SpaceX Falcon 9 v1.1 - SES-8 - DISCUSSION THREAD

[mheney]

Gemini VI-A, not VII.  But yeah, that was the first time a rocket ignited with people on top (and then shut down).

[rickl]

I remember watching the Gemini 6 abort live on TV.  I was just a kid at the time, and had no idea just how hairy it was until much later.


But what I'm interested in is this:  How many times has there been an abort at ignition followed by another launch attempt the same day?  SpaceX did it and succeeded on the F9-1 launch in 2010, and tried to do it again tonight.  I didn't watch F9-1 live, and was absolutely floored when I read about it later.  I had never heard of such a thing.  That was the point where I started paying attention to SpaceX.

[JAC]

Frustration or not, I'd love to see that in extreme slow motion synced with the clock. (The videos NASA made of shuttle launch in slow motion are among my favorite things.)

[Garrett]

bringing over from updates thread:

Case of go fever.

And that's why some of us love a SpaceX launch!

Shouldn't have started the terminal count if their data review wasn't complete.

And yet, if they had completed the review before T -1min, then a launch would have been possible. What your asking for is that they do what others do, no "rocking the boat" and all that. Stick with the status quo, and all will be fine. What a great future that will bring. 

Who's watching the rocket? And who is reviewing data?

Seriously? They probably have well over a hundred core rocket engineers, and only a handful are actually on console during launch. Also, the rocket is well capable of watching itself.

[woods170]

bringing over from updates thread:

Case of go fever.

And that's why some of us love a SpaceX launch!

Shouldn't have started the terminal count if their data review wasn't complete.

And yet, if they had completed the review before T -1min, then a launch would have been possible. What your asking for is that they do what others do, no "rocking the boat" and all that. Stick with the status quo, and all will be fine. What a great future that will bring. 

Who's watching the rocket? And who is reviewing data?

Seriously? They probably have well over a hundred core rocket engineers, and only a handful are actually on console during launch. Also, the rocket is well capable of watching itself.

Oh dear. Going head-to-head with Jim? Sure you wanna do this?

[Borklund]

A brave attempt, but even a lobotomy wouldn't get Jim to change his mind re: SpaceX.

[Garrett]

Oh dear. Going head-to-head with Jim? Sure you wanna do this?

Well, I did post while he sleeps over in the US, thereby avoiding a "live" head-to-head 
Also, most of us here enjoy how SpaceX recycles launch countdowns and we remember how such a process has resulted in successful launches in the past. Simply calling it "go fever" annoyed me, especially coming from Jim who usually has better things to say.

[woods170]

A brave attempt, but even a lobotomy wouldn't get Jim to change his mind re: SpaceX.

Jim's opinion on SpaceX is not all-negative. Some of his posts show his appreciation for certain aspects of SpaceX.

[MP99]

I was also worried about a fever vibe.

When the call came that the customer would be OK to launch 20 mins late, I could imagine the thoughts going "you said that over the public net??". Quickly moved to private net.

http://waynehale.wordpress.com/2013/10/31/keeping-eileen-on-the-ground-part-ii-or-how-i-got-launch-fever/

cheers, Martin

[Jim]

Case of go fever.
 Shouldn't have started the terminal count if their data review wasn't complete.
Who's watching the rocket? And who is reviewing data?

They didn't succumb to it, but starting the countdown put some extra pressure on the review team.  It is only human nature. I am glad they didn't go and took the conservative path. 

I am wondering if Spacex shouldn't revaluate the length of the terminal count and try to shorten it. More complex vehicles have done it, the shuttle was 9 minutes vs the 13 of a F9.  It helps with launch window management.   It might have an affect on recycle time, but that is a trade that can be done.

[Jim]

For this Thursdays launch attempt, how many personnel were needed from out of town ( not Florida residents )?

All the spacecraft people.

[clongton]

Technically, it never got to T-0 since the vehicle didn't commit

Sure?
I think they ignited and after aborted. They have hold down clamps for that. Only open with all engine properly working

Merlin engines are ignited at T-3 seconds. Launch commit and liftoff are at T-0.

[Jarnis]

http://waynehale.wordpress.com/2013/10/31/keeping-eileen-on-the-ground-part-ii-or-how-i-got-launch-fever/

cheers, Martin

That reminded me, still waiting for Part 3 of that very interesting story... Patiently...

[Rocket Science]

I had to agree with Jim’s “Go Fever” comment from yesterday. They really wanted to launch and were trying to “cowboy” it... It was a bit gut wrenching to watch and I wonder how much risk the S/C owner was willing to take apart from extending the window. The safe practices Jim is aware of has been developed over several decades and was paid for dearly. I would have been more comfortable if this was a test flight carrying ballast instead.
 
Hopefully they will learn many valuable lessons from the hardware side and procedure side.  Best of luck to all the teams at SpaceX.

[Rocket Rancher]

Could someone refresh my memory; when was the last time a launch team extended an offical launch window set-up in advance with the Range? I vaguely remember we once extended a window to get a critical AF bird up but I don't remember any other. Yes the shuttle example is one where the short window was driven by mission constrainsts and they chose to change the constrainst. And a commerical bird usually have an orbital mechanics related sunlight contrainst for batteries/power issues. Maybe they had constrainst relieve from the customer already in their back pocket (in reserve), same with the Range support. I don't know the answer to that one.

But it just seem so odd that the team arbrarily wanted to change things on the fly (to cowby it). And yes, they will learn and susceed.  But is it worth learning from your own mistakes when re-inventing the wheel or to learn from the mistakes and experience of others from the past?

[Herb Schaltegger]

For what (little) it is worth, I'm glad they called the abort and ultimately didn't launch. As a slight aside, when the customer gave the okay to extend the window by 20 minutes, can anyone give an explanation of the implications to the payload?

[AnjaZoe]

But it just seem so odd that the team arbrarily wanted to change things on the fly (to cowby it). And yes, they will learn and susceed.  But is it worth learning from your own mistakes when re-inventing the wheel or to learn from the mistakes and experience of others from the past?

In my perception the main difference of the Space-X approach to the rest of the industry is "learning on the fly" instead of the "first time right" approach, coupled with a bit of "we know better than the rest" attitude.

I'm also a bit astonished about the position of SES, which is quite contrary to how they deal with other suppliers. I guess they want really hard to widen the launcher market by supporting the new kid on the block. And of course we don't know what deals they have made behind closed doors as well.

Zoe

[LouScheffer]

Case of go fever.
 Shouldn't have started the terminal count if their data review wasn't complete.
Who's watching the rocket? And who is reviewing data?

They didn't succumb to it, but starting the countdown put some extra pressure on the review team.  It is only human nature. I am glad they didn't go and took the conservative path. 

I am wondering if Spacex shouldn't revaluate the length of the terminal count and try to shorten it. More complex vehicles have done it, the shuttle was 9 minutes vs the 13 of a F9.  It helps with launch window management.   It might have an affect on recycle time, but that is a trade that can be done.

Minus the clock-is-ticking psychological effects, the procedure seems pretty good.  Say there are 33 minutes left in the launch window.  They get 20 minutes to look for big problems, if none are obvious then start the count at 13:00.  Then they get 12 minutes more to look for more subtle errors, meanwhile the count is proceeding and checking for other problems (serving as an additional WDR, in effect).   At T-1 minute, they need to decide, but they have had the maximum time possible to make the decision, and meanwhile verified that every thing else is OK.

The psychological factor is real, but there is experience in dealing with this professionally.  For example, pilots start instrument approaches assuming they will work, working down their last 10 minute or so checklist.  But at the very last minute (200 feet or so) they make a final check if they can see the runway.  If not they abort.   While there is lots of economic pressure to land flights on time, and there are certainly examples of "landing fever", commercial pilots seem to cope with this OK.  A lot of this is that everyone (management, other pilots, passengers, oversight) is on board with a last minute safety-first abort.

[meekGee]

Case of go fever.
 Shouldn't have started the terminal count if their data review wasn't complete.
Who's watching the rocket? And who is reviewing data?

They didn't succumb to it, but starting the countdown put some extra pressure on the review team.  It is only human nature. I am glad they didn't go and took the conservative path. 

I am wondering if Spacex shouldn't revaluate the length of the terminal count and try to shorten it. More complex vehicles have done it, the shuttle was 9 minutes vs the 13 of a F9.  It helps with launch window management.   It might have an affect on recycle time, but that is a trade that can be done.

Minus the clock-is-ticking psychological effects, the procedure seems pretty good.  Say there are 33 minutes left in the launch window.  They get 20 minutes to look for big problems, if none are obvious then start the count at 13:00.  Then they get 12 minutes more to look for more subtle errors, meanwhile the count is proceeding and checking for other problems (serving as an additional WDR, in effect).   At T-1 minute, they need to decide, but they have had the maximum time possible to make the decision, and meanwhile verified that every thing else is OK.

The psychological factor is real, but there is experience in dealing with this professionally.  For example, pilots start instrument approaches assuming they will work, working down their last 10 minute or so checklist.  But at the very last minute (200 feet or so) they make a final check if they can see the runway.  If not they abort.   While there is lots of economic pressure to land flights on time, and there are certainly examples of "landing fever", commercial pilots seem to cope with this OK.  A lot of this is that everyone (management, other pilots, passengers, oversight) is on board with a last minute safety-first abort.

I think the trick was, when you lay out the plan to proceed while waiting for results of the analysis, right at that point, you make it very clear that a scrub is the default option, and we're doing this just so that IF by some chance they get the analysis done in time, we can avoid a scrub.

There's risk in everything, and you also don't want to scrub if the rocket is healthy.

If this becomes a habit, then one day psychology will win and they'll "succumb to launch fever" (good phrasing).  But  these are still first launches, and so it's not a habit and it was a good show of efficient and responsible management.

The proof was in the pudding:  Analysis was not done in time, and they backed off.

[rsnellenberger]

Case of go fever.
 Shouldn't have started the terminal count if their data review wasn't complete.
Who's watching the rocket? And who is reviewing data?

They didn't succumb to it, but starting the countdown put some extra pressure on the review team.  It is only human nature. I am glad they didn't go and took the conservative path. 

I am wondering if Spacex shouldn't revaluate the length of the terminal count and try to shorten it. More complex vehicles have done it, the shuttle was 9 minutes vs the 13 of a F9.  It helps with launch window management.   It might have an affect on recycle time, but that is a trade that can be done.

Listening to the count(s), they seem a lot less crisp and "rehearsed" than the ULA or shuttle counts, although they seem to hit the milestones just as well (i.e., tower retract, pressurizations).  The longer terminal count seems to provide some buffer for things like the delays in responding to the launch director.  A shorter terminal count might force the launch team to tighten up the count (is that the benefit you are looking for here?), but I don't know if they'd consider that a plus or minus...