What should be the pressure inside Starship CH4, LOx header tanks during the entire Earth-Mars trip and landing on Mars ?

The main CH4, LOx tanks should be under 6 bar pressure, because that is the pressure the Raptor turbines must be feed with propellant. Even after the main tanks run out off liquid propellant they should keep this inside pressure, because gaseous CH4 and LOx will take the propellant's place (I don't know if Starship main tanks will need to be repressurized during 4 months Mars trip and how often).

But the header tanks will be filled with liquid propellant for the entire trip to Mars up until the landing and during Mars entry & landing, burn Raptors will use only less than ½ of full engine thrust. So could both header tanks be pressurized only for " for example 2 or 3 bar pressure ". I know that CH4 header tank isn't fully separated from CH4 main tank, but let say it will be fully separated like LOx header tank (from LOx main tank).

Or, is thrust that the Raptors will use during Mars entry and landing burn unrelated to the pressure with which Raptor turbines must be feed with liquid propellant and pressure inside all main and header tanks must be equally 6 bar. If yes, than why ?


1 Answer 1


It is my understanding that 6 bar is the design pressure (max pressure) of the main tanks. I've seen a higher design pressure of 8.5 bar reported for human rated missions. Pressure at the engines will inevitably vary.

It is known that Starship uses "balloon tanks" that require some minimum pressure inside them to prevent buckling under structural loads. We'll call this X because as far as I know Spacex haven't published anything. So pressure must be greater than X bar and less than 6 bar at all times

The height of the O2 tank on the starship is about 16m and the density of oxygen is about 1140 kg/m3. Gravitational acceleration is 9.8m/s2 so the pressure within the O2 tank varies by 16 x 1140 x 9.8 = 178,752 Pa = 1.79 bar from bottom to top due to hydrostatic pressure while Starship stands on the pad. The height of the O2 tank in the booster is about double this, so there's about 3.6 bar variation from bottom to top.

When the stack takes off, acceleration is about 0.5g, so the hydrostatic pressure increases by 50%. That is, the hydrostatic pressure at the booster engines jumps by 1.8 bar, from 3.6 bar to 5.4 bar. This does not include the ullage pressure at the top of the tank, which must be added on. Assuming the ullage pressure is 0.6 bar, we get 0.6+3.6=4.2 bar at the booster engines at ignition, and 0.6+5.4=6 bar at launch. Having done this calculation, I immediately wonder if the bottom of the booster has thicker material to handle pressures over 6 bar.

From this we can conclude that the raptors can tolerate a range of inlet pressures. The minimum pressure will be determined by the need to avoid cavitation at the pumps. Typically about 0.1 bar above the boiling point pressure of the liquid being pumped is required for rotodynamic pumps, corresponding to a liquid depth of about 1m, but it can be lower. More info can be found by googling NPSH (net positive suction head). The max pressure is known only to SpaceX but 10-20 bar is easily achievable without any real cost or weight impact.

Performance of the engines at ½ thrust has nothing to do with pump inlet pressure, and everthing to do with chamber pressure. The pumps are designed to deliver 230-300 bar to the chamber at max thrust. There are control valves to regulate the flow/pressure at turndown. Provided the pumps aren't damaged by cavitation (by operating at too low a pressure) or overpressure (extremely unlikely) they will deliver to the combustion chamber whatever pressure is required.

Somebody may know the design pressure of the header tanks (if different from the main tanks) but doubt anyone (not even Spacex) knows what pressure they will run the header tanks at for Mars approach yet. They will be figuring out the pressure profile of the header tanks for earth landing at the moment. On starship pressure is maintained by autogenous pressurization (injecting boiled propellant gas back into the tanks to fill the volume left by the used propellant) and there are a number variables to consider to find the optimum.

  • $\begingroup$ 1/2 Thanks, it was said to me before that Super heavy/Starship main tanks operation pressure will be 6 bar, because that is pressure with which Raptors inlets must be feed (are designed to be feed) with liquid propellant. So just for summarization. Raptors inlets can tolerate pressure differences between about 1.1 bar (0.1 above boiling point pressure) and 10 to 20 bar. Pressure inside booster LOX tank during launch could be about 6 bar at the bottom of the LOX tank (was it confirmed that LOX tank will be below CH4 tank) and 0.6 at the top of booster LOX tank. $\endgroup$
    – David Cage
    Oct 8, 2021 at 19:00
  • $\begingroup$ 2/2 What about pressure inside booster CH4 tank and Starship main tanks. Will all have similar ullage (gaseous LOX,CH4) pressure of 0.6 bar at the top of tanks. With about 3.6 bar difference between bottom and top of booster CH4 tank and 1.8 bar difference between bottom and top of Starship CH4 tank. $\endgroup$
    – David Cage
    Oct 8, 2021 at 19:02
  • 1
    $\begingroup$ @DavidCage O2 tank is below CH4 tank, as can be seen in many places. O2 actually enters the engine through the centre of the mounting flange. 0.6 bar is a guess at the ullage pressure in the booster, working backwards from a tank design pressure of 6 bar and my calculated hydrostatic pressure of 5.4 bar at launch. Only Spacex know the min ullage pressure X and it may vary through the flight profile. As fuel is used up, acceleration increases and the starship imposes more load on the booster structure requiring higher ullage pressure at the same time as hydrostatic pressure is decreasing. $\endgroup$ Oct 8, 2021 at 21:32
  • 1
    $\begingroup$ No reason for ullage pressure to be same in all tanks so it can be safely assumed to be different The only thing reported is 6 bar design pressure, which will be consistent if same thickness steel is used throughout. Header tanks would tolerate higher pressure than the tanks around them (but not lower as they'd likely collapse under external pressure.) CH4 supply is from main tank through header tank & downcomer. It's not clear what the valve arrangement is to keep the CH4 from returning to the main tank, but if it's a simple non-return valve, CH4 header tank pressure would match the main tank $\endgroup$ Oct 8, 2021 at 21:48
  • 1
    $\begingroup$ I've never seen a rotodynamic pump that needs 6 bar Net Positive Suction Head. I'm used to pumps that run at 3000-3600rpm (based on 50-60Hz electric.) Raptor turbopumps spin way faster than that to save weight & get high pressure in small size. Even so, if 1st stage impeller of the pump is designed to be gentle with the fluid, you can keep required NPSH low. NPSH is the additional pressure above the boiling point pressure of the liquid (at pumping temperature) typically around 0.1 bar but always expressed as liquid depth i.e 1m. Credit can be taken for subcooling the liquid in NPSH calculation $\endgroup$ Oct 8, 2021 at 22:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.