What are your ultimate commercial rocket designs?

[Oli]

Skylon.

Otherwise a reusable kerosene booster with an orbital hydrogen lifting body on top (x-33 on a stick, if you will )

[Lobo]

Since Hyperion pointed me to this thread and ask for my normally incredibly verbose opinion, I'll share a thought here:

First, I don’t know a ton about the pros/cons of methalox vs. kerolox.  I do know that they seem to have pretty similar performance, and are relatively close in density/volume.
But by switching from kerolox to methalox, you’ve traded up one of the most easily handled, transported, and stored fuel’s in RP-1 for a mild cryogenic that’s similar to LOX, with those associated difficulties.

But it seems methalox upper stags can get better isp than kerolox.  Raptor is supposed to be about 380s, while RD-0124 (one of the most efficient kerolox upper stages I think) is about 359s.  Hyperion has pointed out some metallurgical advantages of methalox, and seems to like it, so we’ll pull on that thread some.

From our discussion about Titan and Saturn 1 over on the Apollo not being cancelled thread, it seems that core lengths are very easy, but core diameter changes are much harder.  So I think we should pick one common core diameter that can do it all, and then have different core lengths for different options and applications.

There are customers for payloads in the less than 10mt range.  But I think that F9, Antares, Delta IV, Atlas V, etc all have a similar capacity of 10mt.  I think that’s a good place to have your basic building block.  If the LV is cheap, then it doesn’t really matter if it’s a bit oversized.  I think this is why SpaceX ended up bypassing Falcon 1 and Falcon 5, and went to Falcon 9.   There is still Delta II, but I think if you are looking for a “magic bulluet” LV, I’d probably start with that happy medium of about 10mt.

On the upper end, how big do we need to go?  Well, I think if you had a cheap, 30-40mt launcher, you are good for any payload but a NASA HSF payload for the foreseeable future.  And having a wide payload diameter capability is probably almost more important than pure mass.  But, if you could get this LV system up over 100mt, then I think you’d be in good shape for some wild hypothetical of NASA cancelling SLS and coming to us for our dirt-cheap common commercial launcher.  (Although DCCCL is a bit unwieldy as a name.)

I think a 5m diameter core is a good diameter to have a 10mt to 100mt+ range.  Maybe 5.5m.  Build them someplace where they can be loaded onto a transport ship and shipped the Cape for launching.  The Atlantic or Gulf coasts are the most likely.    They can be shipped to VAFB through the Panama Canal when necessary for polar launches.

So I say focus on that.  I think one larger core is cheaper than 3 smaller cores, and fewer cores have better mass fractions that big clusters of cores.  So there can be a balance.  I don’t think the core itself costs much more or less because of size (well, within reason).  I don’t think a 3m core, a 4m core, or a 5m core will really cost much more or less to build once the tooling is in place.  Unless you plan to use existing tooling, you will make the tooling new for your LV regardless of diameter.

Then design the engine like the RD-170/180/190.  1-chamber, 2-chamber, or 4-chamber, depending on the need, with common turbo pumps.  That makes for a modular and flexible engine system too.
Make the MPS the same for all, just a matter of if you are installing 1, 2, or 4 chambers of the engine.  (Except the long core will need a different MPS, but I’ll get to that in a moment.)

Make the single chamber the equal to the RD-190 series (or AJ26-500), about 500 klbs.  A two chamber will be basically an AJ-1-E6 or RD-180 at 1Mlbs.  The 4-chamber version will be basically a 2Mlbs RD-171.

Your basic 5m, “short” core is going to be the rough equivalent of Atlas V, Delta IV, Antares, and Falcon 9.  About 10mt-ish to LEO.  Like I said, that seems to be a pretty useful workhorse size.  It’s probably be about ½ the height of an Atlas Phase 2 core about 15m tall?  I think Atlas Phase 2 is about 30mt tall, and Delta IV is about 40m tall…all in 15mt.  ULA has a short Atlas phase 2 concept with a single RD-180.  The “medium core” could be basically an Atlas Phase 2, but with a 4-chamber single engine rather than two 2-chamber RD-180’s.  The short and medium cores will have the same MPS, so it’s just a matter of installing 2 or 4 thrust chambers into the engine mount. 

The short core would probably be about 1/3 the height of Delta IV, it’s mathalox, and it’d have a common bulkhead.  The “medium” core would be about 2/3 the height of the Delta IV, and the “tall” core would be about the same height as Delta IV or a little taller.

However, the “Tall” core will need two of the quad-chamber engines.  It will need a new MPS with two engines mounts.  It would look very similar to the Dynetics booster, but with two quad-chamber engines rather than two single chamber engines like the F-1’s.   However, a tri-core medium core will be about the equal to Atlas Phase 2-Heavy, or around 70mt.  Only if NASA chooses this family as their HSF launcher would you ever need the “tall” core with two engine MPS.  You plan for that potential upgrade, but defer the actual development of that MPS until such a time there was a customer for it.

The upper stage can have two lengths as well.  A shorter for the smaller variants, and a stretched version for larger variants.  They are 5m methalox too, using all the same tooling and common bulkhead as the boosters.  Like SpaceX’s Merlin’s, the upper stage is powered by the single chamber version of this engine, optimized for vacuum running with a nozzle extension, etc.  It is a 500klbs engine, too powerful for the smaller LV configurations.  So you design it to be able to throttle down to around 20%.  The RD-0124 can throttle down to 30%, so I think 20% is doable.  That gives you a 100klbs engine, similar to the Merlin 1-vacuum.  Maybe you could have a smaller turbopump so the engine would be “permanently throttled down” when used on the smaller variants.  But the engine is just a variant of the booster engine.
When used for larger variants, then use the full 500klbs version.  If there was ever a need for more 2nd stage power than that, then you have a vacuum version of the 2-chamber variant, for 1Mlbs.  That should be all you never need on the upper stage.

So I envision it looking something like this:

10mt to LEO with the short core/1-chamber engine.

25-30 mt to LEO with the medium core/2-chamber engine. (Atlas Phase 2 basically, or Zenit)

35-ish mt to LEO with a long core/4-chamber engine.  The Dynetics booster basically.  Only necessary if contracted for NASA, or some quadzillionaire that wants to go to Mars or something. Otherwise, variants of the short and medium cores will take care of all the commercial and USAF payloads.

20-70mt to LEO, the “medium” core with flexible options of “short” or “medium” outboard boosters.  It can even launch with just one outboard booster.  No reason why not, and the engine gimbals should handle the off-line center of mass just fine.

For NASA, the tri-core “tall” version would have 12Mlbs of thrust, and should get well over 100mt to LEO.  Probably more like 150mt.

No variant has more than two boosters.  Just various core and booster lengths and engine chamber configurations.  This keeps the pad nice and simple, it only needs three ports, and would look like the Delta IV pad. With similar Mobile Service Structure.  Just make the MSS tall enough, and the flame trench strong enough, to handle up to the largest config, right from the beginning.

The single core short and medium version will cover 10-20mt and will cover that vast majority of USAF/DoD and commercial payloads, and those have the best mass fraction.  Add one outboard booster if you need a little more.  And two outboard boosters if you need more yet.  Make them the medium CCB’s for even more power.  Even a big NASA booster could have a “Tall” core with two “medium” boosters, etc.  They will be bottom lift.  MPS to MPS connected, so the upper attach point doesn’t have to carry the load like SLS.  So different booster core lengths could be mixed and matched for core and boosters.

Lastly for the engines,  GG vs. staged combustion?  Don’t really know there, but, if Aeojet thinks they can be competitive with AJ26, AJ26-500, and AJ26-1E6, and the Russians can make RD-180 pretty cost effectively (yea, their labor costs are low, but still) , I think you could do this engine in staged combustion, and just makes lots of them to drive the costs down.  ...With GG, the upper stage needs to work harder because of the lower efficiency.  So I think you can do this as a staged combustion methalox common platform engine.

For now, I will call this LV family the “Lobo-tomoizer” and the engine the “Hog”…as in Harley Hog.  Fast and Loud.

I now open it up to the slings and arrows of everyone on this thread who’ll probably point out reasons it won’t work that I don’t see.  (of course, that’s the fun of it all, eh?)

[Hyperion5]

Since Hyperion pointed me to this thread and ask for my normally incredibly verbose opinion, I'll share a thought here:

First, I don’t know a ton about the pros/cons of methalox vs. kerolox.  I do know that they seem to have pretty similar performance, and are relatively close in density/volume.
But by switching from kerolox to methalox, you’ve traded up one of the most easily handled, transported, and stored fuel’s in RP-1 for a mild cryogenic that’s similar to LOX, with those associated difficulties.

But it seems methalox upper stags can get better isp than kerolox.  Raptor is supposed to be about 380s, while RD-0124 (one of the most efficient kerolox upper stages I think) is about 359s.  Hyperion has pointed out some metallurgical advantages of methalox, and seems to like it, so we’ll pull on that thread some.

From our discussion about Titan and Saturn 1 over on the Apollo not being cancelled thread, it seems that core lengths are very easy, but core diameter changes are much harder.  So I think we should pick one common core diameter that can do it all, and then have different core lengths for different options and applications.

There are customers for payloads in the less than 10mt range.  But I think that F9, Antares, Delta IV, Atlas V, etc all have a similar capacity of 10mt.  I think that’s a good place to have your basic building block.  If the LV is cheap, then it doesn’t really matter if it’s a bit oversized.  I think this is why SpaceX ended up bypassing Falcon 1 and Falcon 5, and went to Falcon 9.   There is still Delta II, but I think if you are looking for a “magic bulluet” LV, I’d probably start with that happy medium of about 10mt.

On the upper end, how big do we need to go?  Well, I think if you had a cheap, 30-40mt launcher, you are good for any payload but a NASA HSF payload for the foreseeable future.  And having a wide payload diameter capability is probably almost more important than pure mass.  But, if you could get this LV system up over 100mt, then I think you’d be in good shape for some wild hypothetical of NASA cancelling SLS and coming to us for our dirt-cheap common commercial launcher.  (Although DCCCL is a bit unwieldy as a name.)

I think a 5m diameter core is a good diameter to have a 10mt to 100mt+ range.  Maybe 5.5m.  Build them someplace where they can be loaded onto a transport ship and shipped the Cape for launching.  The Atlantic or Gulf coasts are the most likely.    They can be shipped to VAFB through the Panama Canal when necessary for polar launches.

So I say focus on that.  I think one larger core is cheaper than 3 smaller cores, and fewer cores have better mass fractions that big clusters of cores.  So there can be a balance.  I don’t think the core itself costs much more or less because of size (well, within reason).  I don’t think a 3m core, a 4m core, or a 5m core will really cost much more or less to build once the tooling is in place.  Unless you plan to use existing tooling, you will make the tooling new for your LV regardless of diameter.

Then design the engine like the RD-170/180/190.  1-chamber, 2-chamber, or 4-chamber, depending on the need, with common turbo pumps.  That makes for a modular and flexible engine system too.
Make the MPS the same for all, just a matter of if you are installing 1, 2, or 4 chambers of the engine.  (Except the long core will need a different MPS, but I’ll get to that in a moment.)

Make the single chamber the equal to the RD-190 series (or AJ26-500), about 500 klbs.  A two chamber will be basically an AJ-1-E6 or RD-180 at 1Mlbs.  The 4-chamber version will be basically a 2Mlbs RD-171.

Your basic 5m, “short” core is going to be the rough equivalent of Atlas V, Delta IV, Antares, and Falcon 9.  About 10mt-ish to LEO.  Like I said, that seems to be a pretty useful workhorse size.  It’s probably be about ½ the height of an Atlas Phase 2 core about 15m tall?  I think Atlas Phase 2 is about 30mt tall, and Delta IV is about 40m tall…all in 15mt.  ULA has a short Atlas phase 2 concept with a single RD-180.  The “medium core” could be basically an Atlas Phase 2, but with a 4-chamber single engine rather than two 2-chamber RD-180’s.  The short and medium cores will have the same MPS, so it’s just a matter of installing 2 or 4 thrust chambers into the engine mount. 

The short core would probably be about 1/3 the height of Delta IV, it’s mathalox, and it’d have a common bulkhead.  The “medium” core would be about 2/3 the height of the Delta IV, and the “tall” core would be about the same height as Delta IV or a little taller.

However, the “Tall” core will need two of the quad-chamber engines.  It will need a new MPS with two engines mounts.  It would look very similar to the Dynetics booster, but with two quad-chamber engines rather than two single chamber engines like the F-1’s.   However, a tri-core medium core will be about the equal to Atlas Phase 2-Heavy, or around 70mt.  Only if NASA chooses this family as their HSF launcher would you ever need the “tall” core with two engine MPS.  You plan for that potential upgrade, but defer the actual development of that MPS until such a time there was a customer for it.

The upper stage can have two lengths as well.  A shorter for the smaller variants, and a stretched version for larger variants.  They are 5m methalox too, using all the same tooling and common bulkhead as the boosters.  Like SpaceX’s Merlin’s, the upper stage is powered by the single chamber version of this engine, optimized for vacuum running with a nozzle extension, etc.  It is a 500klbs engine, too powerful for the smaller LV configurations.  So you design it to be able to throttle down to around 20%.  The RD-0124 can throttle down to 30%, so I think 20% is doable.  That gives you a 100klbs engine, similar to the Merlin 1-vacuum.  Maybe you could have a smaller turbopump so the engine would be “permanently throttled down” when used on the smaller variants.  But the engine is just a variant of the booster engine.
When used for larger variants, then use the full 500klbs version.  If there was ever a need for more 2nd stage power than that, then you have a vacuum version of the 2-chamber variant, for 1Mlbs.  That should be all you never need on the upper stage.

Someone can correct me if I'm wrong, but I can't recall an US engine being detuned like this to work if it was originally the booster engine.  It might save you a huge amount of expense, but might it not be cheaper just to go with the Spacex approach to one engine type on a rocket?  I do love the idea of a modular engine family like that of the RD-170/180/191 family from NPO Energomash. 

I do think the performance compromises (t/w ratio on the engine, possibly lower Isp due to throttling) might come to bite this plan of yours in the foot.  To me this sounds like a situation where we could really use the advice of a propulsion engineer to separate practical from unattainable. 

[Lobo]

Someone can correct me if I'm wrong, but I can't recall an US engine being detuned like this to work if it was originally the booster engine.  It might save you a huge amount of expense, but might it not be cheaper just to go with the Spacex approach to one engine type on a rocket?  I do love the idea of a modular engine family like that of the RD-170/180/191 family from NPO Energomash. 

I do think the performance compromises (t/w ratio on the engine, possibly lower Isp due to throttling) might come to bite this plan of yours in the foot.  To me this sounds like a situation where we could really use the advice of a propulsion engineer to separate practical from unattainable. 

I'm no rocket scientist, so I have no idea if an engine can be detuned like that.  Just throwing some ideas out there.  Maybe the larger combustion chamber would not work properly if there's too little propellant being pushed through it.  I only base it on the fact that most of the RD family can be throttled down to about 30%.  Other engines even moreso like the CECE.
if you can run 10% or 30% of propellent through a larger engine, then I don't see why you couldn't just put a smaller turbo pump on one and "detune" it down, so that the same basic engine came be used for commonality.
As far as a booster engine being used as an upper stage...Merlin 1 and NK33/43 are booster engines used as upper stage engines.

I'm guess it would work, but maybe it wouldn't work well, and would be better to go with a more specific built engine.  *shrug*

[Pipcard]

Is "fully reusable super-heavy two-stage single-core-only methalox rocket with common engines" now considered to be the ultimate commercial design?

[MATTBLAK]

1x 'family,' but 2x versions of a fully reusable, two-stage LOX/RP1 or LOX/Methane booster capable of placing 20 and up to 50-to-60 metric tons into high-inclination, low Earth orbits. Also able to deliver the largest Communications 'birds' to Geostationary orbits. The 50+plus ton rocket would also have the ability to use a large payload fairing for manned Exploration-class spacecraft and propulsion modules.

I guess I'm describing close analogues to Falcon 9 and a 'single stick' equivalent to Falcon Heavy. Actual turnaround and rational reflight rate to be advised by smarter folk than me. BUT: I could easily imagine an 'affordable', high flight-rate booster being used to assemble a manned Mars expedition pretty quickly, launching from two or three pads before a Mars launch window opened. This is not intended as an SLS slam - but such a booster would be an SLS-killer...

[su27k]

Is "fully reusable super-heavy two-stage single-core-only methalox rocket with common engines" now considered to be the ultimate commercial design?

Two leading commercial space companies are both pursuing designs like this, so I guess the answer is Yes.

[dror]

Is "fully reusable super-heavy two-stage single-core-only methalox rocket with common engines" now considered to be the ultimate commercial design?

Its the best design with current technology but I don't think it's ultimate.
If price goes near the price of fuel than ultimately you need to deviate from the rocket equation.
So I think ultimately somthing like a methalox hypersonic jet for first stage will allow the same payload to orbit with much less fuel and shorter turnaround time.

[AncientU]

Is "fully reusable super-heavy two-stage single-core-only methalox rocket with common engines" now considered to be the ultimate commercial design?

Without the 'super-heavy' -- yes IMO.
The market for launching unique design/orbit <10t sats will never go away, so highly efficient smaller launchers (probably as you describe configuration above) will have a place in the space economy for foreseeable future.

[Katana]

Is "fully reusable super-heavy two-stage single-core-only methalox rocket with common engines" now considered to be the ultimate commercial design?

Without the 'super-heavy' -- yes IMO.
The market for launching unique design/orbit <10t sats will never go away, so highly efficient smaller launchers (probably as you describe configuration above) will have a place in the space economy for foreseeable future.

But wouldn't they become obsolete or unneeded like Falcon 1 did? What about the launch industry shifting to a container ship-like business model? ("generic launch" as Space Ghost 1962 calls it, in which it is okay to launch a 1-tonne satellite on a full RLV for 100 tonnes)

Or:
Low cost expendable mini electric upperstage, one per one satellite when launching multiple satellites.

Allows relocating of different satellites to different orbits when using one large launcher.

Elrctric propulsion may have more develop potential than chemical.

[Pipcard]

(That post mentioning "generic launch" was deleted due to the keywords "unique design/orbit")