Very cool AWST article on CEV, Delta IV-H, Atlas V

[danwoodard]

The major question to me is, why is the Astronaut Office so opposed to use of the EELVs? Predictions of reliability derived from analysis ("Man-Rating") can be very precise (99.999%), and yet be wildely inaccurate, as in the case of the Shuttle, since it is impossible to predict all possible modes of failure. The EELVs are flying, eliminating a lot of uncertainty. The SRB-derived CLV would need a new control system; the SRBs have no internal guidance system and no roll vectoring capability, and there is talk of eliminating the HPUs and using pneumatic nozzel actuators. The second stage would be entirely new except for an engine that would be a possible derivative of the J-2, which has been out of production for decades. Development cost would be (at a guess) $3-5 billion. And then we would still have catastrophic failure modes that might or might not allow time for an escape. <br /><br />I've read the Planetary Society report in which Dr. Giffin first advocated the SRB-derived CEV launcher, but even there he did not give any technical reason why the EELVs would not be just as effective.<br /><br />With regard to which EELV is better, while the D-IV heavy hydrogen venting may be dramatic, it appears safe and is certainly simple. The Shuttle, although it doesn't routinely vent as much hydrogen, has had significant hydrogen fires in the flame trench during on-the-pad shutdowns. <br /><br />More important, after the Challanger many said we should never, ever use solids on a manned vehicle again. They have catastrophic failure modes that may or may not allow time to initiate an abort. They cannot be shut down (early in Shuttle their was talk of blowing openings in the case to neutralize thrust, but this introduced other failure modes) The D-IV can experience a failure of ANY of its engines, shut down the affected engine, and remain under control. Moreover, as anyone knows who has watched it done, assembling an SRB is hazardous and expensive. <br /><br />Finally,