I figured that if I’m going to be considered a space pundit, I may as well share a few of my thoughts about the recent NASA budget situation. I would have put my two cents in earlier, if it wasn’t for the fact that “anonymous” over on Jeff Foust’s Space Politics blog beat me to the punch on most of my thoughts (as well as bringing up several I hadn’t thought about). Heck, if I knew how to contact him offline, I’d like to invite to join Ken as a guest blogger here on Selenian Boondocks.
But here are a few of my thoughts on the matter:
- A lot of discussion over the past year has centered on how ESAS was so clever for dealing with “political realities” regarding the Shuttle workforce. And that because this big workforce meant so many jobs in several congressional districts, it meant that the ESAS approach to implementing the VSE was politically impossible to kill. The whole idea that you could somehow make a program so pork-o-licious that it couldn’t possibly be canceled flies in the face of history. If jobs and inertia was all that was needed to keep a program funded, we’d still have the Saturn V. It provided tons of jobs after all. What it didn’t provide was a benefit even close to commensurate with the cost of keeping it around. Not to mention the fact that with the results of the 2006 midterm elections, I highly doubt that Utah jobs matter quite as much to Congress any more.
- The ESAS approach was also touted as a “go as you pay” type program. Basically claiming that it was perfectly capable of adapting to budgetary ups and downs. However, even Griffin himself recently more or less admitted this was a crock. Without large increases in the Exploration budget, Ares I/Orion are not going to fly in time. And the longer that program gets stretched out, the higher a percentage of the program funding is going to be tied up just in keeping the team together and paying fixed overhead. How are they going to retain all those Shuttle employees when the gap between Shuttle operations and Ares I operations starts getting bigger and bigger? While NASA is being forced to “go as you pay”, it isbecoming increasingly obvious that the ESAS approach is a really lousy fit for that funding method. It is front loaded with a lot of development cost filling a “badly needed gap in US launch capabilities”.
- If the Ares I launch manifest from LaunchSpace.com is to believed (and there are some who are skeptical that their launch manifest really is official), they’re now looking at cutting out unmanned test flights in order to avoid having the first manned Ares I flight shoved out too far into the future. But don’t worry boys and girls. In spite of MSFC’s complete and utter lack of recent successful launch vehicle design experience, their top men guarantee us (to four significant figures mind you!) that Ares I will be the safest launch vehicle ever, by a factor of about 8x. I mean, with top men like that working on the problem, who needs test flights? Why not just design it perfectly in the first place? What could possibly go wrong?
- Regardless of which manifest is correct, most of the scheduled Ares I flights between first flight and when Ares V, EDS, and LSAM are ready are crew/cargo flights to ISS. If COTS is successful however, and if NASA actually obeys the law and buys commercial, what exactly will Ares I do for all that time (other than soak up fixed operating expenses)? Without Ares V, EDS, or LSAM, Orion is pretty much stuck in LEO or thereabouts. They might be able to do a bit of a burn and try and simulate a lunar reentry, but other than that, Ares I is going to be mostly sitting around gathering dust for 4 years or more. All while burning through somewhere around a billion dollars a year in fixed expenses, to keep the team around for lunar operations. The sad thing is that this reality alone makes Ares I a severe threat to COTS. It’ll be very easy to argue that “we’ve got this national asset sitting around here that we’re paying for even if it isn’t used, and we have this space station. Why are we buying commercial services when we’re having to pay for Ares I anyway?”
- With the current NASA budget being so far below the budget request for last year, it will make it all the harder for NASA to actually get this year’s budget request. As some have pointed out, the FY 2008 budget for NASA represents a nearly 6% increase over the budget that just got passed, all the while almost all other discretionary spending is increasing at a much slower rate. Being realistic, there’s a good chance that NASA isn’t going to see big Exploration Systems budgetary increases any time soon. Which means even more delays, cost overruns, etc. Now I’m not saying that NASA couldn’t somehow stick with Ares I and Orion, and somehow keep slogging away at them till they get completed. I’m just saying that whatever technical “victory” that would be would be Pyrrhic at best.
- Most of the shuttle workforce that gets talked about is technicians, not engineers. You can’t just transfer a guy from tightening bolts on an SSME to designing the EDS. If you “retain” these people, it means that you’re having to pay for them while at the same time also paying for the engineering work that needs to be done to get Ares I/V built and flying. Now, some of these people have relevant skills that will be useful during the test programs for Ares I/V, but many of them aren’t really needed until Ares I and Ares V start flying. And some of these people can or will have to be retrained (at additional expense) for new tasks since their old expertise is obsolete.
- The argument often gets made that no Congressman is going to vote to end the US manned space program. But the question is, if in 2014 the Atlas V team (and/or SpaceX or RpK) are flying people into orbit and Ares I isn’t, how much water will this argument hold? If the ability to put people and cargo into orbit on domestic commercial launch vehicles becomes a reality before Ares I, why are Congressmen really going to care that much about continuing to fund an overpriced government competitor?
I could go on, but the main thing I take away from this is that the so-called “political realities” we heard about weren’t. And that by spending so much time and money trying to reduplicate existing commercial earth-to-orbit launch capabilities, NASA has likely blown its chance at a return to the moon. If NASA had gone with their original (pre-ESAS) plan, and built a “CEV” that could be flown on either of the existing EELVs (and consequentially on SpaceX’s Falcon IX if that ever flies), we would likely be in much better shape.
The following two tabs change content below.
Jonathan Goff
President/CEO at Altius Space Machines
Jonathan Goff is a space technologist, inventor, and serial space entrepreneur who created the Selenian Boondocks blog. Jon was a co-founder of Masten Space Systems, and is the founder and CEO of Altius Space Machines, a space robotics startup in Broomfield, CO. His family includes his wife, Tiffany, and five boys: Jarom (deceased), Jonathan, James, Peter, and Andrew.
Jon has a BS in Manufacturing Engineering (1999) and an MS in Mechanical Engineering (2007) from Brigham Young University, and served an LDS proselytizing mission in Olongapo, Philippines from 2000-2002.
Latest posts by Jonathan Goff (see all)
- Fill ‘er Up: New AIAA Aerospace America Article on Propellant Depots - September 2, 2022
- Independent Perspectives on Cislunar Depotization - August 26, 2022
- Starbright Response to ISAM National Strategy RFC - July 2, 2022
If NASA had gone with a commercial EELV launcher for Orion, then it would have a system as unsafe as the existing Shuttle and Soyuz. Human crew should not be taking a 2% risk of death every time they fly, this is 2007. NASA have made the right choice to drastically improve the safety and reliability of human access to space.
cIclops,
If NASA had gone with a commercial EELV launcher for Orion, then it would have a system as unsafe as the existing Shuttle and Soyuz. Human crew should not be taking a 2% risk of death every time they fly, this is 2007. NASA have made the right choice to drastically improve the safety and reliability of human access to space.
Wow. You really believe that a NASA center that hasn’t developed a new manned launch vehicle in my entire lifetime is somehow going to make a vehicle that is 8x safer than any other rocket that has ever flown? I’m sorry, but you’ve been turned into a chimney my friend.
Launching on a Delta or Atlas, even by NASA’s admission would likely be an order of magnitude better than Shuttle. Shuttle fails about 2% of the time (as does Atlas or Delta), but an EELV can have an escape tower (just like the Shaft does), and that escape tower provides most of the reliability improvement. However, the escape tower for Ares I is only considered to work like 80% of the time as opposed to over 90% of the time for EELVs that don’t have huge Roman Candles for first stages. The only way they get their 1:2106 loss of crew number (or whichever obviously bogus 4-significant figure number they gave) was by assuming that the odds of a launch vehicle failure on Ares I that required escape being less than 1:400.
That is just BS pure and simple. You cannot predict in advance the reliability of a system like that. They did the same thing with the shuttle and predicted similarly high reliability numbers. SpaceX ran the numbers on Falcon I, “proving” that it was the safest launch vehicle in the world, and then proceeded to drop the first on in the drink.
The reliability analysis methods they used are at best useful for qualitative comparisons, but are very naive and high level.
~Jon
I guess this leaves it to me to pick the dishonest trolling of misspelled cyclops regarding Soyuz then? Let’s see how many Shuttles blew up and how many Soyuzes blew up? Huh? And the only time a Soyuz the booster blew up, the escape system worked fine, saving the crew.
jon wrote: You cannot predict in advance the reliability of a system like that.
Of course quantitative predictions can be made, how accurate they are can only be determined by testing. What other basis is there for comparing the reliability of unbuilt systems?
BTW if you are unable to respond without using personal insults, please don’t bother.
“I would have put my two cents in earlier, if it wasn’t for the fact that “anonymous” over on Jeff Foust’s Space Politics blog beat me to the punch on most of my thoughts (as well as bringing up several I hadn’t thought about). Heck, if I knew how to contact him offline, I’d like to invite to join Ken as a guest blogger here on Selenian Boondocks.”
Thanks for the kind words, Mr. Goff. As much as I’d like to contribute to a blog in a more formal way, my day job requires me to keep my identity hidden, and I doubt your readers (or readers on any blog) are going to appreciate an anonymous author.
But please keep up the good work. I think you’re one of the few engineers I’ve run into who has an appreciation for the budget, policy, and political environment that drives our civil space program, and accurately reflects that environment in his proposed design solutions. The devil is always in the details and a more thorough review might reveal technical roadblocks for some of your concepts. But they are exactly the sort of thing that NASA should have examined more completely in ESAS. If there is ever a future, independent revisit of ESAS, I hope that innovative, bite-sized, and step-wise approaches like your concepts get into the mix.
I’ll try to visit and comment every so often.
Thanks again…
“If NASA had gone with a commercial EELV launcher for Orion, then it would have a system as unsafe as the existing Shuttle and Soyuz. Human crew should not be taking a 2% risk of death every time they fly,”
How do you figure?
Atlas V, for example, has a perfect track record (8 for 8). So if you went with some of the proposals on this blog and used an existing, medium launch vehicle for crew transport, you’d be employing a vehicle with a demonstrated 100-percent reliability and, based on real-world experience, a zero projected probability of loss-of-life per mission, assuming no changes are made to human-rate the vehicle.
Compare that to an Ares 1, a vehicle that only exists in computer files, has no demonstrated reliability and, based on real-world experience, possesses an projected probability of loss-of-life somewhere between zero and 100%.
“NASA have made the right choice to drastically improve the safety and reliability of human access to space.”
How do you know? Ares 1 has not flown.
Ares 1 employs a new, 5-segment version of the Shuttle SRB which has yet to be tested in flight. Dozens of design changes are involved in that new SRB — propellant mixture, propellant grain geometry, different loads, etc. Yet in their reliability and safety calculations, NASA is assuming that the new SRB has the same projected reliability as demonstrated on the old SRB. How do they know?
Ares 1 incorporates a new version of the J2 rocket engine for its upper stage. Again, dozens of design changes will be involved in the new engine — and it hasn’t even passed SRR. Yet in their reliability and safety calculations, NASA is assuming that the new J2-X has the same projected reliability as demonstrated on the old J2. How do they know?
Ares 1 uses solid-rocket motors which, due to their high thrust, are very difficult to get away from in an abort situation (you need even faster solid-rocket motors). No less than Antonio Elias, designer of the Pegasus, acknowledges that this capability has yet to be demonstrated over on the nasaspaceflight.com forums. Yet in their reliability and safety calculations, NASA is assuming that the Ares 1 abort system will work most of the time. How do they know?
You may buy into all that calculated reliability handwaving, but as others have already pointed out, all it gets you are numbers that are inflated by about a factor of five versus experience (project Shuttle loss-of-life at about 1-in-250 flights versus actual loss-of-life at about 1-in-50 flights).
You have to look at the underlying assumptions very carefully (and in the case of Ares 1, there are some huge assumptions) and give equal, if not more, credence to demonstrated reliability in the real-world (such as that demonstrated with the medium EELVs), as well as common-sense engineering (launch abort and solids don’t mix well).
cIclops,
Of course quantitative predictions can be made, how accurate they are can only be determined by testing. What other basis is there for comparing the reliability of unbuilt systems?
Actually, most good engineering classes on reliability will point out that you *can’t* make good quantitative predictions about reliability of a complex system in advance. The best you can do is qualitative. That qualitative information is still extremely useful, but assuming that somehow you were able to accurately predict the reliabilities of all the components, and all their interations, and all the human factors involved, etc is just not very realistic.
What you can learn is basic trends like, single-stick configurations are safer. Launch abort systems can greatly increase reliability for most systems. Solids make launch abort dicier. Being able to reuse/reinspect your booster post-flight (or at least being able to get a lot of vehicle health data) can allow you to catch manufacturing variation thus increasing reliability. Stuff like that.
But saying your vehicle is going to have a 1:2106 chance of killing the crew is not good reliability engineering. First off, notice that they didn’t specificy a confidence interval. And it’s a good thing. They’d have to fly the thing over 6000 times with only one loss of crew event to demonstrate at a 95% confidence interval that they really are that reliable.
The fact of the matter is that almost anyone you speak with who has done good reliability engineering has huge issues with the approach and conclusions taken by the ESAS team. The Ares-I might possibly be slightly more reliable than EELVs, but it definitely is not “provably” 8x better. That’s just not how things work.
~Jon
Anonymous,
Thanks for the kind words, Mr. Goff. As much as I’d like to contribute to a blog in a more formal way, my day job requires me to keep my identity hidden, and I doubt your readers (or readers on any blog) are going to appreciate an anonymous author.
Well, if you’re too busy at the moment for blogging, that’s perfectly understandable. I *should be* too busy for blogging…
As for anonymity, I know a few people who blog anonymously (like Mr X over at ChairForce Engineer) to avoid getting in trouble at work. I don’t think any of us would mind having an anonymous third coblogger–but when it comes to taking risks with one’s job, I’m not going to pressure you.
But please keep up the good work. I think you’re one of the few engineers I’ve run into who has an appreciation for the budget, policy, and political environment that drives our civil space program, and accurately reflects that environment in his proposed design solutions.
Thank you! I’m glad someone thinks that. I’ve always tried to tie together the technical, policy, and economic implications as best as I can. Unfortunately I’m a lot stronger on the former than I am at the latter two.
The devil is always in the details and a more thorough review might reveal technical roadblocks for some of your concepts. But they are exactly the sort of thing that NASA should have examined more completely in ESAS. If there is ever a future, independent revisit of ESAS, I hope that innovative, bite-sized, and step-wise approaches like your concepts get into the mix.
I hope so too. Alas, I’m really not holding my breath for NASA to do the right thing. They seem to be headed for a train-wreck right now, and I’m not sure if it will be salvageable. So, I’m doing my best to think through alternative, commercial oriented approaches to “route around the damage” so to speak.
I’m not sure if I’ve found anything truly solid yet, but I am trying to work that angle.
I’ll try to visit and comment every so often.
Thanks! Comments (critical or supportive) are always welcome. That’s part of how I figure things out.
~Jon
Joe Blow,
Careful about demonstrated reliability numbers. Without confidence intervals, they’re useless. The way Doug Jones of XCOR once put it, the simplest rule of thumb (for a 95% confidence interval) was to assume that the next two flights both failed. Use that to calculate the reliability. So, with 8 successes and no failures, they’ve only really demonstrated an 80% reliability at a 95% confidence interval.
Confidence intervals take into account things like luck, and random distribution of failures. It’s possible that a given vehicle really only has a 98% reliability in reality, but you’ve gotten lucky so far. See the Saturn V for instance.
But the overall idea that demonstrated reliability is more real than “paper reliability” is valid. It’s just way too easy to miss important factors when doing that.
~Jon
jon goff wrote: First off, notice that they didn’t specificy a confidence interval. And it’s a good thing.
Thank you for your reply.
ESAS Section 8 Risk, figures 8-18 to 8-21 give mean LOC/LOM together with the 25% and 95% CI.
Engineers use numbers to analyze complex systems, numbers derived from qualitative understanding and experience. Quantitative methods are necessary to make rational decisions. To reiterate, what other method is there for comparing the reliability of unbuilt systems?
cIclops,
The problem is that component based reliability estimates like they did are notoriously inaccurate. They can give a general, qualitative feel for the relative safety of various approaches (and hence can be used to some extent for comparing and selecting between boosters), but the quantitative numbers are almost useless. I can pretty much guarantee you that Ares I will never fly enough to prove a reliability of even 98% at a 95% confidence interval. In the past, similar studies using similar methodologies have been drastically wrong. They ignore many of the important causes of launch vehicle failure, including human factors.
In other words, predicted reliability should be treated at best as a rough qualitative decision making tool, not as a rock-solid quantitative tool. Assuming that somehow you can predict in advance exactly what the failure modes of a vehicle can be and what the odds of each happening is naive. There are so many assumptions and so many factors left out of such studies that they are quantitatively suspicious.
Ares-I will not have a LOC statistic anywhere near ~1:2100 like they claim. Especially not at a 95% confidence interval. I wouldn’t even be surprised if they managed to kill a crew during the planned operational lifetime of this vehicle.
Assuming that a bunch of people who’ve never designed and fielded a launch vehicle have somehow come up with a revolutionary expendable launch design that is so much more reliable than anything designed by people who’ve been building them their whole life is just naive. Trusting those numbers as though they’re gospel truth is poor engineering.
~Jon
I enjoyed reading your thoughts on the NASA budget – so much so that I abridged a version and recorded it – dripping with sarcasm – as a podcast this morning.
Hope you get a kick out of it.
Mike,
Bravo! Thanks for podcasting that,
you did a much better job than I could’ve done. You have my permission to do the same for other posts whenever you want.
~Jon