SpaceX Raptor LRE Proposed Successor: Project LEET-1337 LRE

[RedLineTrain]

BTW, like many things in the biography, we've already heard about this for a while.

Where did we already hear about this?  Is my reading of L2 deficient?

He tweeted about it.

But I thought that was regarding Raptor rather than 1337?  What has he tweeted regarding 1337, besides some vague references?

I felt like this was indeed a lot of new detail in the biography.

[Robotbeat]

Yes, there are new details. But it was mentioned:

November 16th 2021 Updates: NEW ENGINE

Elon Musk @elonmusk
True, although it will look clean with close out panels installed.

Raptor 2 has significant improvements in every way, but a complete design overhaul is necessary for the engine that can actually make life multiplanetary. It won’t be called Raptor

[Robotbeat]

In fact, we have a whole 22 page thread on the topic from the end of 2021:
https://forum.nasaspaceflight.com/index.php?topic=55225.0

[RedLineTrain]

Yes, there are new details. But it was mentioned:

November 16th 2021 Updates: NEW ENGINE

Elon Musk @elonmusk
True, although it will look clean with close out panels installed.

Raptor 2 has significant improvements in every way, but a complete design overhaul is necessary for the engine that can actually make life multiplanetary. It won’t be called Raptor

Got it.  Yes, I was aware of this tweet.  But no details that I know of really came out until the biography.

[Alberto-Girardi]

A note about price.

If the price of Raptor is 1 million $, and we have no indication saying Musk was faking this number, it is already in a good spot (compared to other engines). But I know, he won't stop until they reach the limit.
In general I was surprised by the price of modern rocket engines compared to other complex machinery, like jet engines. The engines of the 737 MAX are 14.5 millions each. Sure the plane doesn't have 39 of them. But the price is comparable. Ofcourse  there are very big differences, but it isn't a totally out of the blue apples to oranges comparsion. I was initially surprised the prices were so close or maybe I was underestimating jet engine prices.

The BE4 engine costs about 8 milion, the RD180 25. The Merlin less than 1. The numbers are from EDA.

[Nomadd]

 You left out the craziest comparison. The RS-25.

[hplan]

A note about price.

If the price of Raptor is 1 million $, and we have no indication saying Musk was faking this number, it is already in a good spot (compared to other engines). But I know, he won't stop until they reach the limit.
In general I was surprised by the price of modern rocket engines compared to other complex machinery, like jet engines. The engines of the 737 MAX are 14.5 millions each. Sure the plane doesn't have 39 of them. But the price is comparable. Ofcourse  there are very big differences, but it isn't a totally out of the blue apples to oranges comparsion. I was initially surprised the prices were so close or maybe I was underestimating jet engine prices.

The BE4 engine costs about 8 milion, the RD180 25. The Merlin less than 1. The numbers are from EDA.

Those numbers are indeed crazy. But they aren't really comparable.

Elon always likes to talk about 'marginal cost' -- how much more it costs to make one additional engine. So the $1M figure doesn't include research and development, cost of the factory and equipment to build engines, profit, etc. No doubt if they were selling engines separately, the cost would be much higher.

[Robotbeat]

A big thing with Musk, though, is that although he talks about marginal cost, he’s also nearly always pushing for very high production rates where development and factory costs are amortized pretty quickly and the fully burdened cost can approach the marginal cost (or at least like a factor of 2).

SpaceX is producing hundreds of Raptors per year, about 2 orders of magnitude higher than, say, the manufacturer of RS-25.

[Alberto-Girardi]

You left out the craziest comparison. The RS-25.

I did that because, as you say, it is crazy. But being it made only for the government I give them the benefit of the doubt.

They are indeed around a little bit less than 100 millions each going to the latest contracts, but in the old days were like 60.
But incredibly when I read it I was reassured, because I don't  know where I had read they were 170 milions each, which isn't true (for now!).

[Robotbeat]

You left out the craziest comparison. The RS-25.

I did that because, as you say, it is crazy. But being it made only for the government I give them the benefit of the doubt.

They are indeed around a little bit less than 100 millions each going to the latest contracts, but in the old days were like 60.
But incredibly when I read it I was reassured, because I don't  know where I had read they were 170 milions each, which isn't true (for now!).

The higher figure includes the cost NASA paid to restart the production line. And it’s $146 million. $100 million or so is for the add-on contract by itself.

Note that expanding the production line would probably require additional cost as well. https://arstechnica.com/science/2020/05/nasa-will-pay-a-staggering-146-million-for-each-sls-rocket-engine/amp/

[DanClemmensen]

A big thing with Musk, though, is that although he talks about marginal cost, he’s also nearly always pushing for very high production rates where development and factory costs are amortized pretty quickly and the fully burdened cost can approach the marginal cost (or at least like a factor of 2).

SpaceX is producing hundreds of Raptors per year, about 2 orders of magnitude higher than, say, the manufacturer of RS-25.

He also talks of a eventual marginal cost of $250,000. I suspect that this is when they are building the Mars fleet, at some crazy engine production rate. At that marginal cost, an internal true cost of $1 million might be realistic.

[RedLineTrain]

Just listening again to the biography, Chapter 63.  It states that the 1337 engine effort was only a month long in 2022.  They turned back to Raptor 2.

With Raptor now hitting its manufacturing goals (at least rate, if not $250k cost) and overperforming on chamber pressure, I wonder whether the 1337 engine is moot or pushed off into the far future.  Raptor was at least hitting its stride enough to merit a Raptor 3 this year.

[Robotbeat]

A big thing with Musk, though, is that although he talks about marginal cost, he’s also nearly always pushing for very high production rates where development and factory costs are amortized pretty quickly and the fully burdened cost can approach the marginal cost (or at least like a factor of 2).

SpaceX is producing hundreds of Raptors per year, about 2 orders of magnitude higher than, say, the manufacturer of RS-25.

He also talks of a eventual marginal cost of $250,000. I suspect that this is when they are building the Mars fleet, at some crazy engine production rate. At that marginal cost, an internal true cost of $1 million might be realistic.

No, I think they’re probably at around $1 million true internal cost already, not counting dev costs.

They want to get costs to Mars very low.

If you assume the ship is sent one-way to Mars with 9 Raptors and 75 passengers, the engine costs on the ship alone are like $120,000 per person, already half of the whole mission cost. If you assume $250k per engine, that’s just $30k per person.

…that is the relevance to making humanity multiplanetary

Likewise, say you send 100 tonnes of cargo with 9 Raptors one-way. $1 million means $90/kg from the ship engines alone. $250k cost means just $22.50/kg from the engines, much more reasonable.

Maybe, including amortization of engines on the booster and tankers, you get a round 10 tonnes of payload to Mars per engine, maybe 10 Martian settlers per engine.

So if they want about 2 million tons of payload on Mars per synod, that’s 100,000 Raptors per year they’ll want to make.

It’s automotive manufacturing scale. Even if they manage to increase reuse to a factor of 10 roundtrip missions per year or so, still 10,000 engines per year. Higher than all jet engine production on Earth combined.

[Alberto-Girardi]

A note about price.

If the price of Raptor is 1 million $, and we have no indication saying Musk was faking this number, it is already in a good spot (compared to other engines). But I know, he won't stop until they reach the limit.
In general I was surprised by the price of modern rocket engines compared to other complex machinery, like jet engines. The engines of the 737 MAX are 14.5 millions each. Sure the plane doesn't have 39 of them. But the price is comparable. Ofcourse  there are very big differences, but it isn't a totally out of the blue apples to oranges comparsion. I was initially surprised the prices were so close or maybe I was underestimating jet engine prices.

The BE4 engine costs about 8 milion, the RD180 25. The Merlin less than 1. The numbers are from EDA.

Those numbers are indeed crazy. But they aren't really comparable.

Elon always likes to talk about 'marginal cost' -- how much more it costs to make one additional engine. So the $1M figure doesn't include research and development, cost of the factory and equipment to build engines, profit, etc. No doubt if they were selling engines separately, the cost would be much higher.

Thanks for the explanation. This is indeed something I didn't consider. Hopefully raptor gets to the state of scaling that the marginal cost and the total ammortized price are nearly the same. IIUC what Elon says is useful if you want to make the design cheaper, but isn't good if you want to know the actual cost. He presumes they will get to the stage were R&D cost won't be vety significant.

[Alberto-Girardi]

Just listening again to the biography, Chapter 63.  It states that the 1337 engine effort was only a month long in 2022.  They turned back to Raptor 2.

With Raptor now hitting its manufacturing goals (at least rate, if not $250k cost) and overperforming on chamber pressure, I wonder whether the 1337 engine is moot or pushed off into the far future.  Raptor was at least hitting its stride enough to merit a Raptor 3 this year.

I haven't read the biography (maybe I should, but I don't generally read biographies of living people) , but IIUC we can't rule out that this was one of the many things that were talked about by Musk, explored internally at the company and nothig (at least until now) came out. Generally I would never take an old tweet of him as a proof of anything current unless there is evidence that has actually been done or has stayed the same since. I think the issue of landing the ship on the chopsticks and the ships forward flaps and traspirational cooling are examples of this that come to mind, albeit the first is a little bit more grounded.


edit: Is the snippet posted above everything in the biography about this project?

[RedLineTrain]

edit: Is the snippet posted above everything in the biography about this project?

The biography mentioned some deletion/simplification ideas that were batted around, if but briefly, with a cost goal of less than $1,000 per ton of thrust.

1.  Deleting the hot fuel gas manifold
2.  Merging the fuel turbopump with the main chamber injector
3.  Getting rid of skirt for booster

It's a long biography and contains mostly things that we already know.  So it might be boring to you, as you are waiting for the gold nuggets of new information.

[Hog]

When comparing the SLSME (RS25 production restart) we must remember that the service life is now 4 starts/1700 seconds of runtime vs the older Heritage-STS/SSME/RS25D/Block-2 service life requirement of 55 starts/27,000 seconds. IOW Expendable RS25 vs Reusable RS25.

EDIT: The correct attachment is there now.

[Robotbeat]

That sounds like a shockingly low number even for an expendable engine. Let’s say you have an engine acceptance test, a core stage acceptance test (like the green run), and then launch. That’s 3 of the four uses.

If you have to aborted lift-offs or green runs, you’d have to get new engines!

You can see why they deleted the green run even for crewed launches. They just don’t have the cycle life in the engines to afford that level of acceptance testing.

Just so very short-sighted.

[wannamoonbase]

That sounds like a shockingly low number even for an expendable engine. Let’s say you have an engine acceptance test, a core stage acceptance test (like the green run), and then launch. That’s 3 of the four uses.

If you have to aborted lift-offs or green runs, you’d have to get new engines!

You can see why they deleted the green run even for crewed launches. They just don’t have the cycle life in the engines to afford that level of acceptance testing.

Just so very short-sighted.

Maybe an extra restart cost $100M more per engine.

The RS25 is 50+ year old technology at this point, its embarrassingly expensive.

Beyond all the performance numbers and cost, the key to Raptor is going to be longevity. 

Merlin is doing very well but Raptor needs to be even easier to reuse and allegedly handle many more flights.  It may take a long time to get there.

[Robotbeat]

They’re using the low cycle count of RS-25e to push the max thrust to far higher than SSME. You get a few percentage points greater thrust in exchange for an order of magnitude fewer cycles.