MiniRaptor / SC Merlin

[Nilof]

Assuming that the Raptor prototype that was test-fired last September was one-third scale, i.e. around 1000 kN, is there a possibility that we could see a smaller version of the Raptor as a replacement to the Merlin? I remember seeing a lot of discussion about this around 2014 back when we still didn't have a lot of data on Raptor.

Or more generally, is there a chance that the Merlin could be replaced by a staged combustion engine eventually, as SpaceX gains experience with staged combustion? While total liftoff thrust is still more important for first stages than Isp, and available footprint at the bottom of the rocket for nozzles could be a bottleneck, a potential 50 seconds of extra specific impulse from staged combustion/high chamber pressure is pretty significant.

[Hanno]

Unlikely.

All technical considerations aside, the Merlin is a proven, working design with all the logistical, manufacturing and quality control infrastructure in place. The performance bonus of an SC engine is just not big enough to throw it all away and start from scratch.

You need something that current engine just can't do in order to justify the development on a new engine. Like ITS/Mars for the Raptor or the need to get rid of the RD-180 for the BE-4/AR-1.

[Nilof]

True. But if you've gotten to the point where you have a prototype engine that has completed a significant number of test firings, most of the R&D expenses for developing the engine should be behind you. As far as manufacturing goes, I believe that EM mentioned that Raptor would have a lot in common with the Merlin.

With that said, you'd have to essentially design a new rocket around the new engine. That would probably not be as desirable. Though the idea of a Raptor US still seems to be fairly widespread.

[macpacheco]

F9 / FH design was grounded by the requirements of expendable missions + road transportability (optimize for construction cost and ease of ground transportation).
SpaceX now has a way to fully reuse boosters and upper stage reuse is mostly a matter of having lots of spare performance (for deorbit+landing burns plus a lot of TPS for re-entry).
When we add to this the fact that despite of its higher ISP and higher T:W, raptor rockets need to be thicker so road transport need to be given up.
This suggests that even a F9/FH Raptor 2nd stage is a kludge with limited benefit vs a brand new rocket.
The more I think about this, the more I think the best way is go Raptor all the way, to a mini ITS (meaning a big upper stage with lots of engines) or a Raptor Falcon 9 on steroids (single engine upper stage + 7-9 raptors on the booster). Either way we should end up with a boosters that might be reusable 100x without refurb and upper stages that could be reflown dozens of times without refurb too.
A new rocket that's unconstrained by road transport can achieve FH expendable performance with full reuse. In fact it should be designed without any contingency for any expendable usage. Its reuse all the way. It doesn't matter if it costs ~5x as much as a FH.
Trying to squeeze limited raptor usage in between probably doesn't save enough expended upper stages for the redesign cost. Better aim for a rocket that could be a workhorse for several decades instead.
Remember that it costed SpaceX one billion (so far) to figure out reuse. F9 Block I design costed US$ 400 million. Above that, it ties down lots of engineers that could be better used for more permanent ideas. Better to "waste" a few hundred million in 50 or even 100 more 2nd stages being wasted than divert engineers from getting a mini ITS flying sooner.
From a 1/3 thrust Raptor to 75+% thrust its probably a matter of 6-12 months more in testing/development. Better go all the way.
Besides the new workhorse rocket doesn't need to wait until Raptor testing is complete to be designed. ITS design apparently will go into high gear soon.

[MikeAtkinson]

There are multiple paths towards what is probably the medium term aim, ITS + methalox replacement for F9/FH. SpaceX have all the facts to make a rational choice between them, we just argue from ignorance. We will know more about SpaceX's decision in a few weeks when the updated ITS plan is announced.

Today I think they will do something like: ITS (full size, but derated capability), mini-ITS replacement to FH (refuel in LEO), full capability ITS.

Tomorrow I will probably think something different, I've regularly changed my mind on this as there do not seem to be any conclusive arguments based on the known facts.

It is interesting that with refueling in LEO something with a payload a bit smaller than the current F9 could carry out all existing missions, and most future ones (although fairing size would be a problem for some).

[su27k]

When we add to this the fact that despite of its higher ISP and higher T:W, raptor rockets need to be thicker so road transport need to be given up.

I think it has been explained on this forum many times that for SpaceX methane will have better performance than RP-1 even when using the same tank volume, mainly due to different mixture ratio and the big Isp improvement from Raptor over Merlin.

This suggests that even a F9/FH Raptor 2nd stage is a kludge with limited benefit vs a brand new rocket.

I don't think so, seems to me a reusable and refuelable F9/FH Raptor 2nd stage is the 2nd best option for going forward, right after BFS SSTO. The only reason they want to switch first stage to methane is they find out RP-1 makes it impossible to reach their reusability goal, which I think it's unlikely.

[CorvusCorax]

A design for a Methane based fully reuseable "bigger version" of Falcon 9 is already in development.

Over at Blue Origin, they call it "New Glenn".

I'd say either SpaceX already has a Raptor based Falcon successor design (other than ITS) in the drawing boards to bring to the market when F9 doesn't kick it anymore (because, if, eventually, New Glenn enters service and B/O becomes competitive with it - lots of ifs) Or they skip it altogether assuming that ITS could and would serve any LEO and GEO launches that would ever be needed. easily - once in service.

If ITS does what it's designed for, it could easily haul any number of GEO sat where they need to be. All at once if needed (They could do a launch once a quarter, and stuff any sat on board that happens to be payload ready by then - unlike Ariane where dual launches have to be carefully mached, you could just bulk load com-sats, kick them out in actual GEO so all they'l ever need to do on their own is a slot change, and then deorbit and land the carrier)

The one thing ITS would possibly be less than optimal is LEO and vLEO orbits with high inclination - especially if you need to launch many small sats into many different planes.

That's where F9 would shine however, all they need to do is get the upper stage reuseable.

So I'd say, considering the market situation and potential competition, there's likely no need for SpaceX to develop a (Mini-?)raptor based F9 successor anytime soon - unless upper stage reuse on F9 current architecture fails.


That being said, SpaceX does IMHO need a fully reusable launcher for their mega-constellation. I'm curious what's the plan there...

[envy887]

A staged combustion Merlin would be something like ORSC RP-1/LOX. That's not a direction SpaceX is going long-term, and the performance increase isn't nearly enough to warrant the effort.

A scaled-down Raptor might also be unnecessary long term. A single Raptor with a mid-length nozzle (something like 120:1 expansion) would be ideal for a large FH upper stage, or for an all-methalox FH class single stick upper stage, much like New Glenn's single BE-4U 2nd stage.

[sevenperforce]

When we add to this the fact that despite of its higher ISP and higher T:W, raptor rockets need to be thicker so road transport need to be given up.

I think it has been explained on this forum many times that for SpaceX methane will have better performance than RP-1 even when using the same tank volume, mainly due to different mixture ratio and the big Isp improvement from Raptor over Merlin.

Yep, definitely. The increase in specific impulse and in TWR means that even with the fluffier propellant, putting Raptors on a Falcon 9 body (assuming you could move the common bulkhead appropriately) would get you a 40% increase in payload.

[TheKutKu]

Could we stop spreading the myth that LCH4/LOX is much more lighter than RP-1/LOX? According to the astronautix link http://www.astronautix.com/l/loxlch4.html http://www.astronautix.com/l/loxkerosene.html

At Raptor's 3.8 O/F ratio the propellant's density is 0.991 g/cc, that's 97.1% of RP-1's density, with a more common 3.6 O/F ratio it has 96.5% of RP-1's density. Also that's not taking into account that both Methane and Oxygen are subcooled for the Raptor whereas only the Oxygen is for the Merlin.

[envy887]

Could we stop spreading the myth that LCH4/LOX is much more lighter than RP-1/LOX? According to the astronautix link http://www.astronautix.com/l/loxlch4.html http://www.astronautix.com/l/loxkerosene.html

At Raptor's 3.8 O/F ratio the propellant's density is 0.991 g/cc, that's 97.1% of RP-1's density, with a more common 3.6 O/F ratio it has 96.5% of RP-1's density. Also that's not taking into account that both Methane and Oxygen are subcooled for the Raptor whereas only the Oxygen is for the Merlin.

It's not really a myth. Subcooled kerolox is ~1115 kg/m³. Subcooled methalox at 3.8:1 is ~930 kg/m³. There's roughly a 20% bulk density difference.

And the RP-1 in Falcon 9 is also subcooled, just not the extent that the LOX is... the RP-1 is loaded at several degrees below 0 C.

[livingjw]

Could we stop spreading the myth that LCH4/LOX is much more lighter than RP-1/LOX? According to the astronautix link http://www.astronautix.com/l/loxlch4.html http://www.astronautix.com/l/loxkerosene.html

At Raptor's 3.8 O/F ratio the propellant's density is 0.991 g/cc, that's 97.1% of RP-1's density, with a more common 3.6 O/F ratio it has 96.5% of RP-1's density. Also that's not taking into account that both Methane and Oxygen are subcooled for the Raptor whereas only the Oxygen is for the Merlin.

Agree. The overall performance of RP-LOX and CH4-LOX engines are very close. CH4-LOX performance is a little higher but has lower propellant density so the two almost cancel each other out. The performance improvement between the Merlin 1D and Raptor is due mostly to its staged combustion cycle and its high pressure. The real reasons for choosing CH4-LOX are:
- minimal coking,
- autogenous pressurization,
- in situ propellant production,
- more cooling capacity,
- lower cost

[TheKutKu]

Could we stop spreading the myth that LCH4/LOX is much more lighter than RP-1/LOX? According to the astronautix link http://www.astronautix.com/l/loxlch4.html http://www.astronautix.com/l/loxkerosene.html

At Raptor's 3.8 O/F ratio the propellant's density is 0.991 g/cc, that's 97.1% of RP-1's density, with a more common 3.6 O/F ratio it has 96.5% of RP-1's density. Also that's not taking into account that both Methane and Oxygen are subcooled for the Raptor whereas only the Oxygen is for the Merlin.

It's not really a myth. Subcooled kerolox is ~1115 kg/m³. Subcooled methalox at 3.8:1 is ~930 kg/m³. There's roughly a 20% bulk density difference.

And the RP-1 in Falcon 9 is also subcooled, just not the extent that the LOX is... the RP-1 is loaded at several degrees below 0 C.

Do you have a source for theses density numbers? Astronautix links give a higher Density for (non subcooled) Methalox.

[acsawdey]

http://unitrove.com/engineering/tools/gas/liquefied-natural-gas-density gives 0.4484 for CH4 at 93K.

https://www.gpo.gov/fdsys/pkg/GOVPUB-C13-26d428ad4ca587866a90da5f71b4a727/pdf/GOVPUB-C13-26d428ad4ca587866a90da5f71b4a727.pdf gives 1.2817 for O2 at 60K.

(3.6*1.2817+0.4484)/(1+3.6) = 1.1005

Using the astronautix number for un-densified RP-1 with the densified LOX number:
(2.56*1.2817+0.806)/(1+2.56) = 1.1481

Gives 0.9586 as the ratio.

[envy887]

Could we stop spreading the myth that LCH4/LOX is much more lighter than RP-1/LOX? According to the astronautix link http://www.astronautix.com/l/loxlch4.html http://www.astronautix.com/l/loxkerosene.html

At Raptor's 3.8 O/F ratio the propellant's density is 0.991 g/cc, that's 97.1% of RP-1's density, with a more common 3.6 O/F ratio it has 96.5% of RP-1's density. Also that's not taking into account that both Methane and Oxygen are subcooled for the Raptor whereas only the Oxygen is for the Merlin.

It's not really a myth. Subcooled kerolox is ~1115 kg/m³. Subcooled methalox at 3.8:1 is ~930 kg/m³. There's roughly a 20% bulk density difference.

And the RP-1 in Falcon 9 is also subcooled, just not the extent that the LOX is... the RP-1 is loaded at several degrees below 0 C.

Do you have a source for theses density numbers? Astronautix links give a higher Density for (non subcooled) Methalox.

Astronautix give densities for CH4 and LOX, not bulk density for methalox. At 3.8:1 O/F, 424 kg/m³ fuel density, and 1140 kg/m³ oxidizer density, the calculation is:

(424*(1+3.8))/(424*3.8/1140+1) = 843 kg/m³

If you do this: (424+1140*3.8)/(1+3.8) = 990 kg/m³ you are really calculating the bulk density of a VOLUMETRIC O/F ratio of 3.8:1, which corresponds to a MASS O/F ratio of 10.2:1. Fuel ratios in rocketry are always mass ratios.

To check this, think: are you really going to burn 1140*3.8 = 4332 kg of LOX with only 424 kg of CH4? Don't try to do this, you'll end up with a lot of hot leftover oxygen, which is really nasty stuff.

[livingjw]

The higher density propellant is preferred to minimize tank weight which is proportional to volume.

[envy887]

http://unitrove.com/engineering/tools/gas/liquefied-natural-gas-density gives 0.4484 for CH4 at 93K.

https://www.gpo.gov/fdsys/pkg/GOVPUB-C13-26d428ad4ca587866a90da5f71b4a727/pdf/GOVPUB-C13-26d428ad4ca587866a90da5f71b4a727.pdf gives 1.2817 for O2 at 60K.

(3.6*1.2817+0.4484)/(1+3.6) = 1.1005

Using the astronautix number for un-densified RP-1 with the densified LOX number:
(2.56*1.2817+0.806)/(1+2.56) = 1.1481

Gives 0.9586 as the ratio.

That's a volumetric ratio of 3.6:1 bulk density, not mass ratio of 3.6:1 bulk density.

At 3.6:1 mass ratios with the cooled CH4/LOX densities above, the bulk density is: (448.4*4.6)/(448.4*3.6/1281.7+1) = 912.9 kg/m³

The numbers here are correct - use them:
https://forum.nasaspaceflight.com/index.php?topic=42302.msg1642537#msg1642537

[sevenperforce]

The higher density propellant is preferred to minimize tank weight which is proportional to volume.

Yes, but higher specific energy can more than make up the difference. If you shifted the bulkhead on the Falcon 9 first stage and filled it up with densified methalox, and slapped three Raptors underneath, it would have a 40% performance increase.

[TheKutKu]

Could we stop spreading the myth that LCH4/LOX is much more lighter than RP-1/LOX? According to the astronautix link http://www.astronautix.com/l/loxlch4.html http://www.astronautix.com/l/loxkerosene.html

At Raptor's 3.8 O/F ratio the propellant's density is 0.991 g/cc, that's 97.1% of RP-1's density, with a more common 3.6 O/F ratio it has 96.5% of RP-1's density. Also that's not taking into account that both Methane and Oxygen are subcooled for the Raptor whereas only the Oxygen is for the Merlin.

It's not really a myth. Subcooled kerolox is ~1115 kg/m³. Subcooled methalox at 3.8:1 is ~930 kg/m³. There's roughly a 20% bulk density difference.

And the RP-1 in Falcon 9 is also subcooled, just not the extent that the LOX is... the RP-1 is loaded at several degrees below 0 C.

Do you have a source for theses density numbers? Astronautix links give a higher Density for (non subcooled) Methalox.

Astronautix give densities for CH4 and LOX, not bulk density for methalox. At 3.8:1 O/F, 424 kg/m³ fuel density, and 1140 kg/m³ oxidizer density, the calculation is:

(424*(1+3.)/(424*3.8/1140+1) = 843 kg/m³

If you do this: (424+1140*3./(1+3. = 990 kg/m³ you are really calculating the bulk density of a VOLUMETRIC O/F ratio of 3.8:1, which corresponds to a MASS O/F ratio of 10.2:1. Fuel ratios in rocketry are always mass ratios.

To check this, think: are you really going to burn 1140*3.8 = 4332 kg of LOX with only 424 kg of CH4? Don't try to do this, you'll end up with a lot of hot leftover oxygen, which is really nasty stuff.

Thank you for correcting me! That was a large oversight, i take back what i said.

[envy887]

The higher density propellant is preferred to minimize tank weight which is proportional to volume.

Yes, but higher specific energy can more than make up the difference. If you shifted the bulkhead on the Falcon 9 first stage and filled it up with densified methalox, and slapped three Raptors underneath, it would have a 40% performance increase.

What are your assumptions behind this 40% number?

I have plotted in the attached image the bulk density required to achieve the Falcon 9 booster's delta-v with payload, assuming constant volume, stage dry mass, thrust, and outer mold line. Methalox Raptor is above the required Isp to get the same performance as Merlin, but not by nearly enough to get 40% improvement. More like 4% unless they also do a methane upper stage.