~Stephanie Osborn
http://www.stephanie-osborn.com
* * *
A Few Observations on
Reusable Space Hardware,
or Why the Space Shuttle was
an engineering masterpiece and a logistical nightmare.
by Larry Bauer
The National Space
Transportation System, or what people commonly refer to as the Space Shuttle, was
composed of four parts: the Shuttle itself, the three Shuttle main engines, the
External Tank, and two Solid Rocket Boosters. In theory everything but the tank
and its fuel were recoverable.
I was reminded of this when I
saw an article noting the sixth successful landing of a SpaceX Falcon 9 rocket
after the successful launch of the JCSAT-16 commercial communications
satellite.
From a purely economical
aspect it all comes down to what is the cheapest means to launch a payload into
a desired orbit. On the face of it reusing hardware only makes sense. However
there are a good many factors that mitigate against such as assumption.
With NSTS we salvaged
everything except the ET. The Shuttle and engines landed and the SRB shells
parachuted back into the sea for recovery. In theory maximum reuse of critical
hardware. But let's look first at those boosters. They were made in sections
far from the launch site, built and filled in sections so they could be
transported by barge to Kennedy Space Center. From a logistics standpoint it
would have made much more sense to build them as a single tube and fill them
with solid propellant right there at the launch site. However that was not an
option. The state of Florida would allow launches from KSC, but they refused
permission to build the SRBs there. The solid propellant is nasty stuff and the
process of filling the boosters violated too many state pollution restrictions.
So the SRBs were built and filled originally by Thiokol of Brigham City, Utah,
later bought out by ATK.
I will note that the recovery
and remanufacture of the boosters was very cost- and labor-intensive, and there
was always a debate over whether a cheaper throw-away design might have been
more cost effective. I will also point out that a design incorporating a single
continuous tube would have made the failure that caused the Challenger disaster
impossible. Which does not mean something equally tragic might not have
happened, but you cannot have a joint failure if there are no joints.
The high-performance Space
Shuttle Main Engines, known as SSMEs or the Aerojet Rocketdyne RS-25, are an
example where reuse of hardware only makes good sense. These are the pinnacle
of the state of the art for liquid fuel rocket engines. With the help of the
solid boosters, these engines, sucking immense quantities of liquid hydrogen and
oxygen from the External Tank, could lift a combined vehicle and payload weight
of roughly 2060 tons. Bringing these highly intricate and fine-tuned engines
back for reuse only made sense.
And that does lead me to the
subject of a rather controversial opinion of mine. As magnificent an
engineering achievement as the NSTS was — and I spent the majority of my career
at NASA doing ground support to on-orbit experiment operations so I have a
great fondness for the beastie — the reason why it was a huge logistics failure
rests in the numbers. The shuttle itself weighed in at 2030 tons. Its payload
was 30 tons. The NSTS was a true heavy lift rocket, but most of what should
have been useful payload mass to orbit was spent instead on creating the ability
to land sort of like an airplane. A requirement imposed, by the way, by the Air
Force — who withdrew from the project between the time the design was firmed up
and the first launch. I will also observe that every astronaut pilot I've ever
spoken with all described the shuttle as “that flying brick.” [I can confirm
that astronaut description. —Steph]
And all of the above is
explanation as to why we don't see any shuttle-like designs these days. It is ever
so much more efficient to make as much of the upmass be useful payload as
possible, with the crew compartment just sufficiently robust to carry the
astronauts up and get them back to Earth safely.
* * *
Thanks much for that information, Larry! It squares pretty nicely with what I know of the various programs, as well.
BOOK RECOMMENDATION:
A New American Space Plan, by Travis S. Taylor with Stephanie Osborn, available in print and ebook, discusses the history of space exploration, where we are, how we got there, and where we ought to be and be going. You can find a lot more detail on the Space Shuttle, how it worked, the main engines and how the whole system was designed, right in this book. There's also a good bit about the recent efforts to develop commercial space launch systems and why the epithet "commercial" is often a misnomer.
http://www.stephanie-osborn.com
14 comments:
I suspect Thiokol solid motors had more to do with getting the votes from Utah than insurmountable Floridian red tape.
Mr. Bauer nicely details the plethora of profound design flaws that prevented the Shuttle from demonstrating space transport reusability. It was essentially a rebuild-able space transport. However, he seems to believe that the very non-reusable Shuttle backs up his claim that there are a "good many factors that mitigate against" the notion that reusing hardware makes sense. Besides the fact that one should never generalize from a single data point, the Shuttle only proves that one should design a reusable space transport without all those flaws.
It isn't just reuse that lowers costs. Reuse must be combined with fast turnaround with a small ground crew. The Shuttles took at least 2 months and the proverbial standing army of 10000 people to fly again. In late July, SpaceX test fired a "previously flown" first stage three times in three days, each for a full duration 2.5 minutes as in an actual launch. This makes for five full firings counting the launch and the pre-launch test. They plan several more firings with little or no refurbishment between firings. (They had to pause the firings to use the pad to test a booster for an upcoming launch.) Re-flying ten times looks well within reach reach. The first stage is 75% of the cost of a F9 so it's clear in most analyzes that re-flying this many times will lower costs. At his early phase, SpaceX is already offering a 30% reduction in launch prices with a reused first stage. And this is on top of the cuts of 2 to 3 in expendable launch cost that SpaceX offers compared to Arianespace, ULA and other competitors (who claimed for years that ELV costs could not be reduced by more than a few percent).
Perhaps the most fundamental difference between the SpaceX (and Blue Origin) approach to reusability compared to the Shuttle is the ability to do continuous incremental improvements. Short of replacing the ET and SRBs with a fly-back first stage, there was little one could do to improve Shuttle costs. The Falcon 9 first stage and its Merlin engines, OTOH, have been continuously improved and modified to reach the current flyback capability. They will now start re-launching the boosters. If a structural problem is found after x number of flights, they can fix that and continue. There are no deep profound reasons that a Falcon 9 booster cannot eventually be flown 100 times with short turnaround times. (I'll note that the Dragon spacecraft are also incrementally moving towards reuse. A reusable F9 upper stage may eventually be developed but it may be eclipsed by SpaceX's plans for a whole new 2nd-gen launch system.)
Fully and rapidly reusable space transports are essential for real space development. It's a shame that NASA learned the wrong lessons from the Shuttle and is jumping back to the 60s with the SLS/Orion boondoggle. Thankfully we have commercial actors who are making the key breakthroughs that will lead to affordable access to space.
Hobbyspacer, I recommend rereading Larry's post. He's not arguing against "reusable" at all. He's saying that the final design of Shuttle wasn't a good one in many respects because IMHO they tried to combine several concepts, including heavy lift, space plane, and reusable all in one, and in some respects got the "wrong parts" of each concept.
Also it isn't just the Shuttle that mandated the large ground crew for launch. KSC's facility more or less requires a good bit of it. We really need to "museum-ize" a good bit of the launch facilities there (though I hate to say it) and construct much more modern facilities.
That said, it will be difficult to get the funding to do so, and that will likely force NASA to pull the funding from other projects; one thing I discovered when I co-authored A New American Space Plan is that NASA has been running on a budget of approximately $15B/yr continuously since the 1960s, without getting any sort of correction for inflation, all while having more and more duties added onto their charter.
And unfortunately, much of the final design (and the catastrophic problems that came with it) can be laid at the feet of Capitol Hill and those same people who redlined the budget over and over.
Hopefully Larry will post in the morning and add more into his
Mhere, you might be surprised. There was a LOT of red tape, but not in Utah. Getting jobs probably. Getting votes, possibly.
But I'll say what I've said many times: If von Braun had still been around, there's no way he would have let solid rockets be strapped to a manned spacecraft in the first place.
I can also say that everybody I know/knew in the program was gung-ho for DCX when it was the going thing, and I still remember the chorus of disappointed, horrified groans when it tipped over, crashed and burned. And those same people are also excited over SpaceX's crafts.
https://www.jerrypournelle.com/chaosmanor/ssx-dcx-star-wars/
Shootist, oh yeah. Jerry and I are friends. We've discussed the whole DCX thing on a couple of occasions.
Don't forget "the proverbial standing army of 10000 people" was a feature, not a bug for maintaining a massive bureaucracy.
Couple of things here. Sure, NASA has always for its own survival played the "spread the wealth" political game to get as many congressional districts as possible on their side. But I served on the source evaluation board for the Advanced Solid Rocket Motor program, ASRM, intended to replace the SRB design. Ultimately the decision was made to stick with an upgraded and improved SRB rather than an entirely new design. But at the time we looked at all sorts of options and almost thirty years later I still remember the hard line stone wall position that Florida took on processing solid fuels.
On the assumption that I'm against reuse, far from it. It just has to make sense logistically. My background is in Industrial and Systems Engineering, and from a bleeding edge technical aspect I loved that big cantankerous tempermental beastie to death, but given current rocket technology spending an order of magnitude of upmass not on useable payload, but on the vehicle always made my eyeballs itch.
What inspired my small trip down memory lane in the first place was a passing news article that apparently SpaceX was doing reuseables correctly, and more power to them.
"I will also point out that a design incorporating a single continuous tube would have made the failure that caused the Challenger disaster impossible."
Indeed, such a design was available. I once sat next to a fellow from Aerojet, from DC to Portland, who had worked on the design, and tested it. They were to pour the whole motor at Holoman AFB, in Florida, store them at an angle 30 degrees from the horizontal, and very slowly rotate the casing at that angle afterwards, to keep the grain from detaching. This had been done successfully in a year-long storage test. Of the mention about pollution, that didn't apply at an AFB in the early 1970s, since they already had waivers for the Base. What they didn't have was Jake Garn as the Senator from Florida, and Garn was owed big time by Pres. Nixon, as the man who had nominated him at the Republican convention.
We could have had liquid boosters, but the administration didn't want the budget for Shuttle to be projected for a Democratic Congress over $1 billion in any single year. That would require too many political favors to be traded. Solids could allow torturing the numbers till they gave $975 million at the peak budget year, while liquid boosters were placed at $1.25 billion, but with far cheaper operating costs.
So, not only were liquids ruled out (Thiokol didn't make big liquid engines in Utah) but the segmented booster was accepted, because you could not ship a completed solid booster from Utah to KSC.
"If von Braun had still been around, there's no way he would have let solid rockets be strapped to a manned spacecraft in the first place."
Well, Von Braun *was* around, and resigned soon after the solids decision. It was easy to read the rumors that it was a "resignation in protest", but no one would confirm that. He had already been "booted upstairs" to HQ, however, in a capacity that did not include making decisions on future manned spacecraft. Many of the Peenemunde Team were treated in similar fashion after Apollo 11.
The Shuttle is the result of a long trail of unnecessary requirements and assumptions, beautifully implemented for the most part, and kludged when time and budget constraints so required. It also was based on assumptions that didn't pan out because the logistics ate up the budget for developing a replacement. It was supposed to be replaced by a newer and better craft based on lessons learned after about ten years. That never happened.
Sadly, the SLS may be a worse solution to a non-problem with a vastly lower flight rate, one that is perhaps unworkable because each launch will in effect be a test flight with a rusty/green ground crew.
I once suggested that if NASA makes progress on suspended animation/induced comas, they could put the ground controllers and their families into suspended animation in between launches so the problem goes away.
Commercialization and economy of scale are the keys to an active and healthy space program. NASA, like all government agencies, is more interested in empire building than achievement. And you don't achieve economy of scale with one or two launches a year. But one or two a week and things start to pop. Private industry under some government supervision is what will eventually get us into space on a regular basis and keep us there.
That reminds me of something I read about why the US developed computers before the British, despite the large British head start.
The US wanted a computer. The British wanted a development program that was building a computer.
The SLS program is a success whether the rocket ever flies or not. It's achieving its development and employment goals.
SpaceX is different because Musk views the rocket as a necessary step to get him where he wants to go, which is Mars. He doesn't want to be building a rocket, he wants to be using a rocket.
Post a Comment