Multi Stage Raptor Engine

[matthewgreen2000]

Hi all I was just working on a concept of how to make a sea level raptor engine that can convert into a vacuum engine I know there are other changes that need to be made to make this work but It could reduce the amount of engines need for the starship take a look at my concept looking forward to you feedback good or bad 
 

[Silmfeanor]

A working example of this can be found here, an RL-10 variant:
https://twitter.com/torybruno/status/1114979080992985088?lang=en

[Lars-J]

It doesn’t quite work, unfortunately, the optimal geometry for a sea level engine does not match a truncated vacuum nozzle. Nor would it allow raptors to be as tightly packed as desired.

Some RL-10 models have an extending nozzle but it is not designed to be fired when collapsed.

[AC in NC]

Some RL-10 models have an extending nozzle but it is not designed to be fired when collapsed.

What's the purpose of those extending nozzles?  From that Twitter video clip, it appears to be for the purpose of packing the upper stage more compactly vs. having to stretch the "interstage" a accommodate a longer vac bell.

[Lars-J]

Some RL-10 models have an extending nozzle but it is not designed to be fired when collapsed.

What's the purpose of those extending nozzles?  From that Twitter video clip, it appears to be for the purpose of packing the upper stage more compactly vs. having to stretch the "interstage" a accommodate a longer vac bell.

Indeed. And the only recent development to make that shorter interstage trade (I think) was Boeing with the Delta IV. Lockheed Martin designed the Atlas V to use RL-10s with a fixed nozzle. SpaceX also opted for longer interstages instead of a complicated nozzle extensions.

[Robotbeat]

Some RL-10 models have an extending nozzle but it is not designed to be fired when collapsed.

What's the purpose of those extending nozzles?  From that Twitter video clip, it appears to be for the purpose of packing the upper stage more compactly vs. having to stretch the "interstage" a accommodate a longer vac bell.

Indeed. And the only recent development to make that shorter interstage trade (I think) was Boeing with the Delta IV. Lockheed Martin designed the Atlas V to use RL-10s with a fixed nozzle. SpaceX also opted for longer interstages instead of a complicated nozzle extensions.

Long interstages do mean more weight for the first stage, plus a possibly higher risk of recontact. The nozzle pusher separation system SpaceX had to develop might not be required if SpaceX didn't have such a long interstage.

Also, SpaceX showed an extending nozzle on the reusable falcon 9 upper stage concept video.

It also retracts. One advantage of such a system (not really shown here) is ability to use the same engine for sea level and optimized fully for vacuum. And (shown here) the landing legs don't need to be as long.

[Robotbeat]

I actually like extendable nozzle concepts.

Now someone make one that is much large in diameter than the stage it deploys from.

[Lars-J]

Expandable/extendable nozzles don't make much sense for Starship for the following reasons:
1) The VacRaptor nozzle is liquid cooled all the way to the edge of the nozzle, probably a necessity for having a vacuum nozzle clustered together in a skirt (no way to radiate heat to the sides), and this kind of cooling is impractical with expandable nozzles.
2) Making a reliable extension & collapsing mechanism has never been done, and would be very difficult. (best part is no part)
3) A collapsed VacRaptor would take up less space vertically but the same (or more) space horizontally, so it does not help you for packing engines. (you could not replace the center 3 engines with 3 VacRaptors, there is no room)
4) And most importantly... The current VacRaptor can ALREADY be fired at sea level, if necessary. (see video)

[Asteroza]

I actually like extendable nozzle concepts.

Now someone make one that is much large in diameter than the stage it deploys from.

In theory airmat nozzles inflated by turbine exhaust as proposed for J-2 could do that...

Or you can go the Sea Dragon approach and wrapping the nozzle around the first stage and inflating it like those aluminum stovetop popcorn bags...

[davamanra]

Here's a question about the vacuum raptor engine:  Are there any major problems with using a vacuum raptor engine at sea level?  The only possible issues that I can think of are combustion instability and reduced thrust.

The reason I ask this question is that it seems to me that if combustion instability can be addressed, then the three sea level engines could basically be made redundant. 

[ZachF]

Here's a question about the vacuum raptor engine:  Are there any major problems with using a vacuum raptor engine at sea level?  The only possible issues that I can think of are combustion instability and reduced thrust.

The reason I ask this question is that it seems to me that if combustion instability can be addressed, then the three sea level engines could basically be made redundant.

The sea level engines are for landing and they also perform gimbal function.

The RVac is probably pretty close to flow sep, I imagine ISP at sea level is probably only like 290. Might be less of a case with Raptor 2 if it runs at a little higher pressure and they don’t alter ER.

[davamanra]

Here's a question about the vacuum raptor engine:  Are there any major problems with using a vacuum raptor engine at sea level?  The only possible issues that I can think of are combustion instability and reduced thrust.

The reason I ask this question is that it seems to me that if combustion instability can be addressed, then the three sea level engines could basically be made redundant.

The sea level engines are for landing and they also perform gimbal function.

The RVac is probably pretty close to flow sep, I imagine ISP at sea level is probably only like 290. Might be less of a case with Raptor 2 if it runs at a little higher pressure and they don’t alter ER.

This is why I was looking at this.  If you were to first, control the flow separation (obviously the key problem) and then eliminate the sea level engines and then mount gimballed vacuum engines.  Even if the ISP is greatly reduced, you could use all three of these engines to provide enough thrust for landing a Starship with nearly empty propellent tanks.  Also, by eliminating three engines you would reduce the overall mass and complexity of a Starship which would increase the amount of payload it could take into into space as well as reduce its landing mass.

[alugobi]

Geez, I wonder if SpaceX has thought of any of this.

[eriblo]

Here's a question about the vacuum raptor engine:  Are there any major problems with using a vacuum raptor engine at sea level?  The only possible issues that I can think of are combustion instability and reduced thrust.

The reason I ask this question is that it seems to me that if combustion instability can be addressed, then the three sea level engines could basically be made redundant.

The sea level engines are for landing and they also perform gimbal function.

The RVac is probably pretty close to flow sep, I imagine ISP at sea level is probably only like 290. Might be less of a case with Raptor 2 if it runs at a little higher pressure and they don’t alter ER.

This is why I was looking at this.  If you were to first, control the flow separation (obviously the key problem) and then eliminate the sea level engines and then mount gimballed vacuum engines.  Even if the ISP is greatly reduced, you could use all three of these engines to provide enough thrust for landing a Starship with nearly empty propellent tanks.  Also, by eliminating three engines you would reduce the overall mass and complexity of a Starship which would increase the amount of payload it could take into into space as well as reduce its landing mass.

Issues:

Even if flow separation is avoided or controlled this is most likely only the case for high throttle settings so you lose the ability to throttle down during landing.

Gimbaling of a larger/heavier RVac is much slower unless you beef up the actuators (and possibly reinforce the nozzle as well).

You can only fit 1 RVac in place of the three SL engines, so you lose some of the thrust you just added with the 9 Raptor version and you lose landing engine redundancy. If the 9 Raptor version had gained a limited low altitude abort capability you have now lost it again.

The longer center RVac means that you have to extend the skirt and move the outer ones down to avoid plume impingement, adding quite a bit of mass.

Summary: You now have a lower thrust and most likely heavier Starship that can only do suicide burn landings with reduced or no redundancy.

[IainMcClatchie]

Many engines inject gas in the expanding part of the bell.  A Vacuum Raptor could presumably inject some methane into the bell during a landing burn to avoid flow separation while throttled.  It seems it is not needed during liftoff.

How do you calculate the change in exit plane pressure from the addition of gas into the expanding bell?  Does the main exhaust donate significant momentum to the gas as it leaves the bell, or is it nearly a simple case of parallel expansion?

(And if it's parallel expansion, then, in the cases of the F-1 or the Merlin, is the turbine exhaust supersonic when it enters the bell?  And if subsonic, does it expand subsonically, with cross-section going DOWN as pressure goes down and velocity goes up?)

[davamanra]

Here's a question about the vacuum raptor engine:  Are there any major problems with using a vacuum raptor engine at sea level?  The only possible issues that I can think of are combustion instability and reduced thrust.

The reason I ask this question is that it seems to me that if combustion instability can be addressed, then the three sea level engines could basically be made redundant.

The sea level engines are for landing and they also perform gimbal function.

The RVac is probably pretty close to flow sep, I imagine ISP at sea level is probably only like 290. Might be less of a case with Raptor 2 if it runs at a little higher pressure and they don’t alter ER.

This is why I was looking at this.  If you were to first, control the flow separation (obviously the key problem) and then eliminate the sea level engines and then mount gimballed vacuum engines.  Even if the ISP is greatly reduced, you could use all three of these engines to provide enough thrust for landing a Starship with nearly empty propellent tanks.  Also, by eliminating three engines you would reduce the overall mass and complexity of a Starship which would increase the amount of payload it could take into into space as well as reduce its landing mass.

Issues:

Even if flow separation is avoided or controlled this is most likely only the case for high throttle settings so you lose the ability to throttle down during landing.

Gimbaling of a larger/heavier RVac is much slower unless you beef up the actuators (and possibly reinforce the nozzle as well).

You can only fit 1 RVac in place of the three SL engines, so you lose some of the thrust you just added with the 9 Raptor version and you lose landing engine redundancy. If the 9 Raptor version had gained a limited low altitude abort capability you have now lost it again.

The longer center RVac means that you have to extend the skirt and move the outer ones down to avoid plume impingement, adding quite a bit of mass.

Summary: You now have a lower thrust and most likely heavier Starship that can only do suicide burn landings with reduced or no redundancy.

Issue #1:  This is a real possibility and would need to be addressed.

#2:  Although gimballing might be slower, this can be compensated by faster response to attitude inputs, so the nozzle might be moving slower but it will respond earlier.  In this way you wouldn't have to beef up the actuators or reinforce the nozzle.

#3:  I don't know what you are referring to with a 9 Raptor version.  The present design has only 3 SL Raptors and 3 RVAC's.  With the removal of the SL Raptors, there would be plenty of room for the 3 RVAC's.

#4:  There wouldn't be a center RVAC, in this design.  There would only be three RVAC's.  They would be spread out on a new thrust puck where there would be plenty of room, and they would be on the same plane so there would be no plume impingement.  Also less mass because three less engines 

I think you might be thinking of a Falcon 9 with it's 9 Merlin engines and not the Starship with it's 3 SL Raptors and it's 3 RVAC's.

#5:  As it is right now, Starship only needs 2 of it's three SL engines to land.  If you were to use 3 RVAC engines, even with less thrust, the overall thrust would still be more than 2 SL engines, giving you enough thrust and gimballing ability to land. 

[davamanra]

Many engines inject gas in the expanding part of the bell.  A Vacuum Raptor could presumably inject some methane into the bell during a landing burn to avoid flow separation while throttled.  It seems it is not needed during liftoff.

How do you calculate the change in exit plane pressure from the addition of gas into the expanding bell?  Does the main exhaust donate significant momentum to the gas as it leaves the bell, or is it nearly a simple case of parallel expansion?

(And if it's parallel expansion, then, in the cases of the F-1 or the Merlin, is the turbine exhaust supersonic when it enters the bell?  And if subsonic, does it expand subsonically, with cross-section going DOWN as pressure goes down and velocity goes up?)

1. I think the Raptor might already have feature just to keep the nozzle cool during operation, but not for the purposes you describe.  Having said that, if they were to modify the system to incorporate this function, then that could eliminate the flow separation issue at part throttle.  I like it!

2.  Those calculations are beyond me, but I do know that the exhaust enters the bell trans-sonically and then goes supersonic as it expands in the bell.  From there, most of the exhaust travels in the "down" direction due to atmospheric pressure at SL.  Having said that, your idea for injecting fuel into the bell at SL to maintain this "downward" flow and minimize flow separation sounds promising.   

[Tommyboy]

Here's a question about the vacuum raptor engine:  Are there any major problems with using a vacuum raptor engine at sea level?  The only possible issues that I can think of are combustion instability and reduced thrust.

The reason I ask this question is that it seems to me that if combustion instability can be addressed, then the three sea level engines could basically be made redundant.

The sea level engines are for landing and they also perform gimbal function.

The RVac is probably pretty close to flow sep, I imagine ISP at sea level is probably only like 290. Might be less of a case with Raptor 2 if it runs at a little higher pressure and they don’t alter ER.

This is why I was looking at this.  If you were to first, control the flow separation (obviously the key problem) and then eliminate the sea level engines and then mount gimballed vacuum engines.  Even if the ISP is greatly reduced, you could use all three of these engines to provide enough thrust for landing a Starship with nearly empty propellent tanks.  Also, by eliminating three engines you would reduce the overall mass and complexity of a Starship which would increase the amount of payload it could take into into space as well as reduce its landing mass.

Issues:

Even if flow separation is avoided or controlled this is most likely only the case for high throttle settings so you lose the ability to throttle down during landing.

Gimbaling of a larger/heavier RVac is much slower unless you beef up the actuators (and possibly reinforce the nozzle as well).

You can only fit 1 RVac in place of the three SL engines, so you lose some of the thrust you just added with the 9 Raptor version and you lose landing engine redundancy. If the 9 Raptor version had gained a limited low altitude abort capability you have now lost it again.

The longer center RVac means that you have to extend the skirt and move the outer ones down to avoid plume impingement, adding quite a bit of mass.

Summary: You now have a lower thrust and most likely heavier Starship that can only do suicide burn landings with reduced or no redundancy.

Issue #1:  This is a real possibility and would need to be addressed.

#2:  Although gimballing might be slower, this can be compensated by faster response to attitude inputs, so the nozzle might be moving slower but it will respond earlier.  In this way you wouldn't have to beef up the actuators or reinforce the nozzle.

#3:  I don't know what you are referring to with a 9 Raptor version.  The present design has only 3 SL Raptors and 3 RVAC's.  With the removal of the SL Raptors, there would be plenty of room for the 3 RVAC's.

#4:  There wouldn't be a center RVAC, in this design.  There would only be three RVAC's.  They would be spread out on a new thrust puck where there would be plenty of room, and they would be on the same plane so there would be no plume impingement.  Also less mass because three less engines 

I think you might be thinking of a Falcon 9 with it's 9 Merlin engines and not the Starship with it's 3 SL Raptors and it's 3 RVAC's.

#5:  As it is right now, Starship only needs 2 of it's three SL engines to land.  If you were to use 3 RVAC engines, even with less thrust, the overall thrust would still be more than 2 SL engines, giving you enough thrust and gimballing ability to land.

SpaceX is SpaceX'ing again. Starship will get 3 SL raptors, 6 vacuum raptors, and a tank stretch. Source:
https://twitter.com/elonmusk/status/1472059476253548544
https://twitter.com/elonmusk/status/1472060261309759488

[hkultala]

Some RL-10 models have an extending nozzle but it is not designed to be fired when collapsed.

What's the purpose of those extending nozzles?  From that Twitter video clip, it appears to be for the purpose of packing the upper stage more compactly vs. having to stretch the "interstage" a accommodate a longer vac bell.

Indeed. And the only recent development to make that shorter interstage trade (I think) was Boeing with the Delta IV. Lockheed Martin designed the Atlas V to use RL-10s with a fixed nozzle. SpaceX also opted for longer interstages instead of a complicated nozzle extensions.

.. but the original RL10A-4 of Atlas V has considerably smaller expansion ratio (84:1) and worse isp (451s) than RL10B-2 of original Delta IV (280:1, 465.5 s).

Though later CEntaur of Altas V switched to RL10C-1 which has bigger expansion ratio (1:130), but still much less than B-2 had, and much worse isp.

So, compared to Delta IV, Atlas V suffers in performance for not having than extending nozzle.

[eriblo]

Here's a question about the vacuum raptor engine:  Are there any major problems with using a vacuum raptor engine at sea level?  The only possible issues that I can think of are combustion instability and reduced thrust.

The reason I ask this question is that it seems to me that if combustion instability can be addressed, then the three sea level engines could basically be made redundant.

The sea level engines are for landing and they also perform gimbal function.

The RVac is probably pretty close to flow sep, I imagine ISP at sea level is probably only like 290. Might be less of a case with Raptor 2 if it runs at a little higher pressure and they don’t alter ER.

This is why I was looking at this.  If you were to first, control the flow separation (obviously the key problem) and then eliminate the sea level engines and then mount gimballed vacuum engines.  Even if the ISP is greatly reduced, you could use all three of these engines to provide enough thrust for landing a Starship with nearly empty propellent tanks.  Also, by eliminating three engines you would reduce the overall mass and complexity of a Starship which would increase the amount of payload it could take into into space as well as reduce its landing mass.

Issues:

Even if flow separation is avoided or controlled this is most likely only the case for high throttle settings so you lose the ability to throttle down during landing.

Gimbaling of a larger/heavier RVac is much slower unless you beef up the actuators (and possibly reinforce the nozzle as well).

You can only fit 1 RVac in place of the three SL engines, so you lose some of the thrust you just added with the 9 Raptor version and you lose landing engine redundancy. If the 9 Raptor version had gained a limited low altitude abort capability you have now lost it again.

The longer center RVac means that you have to extend the skirt and move the outer ones down to avoid plume impingement, adding quite a bit of mass.

Summary: You now have a lower thrust and most likely heavier Starship that can only do suicide burn landings with reduced or no redundancy.

Issue #1:  This is a real possibility and would need to be addressed.

#2:  Although gimballing might be slower, this can be compensated by faster response to attitude inputs, so the nozzle might be moving slower but it will respond earlier.  In this way you wouldn't have to beef up the actuators or reinforce the nozzle.

#3:  I don't know what you are referring to with a 9 Raptor version.  The present design has only 3 SL Raptors and 3 RVAC's.  With the removal of the SL Raptors, there would be plenty of room for the 3 RVAC's.

#4:  There wouldn't be a center RVAC, in this design.  There would only be three RVAC's.  They would be spread out on a new thrust puck where there would be plenty of room, and they would be on the same plane so there would be no plume impingement.  Also less mass because three less engines 

I think you might be thinking of a Falcon 9 with it's 9 Merlin engines and not the Starship with it's 3 SL Raptors and it's 3 RVAC's.

#5:  As it is right now, Starship only needs 2 of it's three SL engines to land.  If you were to use 3 RVAC engines, even with less thrust, the overall thrust would still be more than 2 SL engines, giving you enough thrust and gimballing ability to land.

What do you mean by TWC responding earlier? It already reacts instantly (control loops are measured in milli- if not microseconds) so a slower moving nozzle means that the TVC has a larger time constant.

I thought you were talking bout the recently re-suggested 6 vacuum Raptor + 3 SL Raptor version. If you you just remove the SL Raptors on the 6 engine version you will drastically lower your payload.

Landing on 3 vacuum Raptors will be riskier (lower thrust, longer lever-arm, likely smaller throttle range and slower TVC response).