SpaceX - Falcon 1 launch

[MaDeR]

Launch scheduled today
4:00pm (PDT) / 7pm EDT / 23:00 (UTC)
Webcast begins ~30 mins before

Webcast: http://www.spacex.com/webcast.php

We can only wait...

[Van Rijn]

They've had a series of holds. I'm hoping they'll get it this time, but we'll see.

[MaDeR]

I doubt it. Bet on scrub for today. Time to sleep, here is 3 a.m. If launches, I will watch on YouTube or whatever. My body had enough.

[Van Rijn]

Yep. It would be nice if they said more on the webcast, but it sounds like it will be a long time before they could launch at this point.

[Van Rijn]

They finally updated that there was a "minor issue." They were hoping to resume in 30 minutes and think they could still do it in this launch window.

Well, we'll see, but they've said that before.

[GeorgeLeRoyTirebiter]

Webcast is back up, 10 minutes to go, everything's looking OK.

[ginnie]

T-6
minutes

[MaDeR]

I am up too, after a hour sleep. Scary. I am real space maniac...

...and very nervous. Imagine what SpaceX team must feel.

5 minutes!

[MaDeR]

Abort. What the...?

[GeorgeLeRoyTirebiter]

At least it aborted before launch.

[ginnie]

I'm putting a DVD on. But will keep an eye on my monitor too...

[MaDeR]

Again. T-11 minutes and countdown. Someone can go to that new chat? Pleasseee? :>

[MaDeR]

Launch!

[clint_dreamer]

Wow that ended abruptly. The announcer said there was an anomaly on the vehicle and signed off quickly. Doesn't look good by his reaction.

[Nano]

"Anomaly in the vehicle"...

[MaDeR]

Anynone noticed wiggling after launch and before termination of video feed? Compare it with video from launch #2.

[jja]

From Wiki: http://en.wikipedia.org/wiki/Falcon_1


At 04:35 UTC, the Falcon 1 lifted off from Omelek Island. The live video feed of the launch was cut off 140 seconds into the flight, at which time it was announced an anomaly was detected on the vehicle. The fate of vehicle, whether it was destroyed or if it successfully entered low Earth orbit, is not known at this time.[21][22].

Members participating that the SpaceX camp have been twittering that there was a loss of the vehicle during the separation stage, which resulted in the engine exploding.

[loglo]

Anynone noticed wiggling after launch and before termination of video feed? Compare it with video from launch #2.

Video of this launch is up on Spaceref. It did seem to wobble a bit towards the end.

[lpetrich]

In a thread on this mission over in space.com, someone commented about what was likely to happen next after the time of loss of signal.

The first stage looked like it was shutting down, meaning that it was time for that stage to separate and for the second stage to start up. This means that something could have gone wrong at that point, as had happened on the previous Falcon 1 flight.

If the rocket made it into orbit, it should be going over Ascension Island just about now; that place has an almost exactly antipodal location relative to the launch site:

Omelek: 9.05 N, 167.74 E
Ascension: 7.56 S, 14.25 W

[Van Rijn]

It didn't make it to orbit. See:

http://www.nasawatch.com/archives/20..._launch_v.html

[lpetrich]

I checked SpaceX's site, and their talk about the Falcon 9's expected reliability now seems a bit silly. They quoted a study stating that the biggest causes of rocket failures are the engines and related systems, stage separation, and avionics.

That study is Space Launch Vehicle Reliability, and some of its numbers are rather interesting. There are usually a lot of initial failures in the development of some new rocket design, but those that survive more than a few years tend to have much fewer failures.

[mugaliens]

I examined the video in detail, and have noted my observations below.

All references to coordinate systems are as follows:

x axis - positive to the right of the video
y axis - positive to the top of the video
z axis - positive in the direction of the rocket's travel
Rotation about any axis is positive using the "right-hand thumbs up" rule. If your right-hand thumb is aligned with an axis, if you curl your fingers, they'll be sweeping from one postive axis to the other, in succession. Thus, positive rotation around the z axis (thumbs up) has your fingers sweeping from the positive x axis to the y axis. If rotation is around the x-axis, fingers will sweep from y to z axis. If rotation is around the y axis, fingers will sweep from the z to x axis.

T+02:49 - First stage cutoff
T+02:51 - Second stage separation; at separation, second stage experiences negative rotation around the y axis, positive rotation around the z axis, and negative rotation around the x axis.
T+02:52 - Negative rotation around y axis causes second stage engine's nozzle to make contact with upper ring of first stage's shroud.
T+02:52 - 02:55 -
T+03:34 - Significant debris appears on the camera plate window
T+03:36 - Debris accumulation stops. Debris still present.
T+04:17 - Second stage engine's guidance corrections become noticeably greater than previously.
T+04:26 - Second stage engine's guidance correction anomaly steadies into a circular/elliptical pattern of negative orbit about the z-axis.
T+04:26 - 04:58 - This pattern grows steadily larger.
T+04:56 - Second stage beging negative rotation around z axis.
T+05:01 - Elliptical warble combined with roll (negative rotation about z axis)

Conclusions:

T+02:51 - Second stage separation; at separation, second stage experiences negative rotation around the y axis, positive rotation around the z axis, and negative rotation around the x axis.

Conclusion: Separation was assymetrical about all three axes.

T+02:52 - Negative rotation around y axis causes second stage engine's nozzle to make contact with upper ring of first stage's shroud.

Conclusion: Impact with nozzle damaged one or more of the nozzle's gymbal sensors or gymbal motors, and possibly induced errors in the inertial guidance system.

T+04:17 - Second stage engine's guidance corrections become noticeably greater than previously.

Conclusion: A malfunction in the guidance system, possibly caused by the T+02:52 event is resulting in unsteady guidance.

T+04:26 - Second stage engine's guidance correction anomaly steadies into a circular/elliptical pattern of negative orbit about the z-axis.

Conclusion: Guidance anomaly worsens throughout fuel burn. Note: The rocket is the most steady with the most fuel, and is least steady with the least fuel.

T+04:56 - Second stage beging negative rotation around z axis.

Conclusion: The guidance anomaly begins to exceed the system's ability to compensate.

T+05:01 - Elliptical warble combined with roll (negative rotation about z axis)

Conclusion: The anomaly exceeds guidance system's ability to compensate, resulting in the subsequent tumble and structural failure of the second stage.

Additional Observations:

1. The instability wasn't sudden, but rather, progressive. Nor did it begin with the collision, as noted in the 1:25 delay between contact and initial instability.

2. The initial instability appeared to be a transition period lasting 9 seconds.

3. The subsequent instability settled into a circular/elliptical wobble about the z axis.

4. The position of the nozzel seemed to lag the instability. Thus, it was engaged in reactionary guidance control.

5. The progressively worsening instability appeared to begin to overburden the guidance system's ability at T+04:56.

6. When gymbal sensors or gymbals fail, even partially, it's usually a sharp demarcation in position, such that an anomaly is noticed immediately.

7. Rockets without fins are inherently unstable. That is, they are dynamically unstable. This requires either very rapid reactive guidance, or good predictive guidance. If it's reactive guidance, the greater the MACH, the more rapid that guidance must be to keep the rocket relatively steady and on course.

8. The second stage, which begins at the top of the wide black band around the top of the first stage, has no fins.

Additional Comments:

1. The collision between the second stage engine and the first stage shround is a fact. Not only can the contact clearly be seen, but the engine nozzle is physically moved in relation to the second stage, but rapidly returns to position before the commencement of second stage burn.

2. This collision may, or may not have lead to the later instability first noted at T+04:17 which later developed into a circular/elliptical wobble at T+04:26.

3. If the collision caused the later instability, it could have done so in three ways:

3a. Damage of the gymbal sensors.

3b. Damage to the gymbal motors.

3c. Damage to the inertial navigation system, principally the directional gyros.

4. The progressive nature of the instability is consistant with two possibilities:

4a. Directional gyro failure

4b. Dynamic instability coupled with a reactive guidance system that is unable to keep up with the progressively greater oscillations

4c. If the oscillations were caused by dynamic instability, that instability was caused by one of two things.

4c1. Increasing dynamic pressures

4c2. Decreasing stability along the z axis. This is normal as fuel is burned and the moment of internia along the z axis is reduced.

Super-Conclusions:

1. It probably wasn't a failure of the gymbal sensors or motors. If that were the case, we would have seen an anomaly much sooner after the second stage nozzle contacted the first stage shroud.

2. It may have involved the directional gyros, but only if they were of the physical, rotating kind (inexpensive, but there's a weight penalty). The collision could have been enough to send a properly stabilized gyro into a destabilizing death spiral, which would be commensurate with the video. Slight jars would not do this, as gyros receive stabilizing inputs to prevent precession. If the jar was serious enough, however, the induced precession may have exceeded the gyro's ability to correct itself. We call the result "tumbling off into the weeds." If the gyros are of the much more expensive ring-laser type, the shock would not have induced precessional destabilization.

3. I suspect that it wasn't a gyroscopic destabilization, but rather, that it was due to the use of an insufficient guidance system, whether reactive or predictive (proactive). Given the lag between the second stage's orientational position and the direction in which the nozzle was pointed, it's clear that the nozzle lagged the necessary correction, and thus was actually causing the wobble as it tried to keep up with the necessary correction. This is indicative of a reactive guidance system, possibly coupled with an error in the amount of dampening used in the correction algorithms.

Conclusive Conclusion (and a bit of a guess):

It wasn't the collision. Rather, the guidance system simply wasn't up to the task of handling the second stage during the progressive dynamic instability encountered during flight.

Recommendations:

1. Use a ring-laser gyro system for three-axis orientation.

2. Use a predictive (proactive) guidance system, not a reactive one.

3. Incorporate orbital/elliptical wobble pattern recognition and propulsion compensation in the guidance algorithms.

4. Use controllers with a little more experience in checklist discipline and military (ie, tense) operations.

Otherwise, nice attempt!

[Nerull]

I examined the video in detail, and have noted my observations below.

All references to coordinate systems are as follows:

x axis - positive to the right of the video
y axis - positive to the top of the video
z axis - positive in the direction of the rocket's travel
Rotation about any axis is positive using the "right-hand thumbs up" rule. If your right-hand thumb is aligned with an axis, if you curl your fingers, they'll be sweeping from one postive axis to the other, in succession. Thus, positive rotation around the z axis (thumbs up) has your fingers sweeping from the positive x axis to the y axis. If rotation is around the x-axis, fingers will sweep from y to z axis. If rotation is around the y axis, fingers will sweep from the z to x axis.

T+02:49 - First stage cutoff
T+02:51 - Second stage separation; at separation, second stage experiences negative rotation around the y axis, positive rotation around the z axis, and negative rotation around the x axis.
T+02:52 - Negative rotation around y axis causes second stage engine's nozzle to make contact with upper ring of first stage's shroud.
T+02:52 - 02:55 -
T+03:34 - Significant debris appears on the camera plate window
T+03:36 - Debris accumulation stops. Debris still present.
T+04:17 - Second stage engine's guidance corrections become noticeably greater than previously.
T+04:26 - Second stage engine's guidance correction anomaly steadies into a circular/elliptical pattern of negative orbit about the z-axis.
T+04:26 - 04:58 - This pattern grows steadily larger.
T+04:56 - Second stage beging negative rotation around z axis.
T+05:01 - Elliptical warble combined with roll (negative rotation about z axis)

Conclusions:

T+02:51 - Second stage separation; at separation, second stage experiences negative rotation around the y axis, positive rotation around the z axis, and negative rotation around the x axis.

Conclusion: Separation was assymetrical about all three axes.

T+02:52 - Negative rotation around y axis causes second stage engine's nozzle to make contact with upper ring of first stage's shroud.

Conclusion: Impact with nozzle damaged one or more of the nozzle's gymbal sensors or gymbal motors, and possibly induced errors in the inertial guidance system.

T+04:17 - Second stage engine's guidance corrections become noticeably greater than previously.

Conclusion: A malfunction in the guidance system, possibly caused by the T+02:52 event is resulting in unsteady guidance.

T+04:26 - Second stage engine's guidance correction anomaly steadies into a circular/elliptical pattern of negative orbit about the z-axis.

Conclusion: Guidance anomaly worsens throughout fuel burn. Note: The rocket is the most steady with the most fuel, and is least steady with the least fuel.

T+04:56 - Second stage beging negative rotation around z axis.

Conclusion: The guidance anomaly begins to exceed the system's ability to compensate.

T+05:01 - Elliptical warble combined with roll (negative rotation about z axis)

Conclusion: The anomaly exceeds guidance system's ability to compensate, resulting in the subsequent tumble and structural failure of the second stage.

Additional Observations:

1. The instability wasn't sudden, but rather, progressive. Nor did it begin with the collision, as noted in the 1:25 delay between contact and initial instability.

2. The initial instability appeared to be a transition period lasting 9 seconds.

3. The subsequent instability settled into a circular/elliptical wobble about the z axis.

4. The position of the nozzel seemed to lag the instability. Thus, it was engaged in reactionary guidance control.

5. The progressively worsening instability appeared to begin to overburden the guidance system's ability at T+04:56.

6. When gymbal sensors or gymbals fail, even partially, it's usually a sharp demarcation in position, such that an anomaly is noticed immediately.

7. Rockets without fins are inherently unstable. That is, they are dynamically unstable. This requires either very rapid reactive guidance, or good predictive guidance. If it's reactive guidance, the greater the MACH, the more rapid that guidance must be to keep the rocket relatively steady and on course.

8. The second stage, which begins at the top of the wide black band around the top of the first stage, has no fins.

Additional Comments:

1. The collision between the second stage engine and the first stage shround is a fact. Not only can the contact clearly be seen, but the engine nozzle is physically moved in relation to the second stage, but rapidly returns to position before the commencement of second stage burn.

2. This collision may, or may not have lead to the later instability first noted at T+04:17 which later developed into a circular/elliptical wobble at T+04:26.

3. If the collision caused the later instability, it could have done so in three ways:

3a. Damage of the gymbal sensors.

3b. Damage to the gymbal motors.

3c. Damage to the inertial navigation system, principally the directional gyros.

4. The progressive nature of the instability is consistant with two possibilities:

4a. Directional gyro failure

4b. Dynamic instability coupled with a reactive guidance system that is unable to keep up with the progressively greater oscillations

4c. If the oscillations were caused by dynamic instability, that instability was caused by one of two things.

4c1. Increasing dynamic pressures

4c2. Decreasing stability along the z axis. This is normal as fuel is burned and the moment of internia along the z axis is reduced.

Super-Conclusions:

1. It probably wasn't a failure of the gymbal sensors or motors. If that were the case, we would have seen an anomaly much sooner after the second stage nozzle contacted the first stage shroud.

2. It may have involved the directional gyros, but only if they were of the physical, rotating kind (inexpensive, but there's a weight penalty). The collision could have been enough to send a properly stabilized gyro into a destabilizing death spiral, which would be commensurate with the video. Slight jars would not do this, as gyros receive stabilizing inputs to prevent precession. If the jar was serious enough, however, the induced precession may have exceeded the gyro's ability to correct itself. We call the result "tumbling off into the weeds." If the gyros are of the much more expensive ring-laser type, the shock would not have induced precessional destabilization.

3. I suspect that it wasn't a gyroscopic destabilization, but rather, that it was due to the use of an insufficient guidance system, whether reactive or predictive (proactive). Given the lag between the second stage's orientational position and the direction in which the nozzle was pointed, it's clear that the nozzle lagged the necessary correction, and thus was actually causing the wobble as it tried to keep up with the necessary correction. This is indicative of a reactive guidance system, possibly coupled with an error in the amount of dampening used in the correction algorithms.

Conclusive Conclusion (and a bit of a guess):

It wasn't the collision. Rather, the guidance system simply wasn't up to the task of handling the second stage during the progressive dynamic instability encountered during flight.

Recommendations:

1. Use a ring-laser gyro system for three-axis orientation.

2. Use a predictive (proactive) guidance system, not a reactive one.

3. Incorporate orbital/elliptical wobble pattern recognition and propulsion compensation in the guidance algorithms.

4. Use controllers with a little more experience in checklist discipline and military (ie, tense) operations.

Otherwise, nice attempt!

I think you examined the wrong video - thats the second test launch, not this one. It never staged at all.

[naelphin]

Things going boom. What SpaceX's real product is!

[Damburger]

I honestly don't think Musk has a very firm grasp of what it takes to put stuff into orbit. I think that in trying to cut through the 'red tape' he sees in government programmes he has also got rid of too much quality control. The expense of a government funded rocket might simply be the actual cost of a sufficiently robust vehicle.

Certainly his optimism about Falcon 9 (a more complex vehicle than his so far 100% failed Falcon 1) seems misplaced. Even if it weren't, the pricing of it doesn't seem to be that much lower than the price of a Proton launch, per kilogram. Hardly the revolution he was originally promising. His bargain-basement rockets seem to be, literally and figuratively, vapourware.

[NEOWatcher]

I think you examined the wrong video - thats the second test launch, not this one. It never staged at all.

Could just be the same problem. It does sound familiar.

Things going boom. What SpaceX's real product is!

That's an awfully large investment just for a fireworks show.

[Nicolas]

It does sound like Mugs reviewed the previous launch, not this one. This one does not show the separation indeed.

[NEOWatcher]

...This one does not show the separation indeed.

I wasn't sure if that (the one on SpaceXs main page) was the only one available. It did cut off rather early. But; I did notice some oscillations, but didn't watch carefully enough to see if it matched Mug's assesment.

[mugaliens]

It does sound like Mugs reviewed the previous launch, not this one. This one does not show the separation indeed.

Ok, that's two of you. Where's the link to the current one? I merely followed the link to SpaceX's Video site and picked what I thought was the latest one.

[Lukas]

I missed the live videocast and was looking for a video too. Just found this one of Launch #3.
It shows the abort in the initial attempt and the launch until the camera cuts out.

http://www.youtube.com/watch?v=2eGiqqoYP5E

have fun mugs.
Nice analysis of #2 btw. Will have to look into it in more detail when I have more time.