When launched, the HST carried five scientific instruments: the
Wide Field and Planetary Camera (WFPC),
Goddard High Resolution Spectrograph (GHRS),
High Speed Photometer (HSP),
Faint Object Camera (FOC) and the
Faint Object Spectrograph (FOS). Since it was too difficult to bring the HST back to earth for repairs the engineers considered everything from replacing the telescope's secondary mirror by sending a spacewalking astronaut into the telescope's optical tube, to installing a circular shade around the opening of the tube, which would reduce the aperture and improving the focus by blocking out the outer regions of the primary mirror. It was eventually determined that with the HST still in orbit that they could replace the WFPC with the improved
Wide Field and Planetary Camera 2 which would incorporate corrective optics. A crisis meeting of the European Space Agency was held in Germany to discuss the issues with the HST. Among the attendees was James H. Crocker, a senior optical engineer at the Ball Aerospace Corporation. While taking a shower in his German hotel one morning he noticed that the showerhead travelled on a vertical rod and it could be clamped on the rod at different heights and angles. The maid had left the showerhead at the base of the rod and positioned flat against the wall, which meant it was taking up very little space until Crocker loosened the clamp and moved it to the position he wanted. The idea came to him that they could mount the required corrective components on such a device that would allow them to be inserted into the tube before being folded out on robotic arms to the required position to intercept the beams of light from the secondary mirror, correct and then focus them on the various scientific instruments. To fit the COSTAR system onto the telescope, one of the other instruments had to be removed, and astronomers selected the
High Speed Photometer to be sacrificed, which was the least important of the four axial detectors. A total of ten correction mirrors with diameters ranging in size from approximately 18 to 24 mm were used as the Faint Object Camera and Faint Object Spectrograph each had two apertures for each of their two measuring channels whereas the Goddard High Resolution Spectrograph only had one aperture for both of its channels. The design was complicated by the need to ensure that the beams of light for the above instruments which were mounted at the end of the telescope tube missed the beams for the new WFPC 2 which was mounted on one side of the telescope tube. In January 1991
Ball Aerospace Corp. was selected by NASA as the prime contractor to undertake the entire development, production and verification of COSTAR, a process which took 26 months. To calculate the required corrections one team calculated the existing error by examining the still in situ tooling that was used to make the primary mirror, while another independent team calculated it using the distorted images that Hubble had transmitted. Both teams came to practically identical measurement results. The correction mirrors that were subsequently produced were then checked for errors by two independent teams. Once complete the entire COSTAR was then tested in the COSTAR Alignment System (CAS). To check for any error in the CAS the COSTAR was then mounted in the specially developed Hubble Opto-Mechanical Simulator (HOMS) which simulated the errors in the faulty primary mirror, to allow an end to end test and thus verification of the output image. The HOMS system was also tested by two independent groups (one from Ball Aerospace and the other from the Goddard Space Flight Center) using different test instruments. The European Space Agency also contributed to the verification process by providing an engineering model of the Faint Object Camera to provide additional verification. ==Installation==