3 ms·
Let's see. The sabre engine has a closed cycle specific impulse of 450, which gives an effective exhaust velocity of 4.4 km/s. LEO requires about 7.8 km/s veloc
by vlasev 12y ago
Let's see. The sabre engine has a closed cycle specific impulse of 450, which gives an effective exhaust velocity of 4.4 km/s. LEO requires about 7.8 km/s velocity while mach 5 is 1.7 km/s. In one case you accelerate from 0 to 7.8 km/s an in the other from 1.7 to 7.8 (for 6.1 km/s delta-v). The first figure gives a mass ratio of 5.9 while the second one gives you ratio of 4.0. You save about 1/3 of the initial mass if you only have to burn on a closed cycle from Mach 5 upwards.
In reality gravity is a tough mistress so actual rockets launching from a pad require about 9.0 km/s delta-v because they lose a substantial amount due to the gravity losses of climbing up from Earth's gravity well. This is especially true for the first few km's of altitude where the thrust-to-mass ratio is quite low. Potentially, the skylon will eliminate most of those gravity losses by effectively launching from 26km altutude at mach 5 (i'm ignoring the relatively modest fuel consumption for reaching this state - the specific impulse is about 3500).
Assuming a delta-v of 9.0 km/s needed for a single stage rocket gives a mass ratio of 7.7. This means that the skylon should save about 48% of the initial mass required by a SSTO rocket launching from sea level.
- sitkack 12y agoIsn't that mass savings mostly in the 330k$ of fuel for the first stage? The economic argument is in the reusable hardware.