Page 1 of 2

### Landing rockets upright is unnecessary?

Posted: Thu Jan 21, 2016 8:40 pm UTC
First of all, I'm no rocket scientist, so I apologize for any ignorance I express. Lately, there's been a lot of news about relaunchable rockets, and this got me thinking.

The problem of landing a rocket upright is ridiculously difficult. This is due to the fact that rockets have long, cylindrical shapes, and the upright position does not minimize gravitational potential energy. It doesn't take much to have the rocket tip over at touchdown.
So why not bring the rocket down horizontally? Structures on the landing site can lift the rocket back into upright position when ready to launch. Intuitively, this seems much easier than wasting all the fuel and computational power with the balancing act. We simply have to put all effort into reducing the acceleration of the falling rocket and worry about orientation after touchdown.

### Re: Landing rockets upright is unnecessary?

Posted: Thu Jan 21, 2016 8:49 pm UTC
Cradarc wrote:The problem of landing a rocket upright is ridiculously difficult. This is due to the fact that rockets have long, cylindrical shapes, and the upright position does not minimize gravitational potential energy. It doesn't take much to have the rocket tip over at touchdown.
So why not bring the rocket down horizontally? Structures on the landing site can lift the rocket back into upright position when ready to launch. Intuitively, this seems much easier than wasting all the fuel and computational power with the balancing act. We simply have to put all effort into reducing the acceleration of the falling rocket and worry about orientation after touchdown.

How would you propose getting the rocket body to the ground in a horizontal orientation when your engine's thrust vector is pointed along the rocket body's axis?

### Re: Landing rockets upright is unnecessary?

Posted: Thu Jan 21, 2016 9:06 pm UTC
A possibilities come to mind. To land gently enough to prevent damage, you pretty much have to use thrust. parachute landings are a not-so-gentle affair. If you are going to use thrust, your thrust needs to line up with your center of mass, or you will go tumbling around. Landing horizontally also means the structure of the rocket has to be able to support its own weight while lying on its side. This isn't nearly as much weight over any given patch of surface as when upright, since the rocket is tall and skinny, but it's still making extra demands of the structure that's been through a lot of stresses, which the engineers might wish to avoid, if only to save weight.

Something with control surfaces that can land like the Space Shuttle is conceivable, but probably awfully expensive and time consuming to develop.

Landing vertically is a lot better, if it can be done reliably enough. There's going to be some trial and error involved in developing the technology.

I wonder about horizontal ocean landings, with huge gas bladders or the like for buoyancy/shock absorption. But I'm not a rocket scientist either

### Re: Landing rockets upright is unnecessary?

Posted: Thu Jan 21, 2016 9:20 pm UTC
thoughtfully wrote:I wonder about horizontal ocean landings, with huge gas bladders or the like for buoyancy/shock absorption. But I'm not a rocket scientist either

Saltwater.

Noap.

### Re: Landing rockets upright is unnecessary?

Posted: Thu Jan 21, 2016 11:02 pm UTC
sevenperforce wrote:How would you propose getting the rocket body to the ground in a horizontal orientation when your engine's thrust vector is pointed along the rocket body's axis?

You don't purposefully rotate it. You simply focus on reducing its acceleration and if it ends up not upright, oh well.
Attach parachutes to the nose and blast the engine. Sure, there will be some tumbling, but the chutes should keep the direction of thrust pointing mostly upwards. Alternatively, conservation of angular momentum can also be used. Have two small boosters on the side that causes the rocket to rotate about the center axis. The rocket body would precess about the vertical.
Have the rocket fall into a semi-elastic tarp which gives way at some threshold, dropping it onto a cushion. At sea, the tarp can seal the rocket as it plunges into the ocean, and then reeled back up.

The idea is to shift more of the work to the platform on the ground rather than the airborne rocket. The rocket should only be responsible for losing as much kinetic energy as possible.

### Re: Landing rockets upright is unnecessary?

Posted: Fri Jan 22, 2016 12:03 am UTC
sevenperforce wrote:How would you propose getting the rocket body to the ground in a horizontal orientation when your engine's thrust vector is pointed along the rocket body's axis?

You don't purposefully rotate it. You simply focus on reducing its acceleration and if it ends up not upright, oh well.

Oh well, indeed.
Notice that the rocket did land vertically, and only the impact of tipping slightly was enough to explode.

### Re: Landing rockets upright is unnecessary?

Posted: Fri Jan 22, 2016 12:51 am UTC
Cylinders are indeed generally better at taking axial loads than lateral ones. Mitigating it by making the impact not be applied as a point load could help, but between the shape's natural strength axially and the fact that aerospace composites are often much stronger at taking loads that can be borne by the faces rather than the core, there'd...be a fairly massive redesign required to get anything approximating the strength to take a horizontal landing. A lot of spaceborne equipment can't even support its own weight in anything other than launch configuration, let alone take the kind of G-forces from a landing, but just to be able to take off you've got to be able to handle some pretty impressive axial loads already.

### Re: Landing rockets upright is unnecessary?

Posted: Fri Jan 22, 2016 12:59 am UTC
brenok wrote:Notice that the rocket did land vertically, and only the impact of tipping slightly was enough to explode.

Yes, because hard concrete. There was no preparation made for the tip, so if the rocket fails to stay upright it's screwed.
Cradarc wrote:Have the rocket fall into a semi-elastic tarp which gives way at some threshold, dropping it onto a cushion. At sea, the tarp can seal the rocket as it plunges into the ocean, and then reeled back up.

DaBigCheez wrote:A lot of spaceborne equipment can't even support its own weight in anything other than launch configuration

That's incredibly scary if what you're saying is true. If the rocket were to be placed on its side, the hull will just burst open?

### Re: Landing rockets upright is unnecessary?

Posted: Fri Jan 22, 2016 2:11 am UTC
DaBigCheez wrote:Cylinders are indeed generally better at taking axial loads than lateral ones

This.

Simple experiment. Take a piece of paper. Roll it up into a cylinder and tape the edges together. Stand it on its end. No problem, right? Now set it on its side. Notice how it collapses. Rockets collapsing is a Bad Thing(tm).

It's of course possible to make them stronger. But orbital rockets run on the very margins of the possible. The Falcon 9 would collapse like a crushed can on takeoff and landing if it didn't have internal pressurant resisting that. Some other rockets are so weak that you can't even transport them on the ground without pressurizing them first. They're basically big thin-skinned ballons. There's no tanks in there, the skin itself is the tank - with just a thin bulkhead separating the fuel and oxidizer.

Any need for extra reinforcement is a huge hit to your rocket. Rockets bear loads well in only one direction - vertical. The thrust axis is also vertical. So that's how you takeoff and land, at least concerning the gigantic first stage, and usually all stages.

### Re: Landing rockets upright is unnecessary?

Posted: Fri Jan 22, 2016 6:52 am UTC
Yes--a lot of rockets have walls that are literally only as thick as those of a soda can. That is how you can get a rocket stage that weighs only five percent of its fuel weight even WITH the engines and electronics included.

As for internal pressure reinforcing a tank, we can look again to the soda can analogy. A full can with gas pressure inside it has a lot of resistance--a grown man can put his entire weight on it--at least in the axial orientation. However, when the can is empty, you can crush it in your fingers with almost no effort.

### Re: Landing rockets upright is unnecessary?

Posted: Fri Jan 22, 2016 2:20 pm UTC
KarenRei wrote:The Falcon 9 would collapse like a crushed can on takeoff and landing if it didn't have internal pressurant resisting that.

I have 5 questions that I the answer to really quickly.

1: How much does a Falcon 9 cost?
2: Can a Falcon 9 launch and/or land without people inside of it?
3: Does a device that instantly releases all of the pressure inside of a spaceship exist?
4: How much gasoline, TNT and/or nitroglycerin can a Falcon 9 carry?
5: How much would you pay for front row seats to the most awesome explosion implosion you will ever see?

### Re: Landing rockets upright is unnecessary?

Posted: Fri Jan 22, 2016 5:51 pm UTC
jewish_scientist wrote:
KarenRei wrote:The Falcon 9 would collapse like a crushed can on takeoff and landing if it didn't have internal pressurant resisting that.

I have 5 questions that I the answer to really quickly.

1: How much does a Falcon 9 cost?
2: Can a Falcon 9 launch and/or land without people inside of it?
3: Does a device that instantly releases all of the pressure inside of a spaceship exist?
4: How much gasoline, TNT and/or nitroglycerin can a Falcon 9 carry?
5: How much would you pay for front row seats to the most awesome explosion implosion you will ever see?

You mean, doing an armed test of the Falcon 9 Anti-Ship Missile?

It's not an easy question to answer because you have to twerk the landing sequence, it's designed to come in empty. I'm not sure how well it could maintain orientation with a large payload. But in terms of the amount of payload that first stage is designed to be able to carry during a launch, it's 110 tonnes (not counting its own mass or the mass of its propellant). Cost without the upper stages would probably be in the ballpark of \$60m.

Launching it with an actual warhead would prove all of their anti-ship tests thusfar well worth it. How meaningful is a 110 tonne warhead? I'll put it this way: The warhead on the infamous Exocet missile used by the Argentinians in the Falkands War was only 165kg

### Re: Landing rockets upright is unnecessary?

Posted: Fri Jan 22, 2016 7:21 pm UTC
Wow, that's mind-blowing. From the sound of things, most rocket boosters are no more durable than a homemade soda bottle rocket. The success rate of rocket launches is remarkable.

### Re: Landing rockets upright is unnecessary?

Posted: Fri Jan 22, 2016 10:19 pm UTC
Rockets are literally designed to only withstand 30-50% more stress than an ideal mission would produce--the engineers cut the margins as close as they dare. For comparison, static structures like buildings and bridges are usually built to withstand five or more times the expected stress.

### Re: Landing rockets upright is unnecessary?

Posted: Sat Jan 23, 2016 10:39 am UTC
For comparison, static structures like buildings and bridges are usually built to withstand five or more times the expected stress.

Steel bridges are typically designed for something like 1.2 times the permanent loads, plus 1.5 times the expected "live" loads. With (large scale) plastic deformation as material limit. The strength of concrete gets derated by a factor 1.5 compared to its nominal strength, to account for the variety between batches. Details may vary, but nothing like a factor 5.

You might encounter such a high factor in some places in a code, then it is usually to cover cases without extensive analysis. Like, if you want to put piles in a river bed without soil analysis or tests from that particular location, you can use a standard value for their load bearing capacity but divided by a factor 4. Or you can avoid a metal fatigue calculation, if you keep stresses below 0.3 times the yield strength.

### Re: Landing rockets upright is unnecessary?

Posted: Sat Jan 23, 2016 2:29 pm UTC
Ok, so I overestimated typical averages. Anyway, rockets are designed with the smallest margins possible that still avoid an unacceptably high failure rate.

### Re: Landing rockets upright is unnecessary?

Posted: Sat Jan 23, 2016 7:01 pm UTC
First of all, the rocket is coming down 99.99% of the way tail-first (all the weight is in the tail). The thing (assuming it goes either to the limit of space like Blue Origin or well past like the Falcon9) is going to hit the atmosphere well above the speed of sound, and won't drop to subsonic for quite some time (note you have to light the retro-rockets to do this).

As mentioned above, rockets really can't handle being sideways very well. Even if they did, it would take complicated and (multiple) heavy landing legs to support the thing sideways, and lots of parachutes (because you need so many points of support). Even without the structural issues, landing sideways is a pain.

Case in point: I've recently tried this in KSP (Kerbal Space Program, for those who missed Randal's shoutouts to this great game). Adding all the landing wheels (landing legs would point funny and *should* be badly aerodynamic at such angles) and parachutes made the rocket asymmetrical. Not a good idea for keeping a rocket on target at hypersonic speeds. Even then, trying to get everything to line up and land seemed a waste of time. My more recent strategy for landing "too tall" rockets is to dump them in the ocean (Kerbin seems to have salt-free oceans for KSP>1.0.2 or so).

### Re: Landing rockets upright is unnecessary?

Posted: Sun Jan 24, 2016 11:53 am UTC
jewish_scientist wrote:
KarenRei wrote:The Falcon 9 would collapse like a crushed can on takeoff and landing if it didn't have internal pressurant resisting that.

I have 5 questions that I the answer to really quickly.

1: How much does a Falcon 9 cost?
2: Can a Falcon 9 launch and/or land without people inside of it?
3: Does a device that instantly releases all of the pressure inside of a spaceship exist?
4: How much gasoline, TNT and/or nitroglycerin can a Falcon 9 carry?
5: How much would you pay for front row seats to the most awesome explosion implosion you will ever see?

1. About 60M\$ By far the cheapest of its class.

2. It is not yet allowed to launch with people, and have never done so. That is changing soon though (well, there'll never be people inside the Falcon 9 during flight, they will be in the Dragon spaceship on top of it).

3. Assuming you mean the rocket booster and not a space ship, yes. That is called the flight termination system. If something goes wrong during flight, explosive charges open up the tanks so the rocket explodes, rather than risk having it fly into inhabited areas.

4. All of the fuel in the rocket is basically gasoline. You could replace the second stage with a huge chunk of TNT if you wanted to. The propellant of a regular second stage would release more energy on malfunction than the TNT, although TNT would make a sharper shockwave. Unstabilized nitroglycerine would be unlikely to get of the ground.

5. If I am witnessing a rocket launch, I much prefer that it is successful, although failures does look spectacular.

### Re: Landing rockets upright is unnecessary?

Posted: Mon Jan 25, 2016 3:32 pm UTC
Partly on the topic of landing sideways....

I've occasionally given some thought to the use of ducted rockets as a means of altitude compensation and increased Isp. Adding ducts to existing rocket design costs more in weight penalty than it adds in efficiency, but this wouldn't be the case if you designed the entire craft around the airflow system.

Specifically, this sort of design would be most efficient if it changed its angle of attack as it accelerated, so that the cross-section changed in shape to match the airflow with respect to speed. Thus, horizontally-oriented vertical takeoff and landing would be the default.

### Re: Landing rockets upright is unnecessary?

Posted: Mon Jan 25, 2016 4:00 pm UTC
sevenperforce wrote:Partly on the topic of landing sideways....

I've occasionally given some thought to the use of ducted rockets as a means of altitude compensation and increased Isp. Adding ducts to existing rocket design costs more in weight penalty than it adds in efficiency, but this wouldn't be the case if you designed the entire craft around the airflow system.

Specifically, this sort of design would be most efficient if it changed its angle of attack as it accelerated, so that the cross-section changed in shape to match the airflow with respect to speed. Thus, horizontally-oriented vertical takeoff and landing would be the default.

What do you mean by "ducts"? Do you mean an expanding nozzle?

https://en.wikipedia.org/wiki/Expanding_nozzle

I can't comment any more on your idea until I'm clearer on what you mean - for example, your talk of angles of attack sounds like you're talking about either bow shock (which is of some, but limited, relevance) or oxygen breathing (totally different topic from altitude compensation). Let it suffice to say that there's been lots of work in theory, and some attempts in practice, to make altitude adjusting nozzles (either automatic compensation, such as aerospikes, or with physical changes in nozzle shape)... many have shown to not generally be worth the effort; some may or may not (such as aerospikes), and it's still yet to be seen.

### Re: Landing rockets upright is unnecessary?

Posted: Mon Jan 25, 2016 4:48 pm UTC
KarenRei wrote:What do you mean by "ducts"? Do you mean an expanding nozzle?

No, I mean a duct around the exterior of the engine that allows the exhaust to mix with an external airflow, reducing the effective exhaust velocity but increasing the mass flow without significant penalty. The thermal energy in the exhaust that would typically be lost is used to heat and expand the airflow. It's not quite the same as a ramjet, because a ramjet depends on using external air as a source of oxygen for combustion, while a ducted rocket (also known as an air-augmented rocket) uses the external air as added reaction mass.

Ducted rockets have been used on solid- and hybrid-fueled air-to-air missiles with great effect, but there hasn't been much success with putting ducts on orbital boosters, simply because the added weight and drag isn't really worth it. You can simply make the same rocket a little bigger for a lot less cost and complexity. But if your goal is durable reusability, a little added weight and complexity for an increase in Isp is a more attractive tradeoff.

Note that a ducted rocket is not an airbreathing engine per se, because it does not rely on air for combustion, only for augmentation of reaction mass. This means it can continue to function well beyond the range at which combustion would become supersonic and will operate outside the atmosphere normally; that's the altitude-compensating aspect.

Change in angle of attack would be used to change the length of the airflow through the engine duct, so that the airflow path is always matched to the airflow speed.

### Re: Landing rockets upright is unnecessary?

Posted: Mon Jan 25, 2016 5:23 pm UTC
sevenperforce wrote:
KarenRei wrote:What do you mean by "ducts"? Do you mean an expanding nozzle?

No, I mean a duct around the exterior of the engine that allows the exhaust to mix with an external airflow, reducing the effective exhaust velocity but increasing the mass flow without significant penalty. The thermal energy in the exhaust that would typically be lost is used to heat and expand the airflow. It's not quite the same as a ramjet, because a ramjet depends on using external air as a source of oxygen for combustion, while a ducted rocket (also known as an air-augmented rocket) uses the external air as added reaction mass.

Very interesting, I'm rather amazed that I haven't stumbled across this one yet. I kind of like it in regards to the fact that it would greatly help with metalized fuels (aluminum, lithium), which require hydrogen that doesn't contribute to the reaction to function as the working gas (since the oxides condense out of the stream at high temperatures, particularly aluminum, and because you can't make use of the heat of condensation without a working gas). They're quite high ISP even with need for the hydrogen; they'd gain hugely from this (not even to mention the (limited) airbreathing potential). And it'd work in other planetary atmospheres to boot, no need for oxygen - for example, a Mars return stage

Still, that's one giant, heavy nozzle :Þ Not as heavy as with a full airbreather, but still. And the rocket equation doesn't play nicely with high dry mass figures. Nonetheless, it's a quite interesting concept.

### Re: Landing rockets upright is unnecessary?

Posted: Mon Jan 25, 2016 7:08 pm UTC
KarenRei wrote:I'm rather amazed that I haven't stumbled across this one yet. I kind of like it in regards to the fact that it would greatly help with metalized fuels (aluminum, lithium), which require hydrogen that doesn't contribute to the reaction to function as the working gas (since the oxides condense out of the stream at high temperatures, particularly aluminum, and because you can't make use of the heat of condensation without a working gas). They're quite high ISP even with need for the hydrogen; they'd gain hugely from this (not even to mention the (limited) airbreathing potential). And it'd work in other planetary atmospheres to boot, no need for oxygen - for example, a Mars return stage

Still, that's one giant, heavy nozzle :Þ Not as heavy as with a full airbreather, but still. And the rocket equation doesn't play nicely with high dry mass figures. Nonetheless, it's a quite interesting concept.

Yeah, an air-augmented design really opens up a lot of additional possibilities regarding chemistry, like hypergolics and fuels with much higher energy density. You can't beat the specific energy of liquid hydrogen in a conventional rocket, but it has extremely crappy energy density, leading to monstrosities like the Shuttle's external tank.

The weight of the nozzle and the added ducting is unpleasant, but there haven't been many significant explorations of this concept as an integral design. Tons and tons of research has been poured into integrated designs for airbreathers, mostly ramjet and scramjets, but I've only the barest conceptual designs for integrated air-augmented rockets, and no designs for orbital boosters using such a system.

Simply slapping a shroud on the outside of a Falcon 9 wouldn't work; you'd have to have a craft designed for it from the ground up. Such a craft would require a specific asymmetric shape in order to have an angle of attack which changed with airspeed while hopefully also taking some advantage of aerodynamic lift. It offers a lot of advantages, though.

### Re: Landing rockets upright is unnecessary?

Posted: Mon Jan 25, 2016 8:07 pm UTC
You'd almost need a shape like this:

3-ellipsoid-section.png (14.18 KiB) Viewed 8284 times

The inside would have very small hypergolic rocket nozzles placed around the perimeter of each half-ellipsoid and angled inward; depending on which nozzles were activated, different flow paths through the gap could be chosen.

### Re: Landing rockets upright is unnecessary?

Posted: Mon Jan 25, 2016 9:14 pm UTC
I like digging up old threads. I'm sure seven won't mind.

### Re: Landing rockets upright is unnecessary?

Posted: Mon Jan 25, 2016 9:35 pm UTC
thoughtfully wrote:I like digging up old threads. I'm sure seven won't mind.

Guilty as charged.

### Re: Landing rockets upright is unnecessary?

Posted: Mon Jan 25, 2016 11:27 pm UTC
sevenperforce wrote:The weight of the nozzle and the added ducting is unpleasant, but there haven't been many significant explorations of this concept as an integral design.

I can't help but wonder if materials technology might deliver this one to us some day (suitable reductions in nozzle weight). For example, ceramic matrix composites look really promising - very high strength to weight ratios, but unlike plastic-based composites like CFRP they're tolerant of extreme temperatures and oxidative environments. They're mainly looking at them for head shields now, but it's probably directly applicable to applications like this. Technical ceramics in general are a field advancing faster than metallurgy.

### Re: Landing rockets upright is unnecessary?

Posted: Tue Jan 26, 2016 10:01 pm UTC
Yeah, there would definitely have to be some advances in materials science for this sort of a design to be realized. An integrated engine is a blessing and a curse in a lot of ways. With a lot more area exposed to high temperatures, you're dealing with a significant weight penalty at the outset, but you have more space to work with for airbreathing or air-augmented designs.

(And yes, thoughfully, I'm resurrecting that thread...but it's not the WORST thing, I don't think, because that whole thread was more a learning experience than a single proposal.)

The ideal SSTO-capable spacecraft could function as an induced-flow airbreather from standstill to transonic speeds, as a partially airbreathing ramrocket from transonic to hypersonic speeds, as a supersonic-flow air-augmented rocket from hypersonic speeds to ultrasonic speeds, and as a pure rocket from ultrasonic speeds into orbit. Simultaneously, you would want to have altitude-adaptive nozzles. Proposed or constructed designs include:
• Vehicles equipped with a single engine using moving parts (e.g., a leading spike or bleed flaps) to switch between discrete modes depending on airspeed
• Vehicles equipped with a single engine using moving parts to change the shape of the intake and engine chamber to transition between modes of oepration
• Vehicles equipped with multiple engines sharing a single flow path (e.g., an afterburner)
• Vehicles equipped with multiple engines in parallel flow paths
• Some combination of the above
The SABRE engine on the Skylon is an example of combining several of these -- a leading spike which starts open to allow airbreathing and then closes to convert to pure rocket operation, as well as the ramjets around the flow path of the central engine. The SR-71 Blackbird's J58 engine had a similar design, where the spike was used along with bleed flaps to shift between turbofan, turbojet, afterburner, and ramjet afterburner configurations. I've also seen examples of multiple engines in parallel flow paths, like the trijet proposed for the SR-72.

The ideal/perfect SSTO engine would be a single engine with a single intake and nozzle which simply changed its entire chamber shape to seamlessly and smoothly transition between each mode of operation. Unfortunately, Transformers aren't real, so that's never going to happen.

However, if an integrated-engine SSTO craft could be designed which changed its pitch with respect to its speed, thus smoothly altering the flow path without any moving parts whatsoever, this could be realized.

### Re: Landing rockets upright is unnecessary?

Posted: Tue Jan 26, 2016 10:48 pm UTC
Instead of a cylinder, what if the rocket was shaped like a D4? That'd make it stable no matter which way it landed. I'm sure there are 125 great reasons why this won't work.

I imagine #1 is explosion.

### Re: Landing rockets upright is unnecessary?

Posted: Wed Jan 27, 2016 1:51 am UTC
sevenperforce wrote:The ideal SSTO-capable spacecraft could function as an induced-flow airbreather from standstill to transonic speeds, as a partially airbreathing ramrocket from transonic to hypersonic speeds, as a supersonic-flow air-augmented rocket from hypersonic speeds to ultrasonic speeds, and as a pure rocket from ultrasonic speeds into orbit. Simultaneously, you would want to have altitude-adaptive nozzles. Proposed or constructed designs include:
• Vehicles equipped with a single engine using moving parts (e.g., a leading spike or bleed flaps) to switch between discrete modes depending on airspeed
• Vehicles equipped with a single engine using moving parts to change the shape of the intake and engine chamber to transition between modes of oepration
• Vehicles equipped with multiple engines sharing a single flow path (e.g., an afterburner)
• Vehicles equipped with multiple engines in parallel flow paths
• Some combination of the above
The SABRE engine on the Skylon is an example of combining several of these -- a leading spike which starts open to allow airbreathing and then closes to convert to pure rocket operation, as well as the ramjets around the flow path of the central engine. The SR-71 Blackbird's J58 engine had a similar design, where the spike was used along with bleed flaps to shift between turbofan, turbojet, afterburner, and ramjet afterburner configurations. I've also seen examples of multiple engines in parallel flow paths, like the trijet proposed for the SR-72.

The ideal/perfect SSTO engine would be a single engine with a single intake and nozzle which simply changed its entire chamber shape to seamlessly and smoothly transition between each mode of operation. Unfortunately, Transformers aren't real, so that's never going to happen.

However, if an integrated-engine SSTO craft could be designed which changed its pitch with respect to its speed, thus smoothly altering the flow path without any moving parts whatsoever, this could be realized.

The biggest concerns I have about projects like Skylon is the capital costs. So much can go wrong on high capital cost, reusability-is-essential projects, especially where you're trying to break many new technological barriers at once. Still, I do think there's real potential for the future, if we can only start redirecting basic research budgets (aka, NASA and similar) from building unoriginal giant rockets that will surely be overpriced.

That said, I actually have my own preferred launch mechanism which I'm working on a CFD simulator for (based on OpenFOAM) - I'm doing it as a simulation rather than a physical test rocket because I don't exactly have the budget to start my own space program at this point in time . It's a caseless rocket wherein the shape of the fuel assembly itself (aka continuuously extruded honeycomb aluminum with the channels thickly lined with paraffin wax and cored with silica-impregnanted open cell polyurethane foam to retain LOX) also doubles as its own expansion nozzle and maintains its shape as it burns down. Aka, the booster mass at burnout is approximately zero; the rocket equation would adore it, yet Al+paraffin+LOX is a nonetheless a very dense, rather high ISP fuel (well higher than LOX/kerosene). And continuous extrusion of (very) cheap materials, like churning out pipe segments... how much cheaper can you get on production than that? Crazy structural strength too (many tonnes of honeycomb aluminum) - you could use them as bridge pylons if you didn't want to use them as rockets. And completely harmless until you fill the channels with LOX - no harder to transport than pipe segments. If the ends are sealed it'll float, too, until you fill it with LOX. If you have enough air in a saltwater-tolerant sealed payload housing at the top of the booster and then fill the booster with LOX, it'd right itself like a spar platform, giving you a zero-cost any-location "launch pad" with minimal transport costs to get there (rockets work fine underwater, so long as they can withstand the pressures... in this case it could even be a boost, as aluminum burns energetically with water). A couple days of saltwater exposure to aluminum doesn't corrode even the most corrosion-prone alloys, there's no expensive hardware to ruin with saltwater, and you've got structural strength to spare.

The main downsides to the approach that I see are: 1) basically zero possibility for self-destruct in the event of a rocket going haywire (seriously, how would you stop many tonnes of honeycomb aluminum moving at tremendous speed? A little explosive charge to pop the tanks as is currently done isn't going to do the trick), 2) No gimballing, unless you want to try to gimbal the entire booster; it would require some combination of aerodynamic surfaces and use of the payload's OMS jets to make trajectory adjustments; and 3) the structural geometry is really tricky, which is why I have to set it up as a optimization problem in the simulation. And I'm only going to trust its conclusions on the large-scale thermodynamics of the exhaust and transport properties - aluminum combustion is very difficult to model in the best of circumstances, and "aluminum partially protected from the combustion by the high specific heat of its wax coating, which then burns off, allowing the aluminum to simultaneously vaporize, elongate/deform, and break into droplets" isn't exactly the best of circumstances. If the large scale simulations seem to produce something promising then maybe I'll scrounge up the money for small-scale testing to better fill in that piece of the puzzle. But one thing is sure about the propellant combination: even in totally-uncontrolled, extremely low pressure, low surface area, totally-not-stoichiometric-ratio situations.... aluminum+paraffin+lox [url src="https://youtu.be/uPK_rSf1WUc?t=24"]most definitely does burn[/url]. In an optimal environment I have no doubt that one could get near a 100% burn with a sufficient reaction rate.

All of that said... as much as I like my "pet project", though, I'm not blind to the potential of other more "serious" semi-traditional approaches being taken, by people with actual resources and far more experience than I . And increasing use of atmospheric gases to augment thrust sure seems to be a direction worth pursuing.

### Re: Landing rockets upright is unnecessary?

Posted: Wed Jan 27, 2016 4:10 pm UTC
That's...dizzingly detailed. But quite interesting.

I supposed you'd have to coat the exterior with some very healthy combustion inhibitor, because you really don't want to somehow ignite the thing from the outside during flight. That would be bad, and you would not go to space today.

The most promising characteristics -- cheap manufacture and burn-itself-away operation -- unfortunately also throw a wrench into the scenario. Because it burns from the bottom, you can't really use it as a strap-on booster. So you're limited to using it as a first stage, and that's problematic because the very aspect that makes it cheap to manufacture also means it has to be constructed completely at the manufacturing plant, rather than being constructed in pieces and assembled at the launch pad. This means your booster size is limited to what you can safely transport. The Falcon 9v1.1 Full Thrust is just about at this limit, so you aren't going to be able to scale it up any further that around there.

Of course, this size may be all you need, due to the high density of aluminum-paraffin-LOX.

Control is a big issue with any solid-fueled booster, but much more in this case. Hybrid rockets allow you to control burn rate by adjusting oxidant flow, but that's not possible here because the oxidant is already inside the channels. Plus, with no nozzle, you can't gimbal. You can always add an endcap to your booster with empty channels pointed in the radial direction rather than the axial direction, with a small LOX tank in the center, so you end up with a hybrid version of your design functioning as your OMS:

Al-LOX.png (25.92 KiB) Viewed 7983 times

You'd probably need a hypergolic additive for your oxidizer, and restart would be tricky, but it's not too terribly complicated.

The main problem there is that your center of mass is changing dramatically as your booster burns away, meaning that thrusting with this OMS would not necessarily produce the same change in pitch early in the flight as it would later in the flight:

Perhaps a better approach would be to do what the Navy did early-on when they had solid-fueled rocket nozzles on missiles that they didn't want to try and gimbal. They had pretty good success at thrust vectoring by injecting a fluid (sometimes fuel, sometimes bleed air) into the propellant stream halfway down the nozzle, thus causing an axial asymmetry that altered the shape and thrust vector of the main exhaust plume:

fluid injection vectoring.png (52.04 KiB) Viewed 7983 times

Obviously, your nozzle burns itself away, so you can't do exactly this. What you could do, however, is leave several of the honeycomb channels around the perimeter empty, and connect them to a small tank of LOX at the top with pressure/flow control. By increasing the flow to one of the channels, you could produce the same effect:

internal thrust vectoring hybrid.png (4.73 KiB) Viewed 7983 times

You'd want to make sure it didn't actually change the shape of the burn surface, but other than that, it ought to work.

As far as air-augmentation goes...I'm interested in seeing what the limits of induced-flow rocket air-augmentation really would be. I'm imagining a launch vehicle with a HUGE cylindrical cross-section and flow path which makes up for its high cross-sectional area by sucking vast amounts of air through...basically the turbofan equivalent of a rocket. We use the atmosphere as the primary reaction mass for re-entry; why not do the same thing for launch as well? Beats trying to coax a vehicle into giving us useful hypersonic lift, that's for sure.

The launch vehicle would end up looking something like a whale shark with a big, gaping mouth....

### Re: Landing rockets upright is unnecessary?

Posted: Thu Jan 28, 2016 10:20 am UTC
I fail to see how this can not be assembled on launch site. You could easily stack them and bolt them together to form your fist stage, or even place them side by side. When assembled tightly the flame will jump from one section to the other without any interruption. Multiple sections will be more expensive, but perhaps there would still be savings wrt normal rocket construction.
As for safety: the multiple sections may be made suitable for separation in case of an emergency. That would increase the surface area immensely, limiting the range. Combined with a sea launch (to keep the close range relatively empty) you may be able to do this safely.

What I question is the certainty that the structure will burn smoothly and will never burn asymmetrically. Any asymmetrical burning, multiplied by the massive force, will create a huge lateral force. You could correct that with the hybrid control module on the top (or a liquid version of it), but there needs to be a lot of capacity in there. Far more than with the customary designs.

### Re: Landing rockets upright is unnecessary?

Posted: Thu Jan 28, 2016 11:14 am UTC
That's a really clever idea about using empty channels for injection of additional fuel or oxidizer in lieu of gimballing - I'm surprised that didn't occur to me, as I'm aware of the historic usage of propellant injection for thrust vectoring. I'll work that into the models and see how well it actually performs.

I really do think you're hitting on something around the concept of rockets with large cylindrical cross sections (or their linear equivalent, parallel planes with a large gap between them). That is, I'm assuming you're thinking of the exhaust coming from the back of the ducting rather than from the central rocket body. You're not only letting lots of air into the space between the ducting and the rocket body, but you also create a mild aerospike effect, as the rocket body drags a low pressure wake behind it in the center during atmospheric operation (the effect would be weakened however by the air entering around it, but that's more than compensated for by the overall increase in rocket efficiency). And in terms of reentry difficulty (if you're thinking of SSTO applications) a key aspect in determining your heat load is the ratio of your mass to your cross section (aka the ballistic coefficient); if your ducting tapers inwards so the air has to compress (aka can't just pass right through), then you've got a large cross section and are basically dragging a parachute whenever you're not burning the rocket. It enlarges your cross section on other axes also (important when just starting reentry, where you want to orient your largest cross sectional axis toward the direction of travel) - yet inside, it's just a void (aka no mass). So it should give a nice low ballistic coefficient, if the outer ducting isn't too heavy. The linear version in particular - from the side it'd present a massive cross sectional area; also, if designed right could present as a wing or lifting body for atmospheric flight. The annular approach would be lighter for a given amount of ducted air, however.

A kind of fun possibility to any scenario where we ever make very long nozzles workable: the "atmosphere" in LEO is predominantly monoatomic oxygen. It's an energetic hypergolic monoprop free for the taking (such as for stationkeeping) - just incredibly sparse. Ram-compress it and it burns. However, the problem with using it is the same as with scramjets, only worse: you're dealing with so-called "frozen reactions", in that the spacecraft is moving so fast that the reaction time is too slow unless you have a very long nozzle; the reaction occurs in the shocks far behind the spacecraft and doesn't contribute to thrust.

### Re: Landing rockets upright is unnecessary?

Posted: Thu Jan 28, 2016 9:21 pm UTC
I'd anticipate two problems with my channel idea...first, reaction control is going to be limited to how fast the pressure differential can travel down the tube, and second, that the differential burn will change the shape of the "nozzle". Both are bad.

I asked this question over on the Space Exploration portal of Stack Exchange and got a pretty good answer. Basically, there's an upper operating limit on air-augmentation, because once the airstream's velocity exceeds your exhaust velocity, you can no longer use your exhaust to push against the airstream, and you're limited to pure rocket operation. That's not a bad thing; it just means air augmentation no longer contributes beyond about 3-4 km/s. As long as your nozzles can gimbal to a straight-down orientation at that speed, you're not losing anything by having a central airflow, especially if you're nearly out of the atmosphere:

gimbal inward cbypass.png (5.19 KiB) Viewed 7842 times

The larger you make the central bypass, the slower you're moving when air augmentation ceases to function. However, since the amount of air you're able to actually suck through (as opposed to what passes through without interaction) decreases as the size of your bypass increases, this will probably sort itself out.

I do like the idea of a linear approach. I really want capacity for horizontally-oriented VTOL, for loading/unloading/fueling reasons, but that might be asking too much.

I wonder if a really really large/long space station could ever take advantage of the LEO atmosphere....

### Re: Landing rockets upright is unnecessary?

Posted: Fri Jan 29, 2016 9:28 am UTC
sevenperforce wrote:I wonder if a really really large/long space station could ever take advantage of the LEO atmosphere....
Probably, but that doesn't mean it would be more powerfull than a Electrodynamic tether.

### Re: Landing rockets upright is unnecessary?

Posted: Fri Feb 05, 2016 11:28 pm UTC
Zamfir wrote:
For comparison, static structures like buildings and bridges are usually built to withstand five or more times the expected stress.

Steel bridges are typically designed for something like 1.2 times the permanent loads, plus 1.5 times the expected "live" loads. With (large scale) plastic deformation as material limit. The strength of concrete gets derated by a factor 1.5 compared to its nominal strength, to account for the variety between batches. Details may vary, but nothing like a factor 5.

For the record, rockets safety factors are usually more like 1.2 times the live loads, on the propellant tanks, but may be as high as 2 on some components in the engines, depending on the designers. They're running really close to the line, and they're made of aluminium alloys that suffer from cyclic fatigue; so much so that usually pressurising them actually changes the shape slightly and work hardens them; and after a certain number of cycles, for example if the launch is scrubbed six or so times, they have to throw away the tank without flying it.

Steel and concrete bridges don't suffer from cyclic fatigue in the same way, and the safety factor is higher anyway.

### Re: Landing rockets upright is unnecessary?

Posted: Mon Feb 08, 2016 5:00 pm UTC
Earth is...interesting. Our combination of atmosphere, mass, and surface gravity makes it just barely possible to get into orbit. Any harder and it would be prohibitively difficult to get to orbit; any easier and reusable SSTOs would be a snap.

Oh well; gives us a reason to push engineering to its limits!

### Re: Landing rockets upright is unnecessary?

Posted: Mon Feb 08, 2016 9:49 pm UTC
I think what some people are suggesting is essentially a scramjet/ramjet where the flame-holder is a rocket engine that provides the initial startup thrust to get it up to ramjet operational speeds then throttles back as the ramjet portion starts providing more thrust.

### Re: Landing rockets upright is unnecessary?

Posted: Tue Feb 09, 2016 2:28 pm UTC
Sockmonkey wrote:I think what some people are suggesting is essentially a scramjet/ramjet where the flame-holder is a rocket engine that provides the initial startup thrust to get it up to ramjet operational speeds then throttles back as the ramjet portion starts providing more thrust.

Well in that instance it would be more about throttling to fuel-rich operation than throttling back per se.

But what I was suggesting isn't a combusting airbreather at all; it's a stovepipe engine with an inside-out aerospike engine that uses the central airflow as a working mass to increase thrust without increasing fuel consumption.

### Re: Landing rockets upright is unnecessary?

Posted: Tue Feb 09, 2016 4:40 pm UTC
Sockmonkey wrote:I think what some people are suggesting is essentially a scramjet/ramjet where the flame-holder is a rocket engine that provides the initial startup thrust to get it up to ramjet operational speeds then throttles back as the ramjet portion starts providing more thrust.

Not exactly. The concept presented doesn't inherently have to deal with any combustion with the air brought in - the air is a working fluid. It could just as well be martian air or venerian air or whatnot.

A rocket's efficiency is at a maximum when all of the heat of combustion and all of the pressure have been fully expanded to ambient temperature and pressure. This would require an infinitely large nozzle, so one does as much as they can with the nozzle that they have. The more gas involved (even gas that doesn't take part in the reaction), the more fully they can approach ambient for a given-sized nozzle. Furthermore, some reactions simply cannot proceed past a given point without a working fluid - for example, combustion of metals yields metal oxides that condense out of the stream at very high temperatures; you simply have to have another gas to take their heat way (including the heat of condensation) if you want to make use of it.. These effects are significant enough that oftentimes the optimum mixture for a rocket with a given nozzle is significantly off of stoichiometric ratios, with the extra gas (often hydrogen, due to the high number of moles per gram) functioning as nothing more than a working fluid. For example, LOX/LH is often burned at a mass ratio of 1:4 rather than 1:8 as stoichiometry would imply (1*O2(=32) + 2*H2(=4) = 2*H2O) - the extra unburned hydrogen doesn't add much mass but it lets you make use of a larger percentage of the combustion energy.

The point of air-augmentation in rockets - whether or not it can also attain some degree of ramjet functionality - is to make use of the air as additional working fluid to increase your rocket's efficiency.