Important facts about the NBC:

1) CCDs: 2 2kX4k Marconi CCDs, 15 micron pixel size. Very high sensitivity
in the blue
 (60-70% @3500, 90-95% @4000A, 85-90% between 4000 and 6000 A, >80% till 6500A). 
Readout time is 42 seconds. Pixel scale is 0.135"/pix. Noise ~ 2.5 ADU, gain
~1.6 e-/ADU.

2) Geometry: the CCDs are rotated 90 degrees with respect to the previous
configuration (rotation is clockwise on the figdisp display).
The gap between them runs  along  the direction of the bar on the slitmask
 (i.e. along the direction of dispersion in a spectrum). The gap is 100+/- 2
pixels wide corresponding to  13.5" on sky. The bar on the slitmask is 9.5" on sky,
and it is centered within the gap. Two extra arcsec are now lost on each side of
the bar. This may have consequences on the slitmask design: you may lose part of the
slits which are close to the bar. 

3) Longslit observations are affected because the center of the slit falls
in the gap. A new pointing origin called "slitb" has been defined: it
will place the object 30" from the inner edge (30" from the gap) of the
right hand chip, the one with higher QE. The object is visible close to the left 
edge of the slit viewing guider camera. 

4) Spectral coverage is 25% more because we have more pixels; we cover
the full LRIS field along the spatial direction (all 8'; the red side
is only 7.3'). 

5) The images are rotated 90 degrees CW with respect to the red ccd.
Slitmask alignment is affected by the presence and the geometry of the gap. 
A new set of IRAF scripts is available for slitmask alignment on the blue
and shell scripts for telescope moves. 
The total prescan area is 204 pixels, the overscan 320. 

6) Below 4000 A, the CCDs uniformity gets worse and a brick wall
pattern becomes clearly visible. More care with flat fielding is required. If you
plan to use the 1200 grism,  it is recommended to take series of flats during the
afternoon (with the internal lamp-- dome lamps will not provide enough UV light).

7) Image quality: the blue camera, with the coma-correcting first camera 
element now in its nominal setting, and with a flat chip for the first
time, achieves pinhole images of 0.23" through its useful wavelength range.
Thus, if the telescope and camera are well-focused, it should be possible to 
to obtain very good images in the blue under good seeing conditions.