In Dec 2016, new calibrations are being obtained for Run 15.
These calibrations do not use the LED data for extrapolation but are done with pi0 fitting. 
Some algorithm improvements have been added.

I have looked at new calibration results and for large cells these calibrations seem to essentially fix the top-bottom rate asymmetry.
On the small cells, there are many problems and more will need to be done.

There is still a large asymmetry in corrections from high voltages between top and bottom.  That is what the new calibration takes out. 
This gives us a complicated trigger acceptance, which is different top and bottom. However, secondary rates should be relatively unbiased with the new calibrations.

So far I have attached new calibration text files for
pAu
day 142
day 145
day 150
day 156

pAl
day 160
day 165

I will be adding to this list in the next days.

my notes are here