[R] Upper limit in nlsLM not working as expected

[Martin Hehn]

Dear John,

this appears to be one of the cases where your algorithm is favourable over the nlsLM one. The fitting is now a lot better with  ssquared being lower compared to any other fit method I was comparing to. 

As I am relatively new to R, it still puzzles me that there are so many packages doing what I want but they seem so hard to track down.

Best regards,
Martin