Engineering question, Re: aerospike engines.

[sojourner]

I have a quick question. Hypothetically how hard would it be to convert a standard rocket engine, say a Merlin Vacuum engine, to use an aerospike instead of the large nobium nozzle currently in use?  If it could be done, would the resulting engine be use able from sea level to orbit?  If so, this would seem to help make F9 second stage return much easier.

[Streetwind]

Regardless of the difficulty of conversion, I doubt it would improve reusability prospects. An aerospike engine has less specific impulse than an equivalent engine with a bell optimized for its intended pressure regime. In other words, an aerospike at sea level is worse than a dedicated sea level nozzle, and an aerospike in space is worse than a dedicated vacuum nozzle.

This means that the stage dV drops, which in turn means it has to use more fuel to push the same payload to orbit, and that in turn means it has less fuel reserves for a potential return and landing. This effect is magnified by the fact that the Falcon 9 as a whole is quite biased towards the second stage in terms of dV distribution, so that the first stage doesn't go as fast and can be recovered sooner. You might as well keep the more efficient vacuum engine and make use of the larger reserves of fuel that it gives you to make recovery easier.

I mean, I get why you suggest this - using the MVac to land propulsively is probably not going to work very well at all, due to flow separation at sea level in the ridiculously overexpanded nozzle. An aerospike would get around this. But I'm not convinced that the disadvantages in other regimes make it worth going for it. It would probably be easier for SpaceX to eject the nozzle prior to landing, and build a new one for the next flight. You wouldn't have 100% reusability, but probably still greater than 95%, and that can easily be worth it.

Edited April 13, 2017 by Streetwind

[p1t1o]

Im not an actual rocket engineer, but it sounds like a completely new engine to me. Almost everything important to the design of a rocket engine would change, necessitating changes elsewhere. The nozzle design affects chamber pressure and temperature. Chamber pressure/temperature affect injector design and turbopump properties. Fuel injection factors affect other upstream plumbing. Etc.

There are not actually that many things that are easier to "convert" to something else than to build new from scratch. Many times it would be more difficult and expensive to do this.

[sevenperforce]

Conversion would be virtually impossible in this particular case.

Aerospike/plug nozzles are designed in one of two ways. Either they have a toroidal combustion chamber which opens in a ring around the nozzle, or they have many small combustion chambers arranged around the perimeter. For linear aerospike engines, it's always the latter.

Toroidal combustion chamber:

Ring of small combustion chambers:

Linear aerospike, many small chambers:

The Merlin engine has a very strong, lightweight combustion chamber enabling high combustion efficiency and high chamber pressure, leading to good sea-level specific impulse. The combustion chamber would have to be completely redesigned in order to be converted into a toroidal one, as combustion and flow in a toroidal chamber is completely different.

Now, it might be possible, in theory, to use the Merlin engines themselves as the smaller thrust chambers arranged around a single aerospike nozzle. But that would be....just ridiculously massive.

[tater]

 

Bottom right... imagine those are Merlins

Or BE-4s.

 

 

Edited April 13, 2017 by tater

[sojourner]

Well, those are all good points on why it wouldn't work. Back to the drawing board....

 

Thanks for the informative posts!

[sevenperforce]

Hey, no one is saying it wouldn't work, per se.

Eight Merlins arranged around a central plug nozzle with an air-augmentation shroud to kick up the thrust would SSTO easily enough.

[tater]

I wanna see some crazy billionaire build Phil Bono stuff  .