2.2 Tests of Autographs A7 and A8 
In reference 1.342:1 several international tests of first order stereoscopic projection instruments have 
been reported. The tests were made according to the grid method and attempted to determine the 
basic accuracy of the projection systems, i.e., to distinguish as completely as possible between regular 
and irregular errors of the reconstructed bundles of rays in different base positions and to estimate the 
irregular errors in terms of standard error of unit weight after least squares adjustment. 
Similar tests have been made regularly in the A7 autograph nr 310 and at time of delivery of the A8 
autograph nr 1427. Typical results are presented below. 
It should be noted that the A7 autograph was delivered in 1953 and has been in continual use—in 
cluding student training—for more than ten years. All tests were made with grid nr 410, which had 
been checked in the StK 824 stereocomparator, see section 2.1. The computations were made according 
to the principles discussed in reference 1.342:1. 
The results of tests of the left projector are shown graphically in fig. 2.2:1 through 4 and of the right 
projector in fig. 2.2:5 through 8 (the Appendix). Two operators made the measurement in the three 
base positions. The standard errors of unit weight for the two operators are very similar and show a clear 
increase with the radii. From a great number of additional test points which were not used in the adjust 
ment operations, residuals were computed and the distributions tested. In all cases the distributions 
were found to be normal on the 5 per cent level. The degrees of freedom were 10 for base zero and 6 for 
base in and out. 
In summary it should be noted that the radial distortion effects in the projectors of the A7 autograph 
are significant and should be corrected together with other distortion effects in compensating plates. 
The standard errors of unit weight indicate a decrease of the weight of image coordinates with increase 
of radial distance. The average value of the standard error of unit weight is about 4.5 «m, which agrees 
well with the value found in reference 1.342:1. It appears that the standard error of unit weight of 
5 am can be used as a basic standard of quality for the A7 autograph. 
In fig. 2.2:9 through 11 in the Appendix results of the tests of the A8 autograph nr 1427 are shown. 
In fig. 2.2:10 the individual residuals are shown. The tolerance of 10 /urn is not exceeded. From a con 
siderable number of similar tests the average of the standard errors of unit weight of modern A8 auto 
graphs is 4 um, which result may be used as a standard of basic accuracy. 
2.3 Weight Variations of Projector Coordinates. A7 Autographs and C8 Ste- 
reoplanigraphs 
Projector tests of A7 autographs and C8 stereoplanigraphs were made according to the grid method 
in connection with the international controlled experiments of Commission IV I.S.P. 1958—1960, 
reported to the London Congress 1960. See reference 1.342:1. A systematic study of the variations of the 
standard errors of unit weight of the projector coordinates was made to see if there is a weight variation 
in the reconstruction of the bundles of rays similar to that which has been found relating to image coord- 
dinates. See fig. 1.61:5 and 1.62:9. 
In fig. 2.3:1 the results of the investigations are shown. There are significant weight variations of the 
projector coordinates, which will exaggerate the weight variations from the image coordinates of the 
photographs when the reconstruction of the bundles of rays in the instruments is investigated. 
2.4 Tests of the Radial Distortion Effects of Correction Glass Plates 
Radial distortion effects in photographs can be corrected by numerical, optical or mechanical means 
when the bundles of rays are reconstructed. In the Wild autographs aspheric correction glass plates are 
frequently used. At times such corrections can be introduced by aspheric glass plates in the printing 
apparatus when printing diapositives from the original negatives. To test the distortion correction effects 
of such glass plates some experiments have been performed.