Reusable Rockets to Orbit

Terraformer · 2013-09-25 13:56:31

As far as disposable upper stages go... they can't be more complex than a modern car, surely? And modern cars manage to turn a profit being sold for $20-30k...

At the moment, I'm imagining a system based on a reusable first stage, sold much like aircraft are, and disposable upper stages which are manufactured en-masse. Selling space access, not launches. The company doesn't operate the fleet, whoever buys them does - we just sell the vehicles and the upper stages, and let them compete on minimising their other costs. Much like how airline companies work. Or bus services. Or the shipping industry. Really, space transport needs to catch up with every other transport industry...

Terraformer · 2013-09-26 10:32:07

How small could we make a ramjet first stage, GW? How large were the ramjet test articles you worked on? Could we make an X-spaceplane that is, say, 10 tonnes fully fuelled, to demonstrate the system?

Then, of course, the most important question - how much would you estimate it costing?

GW Johnson · 2013-09-27 07:20:17

Hi Terraformer:

I think you're asking about an experimental hypersonic ramjet airplane as a demonstrator vehicle. A thing like that could "easily" be done, and would approximate the old X-15 in terms of vehicle size and program costs. I don't have a dollar figure for that, but I'd guess nearer 10's of $M than a $B. It'd have to be done by a skunk works-like outfit, the "big boys" would turn it into a do-nothing "gravy train" just to suck on the government's tit. That's always at least 10 x more expensive, too.

Ramjet combustors with blockage-element flameholders were built and used in missiles up to about 1 m diameter long ago (Bomarc & similar). What's needed for an accelerator is a dump-combustor ramjet to render the flameholding insensitive to varying flight speeds and conditions. Those were flown up to about 0.5 m diameter in ASALM-PTV ca 1980.

Those dump combustors would need to be scaled up to around 1 meter dia for the experimental airplane, and maybe 2-3 m dia for a real TSTO orbital system of significant payload. That's cut-and-try experimental work, there is no reliable science for it, but it can be done. Engineering art, pure and simple, proponents of computational fluid mechanics notwithstanding. (Real flameholding is simply not in their models.)

There are now very few of us left in the US who know that art. I'm one. If Joe Bendot is still with us out in LA, he's another, but he's significantly older than me, and I'm a senior citizen now. If anybody's going to do this, they'd better get on the stick while we're still around. It's engineering art, and we taught no apprentices, we just all got laid off long ago.

The hardest part about such a project isn't the ramjet, it's the configuration layout to avoid disastrous aeroheat damage from shock impingement heating, and still demonstrate a form that could work for the orbital system. Depends on your airframe material, but with the Inconel skins on the X-15, it was Mach 6-ish for the speed above which the airframe gets cut apart by the heat from the shock waves. Mach 6 is also just about the upper limit for using ramjet, too.

GW

Last edited by GW Johnson (2013-09-27 07:22:37)

Terraformer · 2013-09-27 09:34:00

But could you be persuaded to move countries to work on such a project? It would be troublesome to work on it in the US, given ITAR...

Back to an engineering question - wouldn't the sort of payload bay idea that would have been used on Black Colt, carrying the upper stage within a bay and then drifting it out at the apogee, be a plausible way of getting around the problem of having to fly two aircraft mounted together at hypersonic speeds?

GW Johnson · 2013-09-29 12:26:57

Not sure about working overseas vs ITAR. I'm too old to move anymore, anyway.

But, the payload bay idea is a good idea. There are two limitations to consider, both with fundamental impact on the basic design concept. (1) the bay contains rocket pods a lot easier than winged or lifting-body shapes, and (2) store separations from a weapons bay are only known to work up to about Mach 2, which means that a launcher moving high supersonic or hypersonic must exit the sensible air to open its payload bay and launch the second stage it contains.

Exiting the sensible air requires either rocket thrust or coasting with a lot of deceleration, trading speed for altitude, in the face of drag. The most important staging variables are ranked as (a) velocity, (b) steeply-climbing path angle, and (c) altitude at staging. That ranking opposes the notion of stage release at an apogee point, because you've lost all your speed, and that's the most important item.

It's a tough problem. I'm beginning to believe that staging should not occur at ramjet top speed, but a tad later, after a big rocket-propelled pull-up. Not just a pull-up, but a pull-up maintained past the edge of the sensible air. But I have no numbers to back that notion up. That scenario would match the payload bay concept, as long as the second stage fits the bay (item 1 above).

The downside is that more propulsion from rocket raises first stage size and weight. There's a limit to how big an airplane can be, since mass scales as dimension cubed, while strength scales only as dimension squared. Yield and ultimate material stresses do not scale.

GW

Last edited by GW Johnson (2013-09-29 12:31:54)

Terraformer · 2013-09-29 15:39:40

Well, I was thinking that it would be traded for a smaller upper stage, for the same payload to orbit - just shifting more of the delta-V onto the lower stage.

Terraformer · 2013-09-30 06:11:59

Plus, I was thinking that it would - at first at least - be used with an expendable upper stage, with such stages being mass manufactured in the same way cars are today. If the stage costs $50k and can be used to put 500kg into low orbit, that's still dramatically better than anyone else. Of course, it's no good for crew launch, but for servicing the small satellite market?

GW Johnson · 2013-10-02 09:36:40

It would be pretty neat to put a small 1 or 2 man craft up there, using this staged airplane concept. Then refuel it on-orbit with propellants shot up there with a light gas gun. Then you could go service satellites in-situ with it.

Or maybe, use the staged airplane to send up the men. Then refuel via light gas gun an orbital repair tug vehicle, previously launched and left up there. Man the refueled tug, and go service satellites with it. Leave the tug and come home in the rocket pod/spaceplane second stage. Again, and again, and again.

That's what the shuttle did, except we had to launch and recover the entire orbital repair tug (cargo bay and arm and operator's cabin) every single flight. That turned out to be not a very smart way of doing an otherwise very beneficial job.

GW

Terraformer · 2013-10-02 10:13:16

Well, if the Rainbow Dash can take 10 tonnes to 3km/s@60km altitude, that should allow for a reusable hydrogen fuelled spaceplane, perhaps? A mass ratio of 4, 20% inerts, folding wings...?

A disposable upper stage using hydrogen could then put 2 tonnes into low orbit. Two launches, then, could let you service satellites? Or maybe just build a station somewhere convenient, and launch repair missions from there. You could also then serve as a depot for fuel, and whatever else it is that hotels and research stations might wish to buy.

GW Johnson · 2013-10-02 17:34:07

Lessee, 3 km/s to 7.8 km/s, that's 4.8 km/s required, factored up about 5% to 5.3 km/s for gravity losses. LOX-LH2 at around 1000 psia chamber pressure might get you an exhaust velocity of 4.48 km/s. I get a mass ratio near 3.3 to arrive on LEO with pretty-much dry tanks.

That's a propellant fraction of about 70%. For reusability, I'd hesitate to claim an inert fraction any lower than 25%. So, you're looking at a winged device or rocket pod device that carries about 5% dead-head payload.

If it weighed 10 tons at ignition, it can carry about half a ton to orbit. That's the weight of a man in a suit with some water, food, and oxygen for a day or two, and some hand tools. And we'd still need to squeeze a deorbit burn out of this thing.

Pretty minimal, but it is do-able.

GW

RGClark · 2013-10-10 10:37:40

DARPA wants a reusable first stage booster:

US Military Wants New Experimental Space Plane.
By Leonard David, SPACE.com's Space Insider Columnist | September 17, 2013 05:25pm ET
http://www.space.com/22836-military-exp … darpa.html

The X-33 could perform this role even if you replace the composite tanks with metallic ones:

Saturday, October 5, 2013
DARPA’s Spaceplane: an X-33 version.
http://exoscientist.blogspot.com/2013/1 … rsion.html

This can reduce the cost to space to the $2,000 per kilo, or $1,000 per pound range, a major cut in launch costs. If it had been understood that the X-33 could be used in that role, then instead of it just being a demonstration vehicle, we could have a cut in launch costs to that range a decade ago.

In an upcoming blog post I’ll show the same would have been true for the planned DC-X suborbital follow-on, the DC-X2. Then we could have had a cut in launch costs to that range two decades ago.

Bob Clark

RGClark · 2013-10-10 11:29:44

GW's presentation at the 2013 Mars Society convention on a lightweight thermal protection ceramic material is available on Youtube:

Reusable Ceramic Heat Shields - GW Johnson - 16th Mars Society Convention.
http://www.youtube.com/watch?v=3MXYY3jnNr0

This ceramic material is quite light at .03 specific gravity. However, it is tougher than the shuttle ceramic tiles. The shuttle tiles were quite fragile and maintenance intensive. GW's tiles would cut down on this maintenance cost and would have much reduced turnaround time due to thermal protection system maintenance.
I'm thinking it could also be used on the X-33. The X-33's TPS consisted of metallic shingles. There were tougher than the shuttle's silica tiles thus requiring minimal maintenance but they were rather heavy. GW's ceramics would also be more damage resistant than the shuttle tiles, but would be much lighter than the X-33's metallic shingles.
GW on his blog discussed another key advance that would have important applications to the X-33, reducing the weight of the propellant tanks:

Sunday, October 6, 2013
Building Conformal Propellant Tanks, Etc.
http://exrocketman.blogspot.com/2013/10 … s-etc.html

The conformal shape of the tanks on the X-33 made them have quite poor weight characteristics. The oxygen tanks were 4 times as heavy as a cylindrically-shaped tank carrying the same amount of propellant, and the hydrogen tanks were twice as heavy. GW believes it is possible to make them using metals at only a few percent higher than the weight of cylindrical tanks.

With these two weight saving methods in the TPS and the tanks, then you would probably have low enough structural mass to get the full-size VentureStar to have SSTO capability!

Bob Clark

GW Johnson · 2013-10-10 15:47:00

Hi Bob:

If you are thinking of contacting DARPA regarding their spaceplane, maybe we do have some things to offer.

There's my ceramic composite, which even on a high-ballistic coefficient hypersonic glider could be used everywhere but the nose cap and leading edges. We've been discussing conformal tankage, perhaps Al-Li or maybe even SS.

For interior airframe structure, you'll want composites, and I know 3 different ways to make joints in that stuff that are lightweight and reliable. The best of the 3 is also the least labor-intensive, actually.

Plus there is all the vehicle sizeout stuff you and I have looked at. And there's my experiences with ramjet if they want to add airbreather assist to it. (There are almost no American full-capability ramjetters left.)

Let me know what you think.

GW

Terraformer · 2013-10-11 02:33:24

I read what DARPA are looking for, and I was like, "that's what we've been discussing at NewMars!"... a reusable spaceplane that can reach 3km/s and launch an upper stage to carry 1-4klb into low orbit.

GW Johnson · 2013-10-11 08:44:08

Terraformer:

Exactly!!! The trouble will be getting heard to provide input. Government programs have a long history of "not invented here" attitude, and the myopia that only established giant contractors have any credibility. I'm hoping Bob has a contact on the inside at DARPA. Even if he does, it's long odds against us.

I've been looking at the TSTO spaceplane concept for about 3 years now. It's definitely feasible, even without an engine breakthrough like Skylon. The baseline concept is a rocket airplane for both stages, with the first stage incredibly big, which will be the limiting factor for deliverable payload size. You can substantially reduce the size of that first stage by adding airbreathing assist, in particular ramjet, because of its speed range from about Mach 1.6 to about Mach 5-or-6, if you design it right. Gas turbine is limited to about Mach 3.5-ish, although it works from takeoff. Scramjet is just not ready for prime time.

I'm not at all sure vertical launch is the right thing to do with such a lifting craft, because of the required takeoff thrust. Horizontal takeoff lets you use much smaller rocket engines, saving weight, and reducing the massive rate of weight change during takeoff. Doing supersonic climb on ramjet to a pullover/acceleration run in ramjet, saves an enormous weight in rocket propellants you would otherwise have to have on board.

I'm beginning to believe the best next step is to relight the rockets and climb out of the sensible air to a 3 km/s release of the second stage in effective vacuum. That not only reduces the delta-vee required of the second stage, it also solves a whole host of supersonic/hypersonic store-release aerodynamical difficulties.

The second stage could be a non-lifting rocket pod, or a winged/lifting body spaceplane of some kind. Any winged stage should do a runway landing. A rocket pod second stage fits within a cargo bay easier. Having the second stage enclosed in a cargo bay solves a whole host of hypersonic aerodynamic heating and cluster-vehicle safety issues. The tip of such a rocket pod could be a modernized version of the old Mercury or Gemini capsule, with a modern ablative heat shield (like PICA-X) such that the capsule could be reflown several times.

If the second stage is a winged spaceplane, I'd recommend strong consideration be given to sidestepping-entirely the hypersonic glider problem (extremely-intense aeroheating and extremely-poor landing characteristics). You do that by resurrecting an early 1950's concept: fold the wings into the wake behind a fuselage that enters dead-broadside, belly-first. Then they can be straight subsonic-airfoil wings, and you can land at speeds more like 70 knots than 200+ knots, one whale of a lot safer and more reliable.

Sidestep all of the supersonic/hypersonic airloads problem, too. You don't unfold them until you are subsonic, hanging from a chute. Chutes are cheap. Potentially, reusable ceramics could even be used at the stagnation point if the ballistic coefficient is low enough and the belly is blunt enough (no more carbon-carbon nose cap and leading edges).

--GW

Terraformer · 2013-10-11 09:38:10

Gas turbine can be used to Mach 3.5, you say? So using turbine rather than ramjet would reduce the air-assisted delta-V delivered by perhaps 300m/s, but it would deliver it at the start of the flight, so in terms of mass ratio it might not be such a great difference. Plus, the engines can be used for flyback...

So there's a question - takeoff using turbine to M3.5, then rocket boost until you hit 3km/s, or using rockets to take off, transitioning to Ramjet at M1.5, then transitioning back to rockets at M6.

Of course, you could have 3 sets of engines, but that would be rather complex and probably not worth the mass penalty.

RGClark · 2013-10-12 15:30:01

GW Johnson wrote:

Terraformer:
Exactly!!! The trouble will be getting heard to provide input. Government programs have a long history of "not invented here" attitude, and the myopia that only established giant contractors have any credibility. I'm hoping Bob has a contact on the inside at DARPA. Even if he does, it's long odds against us.
I've been looking at the TSTO spaceplane concept for about 3 years now. It's definitely feasible, even without an engine breakthrough like Skylon. The baseline concept is a rocket airplane for both stages, with the first stage incredibly big, which will be the limiting factor for deliverable payload size. You can substantially reduce the size of that first stage by adding airbreathing assist, in particular ramjet, because of its speed range from about Mach 1.6 to about Mach 5-or-6, if you design it right. Gas turbine is limited to about Mach 3.5-ish, although it works from takeoff. Scramjet is just not ready for prime time.
I'm not at all sure vertical launch is the right thing to do with such a lifting craft, because of the required takeoff thrust. Horizontal takeoff lets you use much smaller rocket engines, saving weight, and reducing the massive rate of weight change during takeoff. Doing supersonic climb on ramjet to a pullover/acceleration run in ramjet, saves an enormous weight in rocket propellants you would otherwise have to have on board.
I'm beginning to believe the best next step is to relight the rockets and climb out of the sensible air to a 3 km/s release of the second stage in effective vacuum. That not only reduces the delta-vee required of the second stage, it also solves a whole host of supersonic/hypersonic store-release aerodynamical difficulties.
The second stage could be a non-lifting rocket pod, or a winged/lifting body spaceplane of some kind. Any winged stage should do a runway landing. A rocket pod second stage fits within a cargo bay easier. Having the second stage enclosed in a cargo bay solves a whole host of hypersonic aerodynamic heating and cluster-vehicle safety issues. The tip of such a rocket pod could be a modernized version of the old Mercury or Gemini capsule, with a modern ablative heat shield (like PICA-X) such that the capsule could be reflown several times.
...
--GW

Jeff Sponable was a program manager on the DC-X, and is also a manager on this DARPA program. So he would be very amenable to a DC-X system. I did meet him once at a conference and corresponded with him via email. He would probably recognize my name but I wouldn't say it gives me any leverage in any proposals.
In any case it would seem to me a combination ramjet/rocket engine would not be terribly hard to make; you just close off the intakes during rocket operation. As I recall GW you don't seem to like the idea of a turbojet/ramjet/rocket combo, but since the turbojet/ramjet has already worked with the SR-71 I don't see why adding on the rocket portion would be too difficult since you could just close off the intakes.
This would be especially useful if we could apply it to a lifting body or winged-body with conformal tanks of comparable weight efficiency to that of cylindrical tanks, as you suggested should be possible.

Bob Clark

Last edited by RGClark (2013-10-12 15:51:48)

JoshNH4H · 2013-10-13 17:26:02

Has anyone consulted the T/W figures for turbojet engines? A quick look at Wikipedia suggests that they're extraordinarily poor: The J-58 Turbojet used on the SR-71 has a thrust-to-weight of about 5. This suggests that accelerations will be pretty low (With a L/D of 10, an extremely generous estimate at high mach (for example the Concorde was 7 and the Space Shuttle gets 1, flying hypersonically)) an engine mass of 10% of wet mass results in an acceleration of 0.4 gs at maximum thrust. For a large craft, that corresponds to a lot of engines.

I've always been a fan of flyback boosters, myself. It just seems much simpler-- Make them big and dumb, and stage around 3 km/s so it's not too hard to get back.

RGClark · 2013-10-13 22:22:44

JoshNH4H wrote:

Has anyone consulted the T/W figures for turbojet engines? A quick look at Wikipedia suggests that they're extraordinarily poor: The J-58 Turbojet used on the SR-71 has a thrust-to-weight of about 5. This suggests that accelerations will be pretty low (With a L/D of 10, an extremely generous estimate at high mach (for example the Concorde was 7 and the Space Shuttle gets 1, flying hypersonically)) an engine mass of 10% of wet mass results in an acceleration of 0.4 gs at maximum thrust. For a large craft, that corresponds to a lot of engines.
I've always been a fan of flyback boosters, myself. It just seems much simpler-- Make them big and dumb, and stage around 3 km/s so it's not too hard to get back.

For horizontal takeoff, a lift/drag ratio of 10 to 1 is not even anything special. See the list of L/D ratios here:

http://en.wikipedia.org/wiki/Lift-to-dr … o#Examples

Based on this, the thrust of the turbojet engines can be as little as 1/20th to 1/30th that of the weight of the vehicle and you can still take off. Remember, you're only using the turbojet at low speeds and low Mach numbers. At higher Mach you're using the ramjet and rocket.
The acceleration would be lower in the turbojet realm but keep in mind the extraordinarily high effective Isp's of turbojets, which can be in the range of 2,000 s to 3,000 s and even higher. This is coming from the fact you don't have to burn oxidizer, which is a big component of the propellent in a rocket.
Then this means that though it might take longer to get up to speed than with a rocket, you don't care because you're fuel usage is so low compared to rockets.

The flyback booster concept is what DARPA is aiming for, purely rocket powered, to get a TSTO system. This is doable and not even particularly hard. However, I don't think a turbojet/ramjet/rocket combined engine is particularly hard either and would result in a SSTO vehicle.

Bob Clark

Last edited by RGClark (2013-10-13 22:32:47)

JoshNH4H · 2013-10-14 23:39:25

At liftoff, that kind of T/W is certainly not special at all. But Terraformer was talking about staging at M3.5-- 1.2 km/s. This is considered hypersonic and the thrust-to-weight ratios of crafts are significantly lower in this region. Please recall that your craft needs to hold itself up from lift at all velocities where it operates, and that optimal wing shapes are significantly different at different speeds. Compare the Concorde to the Boeing 737, and compare the SR-71 to both of these. They look different, of course, and the T/W changes as you optimize and design to withstand different speeds. The point is that a L/D of 10 is really pushing it. This publication suggests that at M3 the SR-71 has a L/D of 6. Also remember that your effective Isp goes down as speed goes up, and that if your L/D is 5 and your T/W is .4, you are effectively halving your Vex because half of your thrust power goes to air drag.

GW Johnson · 2013-10-15 06:31:07

Here's my take on these questions: (1) most airliners are sized roughly to takeoff T/W 0.33, because it makes for a practical airplane with the runways and climbout requirements we have in place. (2) the engines that pushed the SR-71 were T/W > 1, but were not the combined-cycle "air turbo ramjets" that everybody thinks. These had 0-25% bypass air from the 3rd or 4th compressor stage directly to the afterburner, so 75-100% of the air still went through a standard low-bypass turbojet core, which is why they were limited to about Mach 3.5 max, even with superalloys. (3) Most of the combined-cycle engines I have seen are serious compromise designs trying to integrate otherwise incompatible characteristics. You can get thrust (mostly) but efficiency really suffers (badly). They tend to be complicated, heavy, and potentially unreliable. Doing parallel-burn with separate engines gets you the same or better thrust, much better efficiency, the same or lower weight, and much better potential reliability. Plus, you have the flexibility of burning both types simultaneously.

GW

GW Johnson · 2013-10-27 22:29:16

Hi guys, I recently completed a look at a launch-rail-launched craft using an integral rocket-ramjet (IRR) booster and a LOX-LH2 second stage, pushing a minimal 2-crew capsule to LEO. The IRR is basically ASALM-PTV technology that I got to work on decades ago. Stage point for this cluster vehicle is M6 at 60,000 feet.

This one is not horizontal takeoff, but it's not exactly vertical-launch, either. I had to pull some very interesting "tricks" to get this to work out, but it does look doable, if somewhat startling. Posted over at http://exrocketman.blogspot.com, as the 10-27-13 article. It's fairly complete and comprehensive.

Enjoy --

GW

RGClark · 2013-11-02 17:21:37

Lockheed wants to develop a hypersonic follow on to the SR-71:

Exclusive: Skunk Works Reveals SR-71 Successor Plan
By Guy Norris ****@aviationweek.com
Source: AWIN First
http://www.aviationweek.com/Article.asp … 31.xml&p=1

Bob Clark

GW Johnson · 2013-11-02 20:34:54

For a different reason, I had to rough out essentially the same plane circa 1985. Mine was turbojet/ramjet parallel-burn propulsion, a similar layout, and designed for M5 at 100-150 kft. I did it from all open sources.

It was so close "to reality" the FBI confiscated all my design notes, but not my sources or my slide rule, because I did not possess the clearances "to know about such a craft". I have often wondered if such a thing ever got built.

It appears in hindsight apparently not. Although it could have been. The delta-wing pulse detonation experimental craft (seen above Groom Lake) of about 1995 apparently led nowhere.

GW

Terraformer · 2013-11-03 03:46:31

It should be clear why Canada is looking attractive as a place to develop new spacecraft... close enough to America to grab their talent, but outside it so their government doesn't screw things up. Plus, plenty of open wilderness, so getting a permit to launch would perhaps be easier?

New Mars Forums

Announcement

#251 2013-09-25 13:56:31

Re: Reusable Rockets to Orbit

#252 2013-09-26 10:32:07

Re: Reusable Rockets to Orbit

#253 2013-09-27 07:20:17

Re: Reusable Rockets to Orbit

#254 2013-09-27 09:34:00

Re: Reusable Rockets to Orbit

#255 2013-09-29 12:26:57

Re: Reusable Rockets to Orbit

#256 2013-09-29 15:39:40

Re: Reusable Rockets to Orbit

#257 2013-09-30 06:11:59

Re: Reusable Rockets to Orbit

#258 2013-10-02 09:36:40

Re: Reusable Rockets to Orbit

#259 2013-10-02 10:13:16

Re: Reusable Rockets to Orbit

#260 2013-10-02 17:34:07

Re: Reusable Rockets to Orbit

#261 2013-10-10 10:37:40

Re: Reusable Rockets to Orbit

#262 2013-10-10 11:29:44

Re: Reusable Rockets to Orbit

#263 2013-10-10 15:47:00

Re: Reusable Rockets to Orbit

#264 2013-10-11 02:33:24

Re: Reusable Rockets to Orbit

#265 2013-10-11 08:44:08

Re: Reusable Rockets to Orbit

#266 2013-10-11 09:38:10

Re: Reusable Rockets to Orbit

#267 2013-10-12 15:30:01

Re: Reusable Rockets to Orbit

#268 2013-10-13 17:26:02

Re: Reusable Rockets to Orbit

#269 2013-10-13 22:22:44

Re: Reusable Rockets to Orbit

#270 2013-10-14 23:39:25

Re: Reusable Rockets to Orbit

#271 2013-10-15 06:31:07

Re: Reusable Rockets to Orbit

#272 2013-10-27 22:29:16

Re: Reusable Rockets to Orbit

#273 2013-11-02 17:21:37

Re: Reusable Rockets to Orbit

#274 2013-11-02 20:34:54

Re: Reusable Rockets to Orbit

#275 2013-11-03 03:46:31

Re: Reusable Rockets to Orbit

Board footer