Reusable Single Stage to Orbit Concept

[aero]

Since you can produce and launch 1's all day long with a much higher chance each 1 will get to orbit, then even if your payload to orbit is less than half that of a TSTO, two or more SSTO's will still put the same amount of gross payload in orbit for less cost per kg.

You guys tickle me to death. You write as though payload has infinite divisibility and one-half plus one-half is always one. Its especially comical when you compare the efficiency of projected future  heavy lift boosters to current vehicles. If you're launching fuel depots, then sure, you can count on a full load, but if you're launching a geosynchronous communications satellite, it masses what it masses, and using a big, efficient heavy lift rocket isn't necessarily the cheapest way to go. I guess my point is illustrated by the above quote, simply by noting that the TSTO can just barely orbit the astronaut, but it is easy to get him to orbit with two launches.  Only who will decide which half to launch first? Or is it, "Hold your breath and don't worry, your life support is already in orbit."

When talking launch vehicles, you can not ignore the mission.

[KelvinZero]

Could a single stage to orbit really be cost effective if the mission was a fuel depot? That would be fantastic.

[mlorrey]

Actually, Titan II was a valid 1.5STO with a substantial payload fraction. It dropped two of its three 1st stage motors after 50% of fuel was consumed.

Titan-II had two engines on its 1st stage and it did not drop them. Did you mean "Atlas"?

Ah yes, sorry I misspoke...

[Jim Davis]

Mike, thanks for the numbers.

Falcon 1e first stage:
Dry mass 5680 kg
Fuel mass 87000kg
payload mass 500 kg

The Falcon planners guide gives those numbers in pounds although that shouldn't affect your argument since we're concerned with ratios.

I do note that you seem to be suggesting that the added mass of the ejector ram jet assembly will be negligible. That is completely unrealistic. But let's assume for sake of argument that it can be done.

Ram ejector average Isp: 1200 sec (mach 0.5-mach 8.5)

This is also completely unrealistic. A hydrocarbon fueled ramjet could only make 1200 s over a small part of that speed range. Even that assumes very good inlet efficiencies which are hardly likely to be obtained from being added to an existing rocket stage that made no such provision for it. But let's assume for sake of argument that it can be done.

The degree the ram ejector boosts total average isp is related to the thrust of the ejector vs the thrust of the Merlin itself.

Indeed. Let us assume that half the thrust comes from the ejector ramjet which would be quite generous. The effective Isp while the ejector ramjet is operating becomes:

Ieff = 1/(0.5/304 + 0.5/1200) = 485 s

For any combined average flight Isp of 375 seconds or greater:
delta-v 9938 m/s

That's probably a generous effective delta v for a vehicle which has to spend so much time in the lower atmosphere to provide air for the ejector ramjet. But let's assume for sake of argument that it can be done.

Now let's see what the effective Isp is when the ejector ramjet is operating over 1/3 of the delta V range.

Ieff = 3/(1/485 +2/304) = 347 s

So even with generous assumptions you don't reach orbit.

You should be able to reduce the dry mass a bit since you don't need the same amount of structural support that the TSTO needs to support the mass of the second stage.

But on the other hand you're dealing with far greater aerodynamic loads that you normally would be and incomparably greater aerodynamic heating loads. I wouldn't count on reducing dry mass.

There are reasons why airbreathing schemes have yet to make it to the launch pad (or runway).

[mlorrey]

Since you can produce and launch 1's all day long with a much higher chance each 1 will get to orbit, then even if your payload to orbit is less than half that of a TSTO, two or more SSTO's will still put the same amount of gross payload in orbit for less cost per kg.

You guys tickle me to death. You write as though payload has infinite divisibility and one-half plus one-half is always one. Its especially comical when you compare the efficiency of projected future  heavy lift boosters to current vehicles. If you're launching fuel depots, then sure, you can count on a full load, but if you're launching a geosynchronous communications satellite, it masses what it masses, and using a big, efficient heavy lift rocket isn't necessarily the cheapest way to go. I guess my point is illustrated by the above quote, simply by noting that the TSTO can just barely orbit the astronaut, but it is easy to get him to orbit with two launches.  Only who will decide which half to launch first? Or is it, "Hold your breath and don't worry, your life support is already in orbit."

When talking launch vehicles, you can not ignore the mission.

Of course, I never said you didn't have to. Didn't John Glenn get into orbit with his life support on one launch of an Atlas 1.5STO? Of course he did.

Now lets examine the Falcon 9 first stage as a potential expendable SSTO with a ram air ejector cowling.

Gross mass: 270 tonnes
Dry mass: 14.73 tonnes
payload: 5 tonnes
Main engine Isp: 304 sec
Ram air ejector Isp: 1200 sec
For any flight average Isp above 375 seconds, delta-v is 9621 m/s, with dV increasing as avg Isp rises.

Furthermore, should the first stage thrust assembly be reconfigured to allow the outer four engines to detach and be recovered with the ram ejector cowling, then payload mass to orbit rises on the order of 4-5 tonnes, nearly the full payload of the TSTO Falcon 9 rocket. Certainly enough to put a Dragon capsule into orbit.

[RanulfC]

Mike Lorry:

Jim Davis had/has a point: The ram air ejector cowling would only generate the given ISP boost for a very narrow amount of time with most operation being off-design and FAR less effective for the majority of it's operation.

It would "seem" to be a better option to simply design the to stage half the engines later in the flight to increase payload capacity.

Having noted that, Jim Davis:
You're assuming a greater heating and aerodynamic load for an ejector-ramjet cowl because it is an "air-breather"? Because of a more depressed trajectory?

This usually isn't the case with ejector-cowlings. While they impose some design loads the trajectory isn't effected because they are assumed to be used only for a short time at relativly low altitude, usually below 80,000ft. (At least in the majority of studies I've seen)
Like the boost-assist SRMs used on many expendable launchers today the ejector cowling would boost the ISP and thrust of a launch vehicle when it's deep in the atmosphere and moving relativly slowly and would be staged relativly soon after launch.

Randy

[Jim Davis]

Having noted that, Jim Davis:
You're assuming a greater heating and aerodynamic load for an ejector-ramjet cowl because it is an "air-breather"? Because of a more depressed trajectory?

Mike's scenario involved the ejector ramjet to provide substantial thrust augmentation at high Isp over the velocity range of M = 0.5 to 8.5. The dynamic pressures necessary would require a more depressed trajectory.

So the answer is yes.

Like the boost-assist SRMs used on many expendable launchers today the ejector cowling would boost the ISP and thrust of a launch vehicle when it's deep in the atmosphere and moving relativly slowly and would be staged relativly soon after launch.

Mike was very much more ambitious than that.

[RanulfC]

Jim Davis wrote:
Mike was very much more ambitious than that.

Which is why we both questioned his assumptions

I'm not making QUITE that much of a leap, only correcting for the information that "I" know

I've noted that there are quite a number of "assumptions" on the operating parameters for air-breathing propulsion as both Launch Assist and Orbital Launch that are based on contextual and situational criteria that don't actually apply in every single case.

Just an FYI.

Randy

[mlorrey]

Having noted that, Jim Davis:
You're assuming a greater heating and aerodynamic load for an ejector-ramjet cowl because it is an "air-breather"? Because of a more depressed trajectory?

Mike's scenario involved the ejector ramjet to provide substantial thrust augmentation at high Isp over the velocity range of M = 0.5 to 8.5. The dynamic pressures necessary would require a more depressed trajectory.

So the answer is yes.

Like the boost-assist SRMs used on many expendable launchers today the ejector cowling would boost the ISP and thrust of a launch vehicle when it's deep in the atmosphere and moving relativly slowly and would be staged relativly soon after launch.

Mike was very much more ambitious than that.

The Falcon 1 first stage already separates from its second stage at mach 8.5 at 130,000 ft. An F1 based SSTO would accelerate faster than this (lower GLOW and higher thrust) and thus achieve this speed at a lower altitude in the same trajectory, so I don't see Jims flatter trajectory claim as valid. Furthermore, given the lack of second stage mass, the structure of the first stage would be capable of handling a significantly higher max Q pressure.

[Jim Davis]

An F1 based SSTO would accelerate faster than this (lower GLOW and higher thrust) and thus achieve this speed at a lower altitude in the same trajectory, so I don't see Jims flatter trajectory claim as valid.

If it's accelerating faster it will not be following the same trajectory. It still needs a certain vertical and a certain horizontal velocity component to achieve orbit. If it is accelerating faster it will gain the vertical component faster and will have to make its turn to the horizontal sooner, i.e. at a lower altitude.

But even setting this aside your scheme requires much thrust augmentation from the ejector ramjet from M=0.5 to M=8.5. This requires following a much higher dynamic pressure profile, i.e. lower profile.

Furthermore, given the lack of second stage mass, the structure of the first stage would be capable of handling a significantly higher max Q pressure.

Not at all obvious. True, it doesn't have the second stage mass but it will still have the propellant mass needed for ascent to orbit. But even setting that aside there is the much higher thermal loads to consider.

There are reasons why these ideas, which have been around a very long time, never make it to the launch pad.

[rusty]

(reconfigured for clarity)
I dont know how many times I have seen this discussed only to fall due to the same basic problems.
Let me lay out the things that any SSTO, let alone RESUSABLE SSTO designers will need to overcome in order to make this idea reasonable, let alone viable:

1. COST COST COST: If its not cost effective, nay, if its not COMMERCIALLY economic (i.e. as in for a commercial launch provider) then it won't work. Note: Don't expect government funding for this. You might get it (DOD side at least), but might isnt enough. Design it to be commercially feasible.
6. KG/$ to LEO economics: How much can it lift? Can it compete?

2. Reusability: Determine a low cost and effective system to protect the stage during rentry such that refurbishing for reuse is quick and easy.
3. Retriveal: Where are you going to land it and what are the consequences of landing there?
5. Saftey: Make it safe

4. Feasibility: Is Reusability really worth it? Or is it too expensive??

A very good list - needless to say that at Reaction Engines we believe that SKYLON ticks all those boxes.

There is a lot to like about SKYLON, but wouldn't a smaller HTHL vehicle be more robust and economically viable? A payload of 1,500-2,500kg is all that's needed for small payloads, experiments or crew rotation.

[93143]

wouldn't a smaller HTHL vehicle be more robust and economically viable? A payload of 1,500-2,500kg is all that's needed for small payloads, experiments or crew rotation.

This has been answered already.  A smaller vehicle is more developmentally risky, not less, because the required mass fraction is harder to get (which also leaves the final product with smaller margins, if it works at all).  Also, there are issues with scaling down the SABRE engine, which is what makes the whole thing work.  So no, it wouldn't be more robust.

The development cost doesn't decrease in direct proportion to the size, and neither do the manufacturing or operational costs, so the value-for-money is less, and a smaller vehicle would be shut out of markets for hardware above its capacity, meaning less business to offset the considerable development costs.  So it wouldn't be more economically viable either.

There's a reason the design iterations are actually making the vehicle bigger...

[mlorrey]

An F1 based SSTO would accelerate faster than this (lower GLOW and higher thrust) and thus achieve this speed at a lower altitude in the same trajectory, so I don't see Jims flatter trajectory claim as valid.

If it's accelerating faster it will not be following the same trajectory. It still needs a certain vertical and a certain horizontal velocity component to achieve orbit. If it is accelerating faster it will gain the vertical component faster and will have to make its turn to the horizontal sooner, i.e. at a lower altitude.

But even setting this aside your scheme requires much thrust augmentation from the ejector ramjet from M=0.5 to M=8.5. This requires following a much higher dynamic pressure profile, i.e. lower profile.

Again, you don't seem to be reading anything I am saying. 130,000 ft altitude is perfectly valid flight envelope for ram/scram air combustion at mach 8.5, therefore, since the Falcon 1 first stage separates at that altitude and speed, it already follows a proper trajectory for optimum use of a ram ejector. There is no need for a lower profile.

Furthermore, given the lack of second stage mass, the structure of the first stage would be capable of handling a significantly higher max Q pressure.

Not at all obvious. True, it doesn't have the second stage mass but it will still have the propellant mass needed for ascent to orbit. But even setting that aside there is the much higher thermal loads to consider.

Not at all, since we are saving so much mass by combusting atmospheric oxygen, all you have is the orbital ascent propellant mass and that mass is sitting at the BOTTOM of the propellant tanks, not above them in the structure as the second stage would.

Secondly, as I have previously shown that the current flight profile is perfectly valid for optimum use of the ram ejector, there will not be much higher thermal loads to consider.

There are reasons why these ideas, which have been around a very long time, never make it to the launch pad.

Actually, the GNOM concept was tested on a scale prototype and proved such a large performance increase that it could throw the same payload as a US missile 50% more massive.  That was the ONLY time that this concept has been actually tested, the US has never tested the idea.

[aero]

For those of us not familiar with the GNOM project, here is an informative link:http://www.astronautix.com/lvs/gnom.htm
It seems there is a lot to be gained by air augmentation.

[mmeijeri]

The development cost doesn't decrease in direct proportion to the size, and neither do the manufacturing or operational costs, so the value-for-money is less, and a smaller vehicle would be shut out of markets for hardware above its capacity, meaning less business to offset the considerable development costs.  So it wouldn't be more economically viable either.

Surely this is not a general principle? You are not suggesting that the participants in the Northrop Grumman Lunar Lander Challenge could just as easily have developed vehicles five times their current sizes? Or that developing a small business jet takes about as much money as a 787?

[93143]

The development cost doesn't decrease in direct proportion to the size, and neither do the manufacturing or operational costs, so the value-for-money is less, and a smaller vehicle would be shut out of markets for hardware above its capacity, meaning less business to offset the considerable development costs.  So it wouldn't be more economically viable either.

Surely this is not a general principle?

No, it applies specifically when considering SSTO.  Very specifically when considering Skylon, but I believe it applies to every SSTO concept that actually has a chance of working.

You are not suggesting that the participants in the Northrop Grumman Lunar Lander Challenge could just as easily have developed vehicles five times their current sizes?  Or that developing a small business jet takes about as much money as a 787?

Straw man.  I didn't say development cost was unrelated to size, just that the relationship isn't 1:1 linear.  Pay attention.

[mmeijeri]

No, it applies specifically when considering SSTO.  Very specifically when considering Skylon, but I believe it applies to every SSTO concept that actually has a chance of working, if the goal is to get costs down.

Can you say more? I can see that not being able to lift current commercial payloads doesn't help, but if you get incremental costs low enough you can tap new markets. It's a multidimensional thing.

Straw man.  I didn't say development cost was unrelated to size, just that the relationship isn't 1:1 linear.  Pay attention.

I didn't say anything about linear. You appeared to me to be implying there wasn't a strong monotonicity, which struck me as odd.

[93143]

No, it applies specifically when considering SSTO.  Very specifically when considering Skylon, but I believe it applies to every SSTO concept that actually has a chance of working, if the goal is to get costs down.

Can you say more? I can see that not being able to lift current commercial payloads doesn't help, but if you get incremental costs low enough you can tap new markets. It's a multidimensional thing.

An SSTO needs to be very careful with its mass fraction.  Basic geometry and physics says it's easier to achieve a good mass fraction with a larger vehicle.  There's probably a minimum size below which SSTO isn't feasible at all.  Skylon doesn't need as aggressive a mass fraction as an all-rocket vehicle would, but it has extra difficulties because the engines can't be scaled down easily.

Reaction Engines seems to have done the analysis, and apparently the prospects for new markets in the low-mass payload range aren't good enough to justify passing up the big-ticket satellite market and trying to shrink their vehicle enough to be able to take smallsats on dedicated flights.

As it is, there's no reason you couldn't piggyback smallsats on a not-quite-maxed Skylon launch - the lab next door to mine launches their stuff that way all the time.  Granted, Canada seems to be getting tired of doing everything this way; hence the CSLV project...

Straw man.  I didn't say development cost was unrelated to size, just that the relationship isn't 1:1 linear.  Pay attention.

I didn't say anything about linear. You appeared to me to be implying there wasn't a strong monotonicity, which struck me as odd.

It's cheaper to develop a small vehicle, but not by as much as the reduction in size.  This is a general principle, and doesn't take into account the additional difficulties in scaling down an SSTO.  Cut the size in half, and you might cut development costs by 40%.  Go to 1/10, and you might save 80%.  It's similar for manufacturing and operations - you save, but not enough to make up for the loss of capacity, so cost per kg goes up.

[mmeijeri]

An SSTO needs to be very careful with its mass fraction.  Basic geometry and physics says it's easier to achieve a good mass fraction with a larger vehicle.

Do you mean because of cube-square effects?

There's probably a minimum size below which SSTO isn't feasible at all.  Skylon doesn't need as aggressive a mass fraction as an all-rocket vehicle would, but it has extra difficulties because the engines can't be scaled down easily.

What is it about the engines that precludes scaling them down easily? I thought that engines typically scaled up badly, twice as large being more than twice as difficult.

Reaction Engines seems to have done the analysis, and apparently the prospects for new markets in the low-mass payload range aren't good enough to justify passing up the big-ticket satellite market and trying to shrink their vehicle enough to be able to take smallsats on dedicated flights.

That may be true, but I was thinking of manned suborbital hops. More than one way to skin a cat and many more plausible ways that won't work out - if only we knew in advance which was which.

It's cheaper to develop a small vehicle, but not by as much as the reduction in size.  This is a general principle, and doesn't take into account the additional difficulties in scaling down an SSTO.  Cut the size in half, and you might cut development costs by 40%.  Go to 1/10, and you might save 80%.  It's similar for manufacturing and operations - you save, but not enough to make up for the loss of capacity, so cost per kg goes up.

I already mentioned engines which I believe contradict this rule. But I can believe there may be systems for which this is true. Even so cost/kg isn't the only variable. ROI demands by investors may be a tougher constraint to live with. And you don't have to achieve a reduction in price by an order of magnitude all at once, merely enough to build up enough market share. I think the suborbital RLV people are being very sensible in how they go about their work which they ultimately want to lead to orbit.

[Jim Davis]

Again, you don't seem to be reading anything I am saying. 130,000 ft altitude is perfectly valid flight envelope for ram/scram air combustion at mach 8.5, therefore, since the Falcon 1 first stage separates at that altitude and speed, it already follows a proper trajectory for optimum use of a ram ejector. There is no need for a lower profile.

You don't get it, Mike. Mach 8.5 at 130,000 ft equates to a dynamic pressure of about 350 psf. Yes, a ramjet can operate at this dynamic pressure but in order to boost the performance of the vehicle by a significant amount it has to develop thrust at least comparable to the existing rocket engine. This is a very demanding requirement and demands operation at much higher dynamic pressures, typically between 1000 and 2000 psf.

Not at all, since we are saving so much mass by combusting atmospheric oxygen, all you have is the orbital ascent propellant mass and that mass is sitting at the BOTTOM of the propellant tanks, not above them in the structure as the second stage would.

Mike, your notional vehicle is going to be following a high dynamic pressure profile from M=0.5 to M=8.5. The Falcon 1-e first stage has about ten times the mass of the second stage. Your notional vehicle will have a mass of about 39,000 lbs at M=8.5. The standard Falcon 1-e will have a mass of about 16,000 lbs. I do not think it immediately obvious that a twice as massive vehicle following a much higher dynamic pressure profile is going to be subjected to lower stresses. It is certainly going to be subjected to much higher thermal loads regardless of how the loads are distributed.

Secondly, as I have previously shown that the current flight profile is perfectly valid for optimum use of the ram ejector, there will not be much higher thermal loads to consider.

Sorry, Mike you're just not getting it. The ejector ramjet is useless if it doesn't contribute the bulk of the thrust. It can't do that while flying the same low dynamic pressure profile as a pure rocket. A very efficient ramjet which only contributes a minute amount of thrust will have negligible impact on vehicle performance. You might as well add a Hall thruster to the Falcon 1-e.

Actually, the GNOM concept was tested on a scale prototype and proved such a large performance increase that it could throw the same payload as a US missile 50% more massive.  That was the ONLY time that this concept has been actually tested, the US has never tested the idea.

But note that GNOM was a 3 stage concept - nowhere near the single stage performance that you claim.

You're claiming that a negligible mass ejector ramjet can built which will develop high thrust at low dynamic pressures and high specific impulse over a very large Mach number range. Sorry, you haven't made the case for any of this.