Hi Streamkite,
There is no physical basis for picking the value of Alpha End and for most airfoils we have found a value of 30 works best. If this leads to unrealistic results for your particular airfoil, I’d recommend using the tool to find a better value via trial and error.
In the TableExtrap column, aspect ratio is span divided by mean chord length or span squared divided blade planform area. Not sure about the limitations of the Cd extension methods - I’d recommend reading the paper by Viterna that describes his method.
And finally, I wouldn’t start by extrapolating your full range from -10 to 90, but rather use -10 to 30 (or even 20) - then compare this to the actual data you have at high angles of attack to tune the parameters in the spreadsheet. Once there’s agreement between the fitted curve and actual data, compare the extrapolated full range using the -10-20 data and the -10-90 data and see what looks to be more physically realistic. I’m sure many people on this forum would be interested in your results if you did such a comparison and are willing to share the outcome.
cheers,
Pat