Patrick, there is no reason opening three threads for the same question
http://communities.ptc.com/message/205711#205711
http://communities.ptc.com/message/205714#205714
and this here.
We had that same or a very similar question of yours some time ago
Richard told you here http://communities.ptc.com/message/190719#190719 and here http://communities.ptc.com/message/194391#194391 that your modelling seems to be incomplete as the fit function provided won't fit your data and you would have to include something additional in you model. While you functions were different there, the conclusion is the same with your model here, I think.
The data you provided here http://communities.ptc.com/message/197133#197133 and here http://communities.ptc.com/message/198781#198781 seem to be more similar to the one you have now.
As I already wrote in the foregoing threads, it would be necessary to have a parameter for scaling in x/t-direction. This would be a factor of x in your function Function() which is not present. So again, with the function you provide, a good fit will not be possible, I guess.
In the attached file I simply added a third parameter ag2 as a factor of x and got a much better fit. I think you should be able to get rid of one of the other parameters, using a fit function with two parameters only as in the foregoing threads, but I didn't bother doing so.
