next previous
Up: A proposed new design


4 Conclusions

Several conclusions can be drawn from the present study. First, there should be large cost savings in constructing a primary surface supported by the ground compared with constructing a conventional steerable paraboloidal surface. Motion of the surface panels is almost entirely a vertical displacement, involving no change in gravitational loading. Thermal effects should be well balanced over the surface, and should be small because of the small vertical dimensions of the structure. The surface should therefore be stable. As a result, panels can be large but lightly constructed. Adjustment to improve their shape will be independent of zenith angle and should give a permanent benefit.

A further result of the stability, and a consequence of having each panel supported by the ground, is that no radome would be needed for mm wavelengths. Also, it may be possible to make accurate mm-wavelength panels with inexpensive materials and little precision machining as sketched in Fig. 10. The feasibility of this type of construction can be determined by building one panel for testing. A disadvantage of the telescope at mm wavelengths is the greater atmospheric loss that results from the very long focal length. However, the loss is very small compared to the potential gain in collecting area. Even a single telescope array element 110 m in diameter (Table 3) would have over 10 times the collecting area of the IRAM 30 m telescope, the largest existing mm-wave instrument. This compares with an estimated 4% loss for one complete atmosphere (a 10 to 12 K atmospheric temperature contribution) from microwave background measurements at $\lambda = 3.3$ mm (Smoot et al. 1985).

The principle risk associated with the telescope is clearly in the positional stability of the balloon. It is believed that the stability problem may be solved by having (i) smaller scale balloon motion rapidly corrected by moving the feed point electronically and (ii) larger scale motion corrected (less often) by moving the diffraction image to follow the balloon (iii) six-axis control of the balloon position and attitude and (iv) a good site without mountains or unevenly heated ground. A computer simulation of balloon motion using existing meteorological data on winds would go a long way in demonstrating the effectiveness of these measures. A second step would be to use the simulation result in actually constructing a balloon with a six-axis control system and testing it at a good site



Acknowledgements

I thank the referee for his comments, particularly in pointing out the change in efficiency that can result from having the diffraction image follow the balloon and the importance of illumination spillover.


next previous
Up: A proposed new design

Copyright The European Southern Observatory (ESO)