Sign in
Remember me · Forgot password?

Shit Elon Says - Transcript - SpaceX Press Conference at the National Press Club

Transcript History  

Hello everyone, thanks for coming on short notice. There's two things I want to talk about today. I'll start off with the landing of the boost stage. I'm happy to confirm that we were able to do a soft landing of the Falcon 9 boost stage, in the Atlantic, and all the data we've received back shows that it did the soft landing and was in a healthy condition after that. It does look like it was - that the stage was subsequently destroyed by wave action. The seas were very heavy. It was like 15 to 20 foot seas. We suspect the stage was destroyed due to the - essentially, the stormy seas, but the data is very clear that - it shows a soft landing, it shows deployment of all the legs, and the stage was in a safe state in the water. We also have a video feed, although the link was very weak. So, for the video feed, we're trying to clean the video feed up and have it be something that - where you could make sort of sense of it - and we're going to clean it up as much as we can on the SpaceX side and then we're going to post it on our website and try to crowd source - to see if people out there can make it look even better. I know there are people out there who are really good at fixing video streams.

I think that's a really huge milestone for SpaceX, and certainly for the space industry. No-one has ever soft landed a liquid rocket boost stage before, and I think this bodes very well for achieving reusability. As people have probably heard me say, I think what SpaceX has done thus far is evolutionary but not revolutionary. I think if we can make - if we can recover the stage in-tact and relaunch it, the potential is there for a truly revolutionary impact in space transport costs. The cost of propellant is actually only about 0.3% of the cost of the rocket, or of a mission. So, if the mission costs $60 million, the cost of propellant is only $200,000. There's potential there for ultimately a hundred fold improvement in the cost of access to space. I think, with the information we've learnt from this flight, we know we can soft land the rocket and we're taking some additional steps with the upcoming flight, which will be a commercial mission for Orbcomm, to have a much greater probability of getting to the stage in time and recovering it. We're securing much bigger boats this time. Unfortunately.. I think we called every boat on or around the east coast above a certain size and it turns out most of the boats that can take really heavy seas are actually in the gulf and elsewhere but apparently not in the greater Florida area. This time we're going to have much greater capability boats and I think we'll also - we kinda got unlucky that we essentially landed the stage in the middle of a big storm. Hopefully this time we'll not have to do that. It'll also be splashing down, or landing in the water, much closer to land than last time. So I think - hopefully, we'll avoid some of the deeper ocean stuff.

"With each successive launch - we have several more launches this year - we expect to get more and more precise with the landing and, if all goes well, I am optimistic that we'll be able to land the stage back at Cape Canaveral by the end of the year." So that's all great stuff and I think, assuming that happens, we should be able to refly the main boost stage in time next year. It's somewhat of a huge day because we've been trying to do this at SpaceX for a long time - it's been twelve years and we finally did it, and now we've just got to bring it back home in one piece. So, are there any questions about that?

[Question about landing site at the Cape] Yeah, we've actually worked with air force range safety to identify several locations at Cape Canaveral where we can land the stage. They've actually been really helpful. At first we were concerned that range safety might be obstructionist but they've actually been more than supportive, and there's several places where we can land. It kinda depends on how tightly we can control the landing point and I think if we can demonstrate tight control there is a lot of places at the cape where we can land.

[Question about recoverable parts of the stage.] The recovery operations were challenging. We actually couldn't get a boat out there for two days. We literally couldn't find anyone willing to go out there. We even called the Coast Guard, and the Coast Guard wasn't willing to go out. The soonest we could get out there was, as I said, two days latter. We actually have been able to find pieces of the interstage. The interstage is the carbon fiber structure that joins the first and second stage. That's certainly something that you would expect to get destroyed by wave action as it's got that big open hole at the top and waves will come in and blow it apart. We've recovered most of the interstage. We recovered a portion of one leg, and there are a bunch of other little bits and pieces. We've not recovered anything of the main aluminum-lithium airframe.

[Question about landing accuracy and number of attempts this year.] In this case we were just trying to get the rocket to go to zero velocity at sea level and we weren't trying very hard to get to a precise location. None the less, we were within a few miles of our target. I think that's pretty good. We could certainly tighten that up massively, with a little bit of effort. [Question about how close you need to be in order to be allowed to land on land] Probably we need to be comfortably within less than one mile radius error. In principle, we should be able to land with the accuracy of a helicopter. Literally - if you've seen the test flights of the rocket, you can see just how precise it is. I mean, it lands to within less than a meter of its target.

[Question about how many more landing on water tests.] We'll only be doing the water landings until we're confident that we can land with accuracy, and then we'll be transitioning to land landing.

[Question about reusing stages and business model.] Our pricing right now assumes no reusability. None of our prices are contingent on that. Any reusability we're able to achieve would only allow us to reduce prices from where they are today. The more often we're able to fly and the more often we're able to reuse the stages and the less work they require between flights, the lower the costs can be. The boost stage is roughly 70% of the cost of a launch. So, if we're able to reuse it and refly it with minimal work between flights, and customers are comfortable with that - and it might take a few years for customers to get comfortable with that - then obviously there's as much as - ultimately - a 70% reduction from where things are today.

[Did you think it was going to work on the very first try?] Well, it's not, strictly speaking, the very first try. The first time we tried to do a soft ocean landing was Falcon 9 flight 6, that was about eight months ago, and that's where we had an issue just before - basically the rocket spun up too much to the high aerodynamic torque coming in from hypersonic velocity. Something we didn't expect is that even small asymmetries on the outer skin of the rocket would cause it to spin up because it's facing such high forces. The nitrogen thrusters on-board the rocket for that flight weren't able to overcome the aerodynamic torque. For this flight we just had, we doubled the thrust of the nitrogen thrusters, and also added a bunch more nitrogen cold-gas propellant. I gave it sort of a 40% to 50% chance of working. I was that percentage I think. I was actually positively surprised by the fact that the legs deployed, the stage landed, it practically sat there for 8 seconds before we lost data. That's a better outcome than I'd expected.

[Question about environmental benefits.] There's no question, compared to the way that rocket flight normally works, where the stages all just come back and crash and there's a bunch of rocket stages at the bottom of the ocean, reusing the stages is much better for the environment. It takes much less energy. You don't have to keep rebuilding your rockets. I think it's - there's certainly a sustainability element there.

[Question about landing vertically and time to turnaround a stage.] We know with certainty that it landed vertically with the legs deployed and in an essentially nominal configuration because we've got the telemetry to show that. There's multiple sources of telemetry. We have sensors on each individual leg. We have multiple inertial sensors, and multiple GPS units on it. They all sort of agree with that conclusion. In terms of it being ready for reflight, if we recover a stage from the ocean it would probably take a couple of months to refurbish it for flight. However, for stages landing back on land near the launch site, in principle we should be able to refly it the same day. So, it's a huge difference. That's why we're really focused on trying to get it back to the launch site. That's what would make the hugest difference on reusability. Obviously, with the space shuttle we had a case where there was a partially reusable vehicle but the space shuttle was not either rapidly nor completely reusable and in-order to achieve a revolutionary improvement in the price of spaceflight, any reusability has to be both rapid and complete.

[Question about demonstrating reflight reliability.] Absolutely. I think what we'll have to do is do a demonstration reflight without an operational satellite on-board. If that demonstration relight works, and some customers may want more than one, then that's the thing that would really, ultimately, convince them.

I should probably transition to the other news item, which is not as - well, the first one is more positive, this one is a bit negative.

SpaceX has decided to file suit and protest the air force EELV block buy. This is a 36 core sole-source uncompeted procurement that was signed earlier this year, that essentially blocks companies, like SpaceX, from competing for national security launches. Essentially, what we feel is that this is not right. "The national security launches should be put up for competition. They should not be awarded on a sole-source uncompeted basis." It just seems odd that if our vehicle is good enough for NASA and supporting a $100B space station and it's good enough for launching NASA science satellites, for launching complex commercial geostationary satellites, and really every satellite you can imagine, there's no reasonable basis for it not being capable of launching something quite simple like a GPS satellite. This really doesn't seem right to us and we've tried every avenue to try to figure out why is this the case. Is there anything we can do besides file protest, and it seems we're essentially left with the only option, which is to file a protest.

The ULA rockets are, basically, about four times more expensive than ours. So, this contract is costing the US tax payers billions of dollars for no reason and, to add salt to the wound, the primary engine used is a Russian main engine - made in Russia. Moreover, the person who heads up Russian space activities is Dmitry Rogozin, who's on the sanctions list. It seems pretty strange. How is it that we're sending hundreds of millions of US tax payer's money, at a time when Russia is in the process of invading the Ukraine, and it would be hard to imagine someway that Dmitry Rogozin is not benefiting personally from the dollars that are being sent there. On the surface of it, it appears there's a good probably of some sanctions violation, as well. We think this deserves to have a spotlight on it. Let the sun shine on this. As they say, sunlight is the best disinfectant. If everything's fine then I guess that's great, but that seems unlikely to me.

[Question about filing.] It's being filed in the court of federal claims and there may be others that join but currently it's just SpaceX. Something I want to be real clear about, this is not SpaceX protesting and say that these launches should be awarded to us. We're just protesting and say that these launches should be competed. If we compete and lose, that's fine, but why were they not even competed? That just doesn't make sense. We've heard statements like, well there's this certification process, like okay well, we're most of the way through that certification process, so far there's been zero changes to the rocket, this is a paperwork exercise. Since this is a large multi-year contract why not wait a few months for the certification process to complete and then do the competition. That seems very reasonable to me.

[Question about the certification process.] Technically we've done nine Falcon 9 flights. Of the exact configuration that the air force wants, we've done four. All four have completed their mission. They obviously worked. So, yeah.

[Question about informing the air force.] We did inform the air force just before the press conference. First of all, I should say that it's not as though we're battling the whole air force, this is not - that's not the case at all. We're on very good terms with the vast majority of the air force. Our concern really relates to a handful of people in the procurement area of the air force.

[Question about air force moving the goal posts.] I guess the goal posts were certainly removed with respect to SpaceX compared to when Boeing and Lockheed competed for the EELV contracts. Boeing and Lockheed, that's before they merged their launch business into United Launch Alliance, were awarded a large number of launches - I think maybe 30 or 40 launches under the EELV program - before doing a single flight of the Atlas V or the Delta IV. When SpaceX came along and said, hey, we'd like to compete we were told that we had to have three flights of the exact configuration that the air force would fly before they would allow us to compete. That seems like - it's a bit like pulling up the drawbridge. It's not quite right - but we did that, we actually did that, and after we did the three launches and completed them, a month later we were told, oh, the air force has done this huge sole-source procurement, uncompeted, to Boeing-Lockheed. We're like, but we just did the thing you asked us to do. That seems pretty wrong. I'll say this - normally when there's a huge multi-billion dollar sole-source procurement, there's a justification given. In this case, there was no justification provided for this.

[Question about Tesla.] No sorry, I can't answer Tesla questions.

[Question about price of doing business with the government.] In terms of Falcon 9 and Falcon Heavy, we think probably it's 2/3 Falcon 9, 1/3 Falcon Heavy and, in fact, in our protest we only are protesting launches that we could do. So we're not protesting launches we couldn't do. We're trying to be as fair and reasonable as possible here. In respect to, wouldn't our costs be just as like their costs, I don't know why their rockets are so expensive. They're insanely expensive. On the order of $400 million per flight, all things considered. In our case, our commercial price is $60 million and we expect roughly a $30 million increase due to air force mission assurance requirements and that seems to be bearing out. So yes, it makes our rocket 50% more expensive, but it's doesn't make it 400% more expensive. I think we have the advantage that our rocket was designed to be built in a factory that was designed in the 21st century, whereas Atlas V and Delta IV were designed in kinda the 90s and, in fact, have a lot of legacy hardware that stretches back to the 70s and 80s, even before if you count the RL10. I think the fact that we have a new rocket designed with new manufacturing techniques is very helpful for the cost. I think we have a number of design innovations in the rocket that are also helpful for cost. Some of it's just elementary things like, if you look at the Atlas V, it uses three types of propellant. It uses solid rocket motors. It uses a kerosene first stage and a hydrogen upper stage. Our rocket, Falcon 9, by comparison is a two stage rocket and both stages just use propellant grade kerosene. Right off the bat, to the first order of approximation, our operational costs of launch are a third of what an Atlas V would be.

[Question about waiting so long to file the protest and state for commercial launch pad.] We only learnt about the big sole-source award in March. It may have been signed in December but it only came to light, interestingly, one day after the senate hearing on EELV launch costs, which seems remarkably coincidental to me. I don't think that's an accident. We've really just had about a month of awareness and we've been somewhat reeling from that news and trying to see, is this real? Is this actually what's going to be the case? When we basically made no progress with discussions with the air force, we thought we have basically no choice but to file the protest. Oh, launch pad. Our primary location is Florida at Cape Canaveral. We've got our pad 40 on the Cape Canaveral side and then pad 39A on the NASA side and we're actually building out 39A with the ability to do the Falcon Heavy. So, probably the first Falcon Heavy launch will be out of the 39A pad which is really an amazing pad with incredible history. It's where Apollo 11 launched from. For the future we expect most launch activity to go out of the Cape Canaveral, Cape Kennedy area. We're also developing a launch pad on the south coast of Texas, near Brownsville. We're waiting on the final environmental approvals for that. We're expecting to get those soon, and we'll probably have that site active in a couple of years. Then, of course, we've got out site at Vandenberg air force base in California for polar launches. As a rough guess, I think we'll have NASA flights will tend to go out of 39A. Air force and intelligence flights out of pad 40. Commercial geosynchronous flights out of the Brownsville location. Then all polar flights, government and commercial, out of Vandenberg. That seems like the logical breakdown.

[Question about Russian engines.] I do think it's very questionable, particularly in light of international events. It just seems like the wrong time to send hundreds of millions of dollars to the Kremlin.

Allright, thanks everyone.

[Question about certification] I think the reasonable thing to do would be to cancel the 36 core contract, wait a few months for certification to complete, then conduct a full competition. I think that would be in the best interests of the american public, not by a small margin but by a huge one.

[Inaudible question.] Well, I think they should.

Old New Diff   Stop


TODO: diff