SpaceX’s Musk and Thompson Q and A

by Braddock Gaskill

SpaceX founder Elon Musk and VP of Development Operations Chris Thompson have answered questions submitted by NASASpaceflight readers.

In the full Q&A text below, Elon Musk reveals initial launch site plans for the Falcon 9, how the STS External Tank effected Falcon material selection, mission control with a crew of six, a redesigned Merlin 1 engine with regenerative cooling, and new details on the innovative first stage tank design.

Last month we gave forum readers an opportunity to submit questions for Chris Thompson, a Boeing veteran now responsible for SpaceX’s development of vehicle structures, manufacturing, and development operations. Mr. Thompson has been rather busy readying a new Falcon 1 first stage for the next launch attempt on February 8th. However, to our surprise, SpaceX founder and CEO Elon Musk has personally answered most of your questions.

What were the design decisions that caused SpaceX to choose an ablative engine chamber for Merlin 1? Is the Merlin 1 to be redesigned to use a regenerative cooling system? And will the Merlin 2 engines replace the Merlin 1’s on some Falcon 9 configurations?
Thank you for your time. -Jim P, Warminster, PA.

‘Ablative was chosen because we thought, incorrectly, that it would have a lower development cost. All future engines will be regen and we will be coming out with a version of Merlin 1 that is regen.

‘No comment on Merlin 2 yet. We will be releasing a spec to the public in a few months. All we can say at this point is that it will be the world’s largest engine, where engine is defined as a single thrust chamber (the only logical definition in my view). The Saturn F-1 was larger, but obviously is no longer in production.’ – Elon.

Info sheet on the Merlin 2 engine proposal, a large RP-1/LOX-powered engine. Merlin 2 would have powered larger rockets beyond the Falcon 9 family, and rivaled the F-1 in terms of size and performance. The design was shelved before any testing occurred, as focus shifted to the Raptor engine. Credit: SpaceX

2. Have your former colleagues at Boeing given you their opinions on the Falcon series? Positive? friendly joking? Critical? -Martin Johnson, Perry, FL.

‘My former colleagues are very supportive of what we are doing at SpaceX. Joking is always part of the interaction, but I have never had any critical responses.’ – Chris.

3. What are some of the innovations that are reflected in lower operational and launch costs? What are some of the innovations that show up in the vehicle configuration? -spacester, USA.

‘Lower cost contributors:

‘Engines: Main engine is a a single shaft pump LOX/RP engine with a pintle injector. Lowest possible cost choice of architecture. Several proprietary innovations in the pump and injector. Upper stage is a pressure fed LOX/RP engine with a pintle injector. Dead nuts simple.

‘Structure: Both stages use a monocoque (no expensive isogrid) aluminum tank with a common dome. No intertank required. First stage makes use of a SpaceX proprietary semi pressure stabilized design (first ever design of this kind). Can stand on the pad unpressurized with a full load of propellant but makes use of the erector clamp for support in high ground winds. With even partial pressurization, it can stand unaided in extreme winds.

‘Interstage is carbon fiber sandwich, but only vacuum oven cured (doesn’t use an expensive autoclave).

Schematic of the Falcon 1 rocket from several angles. Credit: SpaceX

‘Avionics: Only orbital rocket in the world to use 21st century electronics, which are lower cost and higher performance.

‘Launch Ops: The entire vehicle is built and assembled in the factory and sent to the launch site mounted on its mobile launcher/erector. No standing army required at each launch site. ‘Time from rocket arrives at launch pad to launch can be less than a week, depending on satellite readiness.

‘Highly automated and streamlined countdown sequence requires only six SpaceX engineers in the control room and about that number of technicians on the pad. Telemetry is sent via secure Internet link back to HQ, where dozens of engineers can view the data in realtime, but are only needed for the hour prior to launch.

‘Other contributors:

‘- LOX/RP is the lowest cost choice of propellants on every level’- Having the same propellants on both stages significantly reduces operational and development costs’- SpaceX has the ability to make almost any part of the rocket in house, so has a great deal of bargaining power with suppliers. We are never locked in to anyone.

‘- Engineering is colocated with production and launch ops, providing a tight communication loop. As soon as they spot something expensive in production/launch ops, we engineer a lower cost solution.

‘- The Falcon 1 1st stage is designed to be reusable, as are the Falcon 9 1st and 2nd stages. Once we make that work efficiently (it will probably take 3 to 5 years), SpaceX should achieve an order of magnitude reduction in $/lb to orbit.’ – Elon.

Recovery training for the Falcon 1 with a dummy stage. Falcon 1’s first stage was intended to splash down under parachute to be reused, but this was never demonstrated. Credit: SpaceX

4. What were the supplier problems that caused you to move away from Al-li alloys on the Falcon 5, 9 and what alloys will be used instead? – J.L. Clifford, Chesterfield VA

‘All the mergers in the aluminum industry and the uncertainty surrounding the Shuttle ET (by far the biggest user of Al-Li) reshuffled the deck for Al-Li. We tried for two years to order it and only now are receiving some to try out. Until we are sure of having a reliable long term supply of the material, it would be unwise to baseline it into a design.

‘2219 (used in Delta IV) is our primary aluminum alloy, along with some use of 2014 (used in Atlas V). 2219 is not the best strength to weight, but it is easy to work with and has low corrosion.’ – Elon.

5. Given the huge amount of work taken out to create the pad for the Delta 4, have plans been made already for a suitable pad for the future generation of launchers you have in the pipeline? Or do the Falcons require less infrastructure?
-Lt. Col. (edit for security), Tyndall AFB, FL

‘The Falcons are designed to use much less pad infrastructure than other launch vehicles. It is much more efficient to do work in the controlled environment of a factory with maximum availability of resources, than out on the launch pad.

Kwajalein, which has all azimuth capability, will be our primary launch site in the short term. Vandenberg is also an option for customers with polar or near polar orbits, but there is uncertainty around our ability to use 3-W and it would not make financial sense for us to build up another launch pad there (better just to use Kwaj). However, in the long term, we expect to do a lot of launches from the Cape, Vandenberg and Kwaj.’ – Elon.

[Editors note: we asked for clarification, and yes, Elon is planning on Falcon 9 flights from Kwajalein]

An aerial view of Omelek Island in the Kwajalein Atoll, where every Falcon 1 launched from. Plans to expand the site to support the Falcon 9 were shelved, and the facility was abandoned after the final Falcon 1 launch in 2009. Credit: SpaceX

6. Is accommodating a cryogenic upper stage such as the centaur on the falcon V or IX a possibility? -J.L. Clifford, Chesterfield VA

‘That is certainly a possibility, although it would increase costs quite a bit. In the short to medium term, we will be sticking with LOX/RP, but H2 is very likely in the long term, particularly for missions beyond Earth orbit.’ – Elon.

7. For those of us who appreciate how each rocket launch is in itself a major achievement because of the complexity of rocketry, scrubs are part and parcel of launch attempts. But for those who are used to the routine fashion in which rockets are launched, how would you describe to them, in ways that they’ll understand, just how difficult it is to get a rocket launched and carry out its mission, especially a new vehicle? -Simon Manson, Manchester, England.

‘The analogy I’ve use before is to imagine creating a huge software program that can only be tested in little pieces on a computer that is slightly different from what it is supposed to run on. However, when you do run it as a whole on the actual computer for the first time, it must run almost flawlessly without a single significant bug. When is the last time you saw a software program do that?’ – Elon.

8. You manage to run the Falcon 1 from a crew of just around 25 people. As you increase through your range, how much do you expect this crew number will the exponentially rise, for instance for the Falcon 9? Will it still be far less than required on other launch systems? – Gary Hardy, Swindon, England.

‘For a normal launch operation, the crew is actually only about 12. It is double that only because this is the first launch and we want as many eyes on the data as possible.

‘When falcon 9 is ready for launch , there will probably be a modest increase in SpaceX personnel — perhaps to something like 16 or 17 people, but hopefully less. It shouldn’t take many more people to launch a Falcon 9 than it takes to launch a 747. Per my answer above regarding piping in remote telemetry to HQ, we will actually have as many as 50 or 60 engineers checking data for the hour before launch, which is an extremely efficient use of resources.’ – Elon.

9. Could you describe a little about what it is like to work at SpaceX, and how it differs from space industry work elsewhere? -Braddock Gaskill, USA

‘SpaceX is a kick ass company with some of the best people in the industry and a fun place to work. Unlike most companies, every single person is good at what they do and highly motivated. Moreover, since we are a small company, each person at SpaceX is a major contributor to the team and what you do actually matters. When working at a giant conglomerate, you are just a number that contributes to a small part of the collective.’ – Chris. would like to thank SpaceX Communication’s Dianne Molina for facilitating this opportunity, along with Elon Musk and Chris Thompson for their time.

Related Articles