Working On The Detailed Design
As a result of selecting all major partners in 2002, the JWST project began a re-plan phase in order to bring the telescope design, the instrument designs and the science requirements in concord and to make the whole project fit within the previous estimated budget constraints. Phil Sabelhaus replaced Bernie Seery as the NASA Project Manager.
In 2003, the Project had several possible launchers to take the observatory to L2. ESA was willing to provide an Ariane 5 launch vehicle that would allow simplifications in the primary mirror design because of its superior lift capability when compared to the Atlas V. Nevertheless, the effective area of the primary mirror was reduced to 25 square meters (down from the 29.4 in the original TRW proposal). TRW merged with Northrop Grumman and had became Northrop Grumman Space Technology,(NGST!). The number of segments were reduced from 36 to 18. The originally combined NIRCam and guider were split into two separate instruments, with the US providing the NIRCam and CSA providing the guider. The MIRI cryocooler was replaced by a cryostat to eliminate the need for an additional technology development (this decision was reversed in 2005).
In the summer of 2003, the NIRCam and NIRSpec teams chose HgCdTe Rockwell) technology for their NIR detectors. Similarly NASA selected the beryllium-based mirror technology for the telescope's 6.5-meter primary mirror based upon the recommendation of the JWST prime contractor, Northrop Grumman. The technology had been supplied by Ball Aerospace & Technologies Corporation. The production of the beryllium-based mirrors began within a year (Press release).
The Project officially moved from Phase A (preliminary design studies) to Phase B. In Phase B (which was ultimately a four year period, the detailed JWST design would be developed. However, procurements started for some long lead items like the mirror segments. In 2003, the STScI developed the Mission Operations Concept and was the primary author on the individual science instruments operations concept documents.
In early 2003, the Project initiated the initial development of system and subsystem requirements within the context of specific (yet flexible) design. To aid this process, the recently formed Science Working Group (SWG) developed the Science Requirements Document (SRD). The SRD is a detailed description of major science goals for the mission and the implied performance requirements for the observatory and instruments. The latter are incorporated in the Mission Requirements Document (MRD). With these documents and those at the element level (observatory, ISIM, launch, and ground system), the Project passed the mission System Requirements Review (SRR) in December 2003.
In 2004, as the JWST costs and NASA budgets became more mature and initial contracts reviewed, all teams had to adjust their plans and budgets for a new Launch date of August 2011. During the more detailed design and analysis phase in 2004, several technical issues emerged. An example was the potential need to manage the JWST science program in order to minimize the accumulation of angular momentum due to radiation torques on the large sunshield. This issue would finally be resolved in 2007.
In 2004, Brush-Wellman delivered the first Be segment blanks to Axsys for machining and lightweighting. Also in 2004, as a result of the Columbia accident, NASA grounded the shuttle system and halted plans for a routine shuttle mission to repair the Hubble. (SM4, this decision was reversed in 2006 by NASA Administrator Mike Griffin).