While I have not done or seen the trade studies, there is another factor that may mitigate toward the use of kerosene (RP-1) instead of LH2. That is the bulk-density specific impulse. That combines the Isp x the density of the propellant. RP-1 (or similar hydrocarbons) are much denser than LH2. The result is that you can use a much smaller tank for RP-1 than for LH2. Whether the difference makes up for the lower Isp IN THIS APPLICATION, I don't know. But if you look at the Saturn I and IV designs, you will note that von Braun used LO2/RP-1 for the first stages and LO2/LH2 for the upper stages.<br /><br />By eliminating the reusable aspects of the SSME's you also eliminate the requirements for a recovery subsystem, reducing weight penalties. Can you recover components cheaply enough to make it worthwhile refurbishing and reusing them? YES! But the items recovered must weigh enough and be of high enough value to make recovery worthwhile. For example, the SRB's ARE recovered using parachutes (SRB-DSS...SRB Decelerator Subsystem). The intial cost of development of the SRB-DSS was about $24M in 1978 dollars. We paid for the development costs after the first three or four successful recoveries! (Been awhile, and I can't find the figures in my personal files.) (We lost one complete set of boosters due to inadvertant separation of the main deck fittings at the sep ring firing, due to the ringing frequency coinciding with the accelerometer settings...which Martin Marietta warned NASA-MSFC about when they decided to detach the chutes on water impact. Change to the sensing system eliminated this problem on subsequent flights.)<br /><br />The value of reusable SSME's and the complexity, versus the amount saved by recovering and reusing them, probably doesn't compute for the HLLV.<br /><br />Might not be what some want to hear, but it may, indeed be cheaper to throw stuff away than recover and refurb.<br /><br />Ad Luna! Ad Aries! Ad Astra!<br />Trailrider