OD range 0.1-1.4
linearity 0%
no color correction
monochrome mode
Zeiss proprietary lookup table (LUT) 

choose "average boxed densities"
draw box around each spot along x direction and around background near each spot
same for y direction:  

integrated amp = sqrt( (measureed avg intensity - bkg avg intensity) * num of pixels)
e.g., in above image, 1 listed spot (1st order x direction) would be:  222.17 = sqrt( ( 57.72-0.59)*864)

Values calculated at various times on various scanners all from copies of this image

10/28/99  Our scanner immediately after installation
order        lowerleft           lowerright            middle
         x        y              x       y            x        y
 1      1.0      1.0            1.0     1.0          1.0      1.0
 2      0.384    0.415          0.375   0.407        0.379    0.410
 3      0.0738   0.0964         0.0695  0.0911       0.0720   0.0943
 4      0.0363   0.0335         0.0347  0.0310       0.0358   0.319

order        upperleft           upperright
         x        y              x       y
 1      1.0      1.0            1.0     1.0
 2      0.383    0.417          0.371   0.405
 3      0.0746   0.0987         0.0680  0.0910
 4      0.0367   0.0356         0.0345  0.0360
 

9/28/99  Our scanner at factory (by Klaus Neumann)
order        lowerleft           lowerright            middle
         x        y              x       y            x        y
 1      1.0      1.0            1.0     1.0          1.0      1.0
 2      0.398    0.434          0.370   0.410        0.388    0.422
 3      0.887    0.114          0.710   0.098        0.0816   0.107
 4      0.0287   0.0282         0.0247  0.0250       0.0284   0.0290

order        upperleft           upperright
         x        y              x       y
 1      1.0      1.0            1.0     1.0
 2      0.390    0.427          0.383   0.416
 3      0.0831   0.108          0.0785  0.103
 4      0.0281   0.0282         0.0251  0.0269
 

7/16/99  (from Klaus Neumann - 2 different focusses)
order  scan001x scan001y    scan002x  scan002y
       (emgrid1)             (emgrid2)
 1      1.0      1.0           1.0      1.0
 2      0.40     0.436         0.31     0.333
 3      0.086    0.114         0.040    0.054
 4      0.025    0.026         0.015    0.017
 

7/22/99  (followup to Gordon Conf)
order        lowerleft           lowerright            middle
         x        y              x       y            x        y
 1      1.0      1.0            1.0     1.0          1.0      1.0
 2      0.396    0.428          0.387   0.428        0.380    0.412
 3      0.086    0.111          0.080   0.110        0.0764   0.100
 4      0.027    0.027          0.0277  0.0298       0.0258   0.0260

order        upperleft           upperright
         x        y              x       y
 1      1.0      1.0            1.0     1.0
 2      0.397    0.426          0.398   0.432
 3      0.086    0.110          0.0867  0.107
 4      0.0262   0.0250         0.0278  0.0273

7/10/99  PDS1010GM (NYU) (same data as previously provided)
         x         y
1       1.0       1.0
2       0.488     0.490
3       0.114     0.143
4       0.048     0.040     

Acceptance criteria specified at time of the order:

An electron micrograph of a 400 mesh electron microscope grid
(provided by Richard Henderson will be scanned at a 7 um interval.
Scans of this image will be made in the center as well as at each
corner of the platten and equal quality will be expected at each of
these positions as detailed below.

A 512x512 pixel image will be Fourier transformed and the integrated
intensity of four "spots" will be determined both along two
orthogonal directions, after subtracting the local background (using
the "Average box densities" function in the program Ximdisp).  These
directions will be slightly askew to the scan axis to accomodate our
method for integrating the spot intensities (see attached Fourier
transform).  The integrated amplitudes will be calculated as the
square root of these integrated intensities.  Amplitudes will be
normalized by dividing them by the amplitude of the 1st order spot.
The "standard" data will come from the flatbed, PMT-based scanner in
Cambridge made by Joyce-Loebl; this performs roughly the same as the
PDS 1010GM currently in my laboratory (data shown below).

We will require that the normalized amplitude for the 4th order spot
be at least 50% of that from the flat-bed scanner in both directions
and at all locations on the scanner platten (that is a normalized
amplitude of 0.020).

I would really like to convince you to make these scans at the
factory and to send them to me prior to shipment (e.g. by email) so I
can verify the proper functioning of the instrument.  This analysis
will only take me a couple of hours and could save you the expense
and me the inconvenience of replacing the optics (or the whole
instrument) after installation.