Transterrestrial Musings

Comments

Here's my take on the problem, plucked from another work in progress:

The ongoing high cost of launch arises because there are significant and fundamental limitations in chemical propulsion – at least a factor of two increase in effective specific impulse (Isp) is needed. That is to say chemical propellants are limited by their energy density, which is barely enough to carry their own weight into space, let alone the structure of the rocket and a payload on top of that. Engineers have shaved structural margins to their limits and resorted to multiple stages so that a payload can be carried into space at all, and even then only 1-4.2% of the wet mass can be spared. The result is that to make a rocket that reaches orbit, expensive construction techniques to shave off every possible kilogram must be used.

Small structural margins not only increase the cost of conventional expendable rockets, they impede them from evolving towards reusability too. Launch is a violent process and a fragile structure is more susceptible to fatigue. The Space Shuttle as a reusable vehicle needs extensive refurbishment and testing between launches, to the extent that the launcher is disassembled, inspected, refurbished and rebuilt before every launch. For example a hydrogen tank used for the on-board fuel cell is manufactured to burst at 1.5 times its usual operating pressure in order to save mass, but this safety factor of only 1.5 means the tank may last only 100 cycles. Such a failure-prone component must have a more regular fatigue inspection regime, so operational costs go up. In contrast, the fuselage of a pressurized civil aircraft has a safety factor of 2, and for that relatively little extra mass lasts tens of thousands of pressurization cycles.

Though the overall utility of the Space Shuttle is still debated, the history of the program since its inception in 1972 has yielded valuable lessons in the true cost of running a vehicle with overly-tight margins. Twenty years of experience with Space Shuttle ground maintenance has failed to produce efficiencies that reduce cost without reducing reliability. I believe that every chemically-propelled reusable rocket will suffer from the problems of fragility and reusability because they are inherent to the structural and propulsive performance regime within which chemical rockets are feasible. But a truly low cost launch system must be reusable, because without reusability there is a price floor below which launch costs cannot go.

Posted by Kevin Parkin at March 14, 2005 03:01 PM

"But a truly low cost launch system must be reusable, because without reusability there is a price floor below which launch costs cannot go."

Um, for RLVs there is also a price floor below which costs cannot go, and thats the price of fuel. For ELVs this floor is closer to price of fuel+tankage+engines.

Posted by kert at March 14, 2005 03:49 PM

I should add to the above that the shuttle external tanks are figured with a safety factor of 1.25 - so they wouldn't survive many cycles at all if they were reusable.

Fuel is cheap.

Posted by Kevin Parkin at March 14, 2005 04:05 PM

There are multiple factors which make the Shuttle expensive. In the end you can sum it up to the amount of people involved.

I remember reading a study on Shuttle maintenance costs some years ago, IIRC the main maintenance costs were in the systems using hypergolics which required refueling (the level of toxicity required expensive delicate handling and transport), engines and heat shielding, in about that order. The first two issues seem fixable using current technology and upgrades, the last issue seems harder to fix without designing a new ship and using leading edge technology.

Regarding how to increase Isp more for a space launch vehicle, I see one possible solution. Putting the energy for escaping Earth gravity outside the launch vehicle.

First some debunking:
Nuclear is useless. The only options with good enough trust/weight ratios to escape Earth gravity would emit fissile elements in the exaust. Like Project Orion. You also need shielding for the crew which means the vehicle will weigh more.

High energy density fuels (solid hydrogen, free radicals, fluorine based, boron based, etc) have been researched for yonks (since the 1950s) and nothing decent has come up. Some of the new nitrogen based explosives are interesting, but they will not beat LOX/LH2 or LOX/Kerosene in performance IIRC.

So how do you put the energy outside the launch vehicle? Beamed power. Even the elevator proposals seem to use beamed power instead of cables for powering the elevator.

Needless to say, the current experiments in beamed power for space launch are still in the chemical rocket equivalent of the early Goddard experiments. Even if you had major funding, I suspect it would take a generation or two to get anything decent out of it.

This is a long time to wait and the funding will probably not be as forthcoming as it was for chemical rockets. What, with there not being a WWII at the moment which justifies that kind of spending. So no, we will just have to do with the tech we have now for Earth based launches, sorry.

Regarding composites, I think they are a bit overrated. Even without composites you could have done something much better than the Shuttle with current materials, that is for sure.

Posted by Gojira at March 14, 2005 05:10 PM

> A major issue is inspecting structure for fatigue between flights.
> We have quite a bit of experience with aluminum and other metals

Yes, but...

Take a look at what's going on with the T-34s right now. The principles are known, but there's still stuff that can come back and bites you from time to time.

Posted by Edward Wright at March 14, 2005 07:12 PM

> Um, for RLVs there is also a price floor below which costs cannot go,
> and thats the price of fuel. For ELVs this floor is closer to price
> of fuel+tankage+engines.

Actually, it's the same for either one: propellant + labor +capital cost.

The only difference is how you distribute the costs between the three. For ELVs, propellant gets lost in the noise. For a mature reusable system, costs are distributed more or less evenly between the three elements (hence the famous of rule of thumb about operating costs being three times fuel costs).

Posted by Edward Wright at March 14, 2005 07:16 PM

Let's see: According to Astronautix.com
http://www.astronautix.com/props/loxlh2.htm

"The delivered cost of liquid hydrogen in 1960 was approximately $ 2.60 per kg. Large-scale production was expected to reduce the cost to $ 1.00 per kg. In the 1980's NASA was actually paying $ 3.60 per kg."

If my launcher delivers 1 kg of payload for every 10 kg of LH2 spent, and I charge $1000/kg (because I'm a nice guy and give you a 90% discount) then fuel is 3.6% of that cost.

If you're thinking of a launcher that uses LOX too (I'm not), then LOX may cost about $1/kg according to the following highly amusing launcher cost calculator:

http://members.axion.net/~enrique/spacecraftcost.html

What kind of propellants would cost as much as the launcher?

Posted by Kevin Parkin at March 14, 2005 07:44 PM

"Even without composites you could have done something much better than the Shuttle with current materials, that is for sure."

Speaking of which did anything ever come of titanium fuel tanks?

Posted by Kevin Parkin at March 14, 2005 09:11 PM

> What kind of propellants would cost as much as the launcher?

Jet A?

We just burned $12,000 worth last weekend.

Posted by Edward Wright at March 14, 2005 10:01 PM

Mr. Parkin has hit the nail squarely on the head.
25 years of trying has not yielded a rocket powered space launcher with a reliability greater than .98. So how can you even think of building a chemical rocket powered Spaceliner?

BTW, as I understand it, the Boeing 787 is going to have a Japanese designed and built composite wing structure. Which brings up several unpleasent questions.

Posted by E. Bryan at March 14, 2005 11:37 PM

>I should add to the above that the shuttle >external tanks are figured with a safety factor >of 1.25 - so they wouldn't survive many cycles >at all if they were reusable.

It's been a while since my mechanical behavior class, but I don't recall a direct correlation between safety factor and the fatigue lifetime. A safety factor may be used to refer to static loads, but fatigue strength requires considering dynamic loads, such as pressurization cycles.

It's not so much an issue if the material cracks under a dynamic load, but if the cracks propagate. Opening a crack requires adding energy to form the surface area of the crack. Once the crack is opened a new state of dynamic equilibrium is established. Hopefully the load is insufficient to move out of this new equilibrium. Of course, this may not be the case and the crack may open a few microns further every cycle. Eventually in this case the part will fail.

So simply building a part with a safety factor of 2.5 versus 1.25 will not necesarily double the fatigue lifetime. The fracture mechanics of the material must be understood. A lot of time and money was spent exploring the fatigue behavior of aluminum alloys after some unfortunate accidents (see for example the Dehaviland Comet).

Composites, being a new material system, are not as well understood as metals. They also have new failure modes (delamination, fiber pull out, etc.). Composites also suffer from the fact their production processes are inherently less controllable, and have more variability, than those for metals.

Nonetheless, I see no fundamental material limitations to building a fully reusable space vehicle. The problems with the shuttle appear to stem from a design that was overspecified and underfunded. A simpler design, with more constrained performance goals, may well have been more successful.

Posted by Frank Johnson at March 15, 2005 07:41 AM

Gojira wrote:
>First some debunking:
>Nuclear is useless. The only options with good
>enough trust/weight ratios to escape Earth
>gravity would emit fissile elements in the
>exaust. Like Project Orion. You also need
>shielding for the crew which means the vehicle
>will weigh more.

Minor quibble with the first part of this. You don't need an 'ORION'type drive for Earth based launch. We've known this since the mid-60s. There was a competing design to the NERVA called the DUMBO which was quite capable of high thrust/weight and very 'build-able' with then current technology.
We decided to go with NERVA, a political/managment decision rather than an engineering one.

Yes the vehicle would have 'massed' more due to shielding, but even so a Saturn-V class launch vehicle with F-1+ class engine capability could be built for cargo flights.
(And even manned with shielded passenger section)

Randy

Posted by Randy Campbell at March 15, 2005 02:13 PM

This story reminds me of the Nevil Shute novel "No Highway" I read when I was a kid.

Posted by HA at March 16, 2005 09:27 AM

Randy Campbell - do you know where I could find info on the "DUMBO" design?

Posted by Reid at March 16, 2005 10:07 AM

Randy Campbell - do you know where I could find info on the "DUMBO" design?

Posted by Reid at March 16, 2005 10:08 AM

"You quote him as an expert. But there are more experts within the manufacturers and the certification authorities who agree with these procedures."

When I read that, I thought back to Richard Feynman's criticism of NASA's decision process to launch Challenger. He argued that you don't rank all votes equally. The guys who knew about the problem with the Thiokol tanks and were viewed by all the decision participants as knowing what they were talking about, were voting against the launch.

In the future, if I have a choice, I'll choose to fly on a Boeing over an Airbus given Airbus's head in the sand attitude.

Posted by Michael Greene at March 16, 2005 10:20 AM

Harry Dunne: Where's the booze?

Lloyd Christmas: I got robbed by a sweet old lady on a motorized cart. I didn't even see it coming.

Harry Dunne: Oh, no, no.

Lloyd Christmas: Come on, Harry.

Harry Dunne: It gets worse. My parakeet, Petey.

Lloyd Christmas: Yeah?

Harry Dunne: He's dead.

Lloyd Christmas: Oh, man, I'm sorry. What happened?

Harry Dunne: His head fell off.

Lloyd Christmas: His head fell off?

Harry Dunne: Yeah. He was pretty old.

Posted by Dumb & Dumber at March 16, 2005 10:26 AM

Didn't Rutan just point the way to what "reusable" is going to be? Sure the payload is less (then a chemical ground boost), but at what percentage of "less" [then a booster ground boost] is what can make it into orbit comparitively? And if it's that much cheaper, then why not lift in sections & assemble in orbit?

If you have to boost from ground level vs. a drop from altitude, isn't a substantial part of the "ground gravity deficit" put paid to?

Isn't a "drop ship" just another form of booster? Where's the engineering studies here? (Well, I just imagine that Burt's got a lot of 'em on his PC, but does that have to mean that engineering departments at major universities should remain clueless ...and that includes NASA?)

Why are we still wedded to a boost technology that Von Braun & even bloody Goddard would recognize?

Sure ...you wanna boost x-number of tons from ground in a piece into orbit, you do the math, and put enough engine & fuel under it, and if it doesn't blow, it does go. We know how that works.

It just seems ...so ...inelegant.

Well ...go back to how Von Braun suggested we'd make space stations in the first place. Make smaller pieces. Put a lot of them into orbit as cheaply as possible. Make spares. Then assemble them. Build small to build big.

(If you'll forgive the pun ....) This just doesn't seem like rocket science to me.

Posted by brandon davis at March 16, 2005 10:26 AM

Why don’t they conduct a regular cycle of non-destructive test such an ultrasonic and or x-ray at suspected stress accumulation points? This can be easily calculated. It is cheaper than lawsuits after a fatal crash. This would increase the public faith and trust in Airbus. I rather fly Boeing.

Posted by B at March 16, 2005 10:31 AM

Yeah, we are pretty much stuck with chemical fuels, until somebody comes up with a fancy regenerative scramjet. One trick though is to use slushed hydrogen, finely divided solid H2 in liquid, which is like 25% denser.

As far as spaceplane reliability, engine technology has come a long, long way since early iterations of the Shuttle's hydrogen engines. Engine and structure reliability aren't a problem, so thats most of the important stuff right there.

Nuclear for ground launch is out, if there was a launch snafu that hot reactor (that would stay hot even if shut down) would come back down on our heads. Nuclear engines are fine, but only fire them when they are in orbit already.

Posted by Nathan Henry at March 16, 2005 10:37 AM

The problem with Burt Rurtan's little toy rocket Cessna is that it would need a few hundred times more fuel to get into orbit proper with any reasonable payload. Suborbital flight is easy, only needing a small fraction of the energy of real Mach-25 orbital flight.

To get into orbit, SpaceShipOnePointFive wouldn't fit on the back of a 747 comfortably.

The way to beat this problem is not to make the whole operation bigger, but to ramp up the speed of the carrier plane, either with rockets or ramjets, so it can contribute a larger amount of speeds & altitude.

Posted by Nathan Henry at March 16, 2005 10:45 AM

Bring on the Space Elevator

Posted by TallDave at March 16, 2005 10:48 AM

Seems like intelligent structures, especially wings, are going to be needed for the composite-"heavy" planes. Fibre optic nets should be able to pick up on these micro-events far better than any form of periodic inspection.

Posted by Brian H at March 16, 2005 10:52 AM

> Minor quibble with the first part of this. You don't need an 'ORION'type drive
> for Earth based launch. We've known this since the mid-60s. There was a competing
> design to the NERVA called the DUMBO which was quite capable of high thrust/weight
> and very 'build-able' with then current technology.

Do you know what propellant (reaction mass) is used to get that Isp?

Hydrogen. Because LH2 is extremely low density, your atomic rocket is likely going to need larger tanks than a chemical rocket. So the structural problem only becomes harder. (Not to mention all the radiation problems you've introduced, as well as the practical problem of developing the purely *theoretical* Dumbo reactor.)

We don't need atomic rockets, scramjets, space elevators, antigravity, or other unobtanium. We don't need particularly high Isp. Chemical rockets are quite good enough, if the goal is to develop a practical transportation system rather than a permanent research project.

Posted by Edward Wright at March 16, 2005 11:08 AM

"The result is that to make a rocket that reaches orbit, expensive construction techniques to shave off every possible kilogram must be used."

Nope. This is a leftover from launchers being built with an ICBM mentality. (ICBMs needed to be small so they could be protected.)

If you really don't care about mass fraction, you could go the Big, Dumb Booster route that was looked at for post-Saturn launchers. From the summaries that I have read, the BDBs were to be built using no exotic materials, with late 60s shipyard construction techniques, and with only minimal regard for mass. In addition, the lower stage(s) were hydrocarbon-LOX. High Isp engines tend to have low thrust and so are problematic for first stages, hence the SRBs for the shuttle.

Now, for RLVs, mass fraction seems to be important. In fact, there was much debate in the late 80s and early 90s over whether an RLV should use LOX and hydrogen or LOX and Jet A or Propane. Liquid H2 is damn cold (and problematic for composite tanks) and LNG is much denser and gives you smaller tanks/less drag/better mass fraction.

Hook to current events: In the past, Mike Griffin has been a strong proponent of both RLVs and heavy lift boosters. I wonder if NASA will look at using up the shuttle assets as heavy lift boosters after the program is shut down in 2011.

It seems to me that until we get space elevators or the like that RLVs make the most sense for human and other small, high value cargo. A cheap heavy lift launcher would be useful for boosting big items like planetary transfer stages or large masses of fuel.

Posted by ech at March 16, 2005 11:15 AM

There's no reason why the margins on the first stage of a multistage launcher need be as tight as those o the shuttle. The first stage is dropped almost immediately, so the rocket equation doesn't compound any weight increases very much.

BTW, $1/kg sounds awfully pessimistic for LOX, if you're using it in any significant quantity. Is that some special ultra-pure propellant grade LOX?

Posted by Paul Dietz at March 16, 2005 11:25 AM

Yes, I would think that LOX could be purchased for a third or a quarter of that price.

Posted by Rand Simberg at March 16, 2005 11:30 AM

We'll, after reading about the problems with Airbus over the last few days I decided it might be a good idea to check what kind of plane I will be flying on my vacation (Boston to Vegas)next month.

Of course it's an Airbus on America West.

Should I eat the $400 and buy a ticket on Song/Boeing or should I just start my gambling earlier than I thought?

Posted by Matt B at March 16, 2005 11:45 AM

I figured out one time that I have enough energy in my Lanciar to fly into orbit. Assuming I climb at the same rate as at sea level to 100 Kilometers, then accelerate for a few hours at the same rate as I do on the runway.

The problem isn't energy. The problem is efficiently handling the flight environments required.

Winged flight takes significantly less energy to lift a given weight than levitating that weight on chemical powered thrust.

Remember James Bond's "rocket belt" in the 60's? Lifting that weight vertically by accelerating only the mass carried by the rocket belt alone meant that it had an operational duration of about 20 seconds. A similar turbojet version I remember from that era accelerated the mass of outside air to attain lift, which gave it a flight duration of something like 60 minutes.

Lesson: Use rockets only as a last resort at extreem altitudes and speeds. This should be obvious, but apparently because of development costs has been replaced by the philosophy of "fuel is cheap" (while ignoring the fact that the required ramp up in size and weight to hold that fuel is in the end far more expensive).

Rutan has the rare ability to think outside the rocket box that is as old as chineese fireworks. Calculations on this site for launching weight into orbit are limited by engineering concepts first perfected almost 75 years ago by Vaun Braun. It's well past time to move on.

Those who deride Rutan as flying "cessna's" are at risk of looking stupid in a couple of decades.

I wonder how large the rocket would need to be to launch a human into orbit, assuming it was first accelerated to mach 3 and 70k feet by an XB-70 class aircraft? Remember that the Valkyrie held the absolute weight lift record for a time, so likely could have lifted a considerable sized rocket. That's the difference between 75 year old Vaun Braun rocket technology vs. 40 year old cold war high speed aircraft technology.

I imagine that we could do much better today, given an appropriate budget, and unconstrained engineering leadership.

Posted by at March 16, 2005 12:05 PM

Composites are strong and reliable, but if they do fail, then you have a different stress field. The problem is not so much with the composite breaking as it is "unzipping". Once the composite delaminates, the stiffness (which depends on the material being laminated) is much reduced. A strong glass or boron fiber is strong in tension or compression, but very weak in bending.

Ultrasound is essentially a computerized coin tap. Another way of looking through the surface is to heat the part and examine it through a thermal viewer. The delamination creates a thermal discontinuity which shows up.

There is a lot of work being done on Liquid CH4 (Methane) as a rocket fuel. It works well with LOX. Its density is higher than Hydrogen, but less than kerosine. The really good news, there is a bunch of methane in the outer solar system. The Martian atmosphere could be converted to methane with a local power source.

Posted by Don Meaker at March 16, 2005 12:27 PM

A small word on compsites vs. metals.

signs of fatigue on different types of materials are different, and many are based on what I call 'folklore' rather than scientific means. By that I mean that looking at an xray of a part, one knows how to recognize obvious signs of fatique. Another person might not see a less obvious sign. In reality, not all cases of fatique can neccessarily be seen, even using the most sensitive scanning. This is true even for metals that we have a lot of experience with.

Composites are even more difficult as they have less visible signs than metals, and tend to fail catastrophically before they begin to show signs of fatigue.

Posted by duglmac at March 16, 2005 01:06 PM

Do any of the knowledgeable people here have anything to say about the JP Aerospace plan to use three stages of airships to get to orbit?

http://www.jpaerospace.com/atohandout.pdf

It does seem nice in that they just use bouyancy to stay up while slowly accelerating sideways in a near vaccum. Their argument is that they can use high efficiency drives like ion drives.

I don't know if it's feasible, but it sure is low instantaneous energy, and that makes a real change. Of course, the upper stages are huge (they need to be to displace enough "air" at near vacuums).

I must say I absolutely love the idea of having a "space" station in near space, supported entirely by bouyancy.

Posted by P.H. at March 16, 2005 01:14 PM

Noted with interest the "I'll fly Boeing" comments.
Anyone care to enlighten us on the design of and materials in the rudder of the 777?

If as I suspect it is another composite achievement, when should we expect problems to develop?

Posted by Franz Misch at March 16, 2005 01:19 PM

The now retired Concord has a roughly 100,000 KG cargo capability and can travel at Mach 2.5 @ 70,000 ft. What could it launch?

If new technology made diamond "shapable" on a large scale what difference to cargo weights would that make?

Posted by Rob Read at March 16, 2005 01:22 PM

I'm pretty sure the rudder on the triple-7 is aluminum. That jet was developed a bit before composites got really big.

Posted by rosignol at March 16, 2005 03:23 PM

"Why don’t they conduct a regular cycle of non-destructive test such an ultrasonic and or x-ray at suspected stress accumulation points? This can be easily calculated. It is cheaper than lawsuits after a fatal crash. This would increase the public faith and trust in Airbus. I rather fly Boeing."

After we finish eliminating/severely restricting civil actions under the guise of "tort reform" it may not be cheaper to fix problems instead of compensating victims. Anyone remember the cost analysis done by Ford Motor with the exploding Pintos? That can happen again absent financial incentives to make products safer.

Posted by Steve at March 17, 2005 07:54 AM

> After we finish eliminating/severely restricting civil actions under the guise of "tort reform" it may not
> be cheaper to fix problems instead of compensating victims. Anyone remember the cost analysis done
> by Ford Motor with the exploding Pintos? That can happen again absent financial incentives to make
> products safer.

Financial incentives were not absent -- they just didn't justify the outcome you wanted.

So, you think companies should be forced to "make products safer," no matter how much the fix costs and how few lives are saved?

So, if a company has to cancel a major donation to medical research to fix a defect that might save one life, it should do it, right?

The people who might have been saved by that donation will just have to suffer and die, right?

Interesting how trial lawyers are the one industry the left *doesn't* want regulated for safety.

Posted by at March 17, 2005 09:39 AM

Post a comment