1) bad pixels and cosmic rays corrected but non-telluric corrected Object cubes (7
OBs so 7 cubes).
I am not totally sure what do you mean with “in counts (no average counts)”: the
combination of the cubes per OB is included in the Pipeline.
In the Sinfoni Pipeline Manual (latest version 19.5) about the cube coaddition they
say:
Each pixel intensity of the coadded cube has an intensity given by the weighed mean
of the intensities of each corresponding overlapping pixel from the contributing
cubes, the weight being given by the exposure time of each target frame. 

In the Folder OBJ_7 you find:
- 1OB observed on 28/08/2015 (for3_20150825.fits)
- 3OBs observed on 15/09/2015 (for3_20150915_a.fits, for3_20150915_b.fits,
for3_20150915_c.fits)
- 3OBs observed on 16/09/2015 (for3_20150916_a.fits, for3_20150916_b.fits,
for3_20150916_c.fits)

1b) For completeness and in case you wish to combine them in some different way I
attach also all the cubes before being combined.

In the Folder OBJ_21 you find the single bad pixel and cosmic rays corrected Object
cubes (21 in total)

for3_20150825_01.fits
for3_20150825_02.fits
for3_20150825_03.fits

for3_20150915_a_01.fits 
for3_20150915_a_02.fits 
for3_20150915_a_03.fits

for3_20150915_b_01.fits 
for3_20150915_b_02.fits 
for3_20150915_b_03.fits

for3_20150915_c_01.fits 
for3_20150915_c_02.fits 
for3_20150915_c_03.fits

for3_20150916_a_01.fits 
for3_20150916_a_02.fits 
for3_20150916_a_03.fits

for3_20150916_b_01.fits 
for3_20150916_b_02.fits 
for3_20150916_b_03.fits

for3_20150916_c_01.fits 
for3_20150916_c_02.fits 
for3_20150916_c_03.fits

-------

2) bad pixels and cosmic rays corrected but non-telluric corrected NON-sky
subtracted Object cubes per OB
Exactly same procedure as in point 1) just without sky subtraction.

In the Folder OBJ_nosky_7 you find :
for3_nosky_20150825.fits
for3_nosky_20150915_a.fits, for3_nosky_20150915_b.fits, for3_nosky_20150915_c.fits
for3_nosky_20150916_a.fits, for3_nosky_20150916_b.fits, for3_nosky_20150916_c.fits


In the Folder OBJ_nosky_21 all the 21 NON-sky subtracted Object cubes.
-------

3) Night sky cubes per OB: the reduction on the skies is performed exactly as if
they were objects except that of course no sky is subtracted.

In the Folder SKY_7:
sky_20150825.fits
sky_20150915_a.fits, sky_20150915_b.fits, sky_20150915_c.fits
sky_20150916_a.fits, sky_20150916_b.fits, sky_20150916_c.fits

In the Folder SKY_14 all the 14 night sky cubes.

-------

4) bad pixels and cosmic rays corrected Telluric stars:
In the Folder TELL you find the five tellurics:
-25/08/2015: 1Tell (tell_20150825.fits) -1OB
-15/09/2015: 1Tell (tell_20150915_a.fits) - 3OBs - 1Tell (tell_20150915_b.fits)
-16/09/2015: 1Tell (tell_20150916_a.fits) - 3OBs - 1Tell (tell_20150916_b.fits)

The Telluric star was always Hip014898 (B3V C T~17600K)

In the TELL folder you will also find a text file for each night with the seeing
conditions of each Telluric and Object-Sky observation recovered from the header of
the raw data. (The value of -1 is most likely due to issues with the monitoring)

-------

5) *non-telluric* corrected Template star:
Of the three Template star I have reduced only two since for one of them I couldn’t
find any telluric close in time to its observation.
In the Folder TEMP you will find the combined template star cubes and corresponding
tellurics:
- Template star SC33_0357 (obs: 05/04/2015): temp_20150405.fits   -   Telluric:
temp_tell_20150405.fits (Hip085851 - B5II/III D - T~15200K)
- Template star SC41_3304 (obs: 18/06/2015): temp_20150618.fits   -   Telluric:
temp_tell_20150618.fits (Hip087505 - B2II C - T~21000K)


Notes:
- for the cosmic rays and bad pixels correction I have used the 3D version of the
Laplacian cosmic ray identification based on L.A.Cosmic written by van Dokkum (2001)
adapted by Davis (2008).
- I am using the last version of the sinfoni pipeline 2.7.0 with esorex.
-I had a few issues with the Distortion frames but until now I couldn’t find better
distortion frames to use.