Tuesday, March 27, 2007

Space Access '07 - Armadillo Aerospace

John Carmack – Armadillo Aerospace

(Note: Select "Space Access '07" below for all posts on that conference).

John talked about the developments leading up to Armadillo’s Lunar Lander Challenge vehicles, which started well before the LLC was announced. He showed a video on the team. They have spent $3 Million so far, over 6 years. The have a well-equipped shop. Most of the work is from volunteers. He showed some concepts of combinations of Pixel-like vehicles strapped together in a layer. He also extended this to multiple layers or stages. For the Lunar Lander Challenge, they have really tested the landing gear a lot now. It won’t break. They have also tested software updates that make the vehicle easier to control. There will be tests soon on an improved engine. They are looking into an essentially infinite duration engine that would not wear out under use. They have enough performance for suborbital flight. The system is configurable and modular. You just bolt them together. Just build more when you need more. Economies of quantity are better than economies of scale (ie more modules are better than bigger modules). This reminds me of computer concepts of multiple CPU chip architectures, or parallel processing in software. In these cases, depending on the architecture and problem to be solved, there are limitations on how much improvement each new “node” will bring. I’m not sure what the corresponding limitations are in the case of the Armadillo modules. Various configurations of these modular structures built on LLC-like modules are planned.

The vehicle has a slow ascent – about 250 mph. He expects to lose some vehicles. Now the vehicles are more “savable”. They have had about 6 major tests since the X PRIZE Cup in ’06. They spent about 25% of their time on the FAA license, which they don’t consider to be too bad, especially for a bigger team. Maybe they will apply for a commercial launch license. Now they have some commercial customer prospects. By the end of the year they expect to be space capable. (I have to caution that I don’t have any way to really assess these plans, but they seem ambitious to me.) John said the insurance wasn’t too bad either. They are trying lots of engineering trades and really testing a lot. They aren’t trying to be perfect on the first attempt in the traditional rocket development style.

For commerce, they started with T shirts, speaker fees, and sponsorship. Now they expect to win the Lunar Lander Challenge. This year they have the time they need to win the challenge. They are ready to start putting people full-time so they can go faster. They could effectively go twice as fast this route, but not much more. They plan on a vehicle that can take people to space.
They are making steady progress. They have a variety of potential business deals. They expect the business case to be easy. There is no need to earn lots back because they won’t spend lots. The vehicle will have 360 degree views (not portals). Some people will want to take off like a plane, others will want to take off like a Real Rocket”. (laughter).

The plan is to start with 4 modules, test, then go to a 9 module cluster. This gives engine out redundancy. Then they will put a capsule on the top to reach space. An upper stage would be used to reach orbit. There are lots of engine options – different propellants, etc. The above plan gets a minimal satellite to orbit. However, you can keep adding modules. (Note: I’m not sure how strong the materials to hold the modules together would need to be – maybe they need the Centennial Challenge Space Elevator Tether to produce something really strong. I’m not an engineer so I’ll leave it up to the Armadillo team. It sure is an ambitious plan.) For example, you could have 64 engines. They would have to be reliable enough for such a configuration. This is possible – for example, Jumbotron CRTs, golf course sprinklers, etc. When the modular approach hits diminishing returns, they can make the modules bigger. They expect to be able to do a Falcon 5 class delivery.

They want to get into vertical drag racing. If commercial deals don’t come in they will go for those deals instead. Similarly the X PRIZE Cup could be overruled by a commercial deal. Assuming no commercial deal overrides the LLC, they expect to do 20+ flight tests before October, to win the Challenge money, and then plan a 4000 foot flight in Oklahoma.

They are using 3-axis GPS. The new system is cheap. They should be able to land if there’s a problem. There are many opportunities to improve the vehicle. Even without the improvements a 3-stage configuration should get to space. Even if they only get to 150,000 feet, there should be commercial opportunities.

They have a simple capsule planned, but an external company could build a capsule with a lot of safety features.

That was the whirlwind tour of the plans of Armadillo, with the Lunar Lander Challenge prize as an important incentive and source of focus and enthusiasm on their path to a commercial service. We will just have to wait and see if these ambitious plans pan out.

For more on this talk see RLV News (conference notes and conference video update) and Why Homeschool. Also check out Rand's post at Transterrestrial Musings on the talk (including a commentor asking why we don't have more Centennial Challenges, and later discussion and more discussion on the talk.

The Space Access web site has more interesting information about the issues with cheap access to space over the years, up to this year's conference. Mark your calendars for next year's highly recommended conference.

The Armadillo Aerospace site has information on their module development concept, the 2006 Lunar Lander Challenge event, the video they showed at the Space Access '07 event, and much more.