SpaceX Raptor engine - General Thread 2

[JonathanD]

While they may wreck a few prototypes along the way Starship makes only sense if it turns out to be reusable. Cost/engine is always a critical metric if you need so many of them. There will be 42 + 7 on every first + second stage, that's a lot of money, and more if you want to build many ships to go to Mars of which several will stay.

All the more reason to make sure they have a very robust engine before they begin mass production...

[hopalong]

While they may wreck a few prototypes along the way Starship makes only sense if it turns out to be reusable. Cost/engine is always a critical metric if you need so many of them. There will be 42 + 7 on every first + second stage, that's a lot of money, and more if you want to build many ships to go to Mars of which several will stay.

All the more reason to make sure they have a very robust engine before they begin mass production...

I last time I looked, SH has 31 engines, not 42.

[uhuznaa]

While they may wreck a few prototypes along the way Starship makes only sense if it turns out to be reusable. Cost/engine is always a critical metric if you need so many of them. There will be 42 + 7 on every first + second stage, that's a lot of money, and more if you want to build many ships to go to Mars of which several will stay.

All the more reason to make sure they have a very robust engine before they begin mass production...

I last time I looked, SH has 31 engines, not 42.

Right, sorry, it's 31 engines. Still quite a few.

[uhuznaa]

While they may wreck a few prototypes along the way Starship makes only sense if it turns out to be reusable. Cost/engine is always a critical metric if you need so many of them. There will be 42 + 7 on every first + second stage, that's a lot of money, and more if you want to build many ships to go to Mars of which several will stay.

All the more reason to make sure they have a very robust engine before they begin mass production...

Yes, and definitely a good reason not to prioritize T/W in the first versions over that. If all of this works out Starship will be with us for quite a long time, there's still time for optimizing things later. ISP is important (since the second stage will have a less than ideal dry mass anyway, due to being reusable) and reusability and reliability is important since you want to reuse them as often as possible. Next thing to optimize for is cost and the last thing to optimize for is T/W. I expect the 20th ship they will build to shed a LOT of mass compared to the first one...

[FutureSpaceTourist]

Could you tell, how many people working now at Raptor development, except you? If it’s not classified of course.

https://twitter.com/elonmusk/status/1100494266533433344

Rest of SpaceX propulsion still very active, so only ~50 full-time equivalent people right now. That will grow a lot as we enter production. It’s 10X harder (at least) to design engine production system than engine. In automotive, 100X harder.

[philw1776]

While they may wreck a few prototypes along the way Starship makes only sense if it turns out to be reusable. Cost/engine is always a critical metric if you need so many of them. There will be 42 + 7 on every first + second stage, that's a lot of money, and more if you want to build many ships to go to Mars of which several will stay.

All the more reason to make sure they have a very robust engine before they begin mass production...

Precisely.
That's why I think the SuperHeavy will fly, including real payloads to LEO, etc. with less than 31 engines for a while, say 19 or so.  Iterate the design based on real useage.  Then spend the big bucks manufacturing engineering so you can pop 'em out.
Gotta be careful spending the limited cash.

[hplan]

Could you tell, how many people working now at Raptor development, except you? If it’s not classified of course.

https://twitter.com/elonmusk/status/1100494266533433344

Rest of SpaceX propulsion still very active, so only ~50 full-time equivalent people right now. That will grow a lot as we enter production. It’s 10X harder (at least) to design engine production system than engine. In automotive, 100X harder.

Well, that's an interesting tidbit. Anyone can make a FFSC engine, but can you make 400 of them a year, cost-effectively?

It also implies that they aren't anywhere close to being able to make more than a one-off engine every couple of weeks. Enough for a superheavy? Not for a while yet.

And what is the rest of SpaceX propulsion very active with? Changes to the Merlin 1D?

[dnavas]

And what is the rest of SpaceX propulsion very active with? Changes to the Merlin 1D?

I suppose some are on Dracos -- from DM-1 FRR briefing:

[ncb1397]

And what is the rest of SpaceX propulsion very active with? Changes to the Merlin 1D?

It sounds like he is counting staff that produces parts for engines, assembles and tests them. So, they are probably working on building dracos, super dracos, merlins.

[Keldor]

Precisely.
That's why I think the SuperHeavy will fly, including real payloads to LEO, etc. with less than 31 engines for a while, say 19 or so.  Iterate the design based on real useage.  Then spend the big bucks manufacturing engineering so you can pop 'em out.
Gotta be careful spending the limited cash.

Usint 19 engines instead of 31 on the first stage still means they have to produce 70% as many engines overall after you count the second stage ones.  You're not really gaining much.

What you are doing, however, is requiring a second version of Super Heavy since 19 engines won't be enough to lift the fully fueled rocket off the ground.

[Oli]

Not sure why they're aiming for such high chamber pressures, to my knowledge it's detrimental to reusability. I mean it's not SSTO.

[Lars-J]

Not sure why they're aiming for such high chamber pressures, to my knowledge it's detrimental to reusability. I mean it's not SSTO.

You have to view the whole system holistically. Higher engine efficiency allows more mass margin in other areas. It is all a series of trade-offs, and what they believe they can achieve. They must think that the trade-off is worth it, and/or that they have developed an engine that can be reusable at those pressures.

[Prettz]

Not sure why they're aiming for such high chamber pressures, to my knowledge it's detrimental to reusability. I mean it's not SSTO.

Presumably to allow the second stage to hit those payload-to-mars numbers with the other given constraints. And also the other beyond-LEO destinations it'll need to go to.

[gongora]

This thread is not for discussing how many engines are on Super Heavy.  Or how many more Merlins are going to be produced.  There are other threads for that.  Let's keep this thread focused on Raptor.  Just the engine, not the vehicles using it.

[RobLynn]

Not sure why they're aiming for such high chamber pressures, to my knowledge it's detrimental to reusability. I mean it's not SSTO.

I think it is only the thrust chamber that suffers from high pressures, and maybe that is a replaceable component amounting to only a small portion of the overall engine cost.  Or maybe they have figured out a clever solution for the low-cycle fatigue issue caused by enormous heat flux temperature gradient through the walls causing a cycle of thermal strain beyond elastic limits with each engine firing cycle.  Transpiration cooling of thrust chamber might do it.

[WormPicker959]

Not sure why they're aiming for such high chamber pressures, to my knowledge it's detrimental to reusability. I mean it's not SSTO.

I think it is only the thrust chamber that suffers from high pressures, and maybe that is a replaceable component amounting to only a small portion of the overall engine cost.  Or maybe they have figured out a clever solution for the low-cycle fatigue issue caused by enormous heat flux temperature gradient through the walls causing a cycle of thermal strain beyond elastic limits with each engine firing cycle.  Transpiration cooling of thrust chamber might do it.

So, after reading this and recalling a previous discussion I had a thought, and it's probably dumb, so I hope people who know better can tell me so. But here it is:

My understanding is that if an increase in film/transpiration cooling would be necessary to operate with reusability at these pressures, then there's a bit of a trade-off in efficiency, as the extra fuel that's used for the cooling is not combusted. However, most engines run fuel rich - my understanding of this is that this is because methane is lighter than oxygen, so you get slightly higher efficiency by just throwing out extra hot methane molecules out the business end. So what if, instead of running the combustion fuel-rich, you run it more stoichiometrically, and allow for the methane used for film cooling to be the "hot methane thrown out the back end"? I suppose it doesn't get as hot/high of velocity as it would if the combustion were run fuel-rich. In any case, I'm sure most of this reasoning is not sound, just thought I'd throw it out there.

[CorvusCorax]

Not sure why they're aiming for such high chamber pressures, to my knowledge it's detrimental to reusability. I mean it's not SSTO.

I think it is only the thrust chamber that suffers from high pressures, and maybe that is a replaceable component amounting to only a small portion of the overall engine cost.  Or maybe they have figured out a clever solution for the low-cycle fatigue issue caused by enormous heat flux temperature gradient through the walls causing a cycle of thermal strain beyond elastic limits with each engine firing cycle.  Transpiration cooling of thrust chamber might do it.

So, after reading this and recalling a previous discussion I had a thought, and it's probably dumb, so I hope people who know better can tell me so. But here it is:

My understanding is that if an increase in film/transpiration cooling would be necessary to operate with reusability at these pressures, then there's a bit of a trade-off in efficiency, as the extra fuel that's used for the cooling is not combusted. However, most engines run fuel rich - my understanding of this is that this is because methane is lighter than oxygen, so you get slightly higher efficiency by just throwing out extra hot methane molecules out the business end. So what if, instead of running the combustion fuel-rich, you run it more stoichiometrically, and allow for the methane used for film cooling to be the "hot methane thrown out the back end"? I suppose it doesn't get as hot/high of velocity as it would if the combustion were run fuel-rich. In any case, I'm sure most of this reasoning is not sound, just thought I'd throw it out there.

No, no, and no. To both of you.
The main thrust chamber has the lowest pressure of everything. Check the engine diagram in the previous thread. With a main chamber pressure of 30 MPa you end up with a pressure in the hot, oxygen rich preburner of more than twice as much, which them hits the oxygen turbopumps turbine. That's where the highest mechanical stresses in the engine will occur.

The main combustion chamber has thermal stresses, but if you run it stochimetrically, it runs even hotter, requiring again even more cooling. Most carbohydrate engines suffer ISP from running fuel rich, since unburnt fuel is heavier molecules which reach slower exhaust speeds. Their ISP would increase if you added oxygen, but then the temperature becomes too high ( and with that much oxygen you start running "engine rich"  combustion )

The exception is Hydrolox. Hydrogen is lighter than either oxygen or steam. As such by running a hydrolox engine more fuel rich, you incrrase ISP (at reduced thrust) because you effectively just heat up hydrogen and exhaust it. Similar to what a nuclear engine would do, but with a chemical reaction as energy source.

Film cooling has 2 purposes.
1. Cooling the engine walls and prevent convective heat from reaching it, with a layer of colder fluid.
2. Preventing oxidizer and the ongoing combustion reaction from reaching the chamber wall and combusting the metal.

If the film starts reacting itself , it becomes as hot as the rest and you loose both properties.

[RobLynn]

Not sure why they're aiming for such high chamber pressures, to my knowledge it's detrimental to reusability. I mean it's not SSTO.

I think it is only the thrust chamber that suffers from high pressures, and maybe that is a replaceable component amounting to only a small portion of the overall engine cost.  Or maybe they have figured out a clever solution for the low-cycle fatigue issue caused by enormous heat flux temperature gradient through the walls causing a cycle of thermal strain beyond elastic limits with each engine firing cycle.  Transpiration cooling of thrust chamber might do it.

So, after reading this and recalling a previous discussion I had a thought, and it's probably dumb, so I hope people who know better can tell me so. But here it is:

My understanding is that if an increase in film/transpiration cooling would be necessary to operate with reusability at these pressures, then there's a bit of a trade-off in efficiency, as the extra fuel that's used for the cooling is not combusted. However, most engines run fuel rich - my understanding of this is that this is because methane is lighter than oxygen, so you get slightly higher efficiency by just throwing out extra hot methane molecules out the business end. So what if, instead of running the combustion fuel-rich, you run it more stoichiometrically, and allow for the methane used for film cooling to be the "hot methane thrown out the back end"? I suppose it doesn't get as hot/high of velocity as it would if the combustion were run fuel-rich. In any case, I'm sure most of this reasoning is not sound, just thought I'd throw it out there.

That's about right. All or nearly all fuel in thinnest possible layer next to the wall.  Heated by 10's of MW/m² [edit some engines up to 200MW/m²] of combined convection and radiation.  But I am not sure how much heat flux reduction can be achieved by film cooling - it can't stop the radiation component.  As I understand it pretty much all combustion chambers/nozzles are limited to about 50-100 firing cycles by low-cycle fatigue stresses, and the high thermal conductivity high hot-strength copper alloys used (Narloy Z et al) are not easy to work with typically requiring a lot of cold working to attain good strength and not machining nicely.  Recent work on laser sintering of copper alloys might be creating alternative solutions.

Transpiration cooling of the thrust chamber and nozzle walls offers probably the only solution if you can keep the transpiration coolant super clean/filtered, laser drilling of millions of holes is now possible.  But tiny detritus/dirt blocking the pores may be an unsolvable problem from a reliability point of view.  Another possibility might be super-fine (<<1mm) channel walls and thinner chamber walls, as thinner walls reduce the temperature difference across the thickness.  SSME nozzle cooling channels are around 1mm wide.  Historically manufacturing processes weren't up to making finer features, but that is likely changing in more recent times.

[Hominans Kosmos]

if you can keep the transpiration coolant super clean/filtered, laser drilling of millions of holes is now possible.  But tiny detritus/dirt blocking the pores may be an unsolvable problem from a reliability point of view.

As I understand it, Raptor is being designed for operation without the dirt mixing stage in the combustion cycle. Has there been recent change introducing debris as part of the propellant mixture?

[rakaydos]

Not sure why they're aiming for such high chamber pressures, to my knowledge it's detrimental to reusability. I mean it's not SSTO.

I think it is only the thrust chamber that suffers from high pressures, and maybe that is a replaceable component amounting to only a small portion of the overall engine cost.  Or maybe they have figured out a clever solution for the low-cycle fatigue issue caused by enormous heat flux temperature gradient through the walls causing a cycle of thermal strain beyond elastic limits with each engine firing cycle.  Transpiration cooling of thrust chamber might do it.

As I understand the earlier discussions on FFSC vs (F/O)RSC, (and I'm a kerbal trained layman, so correct me if I'm wrong) it's the turbopump pressure that's the limiting factor, with a pressure easily twice or more that of the main combustion chamber.