4 ms·
The ratio of wet and dry masses for a launch vehicle also depends on ISP, thrust, and aerodynamics. You can have an extremely high ISP and thus a lot of theoret
by manicdee 7y ago
The ratio of wet and dry masses for a launch vehicle also depends on ISP, thrust, and aerodynamics. You can have an extremely high ISP and thus a lot of theoretical delta-v but you waste it all because your thrust ratio is below 1 and the vehicle can’t leave the ground.
Over a certain size you can’t build up anymore and have to make the vehicle flat. This is because a certain surface area of exhaust nozzle can only provide so much thrust. So you get to about 100m and then your rocket can’t leave the ground because the mass of propellant is higher than the available thrust.
Then there’s thrust to weight, meaning that to get off the ground and about two minutes into your launch it is far more important to have a lot of thrust than to have high efficiency. The rule of thumb for KSP players is 1.2 thrust to weight on the ground. You will fight gravity losses and end up with about 0.2g acceleration off the pad. This increases gradually as the mass of the vessel is burnt off.
Later in flight the major force to overcome is atmospheric drag. The faster you go, the worse it gets. Fat rockets will have a harder time. So you can’t go too wide, you can’t go too tall, and there are many trade offs to be made along the way.
But the short version is no, you don’t hold v and delta-v constant. There are many more variables to consider when launching a rocket from a standing start where gravity and an atmosphere are involved.
Rockets do not scale linearly.
- jessaustin 7y agoYes of course rocketry is complicated. ITT we're discussing the specific question that parent raised: are a bunch of small-payload rockets more efficient than on big-payload rocket? We assume that actual engineers have done the work of designing rockets that actually reach orbit, whether small or large; that's why we assume equal velocities. Any particular orbit is characterized by a specific orbital energy. Tsiolkovsky certainly considers the effects of gravity; that's the point.