This site, compiled by Dave Lednicer, lists the LS-3 as using the
Wortmann "FX 67-K-170" at the root and a "FX 67-K-150 mod" at the tip.

http://www.ae.uiuc.edu/m-selig/ads/aircraft.html

This agrees with the data in the Thomas book as well as the
directories by Martin Simmons. It was a very popular airfoil on a lot
of gliders from that era.

If you go to this site, you'll find a data file that lists the
coordinates in non-dimensional values.

http://www.ae.uiuc.edu/m-selig/ads/coord_database.html

Now the tricky part....

It may or may not be 17% thick at the root, and typically the airfoils
were scaled in thickness to allow for more uniform spar height
tapering. As a result, you'll probably need to measure the thickness
vs chord ratio at a number of stations and scale appropriately to
match. Also, the tip airfoil is different, and who knows where the
designer started to "blend" into this airfoil along the span...

Now the cool part...and I'm sure I'll get flamed from people who
disagree.... Can't wait....

It really doesn't matter that you get the shape just right for the
full length of the airfoil. The underlying structure is by itself
fairly accurate. You're not going to be sanding away fiberglass. What
does matter is that the profile itself is relatively free of waves and
that the underlying surface quality is quite smooth. Waviness will do
more damage than a slightly in-accurate profile. Hence, you could get
by with just stripping down the gel-coat to the glass, spraying up a
nice uniform thickness of new gel-coat and then spending all of your
time getting rid of chord wise waves. Once that is done, polish up the
final result to a mirror smooth surface and you're done. Technically
you don't even have to polish to mirror smooth, 600 grit sand paper
would yield a smooth enough surface, but your gel coat will last a lot
longer if you go higher. No templates necessary.

-Kevin