
There are longerons and a wide stiffener that is
riveted to the bottom exterior of the floor skin. The wide
stiffener is no problem, but I ran into a problem on the longerons.
When I marked the centerline down the length of both longerons, I
marked the center of it using the whole width of the longeron, just like
before. This gave me a rivet spacing as shown in the left hand
photo above, with the rivet hole too close to the upright part of the
"L." What I should have done is marked a centerline
using just the open area of the longeron. (Previously,
all the longeron riveting has been done from the flat side of the
longeron.) The holes forced the head of the rivet gun too close to
the upright part of the longeron, and the rivets would not lay flat.
I puzzled over this a bit, and thought I could get a
solid rivet set for a rivet gun in there better. A local EAA
chapter member and Harmon Rocket builder (modified RV), David Chasnoff, allowed me to come over to his
workshop and try some solid riveting. I did, and after ruining
several, quickly came to the conclusion that while I was able to
make some perfect rivets, setting 50 or so in each longeron made it
very unlikely that I do that many without messing up something.
My hat is off to all those RV folks who set domed head rivets on .025
aluminum!
The solution was suggested by David, who graciously
filed off the end of one of his rivet pullers that fit in there just
right, shown on the right Make sure you leave room for the rivet puller head when
you mark the longerons, but still preserve edge distances.

Here is the cradle, built according to the new
drawings, in place to support the cabin. The forward fuselage is
on the work table.

And here is the forward fuselage resting in its
cradle. This is an exciting moment - it is really starting to
look like an airplane - soon I'll be able to sit in it and make engine
noises! I think I will add some lumber on the cradle where the
rear and forward fuselage comes together for a little better support..
I have the tail all the way back to the wall. I
think I could disassemble the work table at this point, and move
it all into the single stall of the garage, but that would be
cramped. I will put off doing that for a while. Lots of room on the worktable now.

The cradle rested the cabin floor at 4 degrees as
nominal, and the rear fuse opening was 74 degrees as nominal.

It looks like the sides are going to match up very
well. Right now, it is pushed back as far as it will
go without trimming the rear longerons about 20 mm..

Here is how the front and rear come together - the
rear frame has a cut in it to allow the longeron to pass through, and
a thick doubler is riveted along side it to retain its structural
strength. The holes on the side will be drilled as soon as the
front and rear are aligned properly.

After the cabin floor and rear fuselage are joined
(just cleco'd for now) I made the templates to position the firewall
and instrument panel. Since the floor needed to be placed and
removed several times in the process of joining them, waiting to put
the heavy firewall and templates on makes sense. Also, the aft
firewall stiffener is clamped in place.

Before the holes for the rivets that join the front
and the rear fuselage can be drilled, the front must be aligned to the
rear. Here, a string was tied at the far rear fuselage on the
centerline, then lined by sight across the front of the rear fuselage
and tied to a ladder in front of the firewall. This provided a very
straight line from the rear across the top to the front. Then, a
string was suspended just above this line, with a plumb bob down near
the centerline marks on the firewall. Then, the cradle was
adjusted until everything lined up, and the initial holes on the sides
were drilled and cleco'd.
Jump back to 5th page
Continue to 7th page
|