Concerning instrument response/atmospheric extinction, Christian Buil has been moving away from the reference star at the same air mass method recently.
Essentially he is separating out the true response of the instrument which he reuses, from the atmospheric extinction which he models. Although this is similar to the technique pros use, I am not totally convinced about this for my atmospheric conditions so am currently still using the old reference star as similar air mass method, though I do admit to reusing instrument responses quite freely for H alpha at high resolution with the LHIRES where the atmospheric extinction effect is small. David Boyd was at OHP this year so perhaps he has some updates on this new method?
I think your method of reusing the dispersion equation should work well with the ALPY provided the temperature does not change too much as it seems very stable to me. I would suggest still taking a lamp spectrum and using a lamp line as a single anchor point to avoid any problems with overall shifts. You could also use this lamp spectrum to make sure there has been no change in dispersion eg due to temperature by checking the position of second line in the spectrum