Updates: August 2004 - September 2004
August 2004 - September 2004
Falcon I on the Launch Pad!
The Falcon I actual flight vehicle is now on the SpaceX launch pad at Vandenberg Air Force Base (near Santa Barbara, California). This is a huge milestone for our company and is the result of a monumental effort by everyone here.
Below are a series of pictures showing Falcon being erected by the mobile launcher on to the quadropod hold down fixture. Time to rotate vertical is about 15 minutes, although we should be able to reduce this to only a few minutes once we’re comfortable with the process. The white building on the launch deck (shown in pictures four and five) is the retractable hangar, which provides shelter to technicians for final checkout and satellite integration. It moves back on rails shortly before launch. Note the new carbon fiber inter-stage and the ablative covering the forward portion of the payload fairing.
Falcon on mobile launcher ready to rotate vertical
About halfway to vertical (the telescoping hydraulic ram is now visible)
Vertical and fixed to the hydraulic launch hold down clamps
Perspective looking straight down the flame trench
Oblique view (white structure on the deck is the retractable hangar)
Side view (large tank in the foreground contains the liquid oxygen propellant supply)
The remaining milestones are (not necessarily in this order):
1. Complete the formal flight qualification program for the engines.
2. Receive range safety approval to launch.
3. Do a full vehicle hold down firing on the launch pad as a final verification.
4. Integrate the TacSat-1 satellite.
5. Launch to a 500km orbit
At this point, it looks like the first available launch window that works for both SpaceX and Vandenberg is mid to late January.
Mobile Control Center
We built our control center in a trailer for maximum flexibility, which means we don’t need to spend capital building up this capability at each range. It also minimizes the ongoing overhead costs of having to pay rent or maintenance for a control center when we only use it for a few weeks at a time during launch.
The MCC contains up to twelve personnel and is fully self-supporting with its own generator, UPS and air-conditioning. It contains everything necessary to interface with the Air Force Base, all voice and data com, vehicle telemetry displays, etc.
Inside of Mobile Control Center
Outside of Mobile Control Center
Every update going forward will have the expected launch schedule, but showing only firm contracts for which we have received a deposit. There are a lot of discussions underway with potential customers and I expect that in 2006 we will manifest at least four flights again, with volume picking up significantly above that in 2007. It will take a few years for the satellite market to adjust to the availability of a low cost launch vehicle.
|Customer||Launch Date||Vehicle||Departure Point|
|US Defense Dept (OSD/NRL)||Q1 2005||Falcon I||Vandenberg|
|US Defense Dept (DARPA)||Q2 2005||Falcon I||Marshall Islands|
|Malaysia (ATSB)||Q4 2005||Falcon I||Marshall Islands|
|Bigelow Aerospace||Q4 2005||Falcon V||Vandenberg|
Falcon V Progress
The information on Falcon V that has been on the website for several months is now significantly out of date. Following discussions with customers, we decided earlier this year to make some important changes.
We are switching from a dual Kestrel upper stage to a single Merlin upper stage. This has a major effect on mass to orbit due to improved mass fraction, higher specific impulse and better staging efficiency. The improved mass fraction results from having a pump-fed upper stage with thin tank walls and low helium requirements vs. pressure-fed with thick tank walls and high helium requirements. The higher specific impulse comes from the much higher chamber pressure of Merlin and much higher expansion ratio. The Merlin vacuum specific impulse is expected to be 340s vs. 325s for Kestrel.
Other changes not included in prior numbers will also boost payload to orbit. During the development process, the Merlin engine has show itself capable of at least an 80,000 lb sea level thrust vs. the nominal 72,000 lb. Also, both the inter-stage and fairing will be made from a carbon fiber sandwich composite (vs. just the inter-stage in Falcon I), improving mass fraction. Following completion of the Merlin 1 qualification, we will begin work on a Merlin 2 with a sea level thrust target in excess of 100,000 lb as well as a slight increase in Isp.
The estimated payload to orbit with the Merlin 2 engine, including both an Isp and mass fraction sandbag, is as follows:
|200 km, 28.5 deg||6020 kg / 13,260 lb|
|400 km, 51 deg||5450 kg / 12,010 lb|
|700 km, sun-synch||4780 kg / 10,530 lb|
|GTO, 9 deg||1920 kg / 4,240 lb|
|Escape velocity (0 C3, no kick stage)||1200 kg / 2,650 lb|
The above performance to GTO & escape can be improved significantly and the burnout g load reduced by using a kick stage, such as a Star motor from ATK. Note, the initial version of Falcon V, which is expected to launch in late 2005, will have approximately 20% less capability than the above numbers (official update to the website coming shortly). These performance numbers are applicable to launches occurring in late 2006 when the Merlin 2 engine upgrade is flight ready.
I expect the Falcon V development to be considerably faster and easier than the Falcon I development, because it uses most of the same components. Unlike Falcon I, where we had to develop two complete engines from scratch, Falcon V in its initial version requires only minor adjustments to the existing Merlin engine. The avionics are the same as Falcon I and the airframe architecture on Falcon V is just a wide body version of Falcon I. Also, most of our launch site infrastructure and the environmental permits already accommodate both vehicles.
Noteworthy items on the Falcon V development:
1. The upper stage Merlin engine is in the fabrication cycle and should be complete in a few months, except for the vacuum skirt extension.
2. LOX and RP-1 manifold parts for the five first stage engines are done.
3. Our custom-made friction stir welder, which will be used for the Falcon V tanks, has arrived and is producing excellent welds with literally zero defects.
4. Spin form tooling for the tank domes is on order and expected in January.
The past few months have mostly consisted of testing and retesting the engines. As reported earlier, we noticed cracks developing in the cast aluminum fuel manifold. As a result, we decided to switch to inconel, a nickel-chromium-iron alloy that has several times the strength of cast aluminum and even greater toughness, at the cost of a few extra pounds of mass. The area under the stress-strain curve (definition of toughness) for our inconel manifold is an order of magnitude greater than the aluminum casting.
Testing at this point is focused mostly on ablative life and making sure that we are not at risk of burning through the chamber wall. Depending on how this goes, we may need to increase film cooling to the engine walls and sacrifice a few seconds of specific impulse.
Maximum thrust achieved so far during testing of Merlin is 81,000 lb and in fact the engine has to be detuned to bring thrust down to within specifications for Falcon I. This is quite good news, because we have been able to build that extra thrust into the Falcon V design and boost payload well above numbers currently advertised on our website.
All major launch pad construction is complete, including concrete, steel work, propellant supply tanks, launch mount, etc. We also built a retractable hangar on the upper deck to provide shelter to people working on the rocket during final checkout and integration of the satellite. This rolls back on rails just prior to launch.
As shown above in launch pad pictures, LOX and RP-1 tanks sit on either side of the launch deck, along with helium pressurant and nitrogen purge supply tanks. Pad electronics are in a small room under the launch quadropod. The work remaining consists mainly of finishing the electrical and communication wiring, as well as the plumbing from the supply tanks to the vehicle.
Our pad infrastructure is probably more than an order of magnitude less complicated than most other US facilities and bears a closer resemblance to the current Russian/Ukrainian approach or the early US Thor architecture. In fact, we drew some of our ideas from an old Thor rocket and its mobile launcher that are sitting in a museum at Vandenberg. It is not clear to me why those ideas were abandoned.
We have made significant mass savings in first stage, particularly in the inter-stage structure. By moving from an aluminum design to a carbon fiber composite design, we cut the weight of the inter-stage in half. Now, we started with a poorly optimized aluminum design and then moved to a really optimized carbon fiber/honeycomb design, so ordinarily you would not see that much of an improvement. The icing on the cake here is that it actually costs us less to build the carbon fiber inter-stage than the aluminum one! Long term, we will probably switch our aluminum fairing design to composite as well.
Other areas of first stage mass improvement:
- Parachute recovery system mounts
- Helium bottle mounts
- Pressurant plumbing changed from steel to titanium
- Separation rings
- Thrust skirt
- Removal of an unnecessary ring at the base of the vehicle
- New (and better in every way) pre-valve design
- Thinner slosh baffles
- Thinner tunnel covers
- Thrust frame changed from steel to titanium
- Gimbal joint changed from steel to titanium
Some parts of the main engine have grown in mass, but the net effect is that we are likely to meet our propellant mass fraction target of 94% for the first stage, including the inter-stage and recovery system. This is among the best mass fraction numbers for any boost stage and is particularly good when considering that our first stage weight incorporates a recovery system and increased design margins for reusability.
We integrated our full suite of avionics with the aluminum honeycomb support structure, which located just below the fairing. This was then mounted to the flight vehicle for transfer to Vandenberg. After ironing out a few nits, communication through the umbilical cable from the Mobile Command Center to the vehicle flight computer worked flawlessly. We were also able to obtain a very precise alignment of the vehicle using a theodolite.
Below is a picture of the avionics tray nearing final assembly. The upper half contains vehicle avionics necessary for guidance & control and transmission of telemetry & video. The lower half contains a completely independent flight termination system that will be used by range safety to terminate thrust in an emergency.