Hi Nate,
I took a look at the airfoils profile data of the NASA/LANGLEY LS(1)-0421 AIRFOIL
To be found in
www.ae.uiuc.edu/m-selig/ads/coord_database.html#L.
as an example to test, and, yes, although the data are given to 10 high minus 5 of the chords length there are bumps in the airfoils profile suggesting that the data my have errors.
A laminar airfoil of a typical chord of 1 meter should be constructed with a maximal waviness under 0,1mm and the profile resulting using these data I refer to has waviness above this limit.
I tried smoothing it and found best results when approximating using AKIMA??s cubic splines. When using my approx-interpolations program I can also additionally visually change interpolation points and using this feature I was able to smooth the curve ??satisfactorily??. I am not that firm in aerodynamics as to dare to affirm that these changes I did would not have an effect of the gross aerodynamic properties of the airfoil. But I believe, that they are unimportant. And I believe that smoothing out the waviness is definitely necessary.
As concerns my ??formula?? I see I have not succeed in making my point clear to you.
I try again:
If we have a look at the set of data points for an airfoil
/LANGLEY LS(1)-0421 AIRFOIL
38. 38.
0.00000 0.00000
.00200 .01560
.00500 .02377
.01250 .03599
.02500 .04912
.03750 .05853
.05000 .06606
.07500 .07771
then the entries of the left column are my ??s?? (you call them ??x?? ) and the entries in the right column are our ??y?? and also my ??y?? but I write them y(s) as their value is given by ??s??. ??s?? attains values between 0 and 1.
If you wish to fabricate an airfoil with a profile matching those ??s?? and ??y(s)?? as above AND if you will construct that airfoil as a laminate OVER a foam core, then you should not hot-wire the foam to this same profile, because applying the laminate on top of the foam will change the dimensions. What you have to do is to reduce the dimensions of the foam core by an amount given by the thickness of the material you will apply. So the shape of the ??reduced?? core is not given by the above set of data points s, y(s) but by a different one s, Y(s) and I gave you the formula
Y(s) = y(s) ?? thickness x square root of(1+ y?(s)?)
relating both sets.
As above, s is between 0 and 1
(This formula has to be appropriately understood for s < thickness and for s > 1-thickness, there the ??reduced?? foam core has disappeared).
Inversely if we find an original pattern for hot-wiring a foam core to, say, a GU 25 profile, do not expect it to match exactly to the GU25 data set in aeronautical data bases; they must differ for the reasons I am explaining.
Of course we should pose us the question: And if we do not correct the shape of the foam core as described, does the resulting deviation of the shape of the airfoil from the desired GU 25 shape have any importance? Well, a back of the envelope calculation tells you that we then increase the wing surface by 1,8%, we change maximal thickness from typically 17% to 16,5% this having some consequences on maximal lift coefficient and stall angle and we make a definitely blunter airfoil nose. But again I do not feel aerodynamicist enough to judge.
This and only this I wanted to suggest. I now hope that I succeeded clearly expressing what I wanted to say.
As concerns your comment on not using sandwich composites but a single layer of laminates instead, of course, you easily get strong enough structures with them. Here in Europe we have different AC using this single laminate techniques, Pipistrel??s Sinus and Virus, Remos, CG??s, and many other . You get fuselages for LSA??s with typical weights in the range of 35 kg. BUT forget using the technique of ??moldless composites?? with them, you have to construct a negative mold that is very expensive and makes sense only if you think of laying up a series. Besides this, although you finally get heavier structures with sandwiches, you get a number of advantages also. Impact resistance of a sandwich laminate is orders of magnitude above that of a single laminate and so passenger protection. Thermal protection also. Imperviousness to resonance effects, jerry-canning for example, resistance to twisting and bending forces are orders of magnitude higher with sandwiches and this dispenses of having to go through difficult calculations on oscillation frequencies and damping.
Regards Juan