I just realized that the films I was using for 5.5 and 11 mins are from a different batch. That's why the corresponding curves reach the 0.17 mark earlier than expected which has a visible effect on the resulting estimations for development time and effective film speed. I did not really care about the batch numbers since I never would have guessed that impact would be that strong. However, doesn't this mean that as soon as I purchase new films which are most likely from a different batch, I actually would need to redo the whole calibration process?