June/July Update

(Note: Subscribers to the email list receive the update a few days earlier than it is posted on the website. Email address privacy is always respected.)

Public Viewing of the Maiden Launch
The maiden launch will be open to the public and we will post the exact date on our website in the coming months. Oddly enough, the Air Force designated viewing location from which you will be able to see the Falcon I launch is called “Hawk’s Nest” (no relation to us) and is accessed off of Hwy 1 on Azalea Lane near Vandenberg.

Our current schedule calls for transferring the rocket to the pad in September, performing a short duration vehicle hold down firing and launching as soon thereafter as possible, without compromising safety or reliability. We will not launch until all engineers are two thumbs up, so that date may get pushed back.

Since this is a high energy, orbital flight, Air Force range safety has a minimum keep out radius of at least 2 miles. However, Falcon I is seven stories tall with an 85,000 lb vacuum thrust engine and a long, bright LOX/RP flame tail, so the launch will definitely be worth seeing.

Launch Manifest
Below is our expected launch schedule. The “contracted” launches are those with a signed contract and for which we have received a deposit. Tentative means we have had serious, in-depth discussions with the potential customer and believe that there is a greater than fifty percent likelihood of sale. As you might expect, we have had many discussions with potential customers that are waiting for at least one successful launch before committing. Those are not mentioned here.

Kistler Sole Source Contract
As some who have been following this matter will note, the General Accounting Office (GAO) agreed with SpaceX that the Kistler sole source contract should not have been granted. It is my understanding that NASA has not yet decided their next course of action. To be clear, my concern with the contract was primarily the very negative market signal that issuing a large and (as the GAO has ruled) unjustified sole source contract to a bankrupt company would send. I think it would be a good thing if Kistler emerged from bankruptcy and continued to pursue their launch vehicle development, albeit only on a fair playing field.

The best destination for those now unallocated tax dollars is open access, performance based contracts (a.k.a. prizes) under the NASA Centennial Prize program. If that is done, it will do wonders to invigorate commercial space development and spur new entrants into the orbital space launch business, just as the X Prize has done for sub-orbital.

SpaceX Launch Pad at Vandenberg Air Force Base
A lot of work has been done getting our launch pad ready at Vandenberg. Although we were able to start with the existing concrete foundation of an old Atlas II pad, there was literally nothing else in place. Compared to other launch pads at Vandenberg, we don’t require much, but there is still a lot of work to do to have a professional launch infrastructure.

We do not use a permanent tower on the pad, employing instead our mobile erector/launcher platform. This is stored in a protected environment when not in use, so is not subject to weather damage and corrosion. This makes our maintenance costs very low and minimizes time spent on the launch pad, which in turn reduces our launch operations costs. We can prepare the entire rocket for launch in the controlled environment of our factory, rather than do final assembly at the launch site.

Right now, all the electrical and communications wiring is in process of being installed and connected to the base grid. In addition, there is a lot of plumbing work to connect liquid oxygen, RP-1 kerosene, helium and nitrogen. These feed into the quick disconnect umbilicals on the rocket, which detach as the rocket lifts off. The water deluge for heat and noise suppression on launch will just use standard base water pressure, as Falcon I doesn’t need anything more to meet specifications.

The CAD solid model below shows what Falcon I will look like on the pad with the mobile launcher, before it is erected for flight:

Falcon I before flight

Here you can see an actual picture of Space Launch Complex - Three West (SLC-3W), our home at Vandenberg, under construction. Lockheed, our next door neighbor at SLC-3E, is preparing their Atlas V pad for launch. They have been good, courteous neighbors and hopefully Lockheed sees us the same way.

“Pad sweet pad”

A tremendous amount of progress was made on the technology development front, as described below. Major milestones include: Merlin fully integrated engine firing with flight tanks, stage separation tested, fairing separation tested and most of the flight avionics & antenna pattern tested.

Elon

TECHNICAL UPDATES

Propulsion
Merlin Integrated Engine Test
After many delays and a few setbacks, we have fired Merlin (our main engine) in the fully integrated configuration, with both the thrust chamber assembly, turbo-pump and gas generator attached to a set of flight tanks. This is one of the biggest milestones before launch and the culmination of a tremendous amount of work by the propulsion team.

All the unit testing done on the thrust chamber, turbo-pump, as well as cold flows done on the tanks and associated plumbing paid off in a very smooth integration. The start sequence with the turbo-pump is actually more benign for the thrust chamber assembly than the horizontal test stand, where the chamber is pressure fed and receives a slam start.

With an improved fuel manifold under construction (the old design cracked during testing), this clears the way to enter the engine qualification program, which is somewhat analogous to the beta test period for software. We know the engine works, we just need to make sure it always works and intend to put in the time to ensure that’s true.

Merlin is not just the main engine for Falcon I, but will also serve as the main engines for Falcon V, so it’s all the more important and worth the investment to ensure that the engine is rock solid reliable. However, we are sufficiently confident at this point that we have enough engines in the manufacturing loop for both Falcon I and the first flight of Falcon V. First firing of five integrated engines on flight tanks for Falcon V is scheduled for early 2005.

Merlin integrated engine firing video

Thrust Frame
The engine thrust frame weight has come in significantly better than our initial baseline for Falcon I. This is due in part to switching from steel to high strength titanium and in part to a better design. Although we are spending more than planned on this piece of equipment, we expect to be able to reuse it essentially forever (i.e. thousands of flights), so long as the stage itself is recovered.

The corner fittings are precision machined and then welded under argon to the gun drilled tubes. The whole frame only weighs 74.8 lbs and is shown below going through structural qualification. We loaded it to 150,000 lbs (almost twice maximum flight load) in the axial direction and applied max gimbal and TVC loads. Nine limit and ultimate load cycles were applied with no indications of yield (strains all returned to zero).

Structures
As several studies have shown, there are really two reasons why launch vehicles fail – it’s either engines or separation events. This makes sense, because failure tends to occur when there is a change of state. Engines are in constant change and obviously a separation event is a pretty significant change of state.

Fairing Separation
Our fairing is a biconic, rib-stiffened aluminum structure with a space grade cork ablative on the nose for thermal protection and non-outgassing sound blankets on the inside. The choice of biconic was a balance between ease of hypersonic flow separation prediction and ease of manufacture.

For a separation system, we use dual-initiated, non-explosive separation nuts to hold the halves together. When these fire, a pair of pneumatic pushers rotate each half over a partial hinge, resulting in a very precise separation arc that ensures neither the payload nor the rocket will be touched by the departing fairing. Since there are no explosive bolts used here and the separation itself is gentle, no meaningful shock load is imparted to the avionics or satellite payload.

Fairing separation video (employing a high speed camera and strobe light)

Stage Separation
The stage separation test incorporates the Falcon interstage as a “mass simulator”. Shock accelerometers were placed at the separation plane to measure the shock produced during the event. Additional shock accelerometers were used on the interstage skin in order to measure shock attenuation across vehicle joints. In this case, we do use comparatively high shock explosive bolts, as we can’t yet obtain the non-explosive separation nuts with sufficient strength to hold the stages together under maximum load. However, there is nothing close to the separation plane that is shock sensitive, so this doesn’t affect our payload environment.

The test proved that the stage separation system functions properly and produces a more than adequate impulse to clear the second stage engine from the first stage during flight. We’ve also tested separation with an offset center of mass and one of the separation bolts firing late and all results are positive. If the second stage engine nozzle does hit the interstage on exit, the worst that will happen is that the niobium nozzle will be dented and immediately undent upon ignition. We chose a refractory metal nozzle over a carbon-carbon nozzle for exactly that reason. The latter will crack like a coffee cup on impact (a la the Shuttle wing leading edge).

Stage Separation Video

Avionics, Guidance and Control
As part of the stage separation test, we were also able to test much of the flight avionics, including the flight computer, power relays, wiring harness and inertial measurement unit (IMU). The IMU was mounted on the interstage, providing a very accurate measure of acceleration in all six axes and showing no ill effects from the shock event.

The red relay boards pictured below are the computer's interface to the rest of the rocket (arranged in theater seating). The computer controlling the rocket is the green board at the lower left. We are not 100% certain, but this is probably the most powerful rocket flight computer in the world, since it is the most recently designed and the only one to use current 21st century technology. It is certainly much more powerful than what’s used on the Space Shuttle.

The flight relay boards (FRBs) are the arms, legs, eyes and ears of the flight computer, doing things like issuing the deploy command to the payload, collecting data from the IMU, issuing commands to the pressure controllers, collecting analog sensor data, actuating valves and firing the stage separation bolts. The aluminum enclosure on the upper left houses the boards and the shock mounted cage on the right is what holds them in the deck.

Relay boards, shock mounts and flight computer

We also completed the antenna pattern test at EDO using a radio true mockup of the avionics bay and surrounding sections. The rocket is truly a flying radio station, with two C-band, four S-band, two UHF and two GPS antennas, as well as a beacon antenna on the first stage for location by the recovery ship.

EDO antenna test

In the last few months we also completed design, fabrication and testing of our lightweight electro-mechanical thrust vector control system for the upper stage. For our first stage, we use Moog hydraulic cylinders, but could not find a suitable unit at reasonable cost for the upper stage and were forced to build our own.

This has worked out well as the whole actuator, including motor, heat sink, gearbox, position sensors and attachments weighs only 5.25 lbs (less than even the most expensive alternative) and is built in-line. It is capable of moving a 7500 lbf rocket engine at 4 inches per second with 800 lbs of force.

Upper stage actuator in the thrust vector control test stand

< Archived Updates