Why so many rivets?

I agree with Bob ... the numbers seem very low. However I have little doubt
that a production airplane will have less rivets than most homebuilts.

In general homebuilt airplanes tend to be quite complex in their
construction ... there are more bits in the firewall of an RV-6 than an
entire C-150 fuselage (OK maybe a little exageration but not too far wrong).

There are several reasons for this:

(a) Designers of homebuilts don't have access to production tooling such as
guerin presses to make firewalls in a single piece. Consequently they use
more, simpler to make pieces, which need more rivets.

(b) In general designers of homebuilt airplanes are not as focused on
refining the product for production ... if you work for C, B or P etc then
you need to design lean for production. You also have the facilities, budget
and extensive test programs to help refine the design. The big companies in
some cases design unconservatively and then beaf up the structure where it
fails. Homebuilt designers in general design conservatively and cannot
afford for a part to fail during test.

(c) Few homebuilts are subject to extensive structural analysis and
consequently the designers has no idea what the margins are re rivet shear
and inter-rivet buckling etc so they have no choice but to stick to the
typical minimum rivet spacings.

I think it would be safe to say that you could not produce many homebuilts
economically .. they are simply over built and their crudeness, whilst good
for homebuilders is not suited for volume production .... the number of
rivets is a huge cost driver WRT to cost of production since the labour in
any airplane will be the most significant single cost in its production.
That is why stir friction welding is being used in the Eclipse business jet.

There are some exceptions to my generalisation re homebuilts being more
complex that factory produced airplanes. The MM-I for example is an
elegantly simple airplane with a minimum number of bits - but then Dave Long
was a Piper design engineer. The T-18 is also a very simple airplane ... of
course John Thorp also had a background in the design of airplanes for
production. I think it all comes down to the designers background.







"Bob K." wrote in message
oups.com...
Earlier, Brian Sponcil wrote:

...I recall the builders telling me that
their RV kits have around 10,000 rivets.
Compared to a Piper Comanche at 3,714
and a Warrior at 1,785 that's a heck
of a lot of rivets.

I've seen various RVs, Cherokees, and Comanches. I've kitted rivets for
HP-series sailplane kits. And I just plain do not belive those numbers
for the Warrior and Comanche. Not for relatively conventional riveted
aluminum airplanes with few composite components and no metal-to-metal
bonding. Until it's demonstrated otherwise, I propose that someone has
cooked the books on this one to make for a good story.

Consider the Warrior wing: Let's guess that the rib spacing is a
relatively lean 12" OC. Let's guess that the rivet spacing is an
equally lean 2". The span of the metal stuff (minus fiberglass tips) is
probably about 32". Taking the fuselage out probably leaves room for at
least 14 ribs on a side. The wing area of 170 ft^2 over the span of 35'
yields an average chord of about 58". Since the skins are riveted top
and bottom, I think that there are going to be about 58 rivets per wing
rib. So that yields at least 58*30 skin-to-rib rivets, and that's 1740.

Admittedly, that's a pretty rough estimate, and disregards the
(probably negative) contribution of the flaps and ailerons to the rivet
count. But it's a start. When you factor in the rivets between the spar
and the skin, between the ribs and the spars, and for the many
inspection panel rings, stringers, and other local additions, you see
you can easily exceed the stated rivet count for the wings alone. And
you've still got an entire fuselage and set of tail surfaces to go. And
also the extremely close-pitched rivets around the baseball-stitched
fuel tanks, and other miscellanea.

I'll change my mind if, when I next see a warrior, I see fewer than 28
wing ribs or greater rivet spacing than 2". But until then, I'm not
convinced.
Thanks, and best regards to all

Bob K.
http://www.hpaircraft.com