Joe Wagner's Columns:
Laminated
Balsa Surface
Outlines

For
Scale Models
By Contributing Editor: Joe Wagner vecojoe@juno.com

Lightness combined with strength and resiliency should be THE
goal in all model airplane construction -- and especially for flying
scale types. That's not so hard to accomplish as it may seem.

Several years ago I designed an all-balsa model bridge that my
youngest daughter built & used to win a "bridge-busting" event
at a school Science Fair.

The bridge weighed under
2.7 ounces -- and held 576 POUNDS
before failing. It was all-balsa; glued with cyanoacrylate; and had
no reinforcements of kevlar, fiberglass, carbon fiber or anything
else. That shows what lightweight all-balsa construction can be
capable of -- if you know how to use it to best advantage.

One way to gain great strength with light balsa structure is by
lamination -- in particular, laminated curved outlines. Wingtips
and rudder outlines are the usual applications for this. But with
a little imagination you can find other places in model airplanes
where laminated balsa will provide strength, flexibility, and
lightness that can't be attained in any other way.

Laminated outlines are far from new. They've been employed on
snowshoes and wooden tennis racquets for centuries. Model
airplane designers have been featuring them for years on such
types as Peanut Scale rubber models and the flying surfaces of
Andy Clancy's many "Lazy Bees" and their variants.

The trouble is, the recommended methods of forming laminated
balsa outlines have never worked well for me. Those methods are
messy and imprecise. I found it hard to achieve void-free laminations
with them. Some of my laminated parts looked OK when I removed
them from their soggy & deformed "molds". But when it came time
to shape and sand them to their final contour, gaps in the glue
line often showed their ugly presence. (It's not so much that the
cracks look bad; it's that they're evidence of a weak area that
should have been as strong as the rest of the laminated outline.)


I won't waste time and space describing the lamination methods
that I've found faulty. Instead, here's a procedure that
WORKS.

First, you need a sturdy forming mold. Cardboard won't do. I use
pine lumber salvaged from carpentry projects; a friend of mine
makes his molds from "blue foam" insulation board. Whatever
material you choose should be at least three times as thick as the
part you want to make, firm enough to hold pins, and FLAT.

Now to shape the mold outline. Look at your full-size plan; and
remember that it's NOT necessary to limit your lamination to a
constant width. By that I mean the wing outline for a Spitfire
could be 3/8" wide along the straight leading edge portion,
widening to perhaps 3/4" at the very tip; then widening even more
as the curve moves around to the trailing edge.

When you've decided on its shape, draw a pattern for the
INSIDE
curvature you want for your lamination. Transfer the outline to
your mold blank (it's OK to assemble that from narrow pieces for
something like a large chord wing), then saw the mold accurately
to shape.

If you don't have a bandsaw, an alternative is to take your
project to the woodworking shop at your local high school and
explain to the instructor what you want to accomplish. Chances
are he'll allow one of his better students to cut and sand the
mold to shape for you. In fact, if you ask in the right way, you
may be able to get the school shop to do the whole job for you,
including supplying the material.

Wrap the finished mold's edge area smoothly with Saran or
Monokote backing material. Hold that in place with tape.

Now let's figure what you'll need in the way of balsa for the
laminations. Its thickness depends on the radius of curvature.
The smaller that is, the thinner the balsa you need. As a rule of
thumb, 1/32" balsa will go around as small as a 3/8" radius with
careful handling. 1/16" balsa will bend around a 3/4" radius, and
so on. Check your pattern for its tightest curvature; and make
your lamination thickness about one sixth of the minimum pattern
radius.

Actually, the job's not that scientific. Once you've made a
couple of parts, you can tell "by eyeball" what thickness
laminations to use. And, because the inner radius gets bigger
with each lamination you add to the mold, you can begin with say
1/16" strips and finish with 3/32" for the outside strips.

Now for strip width. For parts such as wingtips, where you need
two identical laminations, cut your balsa strips three times as
wide as the maximum height you need the finished lamination to
meet. Looking at that Spitfire wing again: the height of the
lamination at the leading edge is probably the maximum needed.
(The tips and T.E. will be less.) Say that the max. L.E. height
is 3/8"; then the lamination strip width you require is 1 1/8".
(Make that 1 1/4" just in case.)

For parts you need just ONE of, such as a rudder or a bulkhead,
make the laminating strips at least 1/8" wider than the finished
part's thickness. (A little more is better yet.)

To figure the length of the strips, just wrap a ribbon or the like
around your mold; measure how long that is; then add an inch
or so. (This method is a bit wasteful of material; but the savings
in time and the gain in strength justify that waste to me.)

Find how many strips you need by dividing the maximum planview
width of the laminated outline by the strip thickness. Back again
to the Spitfire wing: say we earlier decided to make the trailing
edge of the laminated outline a full inch wide. We're going to be
laminating with 3/32" balsa; dividing one inch by 3/32" comes out
slightly under 11. Add one "for good measure": you need a dozen
strips.

Cut these from sheet balsa somehow; you know your preferred method
better than I do. Now put them all into a watertight container.
A bathtub is fine; it might be more convenient to use a length of
plastic sewer pipe capped on one end. Even a plastic bag will work
if you have nothing better available. Fill the container with enough
plain water to submerge the balsa strips.
DON'T add ammonia or
detergent to the water; those may badly weaken the glue bond.
Plain tap water is best --but it takes HOURS for that to fully saturate
balsa.

Balsa SEEMS to soak up water fast. In fact, what happens is that
the surface cells get saturated; those then expand and produce an
almost watertight barrier skin within the wood that slows moisture
from penetrating more deeply. (That's what makes balsa rafts and
life preservers stay afloat. Truly
SATURATED balsa hardly floats
at all, as the Kon-Tiki voyagers discovered the hard way...)

I allow 24 hours soaking per 1/16" of balsa thickness. Hot water
penetrates somewhat faster; but personally I use room-temperature
water and schedule my model building tasks to allow plenty of
soaking time for the lamination strips.

Now to do the work! There are two excellent adhesives for balsa
lamination. One is Sig's "Sig-Weld" white glue; the other is
Great Planes "Pro Wood Glue" (an "aliphatic resin" type that's
free of the brittleness of other "yellow glue" brands).

Remove the sopping wet balsa strips from their container.
Squeegee the surface water off one strip by pulling it between two
fingers; then pin one end firmly to the mold about half an inch
short of the "start point" on your plastic-covered mold edge.

Bend the strip around the forming mold, keeping it in tension as
you go. (Pressing the wood down to force it against a tight
radius MAY cause a fracture; but bending while pulling on the free
end avoids that problem.) Pin the other end to the mold after the
bend is complete.

Now brush a wet coat of full-strength glue on the outside surface
of the balsa strip on the mold. The moisture in the wood will
"thin" the glue somewhat and ease the brushing task. Try for a
uniform coat: not opaque -- that's too thick an application; yet
not transparent either (that's too thin).

As soon as the glue coat's complete, squeegee the surface water
off another balsa strip. Unpin the "starting" end of the first
strip; press the second strip firmly down over the glue, and re-
pin. Now pull the second strip around the outline. Glue should
ooze out all around; that's just what you want. It proves ample
glue is between the laminations, and it also gives a good start
for brush-gluing the next wet balsa strip.

Repeat this procedure until you have all the strips in place.
For some curved shapes it may be necessary to clamp the strips
against the outline at various places around the perimeter.
That can be done in various ways. One that works for me is to
use short lengths of plastic cut from a bleach jug, the same width
as the laminations -- and drill holes through the interior of the
mold for 1/4" dowels. Then to clamp each lamination in place as
it's applied, I put a piece of the "jug plastic" on top of the wet
wood strip, and run a rubber band around it onto a dowel sticking
through the mold. Done this way there's little interference with
the strip-by-strip lamination procedure.

By the time the last strip is in place, there's wet glue all over
everything. No problem! That's what the extra material is for.
Let the glue dry. 24 hours is my minimum curing time; and if I
can let the lamination set three days before disturbing it, I KNOW
I won't get any unpleasant surprises when I slip it off its mold.

Now for the final stages. Remove the pins (and any clamps) from
your laminated outline and free it from its mold. Discard all the
tape and glue-encrusted plastic film. Next slide the lamination
back in place on the mold and pin it a few places. With a coarse
sanding block, true up the edges of the lamination, removing
unevennesses and glue blobs until you've got clean, smooth, level
edge surfaces all around.

Now sand (or plane if much material needs removing) the outer
perimeter of the lamination almost to its final contour. Again
you'll be removing excess glue and surface flaws in the wood from
clamping. Again, that's what the extra material was provided for.
And finally, for "matched pairs" of laminated outlines, use a
balsa stripper to split the single "triple thickness" lamination into
a matched pair close to the required final dimension.

Now the lamination's ready to become part of your airplane.
If possible, instead of joining your laminated outline to parts
(such as straight leading and trailing edges) with a butt joint,
take the time to make long, tapered "scarf" joints instead. Those
will transfer stresses (such as impact loads) gradually from one
structural member to the other, rather than abruptly the way butt
joints do.

The whole secret of maximizing strength with minimum weight
in model structures lies in avoiding stress concentrations, and
permitting the forces generated by flight loads (and the
inevitable ground contacts) to spread throughout an area rather
than localizing. That was the secret of my daughter's
"unbreakable balsa bridge". (Parts of that were made from
SOFT
balsa, in order to better distribute the stress loads.)

Laminated balsa parts are one of the very best ways of achieving
great strength and damage resistance at minimum weight. They take
advantage of their "grain running around the perimeter"
characteristics to provide strength through
RESILIENCY rather than
cumbersome, heavy, and stress-concentrating "reinforcements".

Remember: light models can fly both faster and slower than heavy
ones -- and they don't impact nearly as hard.

* * * * * E N D * * * * *
Back

Scale Aero and Details 1998