Space Exploration Technologies Corporation (SpaceX) March-April 2004 Update
March/April Update
First Flight of Falcon V Sold!
SpaceX is pleased to announce that we have a signed contract and a deposit for the maiden launch of Falcon V. The customer is a US commercial company (Bigelow Aerospace) and launch date is set for November 2005 from our launch complex at Vandenberg Air Force Base in California. Further details will be provided by the customer at their discretion in coming months.
Falcon V with Standard Fairing
Slight Price Reduction for Falcon I
We also anticipate announcing additional customers for Falcon I in the next few months. As mentioned in the Aviation Week cover story, the Falcon I price has been reduced to $5.9 million plus range costs that vary by launch location.
The cost reduction is driven by the fact that Air Force range safety is allowing us to use thrust termination rather than explosive termination. In fact, Falcon will be the only US launch vehicle approved for thrust termination. Our continuing aspiration is to find other ways to reduce price without affecting reliability. This is a case where there is actually a very slight improvement in reliability by removing the explosives.
Per our standard policy, any price reductions are made public and are applied retroactively to customers, so a customer can always be sure of having the best price.
Two Major Problems Solved
After a series of hiccups debugging our main vertical test stand, the propulsion team was able to run the Merlin turbopump with the new impeller. This was the third impeller design and third time lucky. There is more detail below in the propulsion section, but suffice to say that we have overcome the final major issue with our Merlin engine required for flight!
This has been a pretty tough development and there is good reason for it. Merlin will be only the second new American (orbital) rocket booster engine to see flight in over twenty-five years. The other engine is the RS-68 from Boeing and the one before that is the Space Shuttle Main Engine.
The other problem solved is the flight termination receiver, where we have worked out a compromise that appears to satisfy all parties, both for first and future flights. It involves some modest additional expenditures and time on our part, but much less than earlier estimates. We’d like to thank Air Force JAG, AF Range Safety and Boeing for working through this with us. Also, Cincinnati Electronics is definitely high on our preferred vendor list.
Falcon User’s Conference
Our first user’s conference took place in late March and seemed to go quite well. Attendees were present from a number of commercial companies as well as the Air Force and NASA. We received a lot of good feedback from customers on satellite processing, as well as areas for improvement in flight environment and mechanical/electrical interfaces.
In particular, we took back the need for a good size clean room with an overhead crane in close proximity to the launch pad. Fortunately, Lockheed, our nearby pad mate at Vandenberg AFB, just gave up such a building and we will be taking it over. They are welcome to borrow a cup of liquid oxygen from us any time ? With a little refurbishment, it will be a first rate satellite processing facility.
More Testimony
Last month, I was asked to testify before the House of Representatives Subcommittee on Space regarding NASA/DoD cooperation and before the President’s Commission on Moon, Mars and Beyond. On May 5, I’m testifying before the Senate Subcommittee on Space regarding the future of space launch vehicles, which will be webcast live.
Protesting the Kistler Deal
As some may have read, SpaceX is protesting the contract offered to Kistler Aerospace by NASA. Although we have a high regard for NASA as a whole, we are against this particular contract. We believe it does not support a healthy future for American space exploration. However, this should be viewed as an isolated difference of opinion between NASA and SpaceX, and we look forward to working with NASA in other areas.
As background for those unfamiliar with the Kistler deal, the approximately quarter billion dollars involved in the contract would be awarded for flight demonstrations and technology showing the potential to resupply the Space Station and possibly for transportation of astronauts down the road.
That all sounds well and good. The reason SpaceX is opposing the contract and asking the General Accounting Office to intervene is that it was awarded on a sole source, uncompeted basis to Kistler instead of undergoing a full, fair and open competition. SpaceX and other companies (Lockheed and Spacehab also raised objections) should have been offered, but were denied the opportunity to compete on a level playing field. We are only asking for a fair shot at meeting the objectives, not demanding that we win the contract.
Giving a quarter billion dollars to a company like Kistler, which declared bankruptcy last year and has not managed to build a complete prototype in eleven years of operation, without even holding a competition sends a very negative signal to the marketplace.
As it is, we already have a means of space station access via the Russian Soyuz and the European Automated Transfer Vehicle. If what is sought is American access, since US tax dollars are paying for this, the deal makes even less sense, since their intent is to operate from Australia and use exclusively Russian engines.
Our suggested solution to this problem, given that the Kistler contract is already performance based, is to open it up and turn the objectives into a prize that can be won by any company. That company might be Kistler, SpaceX, Lockheed or an entity yet to be formed (perhaps by a reader of this newsletter). Given how powerful the $10 million X Prize has been in generating entrepreneurial vigor around suborbital flight, I’m confident that a quarter billion dollar prize would have an incredibly beneficial effect on orbital flight.
Elon
TECHNICAL DEVELOPMENT UPDATES
Propulsion
A Brief History of the Turbopump
It is almost a canon of rocket development that turbopumps are the toughest technical challenge. In fact, a friend of mine recently characterized pump fed launch vehicles as a turbopump development with a rocket attached. It’s not quite that bad, but there is some truth to the comment.
Let me provide a brief history of our pump development and tell you where we went wrong and how we made mistakes:
Right from the beginning, Tom and I expected the turbopump to bite us in the rear (as it turns out, we weren’t disappointed). So the first thing we did, even before we had an office, was initiate the pump development in mid 2002.
In May of 2003, about 12 months after starting development, we had a completely new turbopump designed, built and operating in the test stand. When we were able to bring the pump up to almost full power only three months later, it seemed like we had miraculously sailed through the development with barely a hitch. Soon, we would be able to join the turbopump with the thrust chamber and perform an integrated engine test.
However, in September the pump started showing significant problems, including severe vibration and a few cases where the shaft itself seized up completely. What had actually happened was that we had misdiagnosed the nature of the problem, confusing cause with effect.
Between May and August last year, we had noticed that the seals were rubbing on the shaft and had concluded that this was overloading the bearings, resulting in vibration and occasional seizure. So the fix we put in was to increase clearance between the shaft and the seals and make up for it by increasing the amount of helium purge.
However, after making the changes, problems continued with only modest improvement. That was the point at which we realized that we had the problem backwards. What was actually happening was that the bearings in the liquid oxygen were overheating (despite being -300F, liquid oxygen is a poor coolant). This caused them to melt their housings and become very loose against the shaft. Now that the shaft was not held rigidly in place, it vibrated against the seals, causing them to cut into and eventually seize the shaft.
We redesigned the turbopump to dramatically increase cooling to the LOX bearings and the problems went away. The seal design was also changed to float instead of being fixed, so even with severe vibration the seals would not cut into the shaft metal.
That solved the operating life issues and in December we were able to run the pump at will with no indications of excess vibration or shaft seizure. However, that uncovered one last problem, which took a few iterations to fix.
Now that we were able to run the pump for extended periods, we found that the cavitation margin on the liquid oxygen portion of the turbopump was lower than required. Cavitation is what occurs when the pump starts creating bubbles in the liquid as it spins, resulting in little “cavities” of vapor. Beyond a certain point, the small bubbles become a big bubble and the blades pump a combination of vapor and liquid, which causes the shaft to overspeed.
It took two more iterations on the LOX impeller/inducer design and in April we solved the final problem with the turbopump. In fact, the LOX pump is now significantly overperforming, showing a 43% margin over the minimum design target for suction specific speed (Nss = shaft rpm * Flow^.5 / Net Positive Suction Head^.75). This bodes well for future thrust upgrades of the Merlin engine.
In addition, the last few months of testing were with flight tanks in the vertical configuration, proving out not just the turbopump, but also several structural components and flight valves.
Turbopump in Vertical Stand 1 attached to the thrust frame
Turbopump test video in Vertical Stand 1
(the cloud coming towards you is LOX)
Structures
Stage Separation
We were able to do a stage separation unit test in April, which performed very well. Our current high def video and pictures show too much proprietary detail, so we can’t put that on the web 🙂 The next update will have a video clip for public viewing with the secret sauce obscured.
Our approach is to use pneumatic pushers that are pressure equalized before separation to ensure each pusher is exerting exactly the same force. This is both lighter and more reliably balanced than a spring system.
The two stages are held together by redundant explosive bolts, which are independently wired from both the first and second stage batteries. The first and second stage batteries simultaneously initiate the explosive bolts, ensuring that, even if there is a wiring or power problem from one side, the stages will still separate.
As a further precaution, the upper stage engine nozzle is made of niobium metal sheet, rather than a more brittle carbon-carbon (such as the Space Shuttle leading edge or the RL-10B2 nozzle). If the nozzle hits the inter-stage at separation, it will only dent, which has no meaningful effect on performance, rather than fracture.
Upper Stage
Pictured below is an upper stage on our large five axis mill for some final machining. Since inert mass is a pound for pound trade against payload on the upper stage, it makes sense to machine away all excess material, which we don’t do on our large lower stage. There the trade is about seven pounds of stage mass for one pound of payload.
Note, this is actually two tanks in one, because we use a common bulkhead to separate the liquid oxygen and RP-1 kerosene.
Another Fairing in Production
This picture shows a set of fairing skins in production on the skin
tooling fit check tool. The golden sheen comes from the alodine corrosion
protection coating.
Avionics, Guidance and Control
Engine Computer
The engine computer, which includes our own relay board driven by an FPGA,
was subjected to a thermal test for over a week. Once we fixed problems
with heat dissipation on the simulated output loads, we cycled the unit
between -40 deg. C and +80 deg. C. These temperatures are well beyond
the specified range for some of the components, and we expect far more
moderate temperatures during the flight.
Engine computer, relay board and power supply
(with a little frost)
Hardware-in-the-Loop
Pictured below is our Hardware-in-the-Loop test simulator. The test setup
simulates all the signals that the vehicle needs to believe that it is
flying, such as tank and engine pressures, GPS and gyro data. It also
records the valve and pyrotechnics that are switched on and off by the
flight computer and allows us to verify the software.
Part of the Hardware-in-the-Loop Simulator
Command Trailer
This is our mobile command trailer, which contains personnel critical
to the launch. There will be enough workstations inside for SpaceX, range
safety and customer needs. It will also carry all the communication equipment
for interfacing with the rocket and range infrastructure. The trailer
has its own power generators and a backup UPS to ensure continuous power
during the countdown sequence.
IMU (Inertial Measurement Unit) Test
A common method to align the vehicle’s IMU to “the world”
is by measuring the local gravity vector with the IMU’s accelerometers
and comparing it to the theoretical local gravity vector. The angle between
the 2 vectors is the vehicle’s tilt angle. Here, the IMU was placed
on a sine table, and tilted in 20 arc-second increments – the local
gravity vector was computed/recorded each time. This experiment verified
the fact that our IMU is accurate enough to “self-align,”
as it showed respectabilities of a few arc-seconds and absolute accuracies
well under one arc minute.
The IMU is the little gold soup can object
sitting on the sine table
