Also, it looks like you might have a bug in generating the airfoil shape. An airfoil like the 8412, while extreme, is well-defined and constructible from NACA-style combination of thickness and camber forms; but you show it as having a "kink" on the bottom surface, which is definitely not correct.
841220 is, sadly, not a legal NACA airfoil designator.
There are three major families of procedurally-defined NACA airfoils -- four-digit like the 2412, five-digit like the 23012, and the 6-series like the 64A012.
It looks like this site only expects the four-digit form, and is mis-parsing anything other than that -- which is a pity, because one of the uses of a tool like this is to get a sense of /why/ the 23012 is similar CLmax to the 2412, but with lower moment and a more sudden stall. If nothing else, input validation is necessary!
Yep. If you're generating them yourself, you'll probably want to be careful to follow the NACA method for combining camber and thickness profiles -- basically, running a circle of thickness-profile-defined diameter along the camber line and unioning the areas. The alternative (which is these days sometimes called the Riblett approach) is just to define the airfoil as the camber line plus or minus half the thickness vertically at each point -- which often leads to better airfoils for many purposes, but will disagree with available NACA data, especially w.r.t. nose radius (and thus, critical angle for airfoils which stall near the nose).
If the goal is to allow a user to compare the properties of different airfoils, there's a lot to get right to make sure they actually know what airfoils they're comparing. The alternative is to allow airfoil selection from one of these sites [1][2], which also allow a link to provide comparative analysis.
