics equipment prototypes so I asked her a couple of assembly questions. Rather than an answer, she just told me to bring it to her lab. She then soldered the connectors to the board for me. I have to admit that she did a lot better job than I’d have managed, using tools that I don’t have. Here’s Maria Toscano-Leary soldering the SV hub kit.
Here’s the network hub side by side with Dynon’s version. The main difference is that Gil’s includes the two autopilot connectors directly. The white connector on Gil’s is a provision for the power and ground wires to the autopilot servos. There were no assembly surprises.
I'm still working on the electrical layout and haven't actually decided which I'll use yet. With the panel being both small and intended for day VFR cross-country, I might just go for the lighter one, after I make some sort of enclosure for the sides and bottom of the kit hub. Right now I'm thinking of a wrap of electrical tape....
The photos are also here and here.
|All times are GMT -6. The time now is 10:55 PM.|