Reusable LOX/Kerosene SSTO with drop tanks

Antius · 2016-09-14 07:52:06

This topic is an extension of a discussion undertaken on the stratolaunch thread. I will flesh the concept out as time allows.

The concept is for a semi-reusable launch vehicle burning either LOX/Kerosene or N20/Propane. The former provides a better ISP but is a soft cryogen requiring stainless steel tanks. The latter can use high strength low alloy (HSLA) steels for all tanks and may fit the concept better. The vehicle is designed essentially as an SSTO, but with staging achieved by dropping empty propellant tanks instead of entire stages. On this basis, a vehicle equipped with a single engine will fly from ground level to orbit and the high value parts of the craft will be reusable. The vehicle will consist of a central core propellant tank surrounded by six strap-on propellant tanks. A single pressure fed engine with gimballing and a valve and plumbing module will be attached to the bottom of the central core module, means of attachment tbc. The strap-on tanks and central core will deliver propellant to the plumbing module via flexible lines, which will detach when the empty propellant tank is ejected.

The reusable part of the vehicle will be bullet shaped and will be mounted on top of the central core tank. It will contain an N2O based manoeuvring system, navigation system, communications array and payload bay, basically everything high tech and expensive except the engine. During launch, four of the outer strap-on tanks will provide the first stage burn propellant, ~3000m/s dV and will eject when propellant is exhausted. The other two will provide the second stage, and the central core provides the third stage. The first stage tanks will be pressurised to ~28bar and chamber pressure at take-off will be ~20bar (300psi). Second and third stage tanks will only provide propellant outside of the Earth’s atmosphere and are expected to be charged at a lower pressure, perhaps 20bar, supporting a chamber pressure of perhaps of 12bar.

Upon reaching orbit, the reusable upper module will deploy its payload through shuttle-type payload doors. The engine at the base of the third-stage central core tank will be detached and stowed within the now empty payload bay for reuse. The engine nozzle is expected to be expendable and will be constructed from ablative lined HSLA. The combustion chamber and throat will be film cooled, using either bled propellant or water. Re-entry burn is achieved by small N2O based manoeuvring thrusters on the reusable module. A heat shield is mounted at the base of the bullet module and a parachute is housed within the top section. The reusable module is recovered via ocean splashdown.

The philosophy is based upon the assumption that tankage is relatively heavy, accounting for most of the dry weight of a launch vehicle, but also relatively cheap. HSLA steels cost about $0.5/kg and some have UTS upwards of 600MPa. They are generally considered to be easy to work with and weld. Propellant tanks constructed from these materials are therefore cheap enough to throw away and their purchase cost might add perhaps $10/kg to the delivery cost of payload to orbit. The expensive parts of the launch vehicle are its communication, guidance, manoeuvring and propulsion systems, all of which require precision engineering. These parts account for a small percentage of take-off mass but represent a high proportion of embedded value. They are therefore worth reusing. The tanks themselves are rather like expendable Coca-Cola cans; worth recycling, but not practicable to consider reusing.

By adopting a staging approach, the vehicle can make use of lower performance propellants and pressure-fed engines reducing cost and complexity by orders of magnitude. After each flight, the vehicle can be rapidly turned around by bolting a new heat shield to its base, replacing the engine nozzle and assembling the module and engine onto a new set of tanks.

Last edited by Antius (2016-09-14 07:56:40)

kbd512 · 2016-09-14 17:31:55

SSTO is mostly a pipe dream, but propellant tank staging should actually work.

GW has a pretty good article on why a LOX/RP-1 SSTO is infeasible:

SSTO Trade Studies

Kirk Sorensen has a pretty good article on why NERVA SSTO is infeasible:

SSTO is a bad idea, but NTR SSTO is worse

* LOX is dirt cheap and the complications from working with a mildly cryogenic oxidizer are reasonable.

* Why contend with the complications of using liquid propellants to begin with if we're just going to use low Isp propellants with heavy tanks?

* Reusability only eats through what little mass fraction is allocated for payload.

* Space and cargo management is required in the return vehicle for returning whatever parts of the rocket you want to keep

* Compression molded composite tanks could be a lot cheaper than any form of metal tank and can be manufactured in the span of minutes to hours

* May be possible to replace many, if not most metal parts in the rocket engines with composites

* If it's possible to manufacture most of the turbo machinery and pipes from composites and the completed product cost less than a million dollars, then the entire rocket is essentially trash and can be discarded as such

RobertDyck · 2016-09-14 18:51:47

kbd512 wrote:

Kirk Sorensen has a pretty good article on why NERVA SSTO is infeasible:
SSTO is a bad idea, but NTR SSTO is worse

Kirk Sorensen wrote:

Nuclear-thermal SSTO turns out to be one of the worst ideas anyone has ever come up with, for two simple reasons: hydrogen and the lousy thrust-to-weight ratio of nuclear thermal rockets. Those are the same two reasons that make NTR lousy or marginal for nearly any other space application as well, but this post will focus on the issues surrounding NTR SSTO.

Two simple points:

Timberwind has dramatically reduced reactor mass, resulting in dramatically improved thrust-to-weight ratio
distilled water propellant for ground launch

LH2 is low molecular weight resulting in high Isp, however liquid density is low resulting in large propellant tank. If your vehicle has to return to Earth then you have to protect the propellant tank with a heat shield. That increases vehicle mass. So for an SSTO use water instead. NTR with water has lower Isp than LH2, but still higher than the best LOX/LH2 chemical rocket.

Use LH2 for in-space propulsion. NASA developed NERVA as an upper stage to replace S-IVB, the 3rd stage of Saturn V. The S-IVB stage had 2 jobs: Earth orbit insertion, and TLI. The nuclear stage would not be able to do Earth orbit insertion, because NASA's safety plan was to place the rocket in stable orbit before starting the reactor. If launch failure occurred, nuclear fuel elements would not have any fission fragments, so could be collected by hand. I have suggested Saturn V could have done this with an S-IV stage (smaller than S-IVB) with the same diameter as S-IVB, then the NERVA stage stacked on top, creating a 4-stage rocket. NERVA would be used for Trans-Mars Injection (TMI). With it's extreme Isp, it was ideal for that job. NERVA 2 in 1974 had Isp=825 seconds, in 1991 their computer simulation had Isp=925s. Space Shuttle Main Engine (SSME) had Isp in vacuum of 453s, and NASA has a couple engines optimized for in-space use only with Isp=460s. In-space engines have a large exhaust bell, which can not be launched in air pressure. If you try to do so, the exhaust stream will separate from the engine bell part way down, resulting in overheating the engine bell. The exhaust nozzle extension would literally melt off.

kbd512 · 2016-09-14 22:54:28

RobertDyck wrote:

Two simple points:
Timberwind has dramatically reduced reactor mass, resulting in dramatically improved thrust-to-weight ratio
distilled water propellant for ground launch
LH2 is low molecular weight resulting in high Isp, however liquid density is low resulting in large propellant tank. If your vehicle has to return to Earth then you have to protect the propellant tank with a heat shield. That increases vehicle mass. So for an SSTO use water instead. NTR with water has lower Isp than LH2, but still higher than the best LOX/LH2 chemical rocket.

Agree on point #1. There are various reactor core configurations that can dramatically reduce the dimensions and mass of the core and thus the mass of the core containment and shielding. I personally like the twisted ribbon nuclear reactors that the Russians built. I bet we could improve upon the fuel elements they used and get that Isp back around 1000s.

Disagree on point #2. The Isp of H2O is considerably worse than the Isp of LOX/LH2. The Isp of a NTR producing a given amount of heat decreases as the mass of the molecule accelerated through the core increases. Google the "equipartition theorem".

RobertDyck · 2016-09-15 00:02:48

kbd512 wrote:

Disagree on point #2. The Isp of H2O is considerably worse than the Isp of LOX/LH2. The Isp of a NTR producing a given amount of heat decreases as the mass of the molecule accelerated through the core increases. Google the "equipartition theorem".

Did you take into account the fact exhaust gas is hotter than dissociation temperature for water molecules? Exhaust isn't water, it's mono-atomic oxygen and mono-atomic hydrogen.

Even exhaust from SSME was so hot that exhaust hovered on the edge of dissociation. If it did completely dissociate, that would consume energy. Energy from a chemical rocket comes from the chemical reaction, for a LOX/LH2 rocket it comes from oxidation of hydrogen. So dissociation consumes your energy source. But SSME exhaust had a lot of hydroxyl (OH) and mono-atomic hydrogen, which only combined to form water (steam) past the engine throat. I don't know what proportion of hydroxyl vs water, and how that proportion changed down the exhaust nozzle. But a nuclear thermal rocket operates beyond that.

elderflower · 2016-09-15 03:16:21

for the same thermal output, dissociation absorbs some of the available heat and restricts the temperature of the exhaust so reducing exhaust velocity.

RobertDyck · 2016-09-15 04:11:30

With nuclear, you just produce more energy. Limit is what temperature will melt engine components. The nuclear reactor has far more energy than required. Typically a reactor with minimum critical mass to get the reactor to operate will have enough nuclear fuel to run for years. But launch to orbit takes what? Saturn V took 12 minutes to Earth orbit. Ariene 5 takes 25 minutes to GTO. Falcon 9 Full Thrust first stage engine is rated for 162 seconds, second stage 397 seconds. NERVA used 99+% U-235 with neutron reflectors. Rods rotates: one side absorbed neutrons, the other reflected. So turn control rods to operate the reactor. This minimized uranium mass, but it still required a few years worth of continuous operation of uranium to achieve critical mass. One reason why Americium-242m was proposed. It doesn't require the greatest energy, just the smallest critical mass. But you still have more than enough energy in the reactor. Limit is propellant.

Antius · 2016-09-15 05:02:09

kbd512 wrote:

SSTO is mostly a pipe dream, but propellant tank staging should actually work.
GW has a pretty good article on why a LOX/RP-1 SSTO is infeasible:
SSTO Trade Studies
Kirk Sorensen has a pretty good article on why NERVA SSTO is infeasible:
SSTO is a bad idea, but NTR SSTO is worse
* LOX is dirt cheap and the complications from working with a mildly cryogenic oxidizer are reasonable.
* Why contend with the complications of using liquid propellants to begin with if we're just going to use low Isp propellants with heavy tanks?
* Reusability only eats through what little mass fraction is allocated for payload.
* Space and cargo management is required in the return vehicle for returning whatever parts of the rocket you want to keep
* Compression molded composite tanks could be a lot cheaper than any form of metal tank and can be manufactured in the span of minutes to hours
* May be possible to replace many, if not most metal parts in the rocket engines with composites
* If it's possible to manufacture most of the turbo machinery and pipes from composites and the completed product cost less than a million dollars, then the entire rocket is essentially trash and can be discarded as such

Many thanks. I have looked at the links you provided. GW's analysis suggests that an SSTO is a bad concept with any of the currently available propulsion technologies. The reasons are varied: (1) an expendable SSTO would have poor payload fraction, ruining its cost advantage; (2) a reusable SSTO would have even poorer payload fraction; (3) A reusable SSTO probably wouldn't be reusable many times, due to creep and fatigue in an airframe with inherently low design factors. For the same reasons, it wouldn't be very safe and would be difficult to human rate.

The concept I have suggested is basically a staged rocket, but dropping tanks instead of entire stages, on the admittedly weakly supported assumption that this would reduce operating cost for a reusable vehicle. But the whole idea of reusability as a means of reducing recurring costs may be wrong headed if it eats too much into payload fractions.

The bottom line is any intelligent discussion on reducing the costs of accessing space needs to start with an understanding of what those costs are and where they come from. It can then choose the right solution to fit the requirement. I for one do not have that knowledge. I have always suspected that Arthur Schmitt's minimum cost design ideas were fundamentally sensible, basically very simple expendable rockets, made from cheap materials and typically built to large sizes to spread out fixed costs. But again, I making hunches and guesses on cost drivers that I cannot substantiate.

kbd512 · 2016-09-15 08:21:21

RobertDyck wrote:

Did you take into account the fact exhaust gas is hotter than dissociation temperature for water molecules? Exhaust isn't water, it's mono-atomic oxygen and mono-atomic hydrogen.

What's the chamber pressure going to be like in this hypothetical bi-modal rocket? How big will the chamber, throat, and nozzle be? The fantastically high operating pressures of both the RS-25 and NERVA engines prevent dissociation from factoring into the performance characteristics of either design in any major way. If a NTR is optimized for H2O dissociation, then it won't be anywhere near optimized for H2 acceleration.

RobertDyck wrote:

Even exhaust from SSME was so hot that exhaust hovered on the edge of dissociation. If it did completely dissociate, that would consume energy. Energy from a chemical rocket comes from the chemical reaction, for a LOX/LH2 rocket it comes from oxidation of hydrogen. So dissociation consumes your energy source. But SSME exhaust had a lot of hydroxyl (OH) and mono-atomic hydrogen, which only combined to form water (steam) past the engine throat. I don't know what proportion of hydroxyl vs water, and how that proportion changed down the exhaust nozzle. But a nuclear thermal rocket operates beyond that.

The exhaust from the SSME is 3300C. The exhaust temperatures from a NERVA rocket reached 3000C, which is very near to the highest ever achieved in a solid core NTR. The exhaust temperatures from most models, to include the models studied during Project Timberwind, were around 2500C. There's a reason why LH2 is the propellant of choice for solid core NTR's.

We even have a thread on this here:

Steam nuclear rocket

Unfortunately, the bi-modal H2O / LH2 NTR just doesn't work the way I wish it did.

Edit (and re-edit):

Rob, your best bet for increasing the T/W ratio of a NTR is the LANTR concept. It's like an afterburner for LH2 fueled NTR's and works by combusting LOX with LH2 in the engine's nozzle. Thrust can be increased by between a factor of 4 and 5 and exhaust temps can reach 3600C, which is a little hotter than RS-25 exhaust. I can envision a two stage rocket using a LOX/RP-1 first stage or existing STS technology and a LOX/LH2 NTR second stage that substantially increases the payload fraction by providing high thrust in the lower atmosphere at reduced Isp and then providing higher Isp at reduced thrust in the upper atmosphere and into orbit.

Last edited by kbd512 (2016-09-15 09:26:02)

Antius · 2016-09-15 10:06:59

An interesting article here on the various cost drivers for launch cost and complication in working out what the real cost is.

http://www.astronautix.com/c/costpricea … thing.html

The article suggests that the lion’s share of the cost of present launches results from the fixed costs of ground facilities, engineering and maintenance facilities and crews and a bloated bureaucracy, spread over a ludicrously small number of launches. The actual design of the launch vehicle is actually less important when weighed against the sheer industrial effort required to support any launch programme. The high costs associated with the shuttle were largely due to NASA’s insistence on maintaining a bloated ground support structure with an enormous legacy of engineers and facilities, whose cost ran to $3billion per year. This huge cost was the same regardless of the number of launches carried out and was largely responsible for the huge cost of the vehicle.

This leads me to believe that whilst launch vehicle architecture is not unimportant, the real thing keeping space launch costs high is a critical lack of scale economies. We won’t see cheap access to space until launch volumes increase to the point where launch operations are able to achieve decent scale economies. This won’t happen unless an ambitious nationalistic government decides to devote a sizable portion of its GDP towards an all-out nationalistic space colonisation attempt. When European powers colonised America, a large portion of the GDP of the colonial nations was spent in this way. It will probably take a similar level of effort to achieve the same thing in the new frontier. It is difficult to imagine this in the present political climate.

Last edited by Antius (2016-09-15 10:08:23)

GW Johnson · 2016-09-15 11:39:55

Referring to the drop-tanks idea which retains the same engine throughout: there is an enormous difference in flying vehicle mass between liftoff and orbit insertion. If you keep the same engine, you have either an unnaceptable gee level at insertion, or an unattainable thrust turndown ratio for your engine. This is closer to factor 100 than it is 10.

Most of the liquid engines with turndown that I have ever heard of struggled to get turndowns larger than factor 2 or 3 or thereabouts. You can beat this with multiple engines, and shutting most of them down. But this isn't as lightweight as one single engine, because of the dplicated pump and control assemblies.

The original Atlas booster really was SSTO, although most fail to remember that today. Back then it had no Centaur upper stage, that came later, and is retained today in Atlas-5. Back then they cheated, they dropped two engines on the way up to "lighten ship" enough to reach orbit SSTO, and to achieve gee-limiting thrust turndown with old-time engines that had only small throttle capability.

This rig first started flying about 1955 as an ICBM pushing an early-generation nke warhead of around 5 tons. It could reach orbit pushing a 3-4 ton Mercury capsule, just barely. So they used it. Tough to man-rate, it was, too. Still not fully trustable when Glenn rode the thing the first time in 1962.

The delivered tonnage could not go up until (1) either went 2-stage, as in the Titan-II modified to push Gemini, or (2) add an upper stage, which was the early form of Centaur.

Also, it looks like Antius has found some corraboration of my longstanding contention that enormous logistical tails are the big expense in space launch. The commercial competition in the satellite launch business has forced simplifications, efficiencies, and reductions in those logistical support tails for the successful commercial rockets. That (and only that) is why launch prices have dropped so much the last 20 years.

I fear he may also be correct in tying willingness to commit GDP to space ventures to any practical likelihood of big missions. I see no sign of that either.

RobertDyck is entirely correct to point out that the only propellant ever used in a nuclear rocket was hydrogen. All of them: Phoebus, Kiwi, and NERVA.

Supposedly, the other higher molecular weight possibilities all offer reduced Isp due to the molecular weight. I realize there's a lot of dissociation at high chamber temperatures, but overall, not as much as you might think, or C-D nozzles would fail to work at all. The pressure drop-driven conversion of high internal energy to high kinetic energy does not work with recombination's energy release. So there cannot be that much dissociation, even at LOX-LH2 temperatures.

Most of the time, I saw atomic oxygen-induced corrosion of hot core materials cited as a reason water was unacceptable, more so than just Isp reduction from its molecular weight. They had some problems with hydrogen embrittlement in the steels, too, as I recall.

Most of the time in practical rocket work, I got the best model of real items by assuming shifting equilibrium flow from chamber to throat, and then frozen-composition flow from throat onward. The throat chamber (absolute) temperature is in the vicinity of 80-85% chamber temperature.

Myself, I believe the attractive nuclear thermal rocket is the gas core approach. Too bad it was never really worked on.

GW

Last edited by GW Johnson (2016-09-15 11:55:12)

kbd512 · 2016-09-15 14:18:35

Antius,

The moral of the story is:

* Competent personnel are expensive and maintaining a small army of competent people is even more expensive, so pick a product or service to specialize in and then stick to doing what you're most efficient at doing

* Reuse existing infrastructure wherever possible and refrain from build new launch or manufacturing facilities unless you absolutely have to

* Build the rockets as close to the launch facilities as is feasible to minimize transportation related costs

* Resist the urge to optimize spacecraft and launch vehicles to the nth degree, which is a major part of what necessitates facilities sprawl and small armies of engineers

* Any spacecraft development program that requires a new launch vehicle is a de-facto launch vehicle development program

The StratoLaunch II concept I proposed was a vehicle for servicing as many different requirements with a single asset as is practical, so that serial production of these aircraft would be possible, ensuring a service lifetime of at least 50 years.

US Military Requirements (Replacement for C-5, KC-10, KC-135, B-52):

* Tanker capable of ferrying a squadron of advanced tactical fighters to any continent on the globe in a day

* Tanker capable of refueling three squadrons of advanced tactical fighters in a theater of operations

* Heavy ground vehicle transport capable of delivering a US armored brigade in a day

* Airborne drone carrier for low intensity conflicts (X-36 mini drones; persistent close air support and ground forces airborne defense against helicopters, attack aircraft, and drones)

* Airborne weapons carrier (cruise missiles and guided bombs)

* Airborne communications, weather, and spy satellite launcher

NASA Requirements:

* Earth scientific observation satellite launcher

* Small commercial communications satellite and space technology experiments launcher

* ISS cargo resupply launcher

* ISS crew launcher

Civil Requirements:

* Heavy cargo transport enabling use of pallets that can be loaded away from the aircraft and then pinned into the fuselage for transport

* Commercial space satellite launcher

We're not going to get into space in any cost-effective manner until we admit what's so obvious. The kitchen sink spacecraft solutions require fantastically large and heavy launch vehicles that are so complicated and expensive as to be completely unaffordable, even for governments, to use on a regular basis. Ground launch vehicles should be for delivery of space stations and exploration class spacecraft only. The kind of ridiculously over-engineered taxi service that current spacecraft provide will be far more cost-effectively performed by small rockets designed for minimally sophisticated task, namely taking humans and small quantities of consumables to more sophisticated space stations and space exploration vehicles.

Until we can build a rocket in a day or two, we're never going to have affordable access to space. We basically have standing armies of people dedicated to production and launch of hand-made vehicles that are, for all intents and purposes, one-of-a-kind vehicles.

GW Johnson · 2016-09-15 16:59:51

Subsonic air launch of rockets to orbit doesn't really buy you a whole lot in terms of performance measures, but it does "buy" you some very unique and frustrating flight dynamics problems, as they found out with the Pegasus system. Scaling up or down in payload really does not change that.

The three variables of importance are speed at rocket ignition, path angle at rocket ignition, and altitude at rocket ignition, ranked in that order of importance. Speed is most important, altitude least.

To relieve the rocket of mass ratio requirements and increase its payload fraction, you need at least 2-3 km/s speed at rocket ignition. Mach 6 is 2 km/s; Mach 10 is 3 km/s. Subsonic release is around 0.2 km/s, an order of magnitude (or more) short. There is a really good reason the preferred staging velocity for a 2-stage launch rocket is 2-3 km/s or thereabouts. Pegasus fell far short there.

Path angle at rocket ignition determines whether you can fly a zero-lift ballistic trajectory at minimum drag, or you must add wings and gobs of drag-due-to-lift in order to pull up from near horizontal. The desired path angle is around 45 degrees to go simple ballistic. Vertical launchers inherently follow the ballistic path at min drag. Pegasus had to pull up at high drag with wings, precisely because it lit the rocket flying horizontal.

Altitude is the least important variable. A typical 2-stage launcher will stage around 150-200,000 feet, but it makes little difference of you do this at only 30-40,000 feet. The difference is relatively small compared to the orbit altitude (80+ miles utter minimum, 200+ miles preferred). This small difference is the only advantage that Pegasus bought for all the troubles it incurred.

Shock-impingement heating and hypersonic recovery temperatures being what they are, you will never successfully carry a rocket alongside, on the dorsal surface, or under the belly of a Mach 4 to 6 airplane, DARPA’s new XS-1 concept from Boeing notwithstanding. The best you can hope for is a sharp pullup at release, so your payload rocket need not have wings. The faster flight speed you attempt this, the more likely your carrier aircraft will fold up its wings when there are no hinges. It’s bad enough at Mach 0.7. Worse by far supersonic.

And, the more gigantic its engines must be, even if the wings stay on. Pull up and airspeed drops dramatically, unless you have gobs and gobs of thrust. Basic energy management.

Myself, I think there's just more merit to vertical-launch 2-stage rockets, at this time in history.

But if you can solve (1) how to fly hypersonic re-usably, and/or (2) how to carry your payload internally until you leave the sensible atmosphere to release it, or (3) how to achieve positive separation flying supersonic where aerodynamic forces exceed weight forces by orders of magnitude, then you can solve this air launch-to-orbit problem.

There’s a very good reason military weapons release speeds are limited to subsonic at 485 knots!!! Ever seen a 500 lb bomb roll outboard along the bottom of a wing, up over the tip, and then roll inboard? I have, in the surviving (!!) camera film from the aircraft. It took the horizontal tail off the aircraft when it did finally depart off the trailing edge of the wing.

Tain't easy, that's for sure! I've been trying, myself, for some time now. My qualifications: BS Aerospace Engineering '72, MS Aerospace Engineering '74, PE certification since '79, PhD General Engineering '00, and 20 years' experience in defense weapons new product development engineering and test work with all sorts of rockets and ramjets.

Be my guest.

GW

kbd512 · 2016-09-15 19:41:32

GW Johnson wrote:

Subsonic air launch of rockets to orbit doesn't really buy you a whole lot in terms of performance measures, but it does "buy" you some very unique and frustrating flight dynamics problems, as they found out with the Pegasus system. Scaling up or down in payload really does not change that.

Agreed.

GW Johnson wrote:

The three variables of importance are speed at rocket ignition, path angle at rocket ignition, and altitude at rocket ignition, ranked in that order of importance. Speed is most important, altitude least.

Agreed. The dV increment that the rocket has to overcome to achieve orbital velocity is extreme and air launch from a subsonic aircraft isn't going to help in that department. However, there's a substantial pressure differential between 40K and SL, wouldn't you say?

Falcon 1 (LOX/RP-1) 670kg payload to LEO 38555kg = 57.5kg of vehicle per kg of payload delivered
Pegasus (solid) 450kg payload to LEO 23130kg = 52.2kg of vehicle per kg of payload delivered

That's not much of a difference, but it's something.

GW Johnson wrote:

To relieve the rocket of mass ratio requirements and increase its payload fraction, you need at least 2-3 km/s speed at rocket ignition. Mach 6 is 2 km/s; Mach 10 is 3 km/s. Subsonic release is around 0.2 km/s, an order of magnitude (or more) short. There is a really good reason the preferred staging velocity for a 2-stage launch rocket is 2-3 km/s or thereabouts. Pegasus fell far short there.

I asked the question of you in the StratoLaunch II thread regarding whether or not it is feasible to purchase 2km/s dV using a ramjet attachment / skirt around an aerospike nozzle which could then be jettisoned as the attachment approaches a velocity at which it can no longer efficiently operate.

GW Johnson wrote:

Path angle at rocket ignition determines whether you can fly a zero-lift ballistic trajectory at minimum drag, or you must add wings and gobs of drag-due-to-lift in order to pull up from near horizontal. The desired path angle is around 45 degrees to go simple ballistic. Vertical launchers inherently follow the ballistic path at min drag. Pegasus had to pull up at high drag with wings, precisely because it lit the rocket flying horizontal.

Part of the reason for the four GE90's is that StratoLaunch II needs enough thrust to enable it to zoom to 40K and then eject the rocket on the correct flight path angle. The launch aircraft also needs to be capable of taking off from and landing on ordinary airport runways at MTOW and overload conditions.

GW Johnson wrote:

Altitude is the least important variable. A typical 2-stage launcher will stage around 150-200,000 feet, but it makes little difference of you do this at only 30-40,000 feet. The difference is relatively small compared to the orbit altitude (80+ miles utter minimum, 200+ miles preferred). This small difference is the only advantage that Pegasus bought for all the troubles it incurred.

Would altitude be more important if you could design a first stage engine to perform optimally at a higher altitude? ATK's Thunderbolt weighs 500,000 lbs and delivers a 10,000lb payload. How much more would the rocket weigh if it was launched at SL?

GW Johnson wrote:

Shock-impingement heating and hypersonic recovery temperatures being what they are, you will never successfully carry a rocket alongside, on the dorsal surface, or under the belly of a Mach 4 to 6 airplane, DARPA’s new XS-1 concept from Boeing notwithstanding. The best you can hope for is a sharp pullup at release, so your payload rocket need not have wings. The faster flight speed you attempt this, the more likely your carrier aircraft will fold up its wings when there are no hinges. It’s bad enough at Mach 0.7. Worse by far supersonic.

The shorter stiffer wings of StatoLaunch II should enable it to pull up and punch out the rocket without imparting as much stress to the wings. The SR71 performed supersonic releases of relatively large and heavy drones. However, supersonic release of a rocket massive enough to reach orbit with a useful payload is probably out of the question.

GW Johnson wrote:

And, the more gigantic its engines must be, even if the wings stay on. Pull up and airspeed drops dramatically, unless you have gobs and gobs of thrust. Basic energy management.

StratoLaunch I - 6 PW4056 (56,750lbf / ea at SL) = 340,500
StratoLaunch II - 4 GE90-115B (115,300 / ea at SL) = 461,200

GW Johnson wrote:

Myself, I think there's just more merit to vertical-launch 2-stage rockets, at this time in history.

Some of us just want to expand the corners of the envelope...

GW Johnson wrote:

But if you can solve (1) how to fly hypersonic re-usably, and/or (2) how to carry your payload internally until you leave the sensible atmosphere to release it, or (3) how to achieve positive separation flying supersonic where aerodynamic forces exceed weight forces by orders of magnitude, then you can solve this air launch-to-orbit problem.

Why do you have to nearly leave the atmosphere just to gain the benefits of launching from a higher altitude?

GW Johnson wrote:

There’s a very good reason military weapons release speeds are limited to subsonic at 485 knots!!! Ever seen a 500 lb bomb roll outboard along the bottom of a wing, up over the tip, and then roll inboard? I have, in the surviving (!!) camera film from the aircraft. It took the horizontal tail off the aircraft when it did finally depart off the trailing edge of the wing.

Depends on clearance limits. Hornets are cleared to 600KCAS or something close to that for release of Mk82's. That doesn't mean it's a particularly good idea, but you can do it under certain conditions.

GW Johnson wrote:

Tain't easy, that's for sure! I've been trying, myself, for some time now. My qualifications: BS Aerospace Engineering '72, MS Aerospace Engineering '74, PE certification since '79, PhD General Engineering '00, and 20 years' experience in defense weapons new product development engineering and test work with all sorts of rockets and ramjets.
Be my guest.
GW

I think an airborne launcher is a worthwhile tool to have in our belt and a complementary technology to Falcon 9, New Shepard, and Vulcan.

Last edited by kbd512 (2016-09-15 20:27:40)

Antius · 2016-09-16 05:28:58

GW Johnson wrote:

Subsonic air launch of rockets to orbit doesn't really buy you a whole lot in terms of performance measures, but it does "buy" you some very unique and frustrating flight dynamics problems, as they found out with the Pegasus system. Scaling up or down in payload really does not change that.
The three variables of importance are speed at rocket ignition, path angle at rocket ignition, and altitude at rocket ignition, ranked in that order of importance. Speed is most important, altitude least.
To relieve the rocket of mass ratio requirements and increase its payload fraction, you need at least 2-3 km/s speed at rocket ignition. Mach 6 is 2 km/s; Mach 10 is 3 km/s. Subsonic release is around 0.2 km/s, an order of magnitude (or more) short. There is a really good reason the preferred staging velocity for a 2-stage launch rocket is 2-3 km/s or thereabouts. Pegasus fell far short there.
Path angle at rocket ignition determines whether you can fly a zero-lift ballistic trajectory at minimum drag, or you must add wings and gobs of drag-due-to-lift in order to pull up from near horizontal. The desired path angle is around 45 degrees to go simple ballistic. Vertical launchers inherently follow the ballistic path at min drag. Pegasus had to pull up at high drag with wings, precisely because it lit the rocket flying horizontal.
Altitude is the least important variable. A typical 2-stage launcher will stage around 150-200,000 feet, but it makes little difference of you do this at only 30-40,000 feet. The difference is relatively small compared to the orbit altitude (80+ miles utter minimum, 200+ miles preferred). This small difference is the only advantage that Pegasus bought for all the troubles it incurred.
Shock-impingement heating and hypersonic recovery temperatures being what they are, you will never successfully carry a rocket alongside, on the dorsal surface, or under the belly of a Mach 4 to 6 airplane, DARPA’s new XS-1 concept from Boeing notwithstanding. The best you can hope for is a sharp pullup at release, so your payload rocket need not have wings. The faster flight speed you attempt this, the more likely your carrier aircraft will fold up its wings when there are no hinges. It’s bad enough at Mach 0.7. Worse by far supersonic.
And, the more gigantic its engines must be, even if the wings stay on. Pull up and airspeed drops dramatically, unless you have gobs and gobs of thrust. Basic energy management.
Myself, I think there's just more merit to vertical-launch 2-stage rockets, at this time in history.
But if you can solve (1) how to fly hypersonic re-usably, and/or (2) how to carry your payload internally until you leave the sensible atmosphere to release it, or (3) how to achieve positive separation flying supersonic where aerodynamic forces exceed weight forces by orders of magnitude, then you can solve this air launch-to-orbit problem.
There’s a very good reason military weapons release speeds are limited to subsonic at 485 knots!!! Ever seen a 500 lb bomb roll outboard along the bottom of a wing, up over the tip, and then roll inboard? I have, in the surviving (!!) camera film from the aircraft. It took the horizontal tail off the aircraft when it did finally depart off the trailing edge of the wing.
Tain't easy, that's for sure! I've been trying, myself, for some time now. My qualifications: BS Aerospace Engineering '72, MS Aerospace Engineering '74, PE certification since '79, PhD General Engineering '00, and 20 years' experience in defense weapons new product development engineering and test work with all sorts of rockets and ramjets.
Be my guest.
GW

This would appear to bring us straight back to Sea Dragon. A huge, expendable multistage rocket, built in shipyards from alloy steels.

Design and systems engineering costs are minimised by opting for simple as possible pressure-fed liquid propellant systems and scaling the rocket up to the largest size practically possible.

Transportation becomes a non-issue, as the rocket is sea launched. For large expendable engines working at low chamber pressures, film cooling and ablative cooling can be employed.

Basically, a very simple product built as large as possible and then mass produced.

With the lower dry weight allowed through the use of composite tanks, we might substitute liquified natural gas for liquid hydrogen with about the same payload fraction.

The discarded upper stage propellant tanks would make ideal space station modules. At 22m in diameter, We could ring together perhaps a dozen of them and rotate to produce an O'Neill type rotating colony.

To what size could Sea Dragon practically be scaled up? Could we produce super-boosters lifting 1000's of tonnes into orbit? How could we simplify the vehicle even further to produce a truly minimum cost design today?

Last edited by Antius (2016-09-16 05:31:05)

GW Johnson · 2016-09-16 10:08:10

Kbd512:

I would like to see an airborne launcher myself. What I ran into trying to come up with one is the ~Mach 3.3 limit of gas turbine technology. To go faster than that requires ramjet technology, which means you need a booster to reach ramjet takeover speed. Once you resort to that, you might as well do a launch rocket, because the most practical booster is a rocket. The high speed flight problems with carrying and separating a payload become fairly intractable from Mach 3.5 up.

Even at turbine speeds, there is the store separation issue. The Mig-25 had this. It was capable of Mach 3.3 "clean" but was placarded (by the Russians!) to max 2.8 Mach with external stores. I presume these stores were missiles that powered away positively. The limiting issue was turbulence shed from the external stores creating fatal vibration loads.

I know they launched some D-21 drones from the back of a version of the SR-71. They did it by starting the drone engine while supersonic about Mach 2.5-ish, releasing it, and diving as hard as possible out from under it. They stopped attempting this when a collision between drone and aircraft killed a crew. From then on, the drone was launched subsonically from B-52’s, using a big solid rocket booster slung underneath the drone’s belly to carry it to takeover about Mach 2.5-ish. It was drop, then light the rocket once it fell to a safe distance.

If a Mach 4+ ramjet airplane were otherwise feasible (and it is), there is the shock-impingement problem with parallel-mounted external stores. A shock shed by one item impinges on the other, and greatly magnifies locally the aerodynamic heating at the impingement point and just downstream. This is bad enough in the Mach 4 range, at Mach 5-6 it becomes fatal for all known airframe materials (which at those speeds even on a clean airframe are stainless steels and inconels, not even titanium holds up). This effect nearly cut the tail off the X-15 at Mach 6.67 with the scramjet test article on its ventral fin stub. The experimental protective white ablative coating on that flight proved useless.

Ramjets themselves are much more than a shroud or a ring about something. In practice a shroud ring around a rocket with attempted afterburning inside the ring gets you very little performance benefit, often a negative increment. Most of the time, you cannot achieve afterburning inside the ring.

There are basically two design regimes for ramjet, what I call the "low" and "high" speed regimes. In each, you get better flame stability with a sudden dump flame stabilization scheme than with the older and better-known colander burners or V-gutter baffles. The latter are limited to a much narrower range of air flow velocities inside the engine. But some sort of flame stabilization is a requirement.

Low speed designs have simple pitot inlets and convergent-only nozzles that are not always choked. This pitot inlet is best done as a nose inlet, performance is really crappy if you attempt side-mounted inlets, due to drag. They operate from about Mach 0.7 to about Mach 2+ (theoretically higher, but not practically). Sized properly, performance might be around 800 Isp max, but remember the cycle pressure is low, so also is frontal thrust density quite low. These never accelerate very well on their own. Sizepoint is usually in the vicinity of Mach 1.1, where the throat first chokes and combustor velocities have maximized. You size at the richest mixture you intend to run, usually around 10% rich. Size the inlet just barely not to spill there. Isp performance maximizes near sizepoint, thrust increases slowly throughout the practical speed range. You fuel these with something relatively volatile, usually gasoline.

The high speed designs feature external compression features on the inlets, and C-D nozzles that are always choked. They operate from about Mach 2 to as high as Mach 6-ish. Performance potential is higher at around 1500 sec Isp max, and thrust density is better since the cycle pressures are somewhat higher. These can accelerate supersonic missiles during airbreathing sustain, but are limited to under 45 degree path angles for steady climb. Sizepoint is usually at the "design" shock-on-lip Mach number for the inlet scheme, somewhere in the low stratosphere, which is where supercritical inlet pressure margin is minimized by the coldest air. Again, you size at the richest mixture you intend to run, usually around 10% rich. You size the inlet never to spill air. Side-mounted inlets usually drag-limit to about Mach 4. Nose or chin inlets are much lower drag, and have (accidentally!!!) demonstrated Mach 6. Isp maximizes near sizepoint speed. Thrust increases rapidly as you approach sizepoint, is nearly constant to about Mach 4, then decreases slowly as you speed up. You fuel these with something less volatile, usually one or another form of kerosene. They’re all roughly alike.

Properly-sized engines of either speed range share certain limitations on geometry. About the max throat/combustor area ratio is 65%, limited by flame stability. The only successful designs using larger open throats were using hypergolic magnesium vapor fuel that needed no flameholding. That's a lower energy class of fuels with lower Isp potential.

Typically in a flameholding dump combustor design, the inlet duct(s) area/combustor area ratio is 0.45 to 0.50. There's no way around that in designs that actually work. The inlet capture area (which is that swept out, not the physical opening) is in the neighborhood of 0.25 to 0.30 of the combustor area. The exit area (if there is one) is usually no more than about the same as the combustor area.

Geometries like that do not match well at all with the notion of an augmentor ring. But they do work quite well. They work up past 80,000 feet at Mach 3 to 4. I don’t know how much higher. But I would doubt stable operation is possible above 150,000 feet, even at Mach 4+, because of the low pressures and densities. Pressure affects flame stability. In the thin air, there is no potential for climb or acceleration, because the developable forces are so low (proportional roughly to ambient air pressure), just like turbine or any other airbreather. That’s the “service ceiling” effect.

Antius and Kbd512:

I’m not saying that ramjet aircraft traveling at Mach to 5 to 6 cannot release second stages that go to orbit, because they can. But you have to solve some stupendously-difficult problems with a great deal of finesse to do so. The most difficult of these is the shock-impingement heating problem for all externally-carried stores above about Mach 4-ish. To avoid it, you must go with internal carry, which causes you extreme store-separation risks, plus a whole host of airframe design restrictions, and center-of-gravity travel problems.

Such air launch might still be beneficial enough in terms of staging velocity to be attractive if done from a supersonic turbine aircraft. But you will have to solve the safe supersonic store separation issue. Anything above Mach 2 becomes interesting from a staging velocity benefit viewpoint, and the higher the better. About Mach 3.3 is the limit with practical turbines at this time in history. That was SR-71 and Mig-25’s limit. The XB-70 was limited to about Mach 3.0. Everything else has been limited to about Mach 2.5 or slower. Big supersonic bomber-like aircraft have very little potential for other uses besides launch, though. Those are very expensive machines to build and fly.

In some of the paper studies I did, I could rough out a rocket-boosted ramjet airplane that might reach Mach 5 to 6 in horizontal flight at around 60,000 feet, and have the power to pull up and release an external store at a 45 degree path angle. I did not solve the safe separation problem, nor the shock impingement heating problem. But by careful staging-off of components for separate recovery (such as subsonic wings with the rocket booster), I could achieve that stagepoint from a runway takeoff.

I took a brief look at adding a ramjet-assist pod or pods to a vertical-launch rocket. But those trajectories are leaving sensible air still ascending very nearly vertically at around only Mach 2-ish near 80,000 feet-ish, so the low speed range pitot inlet design was more appropriate. I never saw much benefit for the weight because of the rather low frontal thrust density of this class of designs. Thrust minus drag minus weight for the pod was nearly always negative.

I don’t know much about the giant sea-launched rocket concept. Does it not need to be out of the water to light its engine(s) ?

GW

kbd512 · 2016-09-16 11:37:02

Antius,

The same people who build our submarines can easily build the steel tanks for a pressure fed TSTO. A small nuclear reactor is required to provide the power to distill sea water for LOX/LH2 for the upper stage, along with US Navy personnel to operate the reactor. There are now single step chemical processes for manufacturing kerosene for fuel for the first stage.

Transportation is still an issue because the rocket will be built in a dry dock and then delivered by a tug boat to a floating launch platform which then has to erect and fuel the rocket. If the payloads are propellants or water, then payload integration is less of a problem. This rocket should be used for bulk cargo delivery only so man rating is not required. These rockets will see infrequent use, likely in conjunction with an exploration mission, which also means that costs will be similar to ground launch turbopump fed rockets. No launch pads, crawlers, crawler roadways, or sound suppression systems need to be maintained, so that's a plus.

I figure that the first stage should use a plug aerospike nozzle, rather than a conventional bell which is more likely to be damaged during splashdown, and come equipped with hydraulically actuated fins for deceleration like a snakeye bomb. Perhaps retro-propulsion can be applied just before splashdown.

1t to 5t Air launch for delivery of small satellites and small human transfer vehicles.

10t to 125t Ground launch for delivery of exploration class spacecraft and space stations.

250t to 500t Sea launch for delivery of propellants and water.

If we have manage to get all three of those launch vehicle technologies operational at the same time, then we have a robust launch vehicle program enabling us to go pretty much anywhere we want to within the inner solar system.

elderflower · 2016-09-16 14:54:15

For single stage to orbit, take a look at Reaction Engines' Skylon proposal.

kbd512 · 2016-09-16 19:54:32

elderflower wrote:

For single stage to orbit, take a look at Reaction Engines' Skylon proposal.

How far along are they on the development of SABRE?

elderflower · 2016-09-17 10:29:58

Partial static testing of the engine, I think. I don't suppose they can effectively test the air breathing coolers yet as they probably don't have the means to accelerate it to hypersonic velocities. They will have to build a special test vehicle.

GW Johnson · 2016-09-17 20:39:03

My information may be obsolete, but the last time I heard, they were running some kind of tests on the SABRE engine, and had done some sort of test (probably a bench test of some kind) that convinced them their high speed heat exchange scheme would work. They seem to have accounted for how to prevent ice clogs from the humidity n the air, too.

The airframe shape proposed for Skylon appears to have mid-mounted wings with tip-mounted engines. They seem aware of the air capture and compressor blading temperature problem, because the mission profile calls for airbreathing to Mach 5 at some high altitude although still down in the air, then all-rocket from there to orbit. At Mach 5 and 85,000 feet on a standard day, the total temperature (near static once decelerated in the inlet) is 2252 R = 1251 K, and that's without any air compression enthalpy rise, which is a huge effect. Even turbine blade designs are only good for about 2660 R = 1480 K. You see the problem for real compressor blade materials. You have to build them at least as tough as turbine blades.

What concerns me is reentry, which starts at Mach 25-ish, and ends the hypersonics on a slender shape at about Mach 5. The shocks shed by the tip-mounted engines are going to impinge-upon (and cut completely through) the wings. That's the same shock impingement problem I tried to describe in post 16 just above. I have very serious doubts about the Skylon airframe shape proposal, as a result.

You simply cannot have parallel nacelles and expect to survive, no matter if you used solid silica phenolic heat shield materials (very heavy indeed) on the impingement zone. The X-15 airframe shape could have survived Mach 6+ without that scramjet test article mounted in parallel, but not with it (and almost didn't). Shapes like the SR-71 cannot survive at Mach 6.

There's a very good reason all entry shapes used so far do not feature any parallel-mounted nacelles or any other parallel-mounted structures. Your driving temperature in degrees K for heat transfer is roughly numerically equal to your speed in meters/second all during entry hypersonics, but your film coefficients are an order of magnitude or two higher than you would ever expect (higher than most stagnation correlations), even at only Mach 6 or so, in a shock-impingement zone. The material heats to melting and erodes away in a matter of seconds.

Hypersonic aerothermo is a bitch, ain't it?

GW

Last edited by GW Johnson (2016-09-17 20:46:07)

kbd512 · 2016-09-18 19:08:27

GW,

I get that a functional ramjet is more complicated than just a cowling fitted around the rocket motor, but I want to know if this has been tried before.

Suppose the cowling could move forward and backwards around the rear end of the rocket, using the body of the rocket as the diffuser, to vary the inlet area and pressure using hydraulic pressure from fuel being fed into the combustion chamber. The flame holder is the nozzle of an aerospike rocket engine. The rocket motor would initially accelerate the vehicle to operating speed so the ramjet can take over. Thereafter, the ramjet accelerates the vehicle until it reaches about 2km/s and then the fuel valves that power the ramjet cowling close, pyro separates the ramjet cowling, and the rocket takes over again until orbital velocity is achieved.

If you could gain 2km/s in initial velocity and are around 100K ft above the ground using the ramjet, will a LOX/RP-1 rocket then have enough Isp to reach orbit?

Antius · 2016-09-19 00:56:50

kbd512 wrote:

GW,
I get that a functional ramjet is more complicated than just a cowling fitted around the rocket motor, but I want to know if this has been tried before.
Suppose the cowling could move forward and backwards around the rear end of the rocket, using the body of the rocket as the diffuser, to vary the inlet area and pressure using hydraulic pressure from fuel being fed into the combustion chamber. The flame holder is the nozzle of an aerospike rocket engine. The rocket motor would initially accelerate the vehicle to operating speed so the ramjet can take over. Thereafter, the ramjet accelerates the vehicle until it reaches about 2km/s and then the fuel valves that power the ramjet cowling close, pyro separates the ramjet cowling, and the rocket takes over again until orbital velocity is achieved.
If you could gain 2km/s in initial velocity and are around 100K ft above the ground using the ramjet, will a LOX/RP-1 rocket then have enough Isp to reach orbit?

I calculate the total delta-V to reach orbit from 100,000' and 2km/s to be 6855m/s. Using an ideal LOX/Kerosene rocket with Ve of 3500m/s in vacuum, the mass ratio would be 7.1. If rocket dry structural weight fraction is 10%, then payload fraction would be 4.1%. My simple approximation neglects gravity losses or residual aerodynamic drag and also assumes perfect mixture ratio. So real performance will be lower.

GW Johnson · 2016-09-19 08:33:32

I don't know whether you can use a rocket exhaust stream as a flameholder. I've never seen it done, myself. But if he's still alive out in Los Angeles, Joe G. Bendot would know. He was the ejector ramjet expert, which may be what Kbd512 wants to attempt with his ring around a rocket scheme. Joe would be 90-something today. He and I worked on the evaluation of the gas-generator fed ramjet that was the sustainer of the SA-6 about 4 decades ago.

Antius: what I usually use for TSTO vertical launch estimates is 5% loss for gravity and 5% loss for drag. So, whatever the velocity requirement is, I up it 10% for an initial estimate of required ideal velocity increment for the rocket equation. The loss estimate gets refined from there, usually downward. But that only applies to a zero-lift ballistic trajectory.

GW

kbd512 · 2016-09-19 19:21:48

GW Johnson wrote:

I don't know whether you can use a rocket exhaust stream as a flameholder. I've never seen it done, myself. But if he's still alive out in Los Angeles, Joe G. Bendot would know. He was the ejector ramjet expert, which may be what Kbd512 wants to attempt with his ring around a rocket scheme. Joe would be 90-something today. He and I worked on the evaluation of the gas-generator fed ramjet that was the sustainer of the SA-6 about 4 decades ago.
Antius: what I usually use for TSTO vertical launch estimates is 5% loss for gravity and 5% loss for drag. So, whatever the velocity requirement is, I up it 10% for an initial estimate of required ideal velocity increment for the rocket equation. The loss estimate gets refined from there, usually downward. But that only applies to a zero-lift ballistic trajectory.
GW

I knew someone had thought of this before. Apparently it's called an "air-augmented rocket". I just found out about it half an hour ago. I knew I wasn't asking to try anything that hadn't already been tried at some point.

The Russians started development work on something called the Gnom ballistic missile in the late 1950's using a concept put forth by Boris Shavyrin from Kolomna Bureau of Machine Industry. It used a rocket nozzle combined with a ramjet to dramatically increase mass flow and thus improve Isp. Gnom was to use three solid stages, with a launch weight of 29t, a dV of 7.6km/s, and a range of 11,000km with a 535kg warhead. That'd be absolutely phenomenal performance, even for today. Their air augmented SRBM prototype, the PR-90, delivered a 500kg payload more than 100km using a rocket that only weighs 1500kg! The Isp of the air augmented rocket was 550 seconds (yes, I wrote that correctly)!

An air launched LOX/RP-1 SSTO with a 5t payload should be no problem at all with modern materials and manufacturing methods. I'm sure the engineering isn't simple or easy, but it absolutely can be done and in fact was done with the prototype way back in the 1950's. Apparently the rocket engine nozzle ejector system was considered for air augmentation, which means that Mr. Bendot really would know whether or not it would work, but that work never made it past the conceptual stage of development.

Edit:

The solid rockets in question only had a 120s Isp. The PBAN/APCP SRB's have a 242s Isp. We may not even need to concern ourselves with using liquid propellants if we can make the casings light enough, but I still wonder how light we could make the launch vehicle with LOX and Jet-A. An air-augmented, all-solid TSTO solution could combine the density impulse of solids with the Isp of inefficient ramjets. I realize density impulse is of very little importance for upper stages, but we're trying to stuff this mini rocket in the fuselage of another aircraft so we can carry it to altitude and eject it without incurring a massive amount of drag.

A LOX/RP-1 rocket may be able to go all the way to orbit with a single stage, but I have a feeling we'd run into the thrust and throttling problems that GW already noted so we'd just wind up with a more expensive TSTO that presents LOX and RP-1 storage issues to contend with.

ATK wants to add massive wings and a dual RL-10 powered upper stage. That means LOX/LH2, which is neither inexpensive nor simple to operate even though we're very well versed in LOX/LH2 rocketry at this point. Perhaps LH2 or RP-1 gelled with boron would help reduce the volume and thus inert mass of the stage to some point where liquid upper stages would be more practical for an air launched rocket. I'm just throwing some thoughts out there.

Last edited by kbd512 (2016-09-19 20:17:52)

New Mars Forums

Announcement

#1 2016-09-14 07:52:06

Reusable LOX/Kerosene SSTO with drop tanks

#2 2016-09-14 17:31:55

Re: Reusable LOX/Kerosene SSTO with drop tanks

#3 2016-09-14 18:51:47

Re: Reusable LOX/Kerosene SSTO with drop tanks

#4 2016-09-14 22:54:28

Re: Reusable LOX/Kerosene SSTO with drop tanks

#5 2016-09-15 00:02:48

Re: Reusable LOX/Kerosene SSTO with drop tanks

#6 2016-09-15 03:16:21

Re: Reusable LOX/Kerosene SSTO with drop tanks

#7 2016-09-15 04:11:30

Re: Reusable LOX/Kerosene SSTO with drop tanks

#8 2016-09-15 05:02:09

Re: Reusable LOX/Kerosene SSTO with drop tanks

#9 2016-09-15 08:21:21

Re: Reusable LOX/Kerosene SSTO with drop tanks

#10 2016-09-15 10:06:59

Re: Reusable LOX/Kerosene SSTO with drop tanks

#11 2016-09-15 11:39:55

Re: Reusable LOX/Kerosene SSTO with drop tanks

#12 2016-09-15 14:18:35

Re: Reusable LOX/Kerosene SSTO with drop tanks

#13 2016-09-15 16:59:51

Re: Reusable LOX/Kerosene SSTO with drop tanks

#14 2016-09-15 19:41:32

Re: Reusable LOX/Kerosene SSTO with drop tanks

#15 2016-09-16 05:28:58

Re: Reusable LOX/Kerosene SSTO with drop tanks

#16 2016-09-16 10:08:10

Re: Reusable LOX/Kerosene SSTO with drop tanks

#17 2016-09-16 11:37:02

Re: Reusable LOX/Kerosene SSTO with drop tanks

#18 2016-09-16 14:54:15

Re: Reusable LOX/Kerosene SSTO with drop tanks

#19 2016-09-16 19:54:32

Re: Reusable LOX/Kerosene SSTO with drop tanks

#20 2016-09-17 10:29:58

Re: Reusable LOX/Kerosene SSTO with drop tanks

#21 2016-09-17 20:39:03

Re: Reusable LOX/Kerosene SSTO with drop tanks

#22 2016-09-18 19:08:27

Re: Reusable LOX/Kerosene SSTO with drop tanks

#23 2016-09-19 00:56:50

Re: Reusable LOX/Kerosene SSTO with drop tanks

#24 2016-09-19 08:33:32

Re: Reusable LOX/Kerosene SSTO with drop tanks

#25 2016-09-19 19:21:48

Re: Reusable LOX/Kerosene SSTO with drop tanks

Board footer