The external tank for the shuttle contained tanks of liquid Oxygen and Hydrogen for the main engines to use. These two have similar (though not equal) volumes, but the Oxygen is heavier in an obvious stochastic sense.

Given that, I just assumed that the heavier fuel (Oxygen) tank would be on the bottom and the lighter one would be on top, to make the center of mass lower. It seems like a lower center of mass would make it less likely to tip over during the climb. But that's not how NASA did it. The Oxygen tank was on top. Why?

  • 1
    $\begingroup$ Pressure, maybe? Anyway, the Shuttle had enough of a guidance system to be able to quickly account for any top-heaviness during ascent anyway. $\endgroup$
    – user12
    Jan 14, 2014 at 1:21
  • 1
    $\begingroup$ " to make the center of mass lower". Why would you want to lower the center of mass, when rocket stability greatly improves when the CoM is as far forward from the center of pressure as possible? $\endgroup$ Sep 27, 2021 at 21:30

2 Answers 2


While in the atmosphere, a forward center of gravity contributes to stability. You want the center of gravity to be forward of the center of pressure for aerodynamic stability. A longer distance between the center of gravity and the gimbaling engines gives them more control authority as well.

  • $\begingroup$ Why did the Saturn S-II and S-IV(B) do it the opposite way (LH2 forward of LOX)? $\endgroup$
    – John Auld
    Sep 14, 2022 at 21:22

An additional reason that the forward position is advantageous for the LOX tank is the pressure head created by the distance from the tank to the engines. The turbopumps that drive the engines need sufficient inlet pressure to prevent cavitation. Moving the LOX tank forward allows the weight of the LOX to contribute to that pressure, which reduces the necessary ullage pressure within the tank.

  • $\begingroup$ Isn't this an argument for putting the LH2 tank on top? It's less dense and could benefit more from the bigger "h" in rho * g * h. $\endgroup$ May 13, 2019 at 3:49
  • $\begingroup$ Liquid hydrogen is so....not dense, that it would make very little difference in aiding the turbopump inlet suction pressure $\endgroup$
    – Tristan
    May 13, 2019 at 13:44
  • $\begingroup$ Yeah, it's LH2 that always had NPSP problems on shuttle, so it doesn't make sense to me to say you need to put the LO2 on top to fix its (non existent) NPSP problems. $\endgroup$ May 13, 2019 at 13:53
  • 1
    $\begingroup$ It's not that so much as the fact that the head from the downcomer is enough to supply the required npsp, so the lox tank ullage pressure is only what is necessary for structural requirements $\endgroup$
    – Tristan
    May 13, 2019 at 18:14

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.