Author Topic: Reusability as abort functionality  (Read 14689 times)

Online Silmfeanor

  • Full Member
  • ****
  • Posts: 1164
  • Utrecht, The Netherlands
  • Liked: 217
  • Likes Given: 458
Re: Reusability as abort functionality
« Reply #20 on: 11/03/2013 05:14 PM »
perhaps this should be moved to advanced concepts?
because we are talking about 10+ years here, for sure.

Offline CuddlyRocket

Re: Reusability as abort functionality
« Reply #21 on: 11/03/2013 05:34 PM »
perhaps this should be moved to advanced concepts?
because we are talking about 10+ years here, for sure.

No we're not. Once first stage reusability is demonstrated (which I think will be in the near future), it becomes obvious to ask whether there would be circumstances in which a launcher failure involved LoM but where the first stage is recoverable. If the answer is yes, it's sensible to plan for such situations - the first stage is a piece of equipment costing millions of dollars, even in the rocket industry you don't throw such things away unnecessarily.


...I think planning for some of these scenarios is going to happen sooner than some might think. Basically, as soon as it becomes clear that the F9 first stage is recoverable. If there was a launch where four engines failed, SpaceX is going to want to try and recover the first stage (assuming the centre engine isn't one of them!). You'd simply keep the engines firing to burn off the propellant; launch the second stage and then try to land!

I don't think planning for the kind of failure you cite is worth it. Losing four engines is very unlikely. Instead the effort should go into improving the reliability of the vehicle so aborts become less likely in the first place.

Unlikely things happen! (And there may be other more likely causes of LoM; a four-engine failure was something I came up with off the top of my head.) But it's a question of economics; efforts in improving reliability run into diminishing returns, at some point there is a crossover as to where further resources should be spent. No doubt someone will run the figures. But SpaceX has a Silicon Valley mindset, part of which is the Japanese concept of kaizen or continuous improvement. And the F9 is a research and development vehicle as much as an operational one; best to learn how to do these things before moving on to bigger and more valuable launchers!

Quote
The correct way to go in the event of first stage engine failure is to abort the first stage recovery sequence to recover the margin needed to allow the second stage and payload to achieve orbit.
Really? Or is this another of those items of received wisdom developed in an era of expendable launchers? For example, if it's a cargo Dragon, the first stage could be worth more than the second stage, capsule and cargo combined! Given that the Dragon and cargo are probably recoverable as well, would it not make financial sense to abort the mission; recover the first stage and relaunch the cargo later? (If the payload is a satellite, the opposite might well be true, but horses for courses.)

Offline meekGee

  • Senior Member
  • *****
  • Posts: 7757
  • N. California
  • Liked: 4049
  • Likes Given: 830
Re: Reusability as abort functionality
« Reply #22 on: 11/03/2013 06:07 PM »
My 2c on this often discussed topic:

For a VTVL system, landing involves much the same systems as launch.  So if the launch is not going to plan, odds are that landing is not going to be possible - even if such an idea was entertained at design time.

On top of that, making it even theoretically possible requires that several systems be beefed up (in particular the legs) and so for that very narrow set of circumstances where such recovery can be entertained, you'd be paying a mass penalty on each and every flight.

So no, IMO this is not a good idea.
ABCD - Always Be Counting Down

Offline Lars_J

  • Senior Member
  • *****
  • Posts: 6161
  • California
  • Liked: 664
  • Likes Given: 195
Re: Reusability as abort functionality
« Reply #23 on: 11/03/2013 07:25 PM »
Exactly - it is not practical until we have gas-n-go RLV's which launch from their own legs.

Offline llanitedave

  • Senior Member
  • *****
  • Posts: 2102
  • Nevada Desert
  • Liked: 1280
  • Likes Given: 1487
Re: Reusability as abort functionality
« Reply #24 on: 11/03/2013 07:29 PM »
perhaps this should be moved to advanced concepts?
because we are talking about 10+ years here, for sure.

No we're not. Once first stage reusability is demonstrated (which I think will be in the near future), it becomes obvious to ask whether there would be circumstances in which a launcher failure involved LoM but where the first stage is recoverable. If the answer is yes, it's sensible to plan for such situations - the first stage is a piece of equipment costing millions of dollars, even in the rocket industry you don't throw such things away unnecessarily.


...I think planning for some of these scenarios is going to happen sooner than some might think. Basically, as soon as it becomes clear that the F9 first stage is recoverable. If there was a launch where four engines failed, SpaceX is going to want to try and recover the first stage (assuming the centre engine isn't one of them!). You'd simply keep the engines firing to burn off the propellant; launch the second stage and then try to land!

I don't think planning for the kind of failure you cite is worth it. Losing four engines is very unlikely. Instead the effort should go into improving the reliability of the vehicle so aborts become less likely in the first place.

Unlikely things happen! (And there may be other more likely causes of LoM; a four-engine failure was something I came up with off the top of my head.) But it's a question of economics; efforts in improving reliability run into diminishing returns, at some point there is a crossover as to where further resources should be spent. No doubt someone will run the figures. But SpaceX has a Silicon Valley mindset, part of which is the Japanese concept of kaizen or continuous improvement. And the F9 is a research and development vehicle as much as an operational one; best to learn how to do these things before moving on to bigger and more valuable launchers!

Quote
The correct way to go in the event of first stage engine failure is to abort the first stage recovery sequence to recover the margin needed to allow the second stage and payload to achieve orbit.
Really? Or is this another of those items of received wisdom developed in an era of expendable launchers? For example, if it's a cargo Dragon, the first stage could be worth more than the second stage, capsule and cargo combined! Given that the Dragon and cargo are probably recoverable as well, would it not make financial sense to abort the mission; recover the first stage and relaunch the cargo later? (If the payload is a satellite, the opposite might well be true, but horses for courses.)

Unlikely things happen, but once life and limb are protected, sometimes its more economical just to eat the loss of the unlikely things and just concentrate on continually making them less likely.
"I've just abducted an alien -- now what?"

Offline go4mars

  • Senior Member
  • *****
  • Posts: 3745
  • Earth
  • Liked: 146
  • Likes Given: 3040
Re: Reusability as abort functionality
« Reply #25 on: 11/03/2013 10:22 PM »
(If the payload is a satellite, the opposite might well be true, but horses for courses.)
That's apt cuddly!  Give horses a few days in low gravity, then race them around the circumference.  My kind of future.  Cheers.
Elasmotherium; hurlyburly Doggerlandic Jentilak steeds insouciantly gallop in viridescent taiga, eluding deluginal Burckle's abyssal excavation.

Online douglas100

  • Senior Member
  • *****
  • Posts: 2152
  • Liked: 214
  • Likes Given: 95
Re: Reusability as abort functionality
« Reply #26 on: 11/04/2013 08:58 AM »

Really? Or is this another of those items of received wisdom developed in an era of expendable launchers?

Yes, really. And it isn't "received wisdom." It's common sense as opposed to wishful thinking.

F9R in its presently conceived form (even with recovery of both stages) cannot recover the payload in the event of an abort. To put any effort into abort scenarios when you are going to lose the payload is waste of effort. The effort should go into increasing reliability so the possibility of abort is reduced in the first place. (And that makes it more likely you're going to recover the first stage successfully too.)

Payload abort recovery is a capability for a future vehicle. So we are talking about something which should be in advanced concepts.
Douglas Clark

Offline Jcc

  • Full Member
  • ****
  • Posts: 817
  • Liked: 179
  • Likes Given: 77
Re: Reusability as abort functionality
« Reply #27 on: 11/04/2013 10:21 AM »
Or Skylon.

Offline CraigLieb

  • Full Member
  • ****
  • Posts: 772
  • Dallas Fort Worth
  • Liked: 679
  • Likes Given: 768
Re: Reusability as abort functionality
« Reply #28 on: 11/04/2013 03:29 PM »
Crazy idea number 37:
We seem to assume that a fully-loaded complete stack can't be supported by legs in an abort and CG issues would cause control and tip-over issues. What if there was a specially prepared pit with appropriate fire resistant, padded material (think deep ball pit for the kids play area at pizza joint)?

The pit would have to be deep enough and wide enough to allow a landing vehicle to decend possibly half way up the stack into it and then essentially either rest on the bottom on those legs, or tip to the side somewhat as the legs gave way. The pit would have to have side materials that could support the weight of the stage gently, or mechanisms for gently closing as the stage descends.  This way, legs don't have the whole job to support the recovered vehicle.  This would only be an emergency contingent. I don't know if any stage recoverd in this way would be usable again, maybe just the engines. In addition,  the sides of any pit (in a tip-over) would press against the vehicle 1st stage, and that creates new stresses for which the vehicle isn't designed.  Is that enough to kill the idea?
Colonize Mars!

Offline Lars_J

  • Senior Member
  • *****
  • Posts: 6161
  • California
  • Liked: 664
  • Likes Given: 195
Re: Reusability as abort functionality
« Reply #29 on: 11/04/2013 03:37 PM »
Are you serious?

Offline AJA

  • Full Member
  • ****
  • Posts: 856
  • Per Aspera Ad Ares, Per Aspera Ad Astra
  • India
  • Liked: 135
  • Likes Given: 203
Re: Reusability as abort functionality
« Reply #30 on: 11/04/2013 03:53 PM »
Crazy idea number 37:

LOL.. Entertaining the thought for a second... the "pit" can be the sea. A fully loaded Falcon 9 (either v1.0/ v1.1) - with maximum payload - still has a relative density of (505846 kg+13150 kg)/(pi*((3.7 m/2)^2)*(68.4 m)) ~= 0.7

Offline mlindner

  • Software Engineer
  • Full Member
  • ****
  • Posts: 1887
  • Space Capitalist
  • Silicon Valley, CA -- previously in Ann Arbor, MI
  • Liked: 465
  • Likes Given: 188
Re: Reusability as abort functionality
« Reply #31 on: 11/05/2013 12:39 AM »
Why are we discussing this? This is impossible. A rocket that has a design with legs rated for a full vehicle load won't be making it to orbit with any reasonable payload. Not to mention that after any kind of flight path is flown, returning to launch site becomes impossible because of lack of fuel. Not to mention if a flight path change is attempted while in the atmosphere the vehicle will break up.

Early abort - too heavy
Mid abort - too much dynamic atmospheric pressure from velocity
Late abort - not enough fuel to return
Very late abort (upper stage + payload only) - need heat shield on bottom of stage where the engine is

Can we close this?
« Last Edit: 11/05/2013 12:42 AM by mlindner »
Internal combustion engine in space. It's just a Bad Idea.TM - Robotbeat

Offline CuddlyRocket

Re: Reusability as abort functionality
« Reply #32 on: 11/05/2013 02:29 AM »

Really? Or is this another of those items of received wisdom developed in an era of expendable launchers?

Yes, really. And it isn't "received wisdom." It's common sense as opposed to wishful thinking.

Being trained as a scientist, I am wary of arguments based on common sense, as they frequently turn out not to be sensible even if common. I gave an argument why it may not be as sensible in this case as you suppose, which you signally failed to address.

Quote
F9R in its presently conceived form (even with recovery of both stages) cannot recover the payload in the event of an abort.

If the payload is a Dragon of course it can as the Dragon is designed to be recoverable in the event of an abort; hence the upcoming abort tests. (Granted, these are on a manned Dragon, but the cargo Dragon is a variant.)

Quote
To put any effort into abort scenarios when you are going to lose the payload is waste of effort. The effort should go into increasing reliability so the possibility of abort is reduced in the first place. (And that makes it more likely you're going to recover the first stage successfully too.)

Why is it an either/or?

Quote
Payload abort recovery is a capability for a future vehicle. So we are talking about something which should be in advanced concepts.

I don't agree with the first sentence. And are different ways of using existing vehicles really advanced concepts?

Why are we discussing this? This is impossible. A rocket that has a design with legs rated for a full vehicle load won't be making it to orbit with any reasonable payload.

Nobody's suggesting this. Well, at least I'm not - I assumed that landing legs would be designed for a nearly empty vehicle. This does mean you'd have to consider whether burn off of excess propellant is possible.

Quote
Not to mention that after any kind of flight path is flown, returning to launch site becomes impossible because of lack of fuel.

That would mean reusability is impossible, which seems at odds with what SpaceX are attempting.

Quote
Not to mention if a flight path change is attempted while in the atmosphere the vehicle will break up.

Best not do it in the atmosphere then. Or reduce speed sufficiently first.

Look, I'm not saying that recovery in an abort is easy; just that it may not be impossible, and if not impossible then it's worth looking at because millions of dollars could be saved.

Quote
Can we close this?

If you're not interested just don't read the thread! Why propose short-circuiting things for other people who may be?

Offline Lars_J

  • Senior Member
  • *****
  • Posts: 6161
  • California
  • Liked: 664
  • Likes Given: 195
Re: Reusability as abort functionality
« Reply #33 on: 11/05/2013 04:15 AM »

Really? Or is this another of those items of received wisdom developed in an era of expendable launchers?

Yes, really. And it isn't "received wisdom." It's common sense as opposed to wishful thinking.

Being trained as a scientist, I am wary of arguments based on common sense, as they frequently turn out not to be sensible even if common. I gave an argument why it may not be as sensible in this case as you suppose, which you signally failed to address.

I don't know where you "trained as a scientist" - but if you have some wild idea that goes against common sense, you should back it up with some evidence, instead of complaining that it is too easily dismissed. The burden of proof is on *you*. And if you cannot back it up... well, there isn't much science in your approach, now is there?

Quote
Payload abort recovery is a capability for a future vehicle. So we are talking about something which should be in advanced concepts.

I don't agree with the first sentence. And are different ways of using existing vehicles really advanced concepts?

What don't you agree with? The F9R will have very thin margins for being able to recover its stages. Therefore any anomaly that reduces performance (and would cause an abort) would eat into the very thin recovery margin. Do you dispute this?

Why are we discussing this? This is impossible. A rocket that has a design with legs rated for a full vehicle load won't be making it to orbit with any reasonable payload.

Nobody's suggesting this. Well, at least I'm not - I assumed that landing legs would be designed for a nearly empty vehicle. This does mean you'd have to consider whether burn off of excess propellant is possible.

And what do you think the most efficient way of burning off or venting propellant is? As others have pointed out, it is through a burning engine. And virtually all aborts will involved a propulsion failure. This will negatively impact your ability to burn off propellant - Do you see a pattern forming?

Quote
Not to mention if a flight path change is attempted while in the atmosphere the vehicle will break up.

Best not do it in the atmosphere then. Or reduce speed sufficiently first.

It sounds rather easy when you put it that way. But aborts are most likely to be caused by two problems - guidance and propulsion. Neither will make what you suggest very plausible.

Look, I'm not saying that recovery in an abort is easy; just that it may not be impossible, and if not impossible then it's worth looking at because millions of dollars could be saved.

Just because it may be theoretically possible in a very few situations (care to provide any?), does not make it practical or useful. I see from your profile that you are from London. Do you have earthquake insurance? If not, why?

Quote
Can we close this?

If you're not interested just don't read the thread! Why propose short-circuiting things for other people who may be?

He gave four abort scenarios, why did you not respond to them?
« Last Edit: 11/05/2013 04:18 AM by Lars_J »

Offline meekGee

  • Senior Member
  • *****
  • Posts: 7757
  • N. California
  • Liked: 4049
  • Likes Given: 830
Re: Reusability as abort functionality
« Reply #34 on: 11/05/2013 04:27 AM »

...
Why is it an either/or?
...


Cuddly - I'm all for exploring the envelope.

But you have to address the issue of how a sick rocket is able to perform a maneuver that requires the rocket to be healthy.   There's a very narrow set of circumstances that fits this bill.

Consider:

If you have a GNC issue, definitely no RTLS abort today

If you have a structural issue, definitely no RTLS abort today

If you have a steering actuation issue, no dice.

If you're already moving at a significant speed, but are still heavy, the loads involving RTLS abort are higher than a normal launch (inertia, aero loads etc).

If you're not generating enough thrust to ascend (e.g lost 2 engines), then likely you're descending, and so will crash before you pull of the RTLS abort trick.

...

Sso there's only a very very small set of circumstances in which a rocket is unable to achieve orbit, but is healthy enough to perform RTLS abort.  With a deeper look, since rockets don't have a lot of margin, I suspect that set of circumstances is actually empty. 

Also consider - when things are not going well, you don't always know in real time what the root cause is, so you're not in a good position to make a decision on whether it is possible to RTLS without risking your base.

OTOH - it's just a rocket.   If there's a capsule on top, it can abort safely.  If it's a payload, maybe one day it will be able to abort - but probably the mass is not worth it.  Maybe it will be a per-payload decision.     But the rocket itself?  Let it go.  Not worth it.

----

Here, much shorter:

After a pilot ejects, aircraft are not designed to try to land themselves.  This is not because it is technologically impossible.  It is because there was a reason the pilot ejected, and so the plane is most very likely a goner already.
« Last Edit: 11/05/2013 04:33 AM by meekGee »
ABCD - Always Be Counting Down

Offline Elmar Moelzer

  • Senior Member
  • *****
  • Posts: 2917
  • Liked: 488
  • Likes Given: 733
Re: Reusability as abort functionality
« Reply #35 on: 11/05/2013 01:23 PM »
Hmm, what about recovering only the second stage with the payload in case of a first stage failure. Obviously not as neat as recovering the entire rocket, but maybe more plausible?

Online Silmfeanor

  • Full Member
  • ****
  • Posts: 1164
  • Utrecht, The Netherlands
  • Liked: 217
  • Likes Given: 458
Re: Reusability as abort functionality
« Reply #36 on: 11/05/2013 01:36 PM »
Hmm, what about recovering only the second stage with the payload in case of a first stage failure. Obviously not as neat as recovering the entire rocket, but maybe more plausible?

Same objections.
The 2nd stage will need to have legs to carry a partially full 2nd stage and a fueled payload.
It will need control authority sufficient for such a load
it will require advanced on-the-fly analysis on the exact weight of the stage, propellant, current trajectory, position, CoG and much more.
It will can't be too heavy, or the M1Dvac will not be able to land; or too late, since it can't return. The throttleability of the M1Dvac needs to be sufficient for the loads and landing, which it normally would never do ( if we assume a very close to empty 2nd stage landing on something like superdracos as normal).

Then there is the issue of thermo and aero dynamics on the payload shroud. A normal 2nd stage would never land with it attached. Thus control issues are even greater. The shroud needs to be strengthened to go through a new environment ( the abort path ).
The same goes for the payload. New paths, new shocks, new environments, new loads.

The set of cases where the 2nd stage can recover payloads to the ground is 0.

All of this eats into the payload mass, into the cost of the rocket, and adds new failure paths. Combined this makes the economic possibility of such a change not worth it  and might even make the flying of the rocket not worth it. For each mission where you want this to be a possibility, you will have additional requirements for the rocket and the payload provider. All of these cost big bucks. So, you have a 95%+ succes rate, and all your flights cost 20% more because of these requirements in mass and engineer analysis.  This causes you to lower your flight rate, etc etc etc.

This discussion, I think, is useless.
It's a bit like taking the following seriously:

Imagine that 2 M1D's crap out on the same side of a F9R. SpaceX can easily add another rings of 8 M1D's halfway up the stage, above the legs. If the lower engines crap out, just ignite the next ring, so you can complete the mission! That way there'll be improved safety margins!

Sure it is possible. It'll just never happen, because it makes no sense. The discussion in this thread is on the same path. Dragon can be aborted because it is designed to reenter anyway.

Remember, this is NSF. More engineering based on real-life rockets, less " but they should totally design for this because i think it's possible".

Offline bob the martian

  • Full Member
  • *
  • Posts: 116
  • Liked: 70
  • Likes Given: 31
Re: Reusability as abort functionality
« Reply #37 on: 11/07/2013 03:43 PM »
I was thinking a F9R could use it's ability to land to abort a failed launch.
E.g if an F9R suffers an early stage engine loss, or later double engine loss.
What would be needed to be done to allow it to land with a 2nd stage plus payload on top?

My first thought was it needs to vent a huge amount of propellant to avoid overloading the legs during landing.

Why would you expect to be able to safely land an entire stack on a stage that's undergone multiple engine failures?  The odds of multiple independent engine failures should be infinitesimally low; if more than one engine shreds itself on the same launch, then you almost have to assume something is systemically wrong with the stage and that it is no longer safe to fly.  Not to mention that if you lose the center engine, you're boned on recovery regardless. 

The best-case scenario is recovering the US.  Hopefully it can make the desired orbit (or close enough) on the remaining engines and deploy the payload normally.  Otherwise, it stages early and jettisons the payload (I don't think it will be able to land with the payload and fairing still attached). 

SpaceX have built in a reasonable amount of redundancy in the F9R, but some scenarios are simply unrecoverable. 

« Last Edit: 11/07/2013 03:57 PM by bob the martian »

Offline bioelectromechanic

  • Member
  • Posts: 50
  • Liked: 4
  • Likes Given: 24
Re: Reusability as abort functionality
« Reply #38 on: 11/09/2013 08:20 PM »
I was thinking a F9R could use it's ability to land to abort a failed launch.
E.g if an F9R suffers an early stage engine loss, or later double engine loss.
What would be needed to be done to allow it to land with a 2nd stage plus payload on top?

My first thought was it needs to vent a huge amount of propellant to avoid overloading the legs during landing.

It is a non starter. It would need stronger legs for the added weight of upperstage, payload and fairing.   Vent is a misnomer.  It has to get rid of the propellant fast and guess what?  It can't get rid of it faster than burning it through the engines.  That includes the upperstage propellant.  And then there is  the spacecraft.  Not to mention the that whole vehicle has to be able to take the new aeroloads from the pitch around maneuver.  That means strengthened interstage and fairing.  Also, there now is a live payload, with toxic and hazardous propellant onboard with no ability to command it if using current conops (typically, there is no electrical or rf interaction between spacecraft and LV except for breakwires, and maybe some occasion discreets )

This would require a whole redesign of the vehicle

What about a water recovery, after stage separation?
There might be value in recovering a stage for fault diagnosis and partial reuse.
Carpe diem et vadem ad astra

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 31348
  • Cape Canaveral Spaceport
  • Liked: 9625
  • Likes Given: 299
Re: Reusability as abort functionality
« Reply #39 on: 11/09/2013 08:23 PM »


What about a water recovery, after stage separation?
There might be value in recovering a stage for fault diagnosis and partial reuse.

Not if it is not designed for water recovery.  Not all failures are graceful.
« Last Edit: 11/09/2013 08:24 PM by Jim »

Tags: