The Selig gang is up to I believe volume 5 of their low speed airfoil studies. Here is a link to volume 1.
https://m-selig.ae.illinois.edu/uiuc_lsat/Low-Speed-Airfoil-Data-V1.pdfHere’s a link to volume 5
https://m-selig.ae.illinois.edu/uiuc_lsat/Low-Speed-Airfoil-Data-V5.pdfJust a few thousand pages in the middle. I remember reading volume 1. One of the main points I took away was the degradation in performance due to inaccuracies in construction. If a giving design parameter didn’t perform in the real world the way it did in the wind tunnel or in simulation, it was most likely because it was inaccurately constructed and not because it was a bad design.
I think too often when we incorporate features into our designs that don’t work as expected we fault the idea and not the execution. This is likely what Doug Moon experienced when he installed triangle stock on the leading edge in an attempt to replicate Paul’s leading edge shape and he recognized that after seeing the sketch of Paul’s leading edge in a previous post.
I agree with Doug, this had been a most interesting and educational thread.
Edit: the reference to construction inaccuracies was not from volume 1 above but from the book, Airfoils at Low Speeds, 1989 by Selig, Donovan and Fraser.
5.5 Surface Waviness and Contour Accuracy
So long as an airfoil surface is smooth, that is,free of sharp edges protruding into the boundary layer, surface waviness of the type produced by Monokote over balsa appears to have no measurable impact on performance. For a 12 in chord, the Monokotejbalsa waviness has a peak-to-peak amplitude on the order of 0.02% chord. On the other hand, the warping of the covering in open-bay construction as it stretches over the cells alters the airfoil contour beyond what is normally considered waviness. In such cases, the airfoil really has no single shape and one can expect its performance to be significantly different from the nominal.
Based on our measurements, the best modern-day construction techniques used by modelers are capable of yielding contours accurate to ±0.004 in. For a 12 in chord an error of ±0.004 in is only 0.033% chord. Although at present there is no criterion on the accuracy necessary to meet the nominal performance, all indications are that errors of this order have only a small affect. On the other hand, errors two to four times this amount (which are more common) do begin to effect performance. Error in contour is undoubtedly one of the factors that explains differences in performance between two different models of the same RC sailplane.
The book is available here:
https://m-selig.ae.illinois.edu/uiuc_lsat/Airfoils-at-Low-Speeds.pdf