I made my previous post because I’ve just gotten progressively more annoyed about the general brainlessness that attends discussions of what is needed to launch crew on the vehicle SpaceX is actually developing for that purpose, namely Dragonrider. If an emergency arose that required making a “premature” crewed flight atop a Falcon 9, it is – right this very minute – possible to choose either a standard cargo Dragon or a prototype/test article Dragonrider on which to launch said crew. There is at least one of each sitting in the SpaceX rocket works over on Crenshaw Blvd. right now. I simply think that choosing a possibly incomplete Dragonrider which may, at this point, lack crew couches, cockpit instrumentation, life support system, etc., but which does have a fully built-out, even if untested, abort system, makes more sense than picking a cargo Dragon which definitely lacks everything on the preceding laundry list. I have no inside information about just what, in addition to a functional abort system, the test article Dragonriders will have installed. It just seems to me that a worst case list of deficiencies just puts a test article Dragonrider on par with a cargo Dragon. The latter’s lack of an abort system and the former’s possession of one makes this hypothetical decision a no-brainer of the the first water so far as I’m concerned. It also puts the lie to the oft-expressed notion that Dragonrider doesn’t have an abort system yet. It does, dammit! It may or may not have anything else it needs to get a crew to orbit, but it has an abort system!

Bottom line: you don’t need to spot all of the “Safety, Safety Uber Alles” morons their favorite talking point about Dragon lacking an abort system. I quite share your disdain for these delicate hothouse flowers. To paraphrase Rhett Butler, these people need to be kicked in the balls they haven’t got, and often, and by someone who knows how.

(1) The launch escape system for Dragonrider isn’t ready.

(2) The launch escape system for Dragonrider won’t be ready for some significant time to come.

(3) The launch escape system for Dragonrider is the most critical item on the critical path to a “complete product” as one of you put it.

You explicitly agree about (1) above and implicitly seem to agree about (2) and (3).

With respect to (1), I believe SpaceX still intends to accomplish a pad abort test of its launch escape system this summer and an in-flight abort test of it later in the year. The Dragonrider test article for the first of these tests is almost certainly, then, either completely built by now or in final production checkout. The test article for the in-flight test is probably, at a minimum, well-along in production too and may even be sitting, completed, in some convenient corner of the Hawthorne plant as we speak. In short, the flashy, rocket-y impressively noisy part of the Dragonrider launch escape system is ready. It may be as little as 90 days away from its first major test as an integrated system.

Now a launch escape system is more than just the parts that belch flame when called upon. There must also be a network of sensors and wiring in many portions of the 1st-stage structure and engines to provide physical indications of conditions that might require an abort to be initiated. I am under the impression that these things have been part of the Falcon 9 designs right along. If anyone knows differently, I’d be interested in hearing about it.

So, bottom line: Dragonrider already has a launch abort system that is “ready.” It simply hasn’t been all-up tested yet. Therefore, all of (1), (2) and (3) are, to the best of my knowledge, factually false propositions.

If, because of some extremely near-term emergency, it were considered essential to use a Dragonrider, in something close to its current state of development, to get personnel up to ISS, the people making that trip would almost certainly be doing so on one of the test articles with some last-minute additions (cockpit displays/controls? jury-rigged life support? crew couches?). They would not – repeat not – be looking at launching without any launch abort system, just without one that had been all-up tested yet. I hope you will agree that the latter is a much less worrisome proposition than the former.

If we can get through the remainder of 2014 without such an emergency arising, then the availability and functionality of the Dragonrider launch escape system will no longer be either a critical path item or a subject for crepe-hanging speculation.

There may still be other aspects of Dragonrider’s “complete product” status that would require jury-rigging/McGyvering for an emergency mission even after the launch abort system is no longer any kind of question mark. But I would also think there would be progressively less of this for putative emergency ISS crew missions in 2015, 2016 or early 2017.

Given that Elon and other SpaceX executives have publicly stated a goal of making the first crewed flight of Dragonrider sometime next year, I suspect the majority of the critical path items remaining are of the NASA review and blessing of paperwork piles variety. It would not surprise me in the least, therefore, if all of the significant hardware and software items needed for a complete Mk1 Dragonrider are either already in SpaceX’s hands or will be within a maximum of one year from the present.

Because the people making the program decisions didn’t understand the implication of that pie chart, and because the decisions weren’t rational, instead driven by politics.

The fact they are taking there time and not rushing comsat launches lends credit to the idea that they know better than to take stupid gambles with people’s lives.

Every flight will be gambling with people’s lives, whether they have an abort system or not, and if the mission is important, there is nothing “stupid” about it. You gamble with your life every time you drive to the grocery store.

Claiming launch escape systems aren’t so important also ignores the fact commercial launch isn’t well established, and a accident has far greater consequences than a few lives lost. Like car accidents, people dying on a say a Atlas V or a Falcon 9 will create a backlash towards the companies that made the product involved in that accident. It’s not guaranteed to say the public or Congress will accept the loss of a crew without debate, and it’s also wrong to suppose that somebody won’t spin the story to give commercial crew providers a bad name. I’m pro-commercial crew myself, but I’m not so daft as to think a year or more of time saved is worth risking commercial crew as an industry on.

Accidents happen, and not having reasonable safety measures like a launch escape system in place will damn the product. Look up the Ford Pinto and see how being seen as unsafe tainted it. Sales dropped off, and even now it’s treated like a joke. Ford had other products to sell at the time, but unlike Ford,, commercial crew companies have only one vehicle each. Commercial crew already has has to fight to eek out survival, budget after budget. Skipping tests doesn’t make commercial crew anymore palatable to those that oppose it, and gives them a opening to attack the program. Which isn’t a good idea, seeing as those opposed to commercial crew have made sure that it’s been half-funded from the start.

As to NASA disagreeing with me, if so Mr Simberg, why would they chose to spend time and effort making safety the highest priority for Ares I and expecting the commercial crew companies to test and certify things like launch escape systems? The reason launch and ascent has a lower risk in the pie chart because lunar flight in Constellation could of lasted for more than a week with plenty of guidance, life systems, and spacecraft engine problems having the potential to occur, not because launch and ascent is unimportant. Launch and ascent happens in under 10 minutes usually, and comparing a engine firing during that time to a CO2-scrubber operating the whole mission long, one sees that the scrubber has more opportunities to fail over a far longer duration. Keep in mind engines operating during ascent usually only have to start once, and since most launch vehicles have multiple stages, only few engines in certain launch vehicles are required to fire for 8 minutes or more, most fire for much shorter times. A CO2 scrubber will turn on and off as needed, for as long as needed, potentially needing to operate for a few hours total with dozens of start/shut down cycles. The chart doesn’t tell you the whole story, and certainly isn’t meant to show that launch and ascent is unimportant.

As for SpaceX taking up a lunatic NASA on the idea of going without launch escape systems, I’m optimistic they would know enough to refuse. The fact they are taking there time and not rushing comsat launches lends credit to the idea that they know better than to take stupid gambles with people’s lives.

NASA disagrees with you. Go look at the lunar mission risk pie chart in the ESAS report. Only 3% is associated with ascent. Spending too much on safety of that phase of the mission is a gross misallocation of resources.

SpaceX shows no signs of forgoing a launch escape system to cut costs/time.

Because no one has asked them to. If NASA went to them and said, “We know that the system will be safer with the abort system, but it’s important to get Americans into space on American vehicles now, and we’ll accept the risk,” do you think they’d refuse?

What I’m hearing here is that dying in a large explosion is worse that dying because a parachute or a pressure seal failed. Not so. What I find objectionable is what sounds a little like the assumption that a launch escape system forgives the lack of a more general approach to mission safety. If what you’re doing is developing a launcher from scratch, the money you spend on a launch escape system might be better invested in engineering that reduces the need for such a system. It is therefore simplistic and even somewhat dangerous to say that having a launch escape system determines the completeness of a product.

An automobile really should have an airbag that fills the entire cabin, but instead they chose to make the front ends more safely crushable, and installed more limited airbags. Are those automobiles not “complete” because of this?

Saying it’s the safest phase in a lunar mission is to be ignorant of the very real dangers, and one only has to look to the ill-fated N1 to see what a rocket supposed to launch a lunar-bound ship can do when things go very wrong. The N1 launch attempts created the biggest, or some of the biggest, non-nuclear explosions in history. On a few of those launches the only things that worked correctly was the launch escape system, and that would have made the difference between life and death for a crew. While the smaller rockets used for orbiting astronauts for mere Earth missions don’t have the same explosive power as the N1, they can still blow up, and it’s a issue to deal with by engineering, not ignoring the problem.

Even on better behaved rockets like the Saturn V, ensuring the launch escape system worked was a high priority for engineers of the time. Just to point out things haven’t changed, even in the company trying to shake up the launch industry, SpaceX shows no signs of forgoing a launch escape system to cut costs/time. SpaceX has shown a trend of taking as much time as they need to get things right, like scrubbing launches, even on comsat flights. They seem get the idea that spaceflight demands a certain level of care, and human spaceflight even more so.

Also, statistical speaking, 11 deaths* during entry aren’t significant enough to argue that it is a more dangerous time for crews than launch and ascent, considering the fact that total crewed spaceflights number under 300, and that on these flight under 600 people were flown. To do so shows either a lack of understanding how statistics work, or a deliberate attempt to misuse statistics. For example, 7 people** have died during launch and ascent, and for statistical purposes the numbers are close enough to be equal, however, since we’re still dealing with a small pool of data, the numbers still don’t mean much.

*I’m counting Soyuz-1, Soyuz-11, and STS-107 as entry failures. Your opinion may vary.
**STS-51-L

To the degree that’s true (historically, more people have actually died in entry than during ascent), it’s only true for LEO missions. For lunar missions, ascent is probably one of the safest phases. And only three people (cosmonauts) have been saved by an escape system in the history of human spaceflight. Which is why spending so many billions to make Ares I “safe” while spending nothing on the hardware needed to get beyond LEO was a foolish misallocation of funds on Constellation (a point I make in my book).

When it delays an important mission, yes. What you are saying is that it’s not important to send American astronauts on American vehicles. Dragon will be safer (probably) with a launch-abort system than it is now, but that doesn’t mean it’s not safe “enough” without it.