Thursday, February 17, 2011

How COTS-D Was Killed

Lest we forget, under Mike Griffin NASA awarded to SpaceX an option in their Commercial Orbital Transportation Services contract to develop a crew transport capability. The Space Act Agreement looked like this:

Project Management Plan Review and
Crew Demo 1 System Requirements Review
Financing D1$10,000,000
Crew Demo 1 System Preliminary Design Review$22,420,000
Crew Demo 2 System Requirements Review$25,420,000
Crew Demo 1 Critical Design Review$20,420,000
Crew Demo 2 System Preliminary Design Review$20,420,000
Crew Demo 1 Demonstration Readiness Review$20,420,000
Crew Demo 3 System Requirements Review$25,420,000
Financing 2D$10,000,000
Crew Demo 2 Critical Design Review$18,420,000
Crew Demo 3 System Preliminary Design Review$20,420,000
Crew Demo 1 Mission$15,420,000
Crew Demo 2 Demonstration Readiness Review$18,420,000
Crew Demo 3 Critical Design Review$18,420,000
Crew Demo 2 Mission$8,420,000
Crew Demo 3 Demonstration Readiness Review$18,420,000
Crew Demo 3 Mission$8,420,000

As with all the COTS milestones, SpaceX would not have received these payments until the milestone was completed. The finance milestones were required to demonstrate that SpaceX could fund and complete all the milestones without using the payments from NASA as "seed money".

The COTS-D option was never activated. You may even hear some people at NASA say that it was never "funded", this is wrong. The final nail in the coffin of COTS-D came in the form of an intriguing exchange between Sen. Bill Nelson and then acting NASA Administrator Chris Scolese. Here's the relevant part of the long transcript.

Senator Nelson. In last year's authorization bill, there was guidance to NASA about COTS-D Space Act agreements to develop a U.S. commercial alternative to Soyuz. We authorized $150 million in funding for COTS-D. I noticed that you are
putting $150 million of stimulus funds toward the Commercial Crew and Cargo program, but not actually initiating COTS-D agreements. Why are you not initiating these Space Act agreements?

Mr. Scolese. Well, we are working the commercial program as you defined. There was cargo on it. We have those two contracts with SpaceX and Orbital to do cargo. We had one for COTS-D. I cannot recall a specific--$150 million to go on to COTS-D. We did this year in the stimulus identify $150 million to stimulate a commercial activity, and it is broken into two pieces: $70 million to go off and develop capabilities that any visiting vehicle would need, including commercial vehicles, and that includes developing the human space flight rating requirements, the requirements that you need to be certified for human space flight. As you well know, we build human spacecraft and design them so infrequently that we have to write those requirements down. So part of this is to make those available to everybody, make them understandable to everybody, and that will help not only the commercial providers broadly, but all of us. And then $80 million to stimulate activity for a commercial crew.

[recess for a vote]

Senator Nelson. I want to go back to the question that I had asked you earlier. You described the breakdown of how you intend to program $150 million for Commercial Crew and Cargo. Instead of putting the dollars into the various component pieces that would enable crew capability, would it not make more sense just to invest that in a milestone-based demonstration flight?

Mr. Scolese. We discussed that, and we believe that we need to take a measured approach to developing commercial crew. As you know, again it is a very difficult prospect to develop a crewed vehicle to carry crews safely to and from space, let alone rendezvous and dock with the Space Station. So we are working a measured development where we work progressively from developing the capability to get into space, to conduct the rendezvous and docking with the Space Station, to crew rescue, which can be done without having to worry about crew escape,
all the way up to carrying crew. That is the philosophy that we are working to achieve. To do that, we needed to do some things that broadly help the community that wants to do this, as I mentioned earlier, about developing clear and concise specs and standards so that we can safely put our crew on those vehicles. And further, I think you have seen the annual report of the Aerospace Safety Advisory Panel that had some questions about the detail of our human rating requirements. So that is all part of what we are trying to accomplish, and we believe that will get us a commercial crew capability quicker and safer than if we were to just go off and suggest that we fund a capability.

Senator Nelson. But that was not what the legislation said. The legislation said that $150 million was funding for COTS-D. In this case, you would not even have to pay until the COTS-D partner was able to successfully demonstrate that capability. Is that not right?

Mr. Scolese. It would be dependent upon how we structured it. Of course, we wanted to maximize competition for the vehicle. As you know, there is only one COTS-D provider.

Senator Nelson. Well, when I say "you," I am referring to NASA, and you were not the Acting Administrator at the time. This is an example of where NASA has not followed the legislation. Now, let me ask you this. Would $150 million be enough to demonstrate that capability?

Mr. Scolese. We would have to look at it, but I do not think so, sir.

Senator Nelson. Well, what do you think it would be?

Mr. Scolese. I would have to get back to you on that, but it would be several times that, I would expect, because recall, we have to develop not only the crew portion of it. We have to develop the life support systems, the launch escape systems, the recovery systems. All of those have to be developed and demonstrated, and $150 million does not seem enough to do that.

Senator Nelson. We had a unique opportunity, if NASA had listened and followed the law, we had a unique opportunity this year between the 2009 operating plan and the additional funds provided by the stimulus bill and the development of the 2010 budget to craft a COTS-D plan that would have funded the program at the level that the folks needed. That path was not pursued. NASA did not obey the law. Again, I am not saying it to you because you are the Acting Administrator since January 20, but I want to point this out that sometimes NASA does not want itself to be helped. We have got to get our act together.

And that was the last opportunity for COTS-D. Had NASA obeyed the law and provided the $150M that was allocated in the FY09 budget, SpaceX could have started COTS-D and completed the first seven milestones. When the $150M in stimulus money came in SpaceX would have been on much better footing to claim part of it. Instead, Sen Shelby was able to divert $100M to the development of Ares I, a vehicle that was later scheduled to be cancelled prompting him to insert language into law prohibiting NASA from doing that. Of the remaining $50M, SpaceX received none.

The official position of NASA seems to be that the CCDev program has replaced COTS-D. SpaceX has put in a bid for the new round, which NASA has been prohibited from starting due to the failure of Congress to pass a budget for FY11. Should SpaceX be successful, they intend to start work on the launch abort system which will also allow the Dragon spacecraft to land vertically on land. While I have been assured that the CCDev program will be "milestone based" like COTS, I still have my doubts that it will capture the simplicity of the COTS-D option.

Monday, February 14, 2011

New Space Music Video

An idea that has been bouncing around in my head for a few years..

Great to finally get it out.

Thursday, February 03, 2011

Making Fusion Rockets Relevant

If you read the literature on fusion rockets you probably have a pretty firm idea in mind of what they're good for and when they'll be relevant - in "the future". No good fusion rocket paper is complete without a superconducting magnet here, and a magnetic nozzle there - in fact, these widgets are a primary ingredient of any fusion propulsion design and the more infeasible or untested they are, the better. This seems obvious: fusion rockets are the future because we don't have fusion yet.. right? Actually, no.

Producing nuclear fusion isn't all that hard. Amateurs regularly cobble together desktop fusion devices like the Farnsworth Fusor and other contraptions. The significant hard problem of fusion is getting more energy out of the device than you put into it. The current government backed effort to achieve this is the ITER project who are building a tokamak style device, but many other schemes are also being tried, with significantly less funding.

One of these is the dense plasma focus of hydrogen/boron fuel, a combination called focus fusion. The technique is relatively easy to understand. You take a metal chamber and put a single tubular electrode in the middle, ringed by a number of solid electrodes. Pump all the air out with a vacuum pump and then add the fuel until it is at a few torr. Dumping about two mega-amps of current into electrodes causes a plasma compression called the "pinch" in which nuclear fusion occurs. The result is a stream of electrons in one direction, a stream of ions in the other direction and a whole lot of x-rays, and virtually no neutrons. These happen to be the perfect products for producing electricity and if that's your goal, it means you can do it very efficiently.

The challenge of focus fusion is getting enough power into the device to burn the fuel - typically done with a big heavy bank of capacitors - and containing that heat in the plasma for long enough. Hydrogen / Boron 11 (or pB11 as it is often called) is the hardest fuel to get fusion going, requiring temperatures over 123 keV. As such, dense plasma focus fusion researchers tend to use deuterium instead, which only requires temperatures of 15 keV. The government program uses deuterium/tritium which only requires 13.6 keV, but tritium is a little hard to come by - it has to be made in nuclear reactors - and is strictly controlled. Deuterium can be picked up in rented bottles from your local gas supplier.

Using a dense plasma focus device to produce deuterium-deuterium fusion is pretty simple and requires minimal startup costs - especially if you do your homework and learn from the mistakes of others. Unlike pB11 fuel, D-D fusion produces neutrons. Shielding fusion researchers from neutron exposure is easily achieved with two things: distance and concrete. Measuring neutron output can be as low tech as looking for bubbles in a contained gel, and as high tech as CCD detection of scintillator stimulation. When you're producing neutrons you know you're achieving fusion.

Getting back to rockets, let's look back up at how I finished my first paragraph describing focus fusion: producing electricity [..] if that's your goal. While nuclear-electric propulsion sure is sexy, what if our goal is just to make a good old nuclear thermal rocket? Back in the 60s the US did a lot of great nuclear-thermal rocket work. They were using highly enriched uranium folded into a solid core with liquid hydrogen running through it. They got specific impulse in the 850 s (vac) range and had plans to achieve higher power before being defunded for obvious political reasons. So what might a nuclear fusion thermal rocket look like?

As our goal is to produce heat, not electricity, it makes more sense to use deuterium as our fuel. We only need to produce pulses of electricity to feed into the electrodes to produce fusion, and the most readily available technology to do that with sufficient power density is a compulsator. Much like an alternator, a compulsator is an electromechanical device that converts mechanical rotation to electrical energy in the form of alternating current. A high power rectifying bridge converts that to direct current to feed into the dense plasma focus. Compulsators have been built for railguns which produce more than enough current (and way more than enough voltage). I haven't read much on reducing pulse width (sometimes called "rise rate") of compulsators, but the ~2 microsecond pulses needed for dense plasma focus does seem challenging.

The rest of the rocket cycle is pretty standard. The expansion nozzle is cooled by cycling the fuel through it, this heats the fuel enough for a state change to occur and the expansion is used to turn a turbine which pumps the fuel, and finally the fuel is used to cool the core. The only difference is that the turbine serves double duty by turning the compulsator. A smart engineer will recognize that the rotors of the compulsator could be the turbine. Similarly, although all three components are shown schematically as being on the same drive, there most likely will be gearing involved to keep the pump constant.

Unlike a device for the production of electricity, the dense plasma focus will probably be made from copper. This will absorb the x-rays and transfer the heat to the "fuel" (aka, the coolant, traditionally the propellant-which-isn't-an-oxidizer of a rocket has been called the fuel). The already slow neutrons will pass right through the copper core and be slowed more by the fuel, hopefully enough that they don't hit the outer chamber with enough velocity to make it irradiated or contribute to wear.

Speaking of fuel, most readers familiar with nuclear thermal rockets have probably been thinking about hydrogen this whole time. Although compulsators are certainly more mass efficient than equivalent capacitors and the means to recharge them, they are not known for being light. As a fusion rocket is incapable of spreading radioactive material into the atmosphere, the traditional safety concerns of launching nuclear thermal rockets from the ground does not apply. As such, propellant density is once again important and a hydrocarbon first stage fusion rocket doesn't need strap-on boosters trumping its inherent safety.