plate distance z is approximately 71 mm. The r.m.s. residual over the 
plate width of 52 mm was 91 nm, which could be reduced to 83 nm r.m.s. 
error if an assumed residual distortion correction of (9n- dont) 
nanometres was applied. The maximum value of this correction (for 
n = + 10) was 60 nm, which would imply a systematic angular sensing error 
of only £t 0.15 arc seconds. 
CONCLUSION 
This is an encouraging first observation, but it may need considerable 
investment in precise two-axis measuring equipment before the full 
potentialities of these CENTRAX camera systems can be properly evaluated. 
Given three such cameras operated in conjunction with suitable targeting, 
and given suitable computer algorithms, a precision of 1 um per m may 
ultimately come within reach. 
Acknowledgement is due firstly to Dr J W C Gates who initiated the 
NPL programme in photogrammetry and secondly to Mr G Turner of the 
NPL Glass Workshop who successfully fabricated the CENTRAX prototype. 
REFERENCES 
Burch, J.M., "Optics in Metrology" 1958, ed Mollet, Pergamon Press, London 
p192. 
Downs, M.J. and Raine, K.W., 1979. An unmodulated bi-directional fringe- 
counting interferometer system for measuring displacement Precision 
Engineering 1, 85. 
Dyson, J., 1958 Circular and spiral diffraction gratings Proc. Roy. Soc. 
A 248, 93 
McCleod, J.H., 1954 The axicon - a new type of imaging element J. Opt. Soc. 
Amer. 44, 592. 
Steel, W.H., 1958 Axicons with spherical surfaces "Optics in Metrology" 
ed Mollet, Pergamon Press, London pp 181-191. 
94