I just started learning some basics of rocketry and am struggling to understand the outcomes of a simulator I’m programming. All other things being equal, when I attempt to simulate the altitude over time of a single stage rocket with the below attributes, its peak altitude is only a tiny bit less (~97.5%) than if that rocket were segmented into two stages (also below). Note, both instances have the same total dry mass and fuel mass.

enter image description here

Multi stage :

  • First stage : ISP = 282s, dry mass = 22226kg, diameter = 3.66m, constant thrust = 7600000N, fuel mass = 522629kg, thrust angle = 5 degrees
  • Second stage : ISP = 348s, dry mass = 4000kg, diameter = 1.7m, constant thrust = 934000N, fuel mass = 58069kg, thrust angle = 5 degrees

Single stage:

  • First stage : ISP = 282s, dry mass = 26226kg, diameter = 3.66m, constant thrust = 7600000N, fuel mass = 580698kg, thrust angle = 5 degrees

Is this a reasonable outcome or is my code possibly bugged? I was under the impression multi staging led to significantly greater delta V, and thus max altitude (in this simulation's case) gains. Also, if the entirety of the second stage’s mass were treated as just additional dry mass for the single stage instance, the peak altitude it achieves is nearly identical to the peak altitude of the first stage in the two stage instance. This seems intuitive. But the tiny total performance gain of multi staging over single staging doesn’t for me. If the simulation is wrong, what sort of gains should be expected in a case like this?

  • 4
    $\begingroup$ The Rocket Equation is the one that sums up where the gains for staging come from. With the information you've provided it's hard to diagnose the actual problem with your simulation. I'd start with associating units with your numbers $\endgroup$
    – Erin Anne
    Commented Sep 28, 2022 at 5:40
  • 6
    $\begingroup$ Looking at the graph you seem to have separation very early, the second stage is starting pretty much at ground level and with only modest speed boost which would tend to get much the same peak altitude $\endgroup$ Commented Sep 28, 2022 at 8:15

2 Answers 2


I will argue that your simulation outcome is not reasonable.

Consider what happens if we decouple your multi stage rocket right at launch, making no use of the first stage at all. That is, just using the second stage as a single stage rocket.

Its delta-v:

$$\Delta v = I_{sp} \cdot g \cdot \ln{\left(\frac{M_1}{M_0}\right)} = 348s \cdot 9.81m/s^2 \cdot \ln{\left(\frac{58069kg + 4000kg}{4000kg}\right)} = 9360m/s$$

With a thrust to weight ratio at launch of 1.54

The single stage rocket:

$$\Delta v = 282s \cdot 9.81m/s^2 \cdot \ln{\left(\frac{580698kg + 26226kg}{26226kg}\right)} = 8690m/s$$

With a thrust to weight ratio at launch of 1.28

Thus, the second stage alone should be a more capable rocket than your single stage rocket, before even considering the first stage. There must therefore be something up with your simulation.

  • 3
    $\begingroup$ Thanks for the insight. It should be obviously wrong because of the reason you gave which I guess is why my intuition was telling me something was off, just couldn't tell why. Turns out I wasn't properly accounting for the dropped dry mass of prior stages when computing later stages' dry mass (the whole point of staging... weird). Any ways, thanks for the help / confirming / giving me an intuitive test to apply in the future. $\endgroup$ Commented Sep 29, 2022 at 3:49

I agree with the answer of SE that your results appear to be incorrect.

I suggest that you break your simulation down into discrete testable steps as follows:

  1. Implement thrust only without any angle

    • Consider a single stage rocket

    • Verify that the velocity of the rocket at burnout in your simulation is equal to that calculated by the Tsiolkovsky (rocket) equation

    • Now consider a multi-stage rocket and verify with the multi-stage Tsiolkovsky equation

  2. Implement gravity only

    • Consider a projectile launched with a certain initial velocity

    • Verify with projectile equations (HyperPhysics is a good source)

    • You can move from 1D motion to 2 or 3 dimensions now and verify

  3. Implement thrust with gravity without any angle

    • Consider a single stage rocket

    • Verify that the velocity of the rocket at burnout in your simulation is equal to that calculated by the Tsiolkovsky equation minus $t\,g$ where $t$ is the time since launch and $g$ is $9.81\,\mathrm{m/s^2}$.

  4. Further additions such as thrust angle and aerodynamic forces will make it harder and harder to verify with analytical equations. Therefore, I suggest you implement these additions one-by-one and compare with the previous results to check whether the change is reasonable or not.

  • 1
    $\begingroup$ Ya, thats what I ended up doing. All those additional components like gravity, angle, drag, etc... worked for single staging but didn't seem to for multi-staging. Undoing those components helped me narrow down to the fact the thrust/weight ratio of the second stage (the mass) was much worse than it should have been. And thanks for the hyperphysics link. Those calculators and accompanying visuals are a nice resource. $\endgroup$ Commented Sep 29, 2022 at 3:55

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.