In the landing procedure for the first stage of the Falcon 9 one concern is that :

Merlin 1D throttling at its lowest level still generates more thrust than the empty mass of the first stage. So what are the challenges for throttling lower. For example RD-191 has a throttle range 27%-105% according to Wikipedia. But I am interested only for Merlin 1D engine if it is possible to have this capabilities and what is the work that should be done for this.

If SpaceX have problems for throttling low could be as an option that only the engine at the center of the rocket to be a different Merlin engine (different Merlin 1D version or different new engine at all ) that is produced with the primary purpose to be very low throttle (no matter what is the thrust) ? Even if by achieving the necessary throttling could affect the thrust or Isp of the engine (by not having the good performance of the Merlin 1D FT), the overall performance of Falcon 9 still will be ok. They win in this way a higher possibility to landing successfully the first stage.

Production of a new rocket engine of course has a cost and maybe in economies of scale for the production of Merlin 1D it is not economical because you should produce a different type of engine (in fact this would not influence the cost of the rocket), but it would be very good for SpaceX economy if it helps to lands successfully (slowly and safely). According to SpaceX, Falcon 9 FT now could have a higher total thrust at sea level by 7607 kN (1.71 million pound thrust) compared to previous 6806 kN (1.53 million pound thrust). So even if one of the engines will have a lower thrust or even shut down it, Falcon 9 FT again would have the same performance that was at its previous missions at least around 6800 kN or even more. Probably for the center engine some options could be the previous versions of Merlin engine. So they don’t need to modify one of the Merlin 1D or developing a new version (naming Merlin 1E) for low throttling since they could have difficulties for this, they could use the previous Merlin 1A, 1B, 1C. They have lower thrust Merlin 1A (340 kN) , Merlin 1B (380 kN) , Merlin 1C (350 kN at the beginning or 560 kN the upgraded version) and including even their throttling capability would make possible to thrust in lower levels, in this way making a perfect speed deceleration for the first stage. Merlin 1C is a better option than A and B for reliability, because uses a regeneratively cooled nozzle and combustion chamber. Of course this engines could need some modification to fit properly with Falcon 9 FT but their thrust which is appropriate will be helpful.

So to achieve low throttling of Merlin 1D what could be done, and if this faces difficulties could it be as an option to use a new Merlin engine that would have a low (maybe very low) thrust with/or very low throttling capabilities at the value that is necessary for comfortable landings (no matter what thrust will have), or maybe using the previous Merlin engines with modifications if are necessary to do ?

  • $\begingroup$ Related: space.stackexchange.com/questions/12823/… $\endgroup$
    – called2voyage
    Commented May 5, 2016 at 18:54
  • $\begingroup$ Rocket engines have different designs so even the specific changes that need to do are different. RD-191 can be low throttling and Merlin not because they are different in design and even problems they face are different. Most important part the point it is not only what makes difficult throttling, but the point and the question it is also : what if it is difficult for the Merlin 1D then could be optional and possible a different engine such as previous low thrust merlins or even a new one with low thrust and low throttling a combination that SpaceX hasn't decide yet as option. $\endgroup$
    – Mark777
    Commented May 5, 2016 at 19:26
  • 1
    $\begingroup$ Possible, yes, difficult, no, required or desirable - not really. They could go with Merlin-1D-Vac, same throttlability but lower thrust, so the end thrust would be fairly low - despite being designed for vacuum, the engine would work - but it would be weak. But that's all at cost of launch time thrust and landing time fuel. With the landing procedure perfected that simply won't be needed. $\endgroup$
    – SF.
    Commented May 7, 2016 at 21:51

1 Answer 1


The question presumes that lower landing thrust for Falcon 9 is desirable, which I believe is false.

The landing thrust is a compromise between opposing requirements. With lower thrust, the rocket could make a slower approach, giving more time to correct its trajectory and land closer to the target. But a slower approach means spending more time and thus more propellant fighting gravity. Higher thrust allows a very fast approach, and very short terminal burn, saving fuel.

It appears that the landing attempt on the SES-9 Mission (Flight 22), having very little reserve fuel after the primary mission, attempted to do a 3-engine terminal burn. Video of that landing attempt seemed to show the rocket approaching the edge of the ASDS rather than the center, possibly indicating that the fast approach didn't allow enough time to correct the trajectory. So 3 engines might be too much thrust, although flight 24 managed a similar 3-engine ASDS landing successfully.

Falcon 9's current single-engine throttle limits subsequent to the CRS-6 mission (Flight 17) have proven adequate to land on target; flights 20, 21, and 23 all hit the mark. A lower throttle would spend more fuel and solve no problem.

  • 5
    $\begingroup$ This is correct. Every second spent hovering is 9.8 m/s of delta-v wasted, and the fuel to achieve that delta-v has to be carried throughout the entire launch. Additionally, the further they descend unpowered, the more they shed through aerodynamic drag instead of with rocket thrust. On their last few landings they have done three-engine burns, the most recent one apparently landing hard because thrust on one engine was a bit low. $\endgroup$ Commented Jun 16, 2016 at 14:53

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.