SpaceX long-term stage processing goal = 48 hour turnaround

[Arb]

SpaceX job adverts sometimes yield interesting nuggets information. Here's two from Vehicle Operations Engineer (Launch Engineering)

Identify areas for improvement in ... rapid and reliable processing, and work ... to implement changes to equipment, tooling, operations, and the launch vehicle to serve the company’s long term processing goals (48 hour turnaround from stage arrival to launch, and a 4 hour stage acceptance series in Texas)

Emphasis mine.

Discuss.

[Arb]

Related, they are also still saying (Tooling Manager (Travel Team)):

Help SpaceX achieve its long-term goal of creating the world’s first fully automated launch system capable of rolling the vehicle to the pad, raising it to position, fueling, and executing a full launch sequence in a single hour

[FutureSpaceTourist]

Good finds, thanks. Even with the now reduced 30 mins or so to load fuel, roll out and launch in an hour is pretty ambitious. Coupled with the 48 hr turnaround it's clear SpaceX are aiming to be able to support very high flight rates

I wonder if the 48 hr aim applies to both new and re-used vehicles/stages?

[Kansan52]

I wonder if the 48 hr aim applies to both new and re-used vehicles/stages?

IMHO, yes. Once checked out, a flight proven vehicle should process as smoothly as a new vehicle.

[baldusi]

I wonder if the 48 hr aim applies to both new and re-used vehicles/stages?

IMHO, yes. Once checked out, a flight proven vehicle should process as smoothly as a new vehicle.

Or may be faster, since it will be "proven".

[chalz]

Would 4 hour stage acceptance in Texas just be: getting it off the truck, on the stand, full duration test, back on the truck if all goes well? Do they currently take the engines off the stage to test them individually?

[meekGee]

I understand the quick turn around time goal, but not so much the 4 hour acceptance time goal.

They can't be planning to take each stage back to TX for re-acceptance and still keep 48-hr turn around, since just the drive time is 2 days in each direction.  And of course the stage was just "test fired".

So why would you want to do acceptance testing in 4 hours?  Unless this is for second stages, that still need to be accepted per each flight.

[philw1776]

What's great about over the top, unprecedented aggressive goals like this is it makes engineers think totally "out of the box".  OK, shoot me now because I hate that phrase.  It makes engineers throw out all the preconceived "this is the way it needs to be done" thinking and start with a fresh approach.  Even if they fall short, innovative paradigm breaking approaches will be developed and major improvements made.

[watermod]

For their 4000 some sat configuration I can see the need for lots of launches.  (4000/n-sats per launch)
Is their anything else known to be on their manifest that would require this turn around rate?

 

[Lee Jay]

Related, they are also still saying (Tooling Manager (Travel Team)):

Help SpaceX achieve its long-term goal of creating the world’s first fully automated launch system capable of rolling the vehicle to the pad, raising it to position, fueling, and executing a full launch sequence in a single hour

That one is 7 years old.

http://www.nasaspaceflight.com/2009/01/musk-ambition-spacex-aim-for-fully-reusable-falcon-9/

Musk ambition: SpaceX aim for fully reusable Falcon 9
January 12, 2009 by Chris Bergin

...

“One of the goals I have for the Falcon 9 – which will take us many launches to achieve – is to have the vehicle out of the hanger and into the air in under 60 minutes.”

[wannamoonbase]

What's great about over the top, unprecedented aggressive goals like this is it makes engineers think totally "out of the box".  OK, shoot me now because I hate that phrase.  It makes engineers throw out all the preconceived "this is the way it needs to be done" thinking and start with a fresh approach.  Even if they fall short, innovative paradigm breaking approaches will be developed and major improvements made.

Bingo, that is what I was thinking.  SpaceX and other Elon companies have unrealistic goals but they shoot for unrealistic and settle for revolutionary.

[chalz]

For their 4000 some sat configuration I can see the need for lots of launches.  (4000/n-sats per launch)
Is their anything else known to be on their manifest that would require this turn around rate?

Mars colonisation is the main reason they want this. Plus the long speculated new markets that might materialise with reusable fast turn-around rockets. Nothing on the manifest would need it because they plan multiple launch sites in the interim.

[meekGee]

What's great about over the top, unprecedented aggressive goals like this is it makes engineers think totally "out of the box".  OK, shoot me now because I hate that phrase.  It makes engineers throw out all the preconceived "this is the way it needs to be done" thinking and start with a fresh approach.  Even if they fall short, innovative paradigm breaking approaches will be developed and major improvements made.

Personal nit:  You meant "outside the box", right?

[Im_Utrecht]

Actually it is very simple and Elon said it many times.
He want to build a rocket that can be used as an airplane like a 747.
Otherwise he would consider it a failure.

[philw1776]

What's great about over the top, unprecedented aggressive goals like this is it makes engineers think totally "out of the box".  OK, shoot me now because I hate that phrase.  It makes engineers throw out all the preconceived "this is the way it needs to be done" thinking and start with a fresh approach.  Even if they fall short, innovative paradigm breaking approaches will be developed and major improvements made.

Personal nit:  You meant "outside the box", right?

Yes, of course.
I was going to blame autocorrect but I retyped outside and autocorrect was not to blame.

[meekGee]

What's great about over the top, unprecedented aggressive goals like this is it makes engineers think totally "out of the box".  OK, shoot me now because I hate that phrase.  It makes engineers throw out all the preconceived "this is the way it needs to be done" thinking and start with a fresh approach.  Even if they fall short, innovative paradigm breaking approaches will be developed and major improvements made.

Personal nit:  You meant "outside the box", right?

Yes, of course.
I was going to blame autocorrect but I retyped outside and autocorrect was not to blame.

That never stopped anyone else from blaming it...

[RocketGoBoom]

The limitation won't be their own ability to get the rocket ready.

I think the barrier they will run into is the ability to get the times and dates cleared for the down range. It is not trivial it clean air traffic, boat traffic, etc.

[skybum]

I think the barrier they will run into is the ability to get the times and dates cleared for the down range. It is not trivial it clean air traffic, boat traffic, etc.

Clearing air traffic isn't a problem: aircraft do it for each other all the time, continuously. In the scenario where launch vehicles are treated with aircraft-like turnaround times, they would undoubtedly be treated much like any other aircraft.

Same goes for clearing boat traffic: you don't do it for airplane takeoffs, so why do it for a well-proven reusable rocket stage which you have no intention of dumping in the drink?

[Herb Schaltegger]

I think the barrier they will run into is the ability to get the times and dates cleared for the down range. It is not trivial it clean air traffic, boat traffic, etc.

Clearing air traffic isn't a problem: aircraft do it for each other all the time, continuously. In the scenario where launch vehicles are treated with aircraft-like turnaround times, they would undoubtedly be treated much like any other aircraft.

Um, no. They don't, at least not in the U.S. above 5,000 feet or within specified radii around certain designated metropolitan, military and spaceflight installations. Aircraft (and launching/entering spacecraft) in and around those areas in U.S. territory are subject to FAA flight rules and restrictions. There is a LOT of commercial aircraft travel in south Florida and all along the eastern seaboard of North America.

[AncientU]

The limitation won't be their own ability to get the rocket ready.

I think the barrier they will run into is the ability to get the times and dates cleared for the down range. It is not trivial it clean air traffic, boat traffic, etc.

As long as thinking starts from our tightly constrained existing model of rocket launches (rare, military like, expendible, government controlled, etc.), you are correct that this would be the barrier.  But as noted above, thinking out of the box or outside for some, forces engineers to cast all of that off and find what is physically limiting.  Most of the rest are rules people made up and people can change them.

[Glom]

I think the barrier they will run into is the ability to get the times and dates cleared for the down range. It is not trivial it clean air traffic, boat traffic, etc.

Clearing air traffic isn't a problem: aircraft do it for each other all the time, continuously. In the scenario where launch vehicles are treated with aircraft-like turnaround times, they would undoubtedly be treated much like any other aircraft.

Um, no. They don't, at least not in the U.S. above 5,000 feet or within specified radii around certain designated metropolitan, military and spaceflight installations. Aircraft (and launching/entering spacecraft) in and around those areas in U.S. territory are subject to FAA flight rules and restrictions. There is a LOT of commercial aircraft travel in south Florida and all along the eastern seaboard of North America.

Isn't the point that if they can get to launching in such a casual and reliable way, that will lead to a paradigm shift in how airspace is viewed and it will be designed for permanently active launch corridors with reduced margins because we have so much confidence in the rockets now.

I know we're not there yet.

[rpapo]

As long as thinking starts from our tightly constrained existing model of rocket launches (rare, military like, expendible, government controlled, etc.), you are correct that this would be the barrier.  But as noted above, thinking out of the box or outside for some, forces engineers to cast all of that off and find what is physically limiting.  Most of the rest are rules people made up and people can change them.

Well, they could start by making certain areas immediately east and northeast of the cape as permanent flight restriction zones, much like how nobody's is/was allowed to fly over Area 51.  Restricting boat traffic would be more difficult than air traffic, I think, since you can't just re-route ships to the west of the cape...

[joek]

I think the barrier they will run into is the ability to get the times and dates cleared for the down range. It is not trivial it clean air traffic, boat traffic, etc.

Clearing air traffic isn't a problem: aircraft do it for each other all the time, continuously. In the scenario where launch vehicles are treated with aircraft-like turnaround times, they would undoubtedly be treated much like any other aircraft.

Um, no. They don't, at least not in the U.S. above 5,000 feet or within specified radii around certain designated metropolitan, military and spaceflight installations. Aircraft (and launching/entering spacecraft) in and around those areas in U.S. territory are subject to FAA flight rules and restrictions. There is a LOT of commercial aircraft travel in south Florida and all along the eastern seaboard of North America.

Isn't the point that if they can get to launching in such a casual and reliable way, that will lead to a paradigm shift in how airspace is viewed and it will be designed for permanently active launch corridors with reduced margins because we have so much confidence in the rockets now.

These issues have been recognized and worked for some time, and there is a considerable body of work on the subject; e.g., see (among others):
- Air Traffic Considerations for Future Spaceports, FAA, May 2014
- The FAA’s Current Approach to Integrating Commercial Space Operations into the National Airspace System, FAA, Sep 2013
- Space Transportation Concept of Operations Annex for NextGen, FAA, Aug 2012
- A Tool for Integrating Commercial Space Operations Into The National Airspace System, FAA/AIAA, Aug 2006
- Industry Trends and Key Issues Affecting Federal Oversight and International Competitiveness, GAO, May 2011

edit: add NextGen link.

[shooter6947]

Well, they could start by making certain areas immediately east and northeast of the cape as permanent flight restriction zones, much like how nobody's is/was allowed to fly over Area 51.  Restricting boat traffic would be more difficult than air traffic, I think, since you can't just re-route ships to the west of the cape...

That would be a pretty onerous flight restriction.  With no way to fly over or under, and no way to go around to the east, that would definitely put a damper on aviation traffic.

Would the situation be easier at Boca Chica?  Since that's the edge of the US anyway, and there's much less air traffic in that vicinity, a permanent flight restriction zone could fare better.

[joek]

Well, they could start by making certain areas immediately east and northeast of the cape as permanent flight restriction zones, much like how nobody's is/was allowed to fly over Area 51.  Restricting boat traffic would be more difficult than air traffic, I think, since you can't just re-route ships to the west of the cape...

That would be a pretty onerous flight restriction.  With no way to fly over or under, and no way to go around to the east, that would definitely put a damper on aviation traffic.

Would the situation be easier at Boca Chica?  Since that's the edge of the US anyway, and there's much less air traffic in that vicinity, a permanent flight restriction zone could fare better.

The objective is better integration and management of the entire US national airspace for all uses; see Space Transportation Concept of Operations Annex for NextGen.

I suggest a new thread in the Commercial Space Flight General section if you want to pursue the subject.

[Arb]

Restricting boat traffic would be more difficult than air traffic, I think, since you can't just re-route ships to the west of the cape...

Permanently restricted sea areas are not uncommon. Typically because "military and/or restricted and/or prohibited and/or dangerous and/or you-might-get-blown-up". See Don't go there by Beth Walsh for Chesapeake Bay examples.

[Senex]

I originally posted this over on the "Refurbishment" thread, but it would seem to be at least as relevant here.  And it may serve to get discussion off of airspace restrictions and back on the subject of processing boosters for re-flight.


While speculation is popular and is often justified by the absence of facts, there ARE some facts available. An earlier reference to the X-15 is highly relevant as it endured a very similar flight regime in terms of stresses (arguably greater). 

In an interesting article, space historian David Portree cites a study that looked at the refurbishment costs of the X-15 program that provides probably the most relevant real-world data we have:


"In November 1966, James Love and William Young, engineers at the NASA Flight Research Center at Edwards Air Force Base, completed a brief report in which they noted that the reusable suborbital booster for a reusable orbital spacecraft would undergo pressures, heating rates, and accelerations very similar to those the X-15 experienced."

"The average X-15 refurbishment time was 30 days, a period which had, they noted, hardly changed in four years. Even with identifiable improvements, they doubted that an X-15 could be refurbished in fewer than 20 days.

"At the same time, Love and Young argued that the X-15 program had demonstrated the benefits of reusability. They estimated that refurbishing an X-15 in 1964 had cost about $270,000 per mission.

"Love and Young cited North American Aviation estimates when they placed the cost of a new X-15 at about $9 million. They then calculated that 27 missions using expendable X-15s would have cost a total of $243 million. This meant, they wrote, that the cost of the reusable X-15 program in 1964 had amounted to just 3% of the cost of building 27 X-15s and throwing each one away after a single flight.

My bold.

http://www.wired.com/2013/05/the-x-15-rocket-plane-reusable-space-shuttle-boosters-1966/

At least two implications can be deduced from this reference:

1. There has been much talk on the forums about the effects of "fatigue" with images of much of the booster requiring rework or even replacement.  Three X-15's flew 199 flights — and average of 66 each.  Just because a Falcon 9 flies high and fast does not mean it will be structurally degraded after a few cycles.

2. The X-15 required weeks of refurbishment.  It is safe to assume that most of that related to mechanical equipment.  That was with technology that had literally just been invented.  In a vehicle designed from the beginning with operating economies in mind, using a relatively mature technology, this might be dramatically lower.

[DaveH62]

I don't see a thread on policy implications of re-usability and I'm wondering if there has been much research and thought about the impact of radically reducing the time and cost to get to space.  It seems that 48 hour turn around would have a huge effect on costs and contracting methodology.

[alang]

I think the barrier they will run into is the ability to get the times and dates cleared for the down range. It is not trivial it clean air traffic, boat traffic, etc.

Clearing air traffic isn't a problem: aircraft do it for each other all the time, continuously. In the scenario where launch vehicles are treated with aircraft-like turnaround times, they would undoubtedly be treated much like any other aircraft.

Um, no. They don't, at least not in the U.S. above 5,000 feet or within specified radii around certain designated metropolitan, military and spaceflight installations. Aircraft (and launching/entering spacecraft) in and around those areas in U.S. territory are subject to FAA flight rules and restrictions. There is a LOT of commercial aircraft travel in south Florida and all along the eastern seaboard of North America.

Isn't the point that if they can get to launching in such a casual and reliable way, that will lead to a paradigm shift in how airspace is viewed and it will be designed for permanently active launch corridors with reduced margins because we have so much confidence in the rockets now.

These issues have been recognized and worked for some time, and there is a considerable body of work on the subject; e.g., see (among others):
- Air Traffic Considerations for Future Spaceports, FAA, May 2014
- The FAA’s Current Approach to Integrating Commercial Space Operations into the National Airspace System, FAA, Sep 2013
- Space Transportation Concept of Operations Annex for NextGen, FAA, Aug 2012
- A Tool for Integrating Commercial Space Operations Into The National Airspace System, FAA/AIAA, Aug 2006
- Industry Trends and Key Issues Affecting Federal Oversight and International Competitiveness, GAO, May 2011

edit: add NextGen link.

There are a lot of airline scheduling and flight operational systems that model flights according to a flight 'key' that assumes that only commercial aviation exists. At least in Europe, communication between local ATC, National Air Traffic control, Airports and Airlines has elements of ambiguity. Call signs are allocated by airlines within certain rules and occasional overruled in order to disambiguate them.
Much work has been done between the FAA, Eurocontrol, Airlines and airports to evaluate the idea of a Global Universal Flight Identifier and have real time trajectory planning and airspace modification, however given how little money has been made by airlines since 9/11 I suspect that legislation will be required to get airlines and general aviation to conform to the requirements necessary to deal with increased use of airspace by other players such as drones and rockets.

[mark_m]

SpaceX job adverts sometimes yield interesting nuggets information. Here's two from Vehicle Operations Engineer (Launch Engineering)

Identify areas for improvement in ... rapid and reliable processing, and work ... to implement changes to equipment, tooling, operations, and the launch vehicle to serve the company’s long term processing goals (48 hour turnaround from stage arrival to launch, and a 4 hour stage acceptance series in Texas)

Emphasis mine.

Discuss.

I may be off base here, but I didn't read this as necessarily having anything to do with reuse. I read it as turnaround from stage arrival in Texas, either after completing refurbishment or brand new from the factory, until it's ready for launch.

[chalz]

SpaceX job adverts sometimes yield interesting nuggets information. Here's two from Vehicle Operations Engineer (Launch Engineering)

Identify areas for improvement in ... rapid and reliable processing, and work ... to implement changes to equipment, tooling, operations, and the launch vehicle to serve the company’s long term processing goals (48 hour turnaround from stage arrival to launch, and a 4 hour stage acceptance series in Texas)

Emphasis mine.

Discuss.

I may be off base here, but I didn't read this as necessarily having anything to do with reuse. I read it as turnaround from stage arrival in Texas, either after completing refurbishment or brand new from the factory, until it's ready for launch.

Off base but not run out: it is not specific to reused stages. Stage arrival means arrival at the launch site(wherever that is), not Texas. 'Turnaround' then refers to getting all the components into a functioning rocket with payload. It's true the first stage could be arriving from a landing site 1km away but as a process issue I think they would want it's origin to be unimportant.

Acceptance testing means making sure a new stage works by a bunch of tests (which I am ignorant of) and at least one full duration burn (around 6 mins) of just the stage on the vertical test stand in Texas. Once it has passed it is ready to ship to the launch site.

How acceptance testing will interact with reused stages is not yet clear. We know the Orbcomm stage will not be used again so whether it goes back to Texas or not is not indicative. So far Musk said they have found 'no damage' and on that basis are prepared to static fire the stage. Potentially an easy process to streamline for future stage landings. This would line up with wanting to test a brand new stage in only 4 hours.

None of this speed is much use yet with weeks between launches but is part of the long term aspiration of cheaper, quicker commercial rocket launch. I'm sure they are both years away from being realised too.

[Jim]

I originally posted this over on the "Refurbishment" thread, but it would seem to be at least as relevant here.  And it may serve to get discussion off of airspace restrictions and back on the subject of processing boosters for re-flight.


While speculation is popular and is often justified by the absence of facts, there ARE some facts available. An earlier reference to the X-15 is highly relevant as it endured a very similar flight regime in terms of stresses (arguably greater). 

In an interesting article, space historian David Portree cites a study that looked at the refurbishment costs of the X-15 program that provides probably the most relevant real-world data we have:


"In November 1966, James Love and William Young, engineers at the NASA Flight Research Center at Edwards Air Force Base, completed a brief report in which they noted that the reusable suborbital booster for a reusable orbital spacecraft would undergo pressures, heating rates, and accelerations very similar to those the X-15 experienced."

"The average X-15 refurbishment time was 30 days, a period which had, they noted, hardly changed in four years. Even with identifiable improvements, they doubted that an X-15 could be refurbished in fewer than 20 days.

"At the same time, Love and Young argued that the X-15 program had demonstrated the benefits of reusability. They estimated that refurbishing an X-15 in 1964 had cost about $270,000 per mission.

"Love and Young cited North American Aviation estimates when they placed the cost of a new X-15 at about $9 million. They then calculated that 27 missions using expendable X-15s would have cost a total of $243 million. This meant, they wrote, that the cost of the reusable X-15 program in 1964 had amounted to just 3% of the cost of building 27 X-15s and throwing each one away after a single flight.

My bold.

http://www.wired.com/2013/05/the-x-15-rocket-plane-reusable-space-shuttle-boosters-1966/

At least two implications can be deduced from this reference:

1. There has been much talk on the forums about the effects of "fatigue" with images of much of the booster requiring rework or even replacement.  Three X-15's flew 199 flights — and average of 66 each.  Just because a Falcon 9 flies high and fast does not mean it will be structurally degraded after a few cycles.

2. The X-15 required weeks of refurbishment.  It is safe to assume that most of that related to mechanical equipment.  That was with technology that had literally just been invented.  In a vehicle designed from the beginning with operating economies in mind, using a relatively mature technology, this might be dramatically lower.

Wrong analogy.  X-15 was an aircraft and designed like one.  Aircraft are reusable.  Design standards have yet to be developed for reusable boosters for fatigues mitigation.

[meekGee]

FWIW, in regards to the entire discussion of fatigue, 3 minutes of powered flight time is already plenty for high-cycle-count fatigue due to vibrations and aerodynamic loads.

So the stage is already designed with fatigue requirements.  (As is any high fidelity design...)

Additionally,SpaceX has been gathering data (to characterize the flight environment) over many flights now, so even before retrieving this stage, they already knew quite a bit about how the stage is being loaded.

Plus - designing and testing for fatigue is not something new.  It's been done in airplanes, ship-building, car manufacturing...  it's not magic.

I can't see why this has become such a giant issue.  Just like previous "hurdles" - it will loom large in the forum until suddenly it will be in the rear view mirror and there wouldn't even be a bump felt to witness it was there...

[Jim]

Yes, the 3 minutes of flight and the instrumentation from the many flights prevented the helium bottle from not breaking free.  Also, flight environments is not the same as vehicle loads.

The accident provided a benefit by pointing out how little data they have on the vehicle.

[Coastal Ron]

X-15 was an aircraft and designed like one.

Falcon 9 v1.1 FT is a reusable rocket and designed like one.

Design standards have yet to be developed for reusable boosters for fatigues mitigation.

Outside of SpaceX you mean.  Industry standards, sure.  But that doesn't mean SpaceX hasn't developed design standards for reusable boosters.

I'm not saying I know SpaceX has, just that reusable boosters have been a goal for SpaceX for years, so it's not like it would be something they haven't thought of, or prepared for.  Same for Blue Origin.

But ULA?  I doubt they would have spent much time or energy on developing design standards for reusable boosters, especially since they don't believe recovering and reusing the enter booster makes sense.  And other than Orbital ATK, SpaceX and ULA make up the entire orbital rocket industry that would be creating such standards.

So SpaceX creating the industry standards themselves would not be surprising...

[Jim]

Falcon 9 v1.1 FT is a reusable rocket and designed like one.

no, it is an expendable rocket that is being converted/upgraded to a reusable one. 

[Robotbeat]

Falcon 9 v1.1 FT is a reusable rocket and designed like one.

no, it is an expendable rocket that is being converted/upgraded to a reusable one.

In what sense? v1.1 for certain was designed to be reusable, with leg attachments even on flights where landing wasn't attempted, from the very first flight. The center engine position and octoweb all were carefully arranged for VTVL recovery and subsequent reuse. They failed to recover it until now, but v1.1 (and v1.1 full thrust) was not designed as an expendable rocket.

It's a reusable rocket cheap enough to be expended, but it was not designed as an expendable rocket.

[Dante80]

Yes, the 3 minutes of flight and the instrumentation from the many flights prevented the helium bottle from not breaking free.  Also, flight environments is not the same as vehicle loads.

The accident provided a benefit by pointing out how little data they have on the vehicle.

You keep saying that. The reality though is that SpaceX identified the problem with the telemetry they had after a fault tree analysis, fixed it, and came back to flight 6 months after the failure. RTF means that the failure investigation is concluded, FAA signed it off, and NASA, USAF and major SpaceX customers got briefed about it. This is a fact, isn't it?

Moreover, there have been hints that the rocket had more than 3000 channels of telemetry. In what way could the "very little data on the vehicle" be substantiated?

[Rocket Science]

The X-15 and Falcon have different parameters used in their design. The X-15 was overbuilt for multiple missions in order to collect data from the space environment and on the vehicle itself in various flight profiles: max height, max speed etc…
The Falcon is designed for re usability and for delivering a payload. It has to walk the fine line of being both robust structurally for reuse, yet light enough as to not affect its mass fraction which would defeat its reason to exist as a reusable vehicle.
Lessons learned from previous X-planes contributed to the database that the X-15 designers drew from. Falcon’s designers are at the pointy end of the spear and are writing the reusable booster books as they proceed through the design evolution. It is an experimental vehicle with an operational mission. Very unique in the world of X vehicles. Let’s remember that only after the Shuttle failures did NASA finally acknowledge that they were flying an experimental vehicle in an operational role with lives on the line…The experiment continues… No conclusion can be drawn at this point for re usability…

[AncientU]

Some have concluded that it never was a technical challenge, just an economic one.

Since tech isn't the issue, maybe a 48 hour turn-around is -- or maybe that is not a technical problem either...

[Coastal Ron]

Lessons learned from previous X-planes contributed to the database that the X-15 designers drew from. Falcon’s designers are at the pointy end of the spear and are writing the reusable booster books as they proceed through the design evolution.

Falcon 9 v1.1 also drew from lessons learned from 3 years of v1.0 flights, 2 years of v1.1 non-FT flights, as well as 5 years of manufacturing knowledge about what they were capable of.  As the saying goes, an overnight success years in the making...

No conclusion can be drawn at this point for re usability…

Yep.  Recoverability though, is a different matter...

[Rocket Science]

I agree Ron with most of what you wrote but to just maintain perspective, all those prior flight mentioned also led to a LOV just 7 months ago and even more information for the books they are writing. This last flight with upgrades I view as a reset for them in terms of procedures and processes.
I am appreciative that an organization like SpaceX is willing "to at least try" to quote Elon. I also remember him saying a couple of years back if he doesn't achieve re-usability he might just walk away from it all... (I really didn't believe it when he said it at the time, but I guess we'll see)

[meekGee]

I agree Ron with most of what you wrote but to just maintain perspective, all those prior flight mentioned also led to a LOV just 7 months ago and even more information for the books they are writing. This last flight with upgrades I view as a reset for them in terms of procedures and processes.
I am appreciative that an organization like SpaceX is willing "to at least try" to quote Elon. I also remember him saying a couple of years back if he doesn't achieve re-usability he might just walk away from it all... (I really didn't believe it when he said it at the time, but I guess we'll see)

All true about the failures, except all the issues they've had had nothing to do with reusability.  They designed this rocket (3rd iteration now) for RTLS and reusability.  They even designed the 2nd iteration to be such.  This doesn't guarantee that there's no holes in the design, but making modifications to plug these holes (based on study of the returned core) is a much easier task than standing there scratching your head and going "how in hell are we going to make thing this reusable?"

[Jim]

Yep.  Recoverability though, is a different matter...

Without reuse, it is a wasted effort.

[QuantumG]

Yep.  Recoverability though, is a different matter...

Without reuse, it is a wasted effort.

Why? Just being about to inspect engines that have flown is worth a lot.

[Dante80]

I think that Jim is trying to say the following.

1. There is a lot of effort, cost and penalties involved in the current SpaceX architecture for re-use.
2. If there is no economic re-usability, that effort becomes dissimilar to the benefits that recoverability might bring to the table for future iterations of the architecture.
3. Thus, until economic re-usability is proven, the whole architecture around it can be viewed as a money drain vs the alternative (optimizing for a cheap, mass produced, expendable LV with max performance).

[Jim]

Why? Just being about to inspect engines that have flown is worth a lot.

That can be done on a test stand. 

[QuantumG]

Why? Just being about to inspect engines that have flown is worth a lot.

That can be done on a test stand.

You have to recover the engines first.  Also, inspection of just about everything else on the stage after flight is an opportunity that they haven't had before.

[Vultur]

Some have concluded that it never was a technical challenge, just an economic one

Economic combined with politics ... NASA didn't have the political will/funding either to develop the Shuttle properly, or to develop a new better RLV or partial RLV after the Shuttle's inefficiency/expense became clear. Buran was always going to have an expendable booster IIRC, but anyway it got killed by the fall of the Soviet Union, and Russia hasn't tried reusability since.

And no private company that's tried (until just now) has had the funding to get very far.

I'm not sure when the software to do the sort of unmanned VTVL SpaceX & Blue Origin have done became available, but something like a 'flyback first stage' with a pure rocket second stage probably could have been done 40+ years ago.

[nacnud]

Buran was supposed to develop into a fully reusable design, orbiter, boosters and ET. At least that was claimed at the time.

[Jim]

Why? Just being about to inspect engines that have flown is worth a lot.

That can be done on a test stand.

You have to recover the engines first.  Also, inspection of just about everything else on the stage after flight is an opportunity that they haven't had before.

don't even need to fly the engines.  Just run them on the test stand.    That will provide the data needed.

[Donosauro]

I ... remember [Elon] saying a couple of years back if he doesn't achieve re-usability he might just walk away from it all... (I really didn't believe it when he said it at the time, but I guess we'll see)

Yes, but he didn't say, did he, that he would give up if SpaceX's current approach failed to achieve viable reusability? After all, this is at least the second approach they've taken: parachute recovery was tried, unsuccessfully, with Falcon 1. Perhaps SpaceX has one or more approaches, or variations on the current one, in case this one fails.

[joek]

...parachute recovery was tried, unsuccessfully, with Falcon 1...

And Falcon 9.

[Robotbeat]

How are there STILL people who think that it might be impossible for a Falcon 9 booster to be reused (i.e. even after tweaks to the design)?

The "Anti-rocket-reusability Law of the Universe" hypothesis. I just don't get this.

[meekGee]

It's been pointed often that in order to understand what happens to a rocket in flight, you need to fly it.

Not just the engines.  Take for example, oh, I don't know, the struts that hold the He bottles to the tank.

Wouldn't it be nice to inspect their ball ends after flight to look for almost-failures?   Might be useful in avoiding a future failure!

(Of course, Dr. Evil, that too, has already happened)

Point is, a returned stage automatically allows them to increase the reliability of the design.

How does any of the expendable rocket vendors know if there are near-failures in their rockets?

[QuantumG]

don't even need to fly the engines.  Just run them on the test stand.    That will provide the data needed.

Flight stresses can't be tested on the ground.

[llanitedave]

Yes, the 3 minutes of flight and the instrumentation from the many flights prevented the helium bottle from not breaking free.  Also, flight environments is not the same as vehicle loads.

The accident provided a benefit by pointing out how little data they have on the vehicle.

Why? Just being about to inspect engines that have flown is worth a lot.

That can be done on a test stand.

These two posts look very contradictory to me.

[QuantumG]

These two posts look very contradictory to me.

It's almost like he got me to say what he wanted to say and then stopped talking. It's a great strategy.

[gregpet]

Could SpaceX remove redundancy as well as redesign over-engineered parts as more stages are returned and studied?

Could you eventually take significant weight off of the Falcon given reusability?

[mikelepage]

Seems like time to recommend everyone watches this talk again (Jeff Greason 2014 ISPCS Keynote):


As far as SpaceX goes, I think it makes sense that there will have to be a separate pricing structure for reused boosters because the risk is qualitatively different and thus far unknown.  I'm assuming every returned booster will have at least one qualification test (static fire without payload), and then a pre-flight static fire on the pad before it returns to flight service.

Then if they were being systematic about it, I'd imagine the 2nd returned booster would be reused once, and then taken apart and inspected, the 3rd returned booster reused twice, and then taken apart and inspected, and so on, until they have empirical data for how each component degrades over time with n flights.  Upgrade components accordingly for F9 v1.3.  Once they have a handle on that, they could start reusing them to destruction (with non-critical payloads) so they can gauge the max number of flights for which any booster can reasonably be used.

Only after they've done all that do I think they can switch to a commoditised service model where the fee you pay is for getting a payload to orbit, without distinguishing between new and used stages.  In the meanwhile, the only downside risk is that (with non-critical payloads), they'll discover some new phenomena that causes reused stages to blow up much more frequently than they were expecting.

Jeff Greason's so-called "valley of death" is being crossed

[macpacheco]

Could SpaceX remove redundancy as well as redesign over-engineered parts as more stages are returned and studied?

Could you eventually take significant weight off of the Falcon given reusability?

Its more likely SpaceX will add redundancy structural/thermals/wear margins to allow for more reuses.

[Rocket Science]

How are there STILL people who think that it might be impossible for a Falcon 9 booster to be reused (i.e. even after tweaks to the design)?

The "Anti-rocket-reusability Law of the Universe" hypothesis. I just don't get this.

I can only speak for myself Chris, but first he has to actually recover, re-fly it on multiple occasions and then demonstrate the economics of doing so. My opinion of Elon has always been is “don’t tell me, show me…” and when he does I will always commend he for doing so. Even more, is that he "is not" doing it for the most part at the taxpayer expense as with other X programs. He has earned "a well done and carry on" in my book.

[guckyfan]

Could SpaceX remove redundancy as well as redesign over-engineered parts as more stages are returned and studied?

Could you eventually take significant weight off of the Falcon given reusability?

Its more likely SpaceX will add redundancy structural/thermals/wear margins to allow for more reuses.

You are making a big assumption here. The assumption that they have not added margines to cover those contingencies. I assume they did add margins and may find they may need to increase those margins on some points and may be able to shed margins on other points. I make this assumption because they were building the stage for reuse. After all we do know that the second stage shed margin after they decided not to reuse it.

[woods170]

Why? Just being about to inspect engines that have flown is worth a lot.

That can be done on a test stand.

You have to recover the engines first.  Also, inspection of just about everything else on the stage after flight is an opportunity that they haven't had before.

don't even need to fly the engines.  Just run them on the test stand.    That will provide the data needed.

I beg to differ. There are quite a few variations in conditions that an engine does not see on a teststand. First flight of Ariane 5 ECA  is just one example that can testify to that.

[JasonAW3]

Why? Just being about to inspect engines that have flown is worth a lot.

That can be done on a test stand.

You have to recover the engines first.  Also, inspection of just about everything else on the stage after flight is an opportunity that they haven't had before.

don't even need to fly the engines.  Just run them on the test stand.    That will provide the data needed.

I beg to differ. There are quite a few variations in conditions that an engine does not see on a teststand. First flight of Ariane 5 ECA  is just one example that can testify to that.

Look, Engineering is a science, but that doesn't mean that there isn't an element of both art and luck to it.

No matter how well you plan for a situation in spaceflight, sooner or later something is going to happen that was both inobvious and sneaky that will totally ruin your day.  The Falcon 9 explosion from last year for an example.

From the visual info that I got from the pictures of how the Helium tank was set up in the LOX tank, it looks like there was plenty of redundancy that should have allowed the loss of a couple of struts on the flight.  This was obviously not the case, in retrospect.  This has since been corrected.

Whether or not the first stage can handle multiple launches is yet to be seen.  Like I said in another thread.  We're effectively in the Barnstorming stage of developement of reusable launch vehicles.  The Shuttle was the first, mostly reusable craft.  The Falcon 9 is the next.

[Jim]

I beg to differ. There are quite a few variations in conditions that an engine does not see on a teststand. First flight of Ariane 5 ECA  is just one example that can testify to that.

We are talking about engine reusability.  All that can be learned from the test stand.

[meekGee]

Why? Just being about to inspect engines that have flown is worth a lot.

That can be done on a test stand.

You have to recover the engines first.  Also, inspection of just about everything else on the stage after flight is an opportunity that they haven't had before.

don't even need to fly the engines.  Just run them on the test stand.    That will provide the data needed.

I beg to differ. There are quite a few variations in conditions that an engine does not see on a teststand. First flight of Ariane 5 ECA  is just one example that can testify to that.

Look, Engineering is a science, but that doesn't mean that there isn't an element of both art and luck to it.

No matter how well you plan for a situation in spaceflight, sooner or later something is going to happen that was both inobvious and sneaky that will totally ruin your day.  The Falcon 9 explosion from last year for an example.

From the visual info that I got from the pictures of how the Helium tank was set up in the LOX tank, it looks like there was plenty of redundancy that should have allowed the loss of a couple of struts on the flight.  This was obviously not the case, in retrospect.  This has since been corrected.

Whether or not the first stage can handle multiple launches is yet to be seen.  Like I said in another thread.  We're effectively in the Barnstorming stage of developement of reusable launch vehicles.  The Shuttle was the first, mostly reusable craft.  The Falcon 9 is the next.

When people say "can handle reflight", it does not mean there are no issues that need to be addressed after examining this returned stage.

This stage ("1.2") is the best they could do from a design standpoint in order to make it reusable, without actually having a returned core.

It might be good enough. It might need some tweaks.  It most certainly has  extra margins (by the very definition of "margin")

My bet is that the tweaks are minor.

Between what they've learned about propulsive reentry and the about this stage though, they're in a unique position to get to rapid reusability, and given their business model, they can do it very quickly now.

[Jim]

Flight stresses can't be tested on the ground.

What "flight" stresses" wrt the engines?  The engines produced stresses greatly out weigh any produced by flight.  We are not talking about the airframe or the rest of the vehicle.  Putting a engine in a test stand and running it multiple times and for longer durations is going to provide more data on the robustness of the engine than from a few minutes on a few flights.

[Jim]

Yes, the 3 minutes of flight and the instrumentation from the many flights prevented the helium bottle from not breaking free.  Also, flight environments is not the same as vehicle loads.

The accident provided a benefit by pointing out how little data they have on the vehicle.

Why? Just being about to inspect engines that have flown is worth a lot.

That can be done on a test stand.

These two posts look very contradictory to me.

No, they aren't.
a.  I was only talking about the engines and just having the stage return just for inspection (not reuse). There is a difference between engine and airframes.   Engines can be operated on a test stand.  A launch vehicle can't.   
b.  The mods to the vehicle and flight regime of the boost back, entry and landing negate any benefits of booster return for just inspection (and not reuse).

[Senex]

I originally posted this over on the "Refurbishment" thread, but it would seem to be at least as relevant here.  And it may serve to get discussion off of airspace restrictions and back on the subject of processing boosters for re-flight.


While speculation is popular and is often justified by the absence of facts, there ARE some facts available. An earlier reference to the X-15 is highly relevant as it endured a very similar flight regime in terms of stresses (arguably greater). 

In an interesting article, space historian David Portree cites a study that looked at the refurbishment costs of the X-15 program that provides probably the most relevant real-world data we have:


"In November 1966, James Love and William Young, engineers at the NASA Flight Research Center at Edwards Air Force Base, completed a brief report in which they noted that the reusable suborbital booster for a reusable orbital spacecraft would undergo pressures, heating rates, and accelerations very similar to those the X-15 experienced."

"The average X-15 refurbishment time was 30 days, a period which had, they noted, hardly changed in four years. Even with identifiable improvements, they doubted that an X-15 could be refurbished in fewer than 20 days.

"At the same time, Love and Young argued that the X-15 program had demonstrated the benefits of reusability. They estimated that refurbishing an X-15 in 1964 had cost about $270,000 per mission.

"Love and Young cited North American Aviation estimates when they placed the cost of a new X-15 at about $9 million. They then calculated that 27 missions using expendable X-15s would have cost a total of $243 million. This meant, they wrote, that the cost of the reusable X-15 program in 1964 had amounted to just 3% of the cost of building 27 X-15s and throwing each one away after a single flight.

My bold.

http://www.wired.com/2013/05/the-x-15-rocket-plane-reusable-space-shuttle-boosters-1966/

At least two implications can be deduced from this reference:

1. There has been much talk on the forums about the effects of "fatigue" with images of much of the booster requiring rework or even replacement.  Three X-15's flew 199 flights — and average of 66 each.  Just because a Falcon 9 flies high and fast does not mean it will be structurally degraded after a few cycles.

2. The X-15 required weeks of refurbishment.  It is safe to assume that most of that related to mechanical equipment.  That was with technology that had literally just been invented.  In a vehicle designed from the beginning with operating economies in mind, using a relatively mature technology, this might be dramatically lower.

Wrong analogy.  X-15 was an aircraft and designed like one.  Aircraft are reusable.  Design standards have yet to be developed for reusable boosters for fatigues mitigation.

It's not just an analogy — it's an existence proof.  I posted the above in response to the continuing attitude on the part of some people that reusability is debatable/unlikely/impossible. 

Knowing far less than we do today, North American built a cranky, "Rube Goldberg" of a rocket-plane.  They had very little data to base it on.  It flew in a similar environment and flight regime as a first stage booster.  In some ways what they did was harder — it had to carry a pilot.  And they managed to fly 3 vehicles a total of 199 times. 

"Aircraft are reusable." because we have made them reusable!  If we hadn't gone through a generation when civil rockets were based on ICBM's (that you didn't WANT to come back!) we might have taken a different path and been in a different position today.

NOT reusing is untenable.  One day, in the not-too-distant future, we will look back and shake our heads at the idiocy of "single use rockets."

Some days I feel like I am at the 1905 Buggy Whip Convention . . .  "There's no way those auto mobiles is gonna catch on."

[Jim]

NOT reusing is untenable.  One day, in the not-too-distant future, we will look back and shake our heads at the idiocy of "single use rockets."

The issue isn't reuse.  The shuttle did it and so does X-37.  It is issue is cost effectiveness.

[LouScheffer]

I beg to differ. There are quite a few variations in conditions that an engine does not see on a teststand. First flight of Ariane 5 ECA  is just one example that can testify to that.

We are talking about engine reusability.  All that can be learned from the test stand.

A lot, yes, but not all.  Apollo 6, for example, had a vibration problem that only occurred in vacuum (since on the test stand condensation on the cold bellows provided adequate damping).   A similar problem could easily result in a fatigue lifetime that is much less in flight than it is on the test stand.

[Senex]

NOT reusing is untenable.  One day, in the not-too-distant future, we will look back and shake our heads at the idiocy of "single use rockets."

The issue isn't reuse.  The shuttle did it and so does X-37.  It is issue is cost effectiveness.

Agreed.  And I still stand by what I said.  The economics of reuse will overwhelm "disposable." 

[Jim]

It's not just an analogy — it's an existence proof.  I posted the above in response to the continuing attitude on the part of some people that reusability is debatable/unlikely/impossible. 

Knowing far less than we do today, North American built a cranky, "Rube Goldberg" of a rocket-plane.  They had very little data to base it on.  It flew in a similar environment and flight regime as a first stage booster.  In some ways what they did was harder — it had to carry a pilot.  And they managed to fly 3 vehicles a total of 199 times. 

wrong.  They had previous experience and data to base the design.  There were earlier X-planes.  They had a good grasp of the loads and environments involved.  Most of the vehicle issues were in the control systems and not the airframe.

[Jim]

The economics of reuse will overwhelm "disposable." 

Based on what data?

[Jim]

A lot, yes, but not all.  Apollo 6, for example, had a vibration problem that only occurred in vacuum (since on the test stand condensation on the cold bellows provided adequate damping).   A similar problem could easily result in a fatigue lifetime that is much less in flight than it is on the test stand.

That is an integration issue.  And something that would be not be found by returning the vehicle.

[Senex]

Wrong?  Is it wrong to say that they knew less than we know today?  60 years less?

Disposable works for kleenex vs hankies.  How can anyone defend throwing away a $60 m plus precision machine (fill in 747 analogy).

There will be issues.  Yes.  The Comet jet airliner's window openings cracked.  And they solved it and moved on.

Are you saying the problems are going to be economically unsolvable?

[LouScheffer]

A lot, yes, but not all.  Apollo 6, for example, had a vibration problem that only occurred in vacuum (since on the test stand condensation on the cold bellows provided adequate damping).   A similar problem could easily result in a fatigue lifetime that is much less in flight than it is on the test stand.

That is an integration issue.  And something that would be not be found by returning the vehicle.

Nope, it was an engine problem.  See http://heroicrelics.org/info/j-2/augmented-spark-igniter.html for details.  And you could certainly find problems like this by looking for parts on the returned engine that are stretched, bent, or otherwise deformed in ways you have not seen on the test stand.

[abaddon]

Could SpaceX remove redundancy as well as redesign over-engineered parts as more stages are returned and studied?

Could you eventually take significant weight off of the Falcon given reusability?

Its more likely SpaceX will add redundancy structural/thermals/wear margins to allow for more reuses.

It is most likely that SpaceX will do both, in different areas.

[abaddon]

I believe Jim is saying that he expects the Merlin 1D's to be fully reusable without refurbishment.  Because all that is needed is test stand testing and SpaceX has clearly done plenty of that.

[Jim]

Nope, it was an engine problem.  See http://heroicrelics.org/info/j-2/augmented-spark-igniter.html for details.  And you could certainly find problems like this by looking for parts on the returned engine that are stretched, bent, or otherwise deformed in ways you have not seen on the test stand.

Not really, would be hard to discern whether the issue came from the launch or the return.

[sublimemarsupial]

Nope, it was an engine problem.  See http://heroicrelics.org/info/j-2/augmented-spark-igniter.html for details.  And you could certainly find problems like this by looking for parts on the returned engine that are stretched, bent, or otherwise deformed in ways you have not seen on the test stand.

Not really, would be hard to discern whether the issue came from the launch or the return.

Who cares whether it came from the launch or the return? It would be a problem that could be readily identified after flight via inspection and fixed, making the reusable vehicle more reliable long term

[Jim]

Who cares whether it came from the launch or the return? It would be a problem that could be readily identified after flight via inspection and fixed, making the reusable vehicle more reliable long term

All this is in the context whether returning the booster is useful without the intent of reuse.

[sublimemarsupial]

Who cares whether it came from the launch or the return? It would be a problem that could be readily identified after flight via inspection and fixed, making the reusable vehicle more reliable long term

All this is in the context whether returning the booster is useful without the intent of reuse.

Ok change reusable for returnable, vehicle reliability is still imroved if there is a chance the defect was from ascent - which for this problem it was.

[RDMM2081]

WRT to engine re-use, aren't we missing the fact that the engines on an F9 do experience something that is not able to be replicated on the test stand?  Specifically, the supersonic retro burn.  What effect does engine startup and ignition in a supersonic environment have on engine components?  Well, now there are engines back on land which have experienced this and the engineers can check them out and see!

[Senex]

The economics of reuse will overwhelm "disposable." 

Based on what data?

You want data.  I respect that.  That was the purpose of my original post — to identify another data point.

Look at the data points we have regarding reusable boosters:

• The Shuttle, both the Orbiter and SRB's
• DC-X
• Masten, Armadillo, etc.
• XCor's rocket planes
• You mention the X-37
• Falcon 9, Blue Origin (They're not talking.  That's why we are having this discussion.)

DC-X, Masten, Armadillo, the rocket planes . . . different flight regime.  The X-37 has more in common with a reusable Dragon capsule than a booster.

Of these, only the Shuttle components operated in a relevant flight regime.

And the X-15:

"In November 1966, James Love and William Young, engineers at the NASA Flight Research Center at Edwards Air Force Base, completed a brief report in which they noted that the reusable suborbital booster for a reusable orbital spacecraft would undergo pressures, heating rates, and accelerations very similar to those the X-15 experienced."

      - from:  Survey of Operation and Cost Experience of the X-15 Airplane as a Reusable Space Vehicle, NASA Technical Note D-3732, James Love and William Young, November 1966.

And they concluded that re-use was economically very practical.

Again, I encourage people to read David Portree's article in Wired referencing the study:

http://www.wired.com/2013/05/the-x-15-rocket-plane-reusable-space-shuttle-boosters-1966/

[LouScheffer]

Nope, it was an engine problem.  See http://heroicrelics.org/info/j-2/augmented-spark-igniter.html for details.  And you could certainly find problems like this by looking for parts on the returned engine that are stretched, bent, or otherwise deformed in ways you have not seen on the test stand.

Not really, would be hard to discern whether the issue came from the launch or the return.

Who cares whether it came from the launch or the return? It would be a problem that could be readily identified after flight via inspection and fixed, making the reusable vehicle more reliable long term

This was not Jim's point.  Others were claiming that inspecting a returned vehicle could lead to better quality engines, even if the vehicle is never re-used.  Jim was pointing out that in that scenario, you don't care about margins on the way down, just on the way up, and it might be hard to tell where the margin excursion happened.

In practice, if a margin is found to be exceeded, but not to failure, it would be subject to additional analysis to see if it happened on the way up, the way down, or both.  There is at least some chance it happened on the way up, and hence it's possible that examining landed engines could find flaws that are relevant to even expendable use.  How common this is remains to be seen.

[JamesH]

So, according to most, even Jim, the engines are reusable without major refurb, because they were designed that way, and have been thoroughly tested on the stand.  Cool. Something we can leave out of the refurb costs (except for expected refurb costs already identified, expected to be minimal)

That leaves the airframe, avionics and plumbing.

That's good, isn't it? Those are the 'less' complicated parts of the craft, can be relatively easily examined, and are also the least expensive to replace. Electronics can self test (to get most possible problems), tanks can be pressure tested, legs can be checked, grid fins can be examined.  Sounds good to me.

Even if the whole airframe is a total loss (seems unlikely, as it's designed to be reusable), there are still those 9 engines that ARE reusable (because Jim said so).

[Jim]

The engines still have to be inspected since they now have experienced a reentry, supersonic retro burn and landing.

[Lars-J]

The engines still have to be inspected since they now have experienced a reentry, supersonic retro burn and landing.

And here I thought I just read from your elsewhere that this stage provided no new information about the engines, and that all testing could be done on the ground. 

[chipguy]

Falcon 9 v1.1 FT is a reusable rocket and designed like one.

no, it is an expendable rocket that is being converted/upgraded to a reusable one.

It is a reusable rocket initially deployed to service an expendable business model while
the kinks in the reusable regime of operation are engineered out.

By your logic could ULA "convert/upgrade" Atlas V to a reusable rocket?

[Herb Schaltegger]

The engines still have to be inspected since they now have experienced a reentry, supersonic retro burn and landing.

And here I thought I just read from your elsewhere that this stage provided no new information about the engines, and that all testing could be done on the ground. 

In the absence of plans to re-fly them operationally, he's right.

[Jim]

It is a reusable rocket initially deployed to service an expendable business model while
the kinks in the reusable regime of operation are engineered out.

By your logic could ULA "convert/upgrade" Atlas V to a reusable rocket?

No, the first version of Falcon 9 was expendable and not reusable.

[sunbingfa]

Reusability aside, the short turnover time may be attractive to military customer, who has been pushing the short notice launch with little success. Unless XS-1 has huge success within the next 3-5 years.

[nadreck]

Falcon 9 v1.1 FT is a reusable rocket and designed like one.

no, it is an expendable rocket that is being converted/upgraded to a reusable one.

No, with all due respect the whole reason why (see the thread: http://forum.nasaspaceflight.com/index.php?topic=37390.0) there is a controversy between the statement from Elon that one reflight of a Falcon Core justifies its reusability vs George Sowers saying you need 20 in the ULA universe is because the F9 was designed to be reusable.

If I went to Ford and asked them how much cheaper they could make an expendable car than say a Ford Focus but with similar performance, after they stopped laughing, they could probably settle down to engineer something but the design and manufacturing setup costs would probably dictate that you needed a market of millions of units before you could get the price much lower than half that of a Ford Focus and it would still be at least as reusable as 3rd party refilled print cartridges.

Reuse was being designed in (it didn't work for the whole system but components were being designed for it) in the Falcon 1 even. So what is the hard to believe part is that SpaceX did all this designing in for something that still has a unit cost to manufacture below the same class of launchers that never had reuse considered for any part.  Even more amazing is that it took them a tiny fraction of the development costs (adjusted for today's dollar) to do the same thing.

The sad thing is that ULA has convinced itself that taking all their existing engineering and adding reuse to it is the same process as SpaceX starting from scratch and designing systems for reuse.  ULA keeps carrying its own 'expendable experience' baggage forward with the Vulcan and worse seems to presume that SpaceX has the same baggage which is not the case.

[Robotbeat]

This whole discussion reminds me of something I read on Twitter.

There are two reasons something new (supposedly) can't work:
1) It's been done before.
2) it hasn't been done before.

[Dante80]

The economics of reuse will overwhelm "disposable." 

Based on what data?

You don't need data, just common sense.

It could happen in 5 years, or 20, or even a century. It is going to happen though (barring a breakthrough in exotic space transport methods or a thermonuclear war).

[Jim]

Falcon 9 v1.1 FT is a reusable rocket and designed like one.

no, it is an expendable rocket that is being converted/upgraded to a reusable one.

No, with all due respect the whole reason why (see the thread: http://forum.nasaspaceflight.com/index.php?topic=37390.0) there is a controversy between the statement from Elon that one reflight of a Falcon Core justifies its reusability vs George Sowers saying you need 20 in the ULA universe is because the F9 was designed to be reusable.

snip

Reuse was being designed in (it didn't work for the whole system but components were being designed for it) in the Falcon 1 even. So what is the hard to believe part is that SpaceX did all this designing in for something that still has a unit cost to manufacture below the same class of launchers that never had reuse considered for any part.  Even more amazing is that it took them a tiny fraction of the development costs (adjusted for today's dollar) to do the same thing.

Wrong.  Just adding a parachute does not qualify as designing for reuse.

Other than tank diameter and domes, there is little hardware that the early F9 first stages share with the current ones.  Thrust structure, avionics, propulsion system, pressure systems, are all different.  Even launcher.

[schaban]

Wrong.  Just adding a parachute does not qualify as designing for reuse.

Other than tank diameter and domes, there is little hardware that the early F9 first stages share with the current ones.  Thrust structure, avionics, propulsion system, pressure systems, are all different.  Even launcher.

I think engines and perhaps other parts were designed to withstand salt water, no?

[abaddon]

Jim, the original quote you are responding to was "Falcon 9 v1.1 FT is a reusable rocket and designed like one."  Why are we discussing the original F9?

[Jim]

You don't need data, just common sense.

Not true.  Reusable does not always mean cheaper.

[nadreck]

Falcon 9 v1.1 FT is a reusable rocket and designed like one.

no, it is an expendable rocket that is being converted/upgraded to a reusable one.

No, with all due respect the whole reason why (see the thread: http://forum.nasaspaceflight.com/index.php?topic=37390.0) there is a controversy between the statement from Elon that one reflight of a Falcon Core justifies its reusability vs George Sowers saying you need 20 in the ULA universe is because the F9 was designed to be reusable.

snip

Reuse was being designed in (it didn't work for the whole system but components were being designed for it) in the Falcon 1 even. So what is the hard to believe part is that SpaceX did all this designing in for something that still has a unit cost to manufacture below the same class of launchers that never had reuse considered for any part.  Even more amazing is that it took them a tiny fraction of the development costs (adjusted for today's dollar) to do the same thing.

Wrong.  Just adding a parachute does not qualify as designing for reuse.

Other than tank diameter and domes, there is little hardware that the early F9 first stages share with the current ones.  Thrust structure, avionics, propulsion system, pressure systems, are all different.  Even launcher.

The parachute was the wasted part of the design for reuse. I don't know about the kestrel but certainly the Merlin was designed with reuse in mind as was the majority of the vehicle. We will probably never know how much of that was clearly a waste of effort like the parachute that was replaced with a design with more potential for reuse. But from day one SpaceX was designing everything that they felt they could design with reuse in mind. On the Falcon 9 they have since given up on that for the 2nd stage but that doesn't mean that many of the systems in it weren't still designed with reuse in mind.

[rcoppola]

I think it's a good idea to revisit the 2 posts that initiated this thread. There are dozens of threads concerning reuse, etc..

The 48 hour rule, while it certainly has implications for future returned cores, is not predicated on such. Neither is the one hour pad processing. 

I find the technical aspects of how you could essentially create a streamlined, low personnel, automated, quick processing and launch flow, extremely interesting. I'd much rather here form people like Jim and others who actually have intimate knowledge in this area on how that could be realistically achieved, rather then re-use, good/bad, cheap, not, always planned, only recently developed, etc..rehash.

[nadreck]

I think it's a good idea to revisit the 2 posts that initiated this thread. There are dozens of threads concerning reuse, etc..

The 48 hour rule, while it certainly has implications for future returned cores, is not predicated on such. Neither is the one hour pad processing. 

I find the technical aspects of how you could essentially create a streamlined, low personnel, automated, quick processing and launch flow, extremely interesting. I'd much rather here form people like Jim and others who actually have intimate knowledge in this area on how that could be realistically achieved, rather then re-use, good/bad, cheap, not, always planned, only recently developed, etc..rehash.

I don't see anything from Jim on this thread about how to realistically achieve a streamlined, low personnel, automated, quick processing and launch flow.

I do see a lot of statements from him about reuse and design intent, and given my nature and the nature of how I participate on this board, I can't help but feel an urge to correct something I see as misrepresenting the design and design intent of what we are discussing.

[mvpel]

After watching the video of the semi-tractor scooching around trying to reverse direction in order to back the stage into the hangar at 39A, I expect that for the 48-hour target they'll be designing a booster cradle that can be towed from either end, or at least making sure to mount it on the cradle engines-first at the landing pad next time.

[JasonAW3]

You don't need data, just common sense.

Not true.  Reusable does not always mean cheaper.

Short term, you're probably correct.  The technology to produce reliable, reusable rockets is currently in the developmental phase.  While the Shuttle was supposed to be a faster turn around and highly reliable, the refurbishing costs far outweighed any potentile benifit from reusability.

  The Shuttle 2 was supposed to incorporate lessons learned from the original STS system and develop a whole new SSTO type design.  It got canceled not just due to the Challenger disaster, but because the Technology just wasn't quite there yet. (Cost was also a factor, but not quuite as prominant as most seem to think).

     The Falcon 9 is trying to incorporate a number of lessons learned from the DC-X and the Shuttle to produce a reusable launch system.

     Yes, the Tech isn't quite there yet, to make this a fully reusable system, but that's what technological development is for.  Each launch that SpaceX makes gains them more information.  Each stage recovered will further their knowledge base and provide them with the knowledge to be able to produce a fully reusable system.

     Yes, they'll still have to expend such things as payload shrouds and interstage segments and Dragon Trunks, but it will still be FAR cheaper than expending entire rockets, again, once the technology is more mature.

     And Jim, saying that it can't be done is much akin to saying that going to the moon can't be done, or that heavier than aircraft are impossible.  Yes, it will ALWAYS be more complex than fliyoing an aircraft, (Although, I'd love to see myself proven wrong on this) but I have little doubt that  it will happen, and maybe even in the next few years.

     Heck, just a few years ago, no one thought recovering the booster stage from either a orbital or suborbital flight was even possible.  Yet, SpaceX and Blue Origin, respectively, have done so.

     So, I'm pretty much of the opinion that ANYTHING is possible, it just takes time, knowledge and ingenuity to make it happen.  (Yeah, and a lot of money as well).

[Jim]

I do see a lot of statements from him about reuse and design intent, and given my nature and the nature of how I participate on this board, I can't help but feel an urge to correct something I see as misrepresenting the design and design intent of what we are discussing.

And you are the one misrepresenting the original design of the Falcon 9.
Did the original Falcon 9 have attachments for legs, guidance for first stage, first stage inflight restart capability, landing sensors, robust thrust structure, etc

[JasonAW3]

I do see a lot of statements from him about reuse and design intent, and given my nature and the nature of how I participate on this board, I can't help but feel an urge to correct something I see as misrepresenting the design and design intent of what we are discussing.

And you are the one misrepresenting the original design of the Falcon 9.
Did the original Falcon 9 have attachments for legs, guidance for first stage, first stage inflight restart capability, landing sensors, robust thrust structure, etc

Jim,  the Falcon 9 was developed to make sure that SpaceX could launch payloads into space first and formost, with making sufficent money to continue to develop the system.  The incremental addition of components was, again, to see ifbeing able to land a booster was possible.  You'll note that very little of the original design of the Falcon 9 version 1, visually, actually remains in the current version, except the actual diameter of the stage.

     It's likely that it will change even further as improvements are incorporated into the design.

     With the Falcon Heavy, I'm not sure that they'll be able to recover the center booster, although I suspect the strap-ons will likely make it home.  If all three DO make a safe landing, that will be another major milestone to full reusability, as the center booster will be coming from both a higher altitude and velocity than either of the strap-ons, or even any Falcon 9 first stage yet.

     With respect, Jim, you ARE right that the Falcon 9 V1.1 FT is a VERY different bird than the Falcon 9 V1, but that's as it should be.

      Could you imagine how different things would be if they'd have developed the S-IC had have been able to boost back and land?  Bringing back the S-II might have been do-able as well, but we'll never really know.

     But I figure, if the major aerospace firms in the 60's and 70's thought it was possible to make reusable launchers, then just maybe they were right?

[nadreck]

I do see a lot of statements from him about reuse and design intent, and given my nature and the nature of how I participate on this board, I can't help but feel an urge to correct something I see as misrepresenting the design and design intent of what we are discussing.

And you are the one misrepresenting the original design of the Falcon 9.
Did the original Falcon 9 have attachments for legs, guidance for first stage, first stage inflight restart capability, landing sensors, robust thrust structure, etc

It had provisions for some of those, it tested some of those, and as a result of all the work on the 1.0, the 1.1 which was closer to recoverable was able to test supersonic retropulsion on its first flight. I would argue that testing restarts on the 2nd stage (and at McGregor on 1st stage engines) during the V1.0 era was part of that. If the F9 was designed to be expendable, you would not have had the V1.1 able to do all it did without a lot larger a design effort.

The first grasshopper was a V1.0 (it first flew more than a year before the first V1.1 launch which demonstrated guidance of a sort on a V1.0 frame) so legs (not the ones used for V1.1 mind you, but legs all the same, and that is better than a discarded tech like parachutes right?) were tested.

As early as the national press club talk on November 29 2011:

So, the pivotal breakthrough that's necessary, that some company has got to come up with, to make life multi-planetary is a fully and rapidly reusable orbit class rocket. This is a very difficult thing to do because we live on a planet where that is just barely possible. If gravity were a little lower it'd be easier, but if it was a little higher it would be impossible. Even for an expendable launch vehicle, where you don't have to have any recovery, after a lot of smart people have done their best to optimize the weight of the vehicle and efficiency of the engines and the guidance systems and everything, you get maybe 2 to 3% of your liftoff weight to orbit. That's not a lot of room for error. If your rocket ends up being just a little bit heavier, you get nothing to orbit, and this is why only a few countries have ever reached orbit.

Now you say, okay, let's make it reusable, which means you've got to strengthen stages, you've got to add a lot of weight, a lot of thermal protection, you've got to do a lot of things that add weight to that vehicle, and still have a useful payload to orbit. Of that meager 2 to 3%, maybe if you're really good you can get it to 4%, you've got to add all that's necessary to bring the rocket stages back to the launch pad and be able to refly them, and still have useful payload to orbit. It's a very difficult thing. This has been attempted many times in the past, and generally what's happened is when people concluded that success was not one of the possible outcomes, the project's been abandoned. Well, some government projects kept going, even when success was not one of the possible outcomes, unfortunately, but then eventually they get cancelled. So it's just a very tough engineering problem.

It wasn't something that I thought - I wasn't sure it could be solved for a while, but then, just relatively recently - in the last 12 months or so - I've come to the conclusion that it can be solved, and SpaceX is going to try to do it. Now, we could fail. I'm not saying we're certain of success here, but we're going to try to do it, and we have a design that, on paper, doing the calculations, do the simulations, it does work. Now, we have to make sure those simulations and reality agree, because generally when they don't, reality wins. That's yet to be determined, and the simulation that you may have seen in the lobby coming in, which will be posted to our website right around now, will show you a simulation of what we plan to do.

http://shitelonsays.com/transcript/npc-luncheon-with-elon-musk-2011-09-29

Yes it was being designed in from the start

[Dave G]

And you are the one misrepresenting the original design of the Falcon 9.
Did the original Falcon 9 have attachments for legs, guidance for first stage, first stage inflight restart capability, landing sensors, robust thrust structure, etc

The original intent was to recover the Falcon 9 first stage using parachutes, just as they tried to do with Falcon 1.  When that didn't work, they added the stuff you mention above.  When that didn't work, they added grid fins. In other words, they just kept working at it until they were successful.  They never gave up.

Will refurbishing the stage be cost effective?  I think that question misses the main point.  If it isn't cost effective initially, SpaceX will keep working at it, constantly tweaking their design and processes until it does become cost effective. And they won't stop there. They'll continue to tweak things, constantly optimizing for cost and reliability.  This is the basis on which the company was formed, and they won't give up until they get there.

[Jim]

I do see a lot of statements from him about reuse and design intent, and given my nature and the nature of how I participate on this board, I can't help but feel an urge to correct something I see as misrepresenting the design and design intent of what we are discussing.

And you are the one misrepresenting the original design of the Falcon 9.
Did the original Falcon 9 have attachments for legs, guidance for first stage, first stage inflight restart capability, landing sensors, robust thrust structure, etc

It had provisions for some of those

It had no provisions for none of those.  That is why there was a V1.1.   The first F9 version was for COTS and to make money to support the rest of the development.  F9 came from the F5 to make money.

[nadreck]

I do see a lot of statements from him about reuse and design intent, and given my nature and the nature of how I participate on this board, I can't help but feel an urge to correct something I see as misrepresenting the design and design intent of what we are discussing.

And you are the one misrepresenting the original design of the Falcon 9.
Did the original Falcon 9 have attachments for legs, guidance for first stage, first stage inflight restart capability, landing sensors, robust thrust structure, etc

It had provisions for some of those

It had no provisions for none of those.  That is why there was a V1.1.   The first F9 version was for COTS and to make money to support the rest of the development.  F9 came from the F5 to make money.

So the structure of Grasshopper was different, it had no guidanace, no legs?

Engines meant for an F9 1.0 first stage only fired once?

The 2nd stage never restarted an engine in V1.0?

[Jim]

I would argue that testing restarts on the 2nd stage (and at McGregor on 1st stage engines) during the V1.0 era was part of that. If the F9 was designed to be expendable, you would not have had the V1.1 able to do all it did without a lot larger a design effort.

That would be a wrong  Restarts on the second stage are needed for the basic GSO mission.
Grasshopper was not a basic F9, it was highly modified (battleship).  that could be done to any rocket with a deep throttle engine.

[nadreck]

I would argue that testing restarts on the 2nd stage (and at McGregor on 1st stage engines) during the V1.0 era was part of that. If the F9 was designed to be expendable, you would not have had the V1.1 able to do all it did without a lot larger a design effort.

That would be a wrong  Restarts on the second stage are needed for the basic GSO mission.
Grasshopper was not a basic F9, it was modified.  that could be done to any rocket with a deep throttle engine.

And for reuse, there were not GSO missions with the 1.0 so they didn't need to test it there but they did anyway.

Grasshopper's shell was not modified, but the engine array was; and legs were added, but if legs could be added that meets what I said that V1.0 had provisions for some of those things and tested some.

[Jim]

1.  So the structure of Grasshopper was different, it had no guidanace, no legs?

2.  Engines meant for an F9 1.0 first stage only fired once?

3.  The 2nd stage never restarted an engine in V1.0?

1.  Highly modified, not stock  Supports my point.
2.  not a relevant point, engines for ELV's can and do fire more than once.  See acceptance testing.
3.  not a relevant point but no.

[Jim]

Grasshopper's shell was not modified, but the engine array was; and legs were added, but if legs could be added that meets what I said that V1.0 had provisions for some of those things and tested some.

Grasshopper had a frame that supported the legs.  The legs were not attached directly to the vehicle as the current one.
http://i.imgur.com/D0FtCP3.jpg
Like I said, you could do that to any existing vehicle.

[nadreck]

1.  So the structure of Grasshopper was different, it had no guidanace, no legs?

2.  Engines meant for an F9 1.0 first stage only fired once?

3.  The 2nd stage never restarted an engine in V1.0?

1.  Highly modified, not stock  Supports my point.
2.  not a relevant point, engines for ELV's can and do fire more than once.  See acceptance testing.
3.  not a relevant point but no.

1. It was still the same tank structure as V1.0 which you claimed up thread had to be upgraded with V1.1 for reuse. Grasshopper flew 8 times using the same body as any of the other 5 V1.0 cores that flew.

2. some can some can't, but the point is that even if the restart 'kit' wasn't added to the 1st stage, the engines were demonstrating the capability in the 1C version and at McGregor they were tested for much longer to test for eventual reuse even though the first recovered engine was in fact two generations later (1D FT, after 1D)

3. Again you assert stuff and it is inaccurate, the 1st COTS flight did a restart of the 2nd stage and I assert this was for testing purposes to prove engine restart capability for both recovery AND for GSO which would only take place with V1.1

[QuantumG]

We are not talking about the airframe or the rest of the vehicle.

I am.

[Kansan52]

The goal from the start, F1 and on, was for reuse. It wasn't until F9.1 they had a vehicle that could actually try to be reusable. And the lesson of reuse hasn't be learned yet since we do not know the practicality of relaunching a F9FT.

[Jim]

1. It was still the same tank structure as V1.0 which you claimed up thread had to be upgraded with V1.1 for reuse. Grasshopper flew 8 times using the same body as any of the other 5 V1.0 cores that flew.

2. some can some can't, but the point is that even if the restart 'kit' wasn't added to the 1st stage, the engines were demonstrating the capability in the 1C version and at McGregor they were tested for much longer to test for eventual reuse even though the first recovered engine was in fact two generations later (1D FT, after 1D)

3. Again you assert stuff and it is inaccurate, the 1st COTS flight did a restart of the 2nd stage and I assert this was for testing purposes to prove engine restart capability for both recovery AND for GSO which would only take place with V1.1

1.  Again, not unique to falcon. any launch vehicle could have flown on the landing legs frame.
2. Wrong all can because they are test fired before delivery.
3.  Look no further than your post for wrong assertions.   There was no restart on COTS 1 because the stage was spinning. All upper stages have restart.  It is a necessary capability.

[Jim]

We are not talking about the airframe or the rest of the vehicle.

I am.

http://forum.nasaspaceflight.com/index.php?topic=39195.msg1470263#msg1470263

[sublimemarsupial]

1. It was still the same tank structure as V1.0 which you claimed up thread had to be upgraded with V1.1 for reuse. Grasshopper flew 8 times using the same body as any of the other 5 V1.0 cores that flew.

2. some can some can't, but the point is that even if the restart 'kit' wasn't added to the 1st stage, the engines were demonstrating the capability in the 1C version and at McGregor they were tested for much longer to test for eventual reuse even though the first recovered engine was in fact two generations later (1D FT, after 1D)

3. Again you assert stuff and it is inaccurate, the 1st COTS flight did a restart of the 2nd stage and I assert this was for testing purposes to prove engine restart capability for both recovery AND for GSO which would only take place with V1.1

1.  Again, not unique to falcon. any launch vehicle could have flown on the landing legs frame.
2. Wrong all can because they are test fired before delivery.
3.  Look no further than your post for wrong assertions.   There was no restart on COTS 1 because the stage was spinning. All upper stages have restart.  It is a necessary capability.

Sounds like you were thinking of F9 flight 1, but even then they did attempt a restart. It failed because the stage was spinning and the prop inlets were uncovered but they did attempt it.

As I recall in 2013 and 2014 you were claiming v1.1 had nothing to do with reuse, and it was purely because 1.0 didn't have enough performance. Funny how stories change.

[nadreck]

1. It was still the same tank structure as V1.0 which you claimed up thread had to be upgraded with V1.1 for reuse. Grasshopper flew 8 times using the same body as any of the other 5 V1.0 cores that flew.

2. some can some can't, but the point is that even if the restart 'kit' wasn't added to the 1st stage, the engines were demonstrating the capability in the 1C version and at McGregor they were tested for much longer to test for eventual reuse even though the first recovered engine was in fact two generations later (1D FT, after 1D)

3. Again you assert stuff and it is inaccurate, the 1st COTS flight did a restart of the 2nd stage and I assert this was for testing purposes to prove engine restart capability for both recovery AND for GSO which would only take place with V1.1

1.  Again, not unique to falcon. any launch vehicle could have flown on the landing legs frame.
2. Wrong all can because they are test fired before delivery.
3.  Look no further than your post for wrong assertions.   There was no restart on COTS 1 because the stage was spinning. All upper stages have restart.  It is a necessary capability.

1. Upthread you asserted that the V1.0 body structure could not support recovery/reuse - the grasshopper did

2. Not all can run for 30 minutes or longer, and the testing that was being done at McGregor was meant to qualify the engine from early on (at least as early as the 1C if not the original Merlin) for repeated use.  I can't quickly find a reference to a liquid engine that can only be run once, but I remember reading of them somewhere here.

3. "all upper stages have restart" ROFL how about trying to restart a CASTOR or STAR!!! and here is one quick reference to the restart of the COTS1:  https://en.wikipedia.org/wiki/SpaceX_COTS_Demo_Flight_1#Second_stage

[Jim]

1. Upthread you asserted that the V1.0 body structure could not support recovery/reuse - the grasshopper did

2. Not all can run for 30 minutes or longer, and the testing that was being done at McGregor was meant to qualify the engine from early on (at least as early as the 1C if not the original Merlin) for repeated use.  I can't quickly find a reference to a liquid engine that can only be run once, but I remember reading of them somewhere here.

3. "all upper stages have restart" ROFL how about trying to restart a CASTOR or STAR!!! and here is one quick reference to the restart of the COTS1:  https://en.wikipedia.org/wiki/SpaceX_COTS_Demo_Flight_1#Second_stage

1.  it can't.  It is supported by a frame that takes the landing loads.

2.  Those with ablative thrust chambers but those are pressure fed and not really "engines"

3.  Not relative to this discussion. Those are SRM's and not upper stages  (second stages) and still require another stage or the spacecraft to finalized the orbit.  They also don't do GTO missions.   Antares is designed around the ISS delivery mission and would need a 3 stage for other missions.

[nadreck]

1. Upthread you asserted that the V1.0 body structure could not support recovery/reuse - the grasshopper did

2. Not all can run for 30 minutes or longer, and the testing that was being done at McGregor was meant to qualify the engine from early on (at least as early as the 1C if not the original Merlin) for repeated use.  I can't quickly find a reference to a liquid engine that can only be run once, but I remember reading of them somewhere here.

3. "all upper stages have restart" ROFL how about trying to restart a CASTOR or STAR!!! and here is one quick reference to the restart of the COTS1:  https://en.wikipedia.org/wiki/SpaceX_COTS_Demo_Flight_1#Second_stage

1.  it can't.  It is supported by a frame that takes the landing loads.

2.  Those with ablative thrust chambers but those are pressure fed and not really "engines"

3.  Not relative to this discussion. Those are SRM's and not upper stages  (second stages) and still require another stage or the spacecraft to finalized the orbit.  They also don't do GTO missions.   Antares is designed around the ISS delivery mission and would need a 3 stage for other missions.

1. the frame does not extend beyond the base and that frame was attached to a V1.0 body

2. sure just like the Castor and Star series upper stages I mentioned aren't really upper stages 'cause they don't fit what you said previously about upper stages

3. see number 2, see https://en.wikipedia.org/wiki/Star_48 it is an upper stage for geosynch missions,  note THAT THE COTS-1 upper stage did restart!!!

[Chris Bergin]

No, no, no - this is not a Jim Q&A thread. Jim, I'm sure you've made the point you're trying to make. Everyone else, move along and get back to making your own points, which are as valid on a discussion forum as any one elses.

Anyone not happy about that, consider going to the next Jim seminar, but book soon - tickets are going fast!

[llanitedave]

I would definitely like to see more discussion about the ground equipment needed for fast turnaround.

[rcoppola]

Indeed. I'm very intrigued by how you could automate leg, 2nd stage and fairing (payload) attachment all the way through to an automated pully-type system that then takes the complete rocket/TE up the ramp, link to hydraulic pistons, rotate vertical, lock, auto attach/lock fuel/electrical, initiate fueling, Handoffs, TE retraction and off it goes.

Off course there's much more involved and there'd have to be some personal touches but that would be an epic Rube-Goldberg-Machine...sort of..

Has anything like that entire sequence ever been automated before to such a degree in this industry? What can they take from the car industry and what Tesla and/or BMW is doing with automation?

[Jim]

Car industry is not really applicable.  Cars are build on the order of 48 per hour vs 1 per 48 hours.  Don't need that complexity.  The stages are brought in by trailer into the hangar.  Slings are attached and the stages are lifted off the trailer onto support stands.  it just "takes" a little push of the stands moving them closer to mate the stages.  Manlifts and access stands are used to get to the mating areas to make mechanical, electrical and fluid connections.  Testing commences.   Once complete, the whole vehicle is lifted by crane, the support stands are moved out of the way and the launcher is rolled in.  The vehicle is lower onto the launcher and again connections are made and tested.

[guckyfan]

The legs are the single largest pieces of equipment to be attached. An industrial robot can handle them. Maybe humans still make the connections. One step on preparing preflown stages could be handling them with the legs on. Getting the legs off for transport and remounting them got to be one of the more complex, manual labor processes in the flow.

Otherwise it would mostly be automated testing sequences, I imagine.

As 48 hours turnaround won't be needed for a long time I imagine it would mostly be reduction of labour cost as a goal.

[JamesH]

The legs are the single largest pieces of equipment to be attached. An industrial robot can handle them. Maybe humans still make the connections. One step on preparing preflown stages could be handling them with the legs on. Getting the legs off for transport and remounting them got to be one of the more complex, manual labor processes in the flow.

Otherwise it would mostly be automated testing sequences, I imagine.

As 48 hours turnaround won't be needed for a long time I imagine it would mostly be reduction of labour cost as a goal.

Is it worth it? There are only four legs to remove, optimising something that is relatively fast to do anyway might not be cost effective.

Better to leave them on? Why do they take them off anyway?

[JBF]

The legs are the single largest pieces of equipment to be attached. An industrial robot can handle them. Maybe humans still make the connections. One step on preparing preflown stages could be handling them with the legs on. Getting the legs off for transport and remounting them got to be one of the more complex, manual labor processes in the flow.

Otherwise it would mostly be automated testing sequences, I imagine.

As 48 hours turnaround won't be needed for a long time I imagine it would mostly be reduction of labour cost as a goal.

Is it worth it? There are only four legs to remove, optimising something that is relatively fast to do anyway might not be cost effective.

Better to leave them on? Why do they take them off anyway?

Our current guess is that the deployment cylinders are one shots and mechanically lock after deployment.

[JamesH]

The legs are the single largest pieces of equipment to be attached. An industrial robot can handle them. Maybe humans still make the connections. One step on preparing preflown stages could be handling them with the legs on. Getting the legs off for transport and remounting them got to be one of the more complex, manual labor processes in the flow.

Otherwise it would mostly be automated testing sequences, I imagine.

As 48 hours turnaround won't be needed for a long time I imagine it would mostly be reduction of labour cost as a goal.

Is it worth it? There are only four legs to remove, optimising something that is relatively fast to do anyway might not be cost effective.

Better to leave them on? Why do they take them off anyway?

Our current guess is that the deployment cylinders are one shots and mechanically lock after deployment.

Not beyond the capabilities of SpaceX to improve on that then.

[meekGee]

The legs are the single largest pieces of equipment to be attached. An industrial robot can handle them. Maybe humans still make the connections. One step on preparing preflown stages could be handling them with the legs on. Getting the legs off for transport and remounting them got to be one of the more complex, manual labor processes in the flow.

Otherwise it would mostly be automated testing sequences, I imagine.

As 48 hours turnaround won't be needed for a long time I imagine it would mostly be reduction of labour cost as a goal.

Is it worth it? There are only four legs to remove, optimising something that is relatively fast to do anyway might not be cost effective.

Better to leave them on? Why do they take them off anyway?

Our current guess is that the deployment cylinders are one shots and mechanically lock after deployment.

I agree with the guess, but that doesn't mean they can't be folded in place.

Each leg has three structural attach point (all ball joints), and maybe gas lines or electric connections. (I hope each leg has its own internal bottle).

Saving 12 demate/mate operations like that is a major time saver IMO.  The legs are heavy and awkward to handle.

I hope to see them folded on the spot.

[guckyfan]

Each leg has three structural attach point (all ball joints), and maybe gas lines or electric connections. (I hope each leg has its own internal bottle).

Saving 12 demate/mate operations like that is a major time saver IMO.  The legs are heavy and awkward to handle.

I hope to see them folded on the spot.

I hope that too.

However presently the means of transporting the stage requires the legs gone. That's necessary for size restrictions when transported on public highways. It would not be required for moving in the cape area, even from the present mooring location of the ASDS. But new methods of transporting the stage would be needed. I don't know if an ASDS could land the stage at Vandenberg? Probably not because the west coast ASDS is not homed there. So at the west coast the legs would need to be removed for transport on barge landing.

[D_Dom]

Delta Mariner is not home ported at Vandenberg either. Stage deliveries are standard practice. Could be nominal processing.

[meekGee]

Each leg has three structural attach point (all ball joints), and maybe gas lines or electric connections. (I hope each leg has its own internal bottle).

Saving 12 demate/mate operations like that is a major time saver IMO.  The legs are heavy and awkward to handle.

I hope to see them folded on the spot.

I hope that too.

However presently the means of transporting the stage requires the legs gone. That's necessary for size restrictions when transported on public highways. It would not be required for moving in the cape area, even from the present mooring location of the ASDS. But new methods of transporting the stage would be needed. I don't know if an ASDS could land the stage at Vandenberg? Probably not because the west coast ASDS is not homed there. So at the west coast the legs would need to be removed for transport on barge landing.

Yes - in-place leg folding goes hand I  hand with never leaving the cape.

[RoboGoofers]

Would it be impractical to put dollys under the ends of the legs and roll it around? Not back to a building, but to a strongback at the edge of the landing pad, where they could detank and drop to horizontal.

[Dave G]

OK, I'm not sure I remembered this correctly, but I thought I heard something about the possibility of SpaceX using chilled RP-1 in the future, something about increasing both RP-1 and LOX density through chilling.

If I have this right, could chilled RP-1 also make it easier to refurbish the stage? Would chilled RP-1 make the bottom part of the stage look like the middle - all white after landing?

Also, for future SpaceX launchers, would methane make refurbishing easier due to chilled fuel?

Just a thought.

[RoboGoofers]

OK, I'm not sure I remembered this correctly, but I thought I heard something about the possibility of SpaceX using chilled RP-1 in the future, something about increasing both RP-1 and LOX density through chilling.

If I have this right, could chilled RP-1 also make it easier to refurbish the stage? Would chilled RP-1 make the bottom part of the stage look like the middle - all white after landing?

Also, for future SpaceX launchers, would methane make refurbishing easier due to chilled fuel?

Just a thought.

the RP1 was chilled in the last flight, to around ~20 degrees (though i can't remember if it's C or F). it can't go much lower since it would make the RP1 too viscous, and i assume that you need to maintain a proper ratio with the chilled LOx, which can't get any cooler.

methane would make refurb easier, if only because burning it doesn't produce soot.

[wardy89]

Would it be impractical to put dollys under the ends of the legs and roll it around? Not back to a building, but to a strongback at the edge of the landing pad, where they could detank and drop to horizontal.

I have been wondering the same sort of thing, although in the form of a strongback that could be taken to the rocket. Capable of recovering the stage back to horizontal, practical enough to allow for inspection, 2nd stage integration and payload integration then launch without the 1st stage having to be lifted off the strongback.
My thinking is that it would reduce the number of times you have to lift and load a stage, reducing the number of steps between landing and relaunch.

Taking the current F9 design and looking at pictures of the recovered stage i imagine a strongback capable of lowering a recovered stage to horizontal would need to be narrower than currently used at launch, so it fits between the extended legs and leaves room for them to retracted back to the closed position or be taken off.
 

[bstrong]

Taking the launch strongback all the way to the landing pad doesn't sound like a great idea to me. If you had a "stage manipulator" that can grab a stage, move it, and change it's orientation without requiring any manual rigging, you'd save a ton of labor and time, and it would be useful for other purposes as well (grabbing the stage off a truck, sticking it into a storage slot, etc).

I'm reminded of how forklift drivers shuffle around 40ft yachts in dry-stack storage like it's no big deal. If you had to use cranes, the whole business wouldn't work due to the extra time and labor. You need the equivalent for rocket stages (including the big storage shed with stacked slots for stages).

Edit: The piece of equipment I was imagining is called a reachstacker and is commercially available in sizes large enough to handle a stage (see attachment 2). Some have ability to tilt the payload. You just need a custom grappler. You could put one on the barge once you're confident enough that you're not going to set it on fire.

[Jim]

The first stage is too long for handlers such as those.  Also, rigging is required because built in attach points on the front end of the stage would add a lot of mass (the aft end with the thrust structure has ready made attachments )

[rpapo]

I think that a lot of people in this forum don't quite get just how BIG that stage is.  That forty foot yacht is only a quarter as long, and many times less weight, despite the fact that the stage is very light for its size.

[bstrong]

I think that a lot of people in this forum don't quite get just how BIG that stage is.  That forty foot yacht is only a quarter as long, and many times less weight, despite the fact that the stage is very light for its size.

I wasn't saying you could actually grab it with a forklift. Just using that example of efficiently shuffling around large items (in timeframes measured in seconds, not hours) using equipment that doesn't require rigging.

You can buy a reachstacker with 100,000 lb capacity (forklifts, too), which is more than enough for a stage, I believe. You couldn't use the one in the photo I attached unmodified, but with a 3x longer crossbeam and some custom grapplers on the ends that close to hold the stage in the same way as the rotisserie rings, I don't see why it couldn't be done. These companies crank out customized equipment all the time.

My point is that you have to stop thinking about handling stages one at a time and treating them like they're priceless artifacts. To support the flight rates that justify 48 hour processing, you'll have a large industrial facility with storage for 10-20 stages (probably racked vertically, somehow) and the capacity to process several in parallel. You'll be shuffling them around constantly, and you'll use techniques that more closely resemble the way large objects are shuffled around in other high-throughput industrial settings than what they're doing today.

[AncientU]

Each leg has three structural attach point (all ball joints), and maybe gas lines or electric connections. (I hope each leg has its own internal bottle).

Saving 12 demate/mate operations like that is a major time saver IMO.  The legs are heavy and awkward to handle.

I hope to see them folded on the spot.

I hope that too.

However presently the means of transporting the stage requires the legs gone. That's necessary for size restrictions when transported on public highways. It would not be required for moving in the cape area, even from the present mooring location of the ASDS. But new methods of transporting the stage would be needed. I don't know if an ASDS could land the stage at Vandenberg? Probably not because the west coast ASDS is not homed there. So at the west coast the legs would need to be removed for transport on barge landing.

Yes - in-place leg folding goes hand I  hand with never leaving the cape.

Once a Retriever engages the stage and supports its weight from the Octoweb, the legs can physically be retracted.
I suspect that a vacuum pump-out of the actuating gas could be the motivating force to pull legs back up along the stage -- after the latches are released, that is.  When legs are locked in place, the stage is moved to horizontal (using the above strongback) and driven back to hanger/HIF.  In the hanger/HIF, the stage is rigged with slings and bridge cranes as a new stage would be, except a legs-on method would be needed for the aft end.  Actuating gas is recharged along with other expendable fluids.

This type of retrieval, with legs on, could go from landing/safeing to hanger/HIF in a couple hours.

[Jim]

1.  You can buy a reachstacker with 100,000 lb capacity (forklifts, too), which is more than enough for a stage, I believe. You couldn't use the one in the photo I attached unmodified, but with a 3x longer crossbeam and some custom grapplers on the ends that close to hold the stage in the same way as the rotisserie rings, I don't see why it couldn't be done. These companies crank out customized equipment all the time.

2.  My point is that you have to stop thinking about handling stages one at a time and treating them like they're priceless artifacts.

3.  To support the flight rates that justify 48 hour processing, you'll have a large industrial facility with storage for 10-20 stages (probably racked vertically, somehow) and the capacity to process several in parallel. You'll be shuffling them around constantly, and you'll use techniques that more closely resemble the way large objects are shuffled around in other high-throughput industrial settings than what they're doing today.

1.  The stages are too long to be handled that way.   Would require a huge building just for maneuvering. Towing/movement lengthwise will be SOP no matter what the flight rate. 

2.  Wrong.  One ding and that takes the stage out the flow and maybe out of the fleet. 

3. Nonsense.  7 or so is enough.   And they won't be stacked like cordwood.  Stored stages don't make money.   A low hangar with as many bays as needed.  Doors  will be on each end of the bays, so stages can towed/driven through.

[DAZ]

4 small motorized self-propelled lifting units, one for each leg.  Lifting units attach themselves to the bottom of each leg.  Cable is then attached between the lifting units (one cable to each unit left and right) and cables are then snubbed tight.  Lifting units then lift the stage legs off the ground several inches.  Because of the cables the stage can no longer sag.  At this point the stage can even be leveled by adjusting the cable lengths.  This arrangement would seem to be faster and safer than using large aircraft jacks.

The same arrangement could be used on the ASDS.  On the barge the units would move and attach themselves to pad eyes for transport back to shore.  On the landing pad the units would move to a fixed Detanking/safeing and lowering unit.  Think simplified fixed strong back.  The legs would then be removed and the stage lowered.  It could then be easily loaded onto a standard F9 Road transporter.  A similar arrangement could be used dockside for the barge.  This too would seem to be to me much faster and safer than using a crane.

[bstrong]

1.  The stages are too long to be handled that way.   Would require a huge building just for maneuvering.

A facility that supports multiple launches a day is going to be enormous, no matter what. And besides, Elon loves huge buildings. Across all his companies, he has quite a collection of buildings that seemed unreasonably large when he built or acquired them.

Towing/movement lengthwise will be SOP no matter what the flight rate. 

I agree. Load it onto a flatbed truck for movement. No need to design a new vehicle that can both grab and drive long distances.

2.  Wrong.  One ding and that takes the stage out the flow and maybe out of the fleet. 

Everything may be need to be robotic if human operators aren't up to the task. That negates my argument for off-the-shelf equipment, but not the broader point that a large number of cores are going to be in constant motion, and it's best to be able to grab and manipulate them, even if it requires huge machines to do it.

3. Nonsense.  7 or so is enough.   And they won't be stacked like cordwood.  Stored stages don't make money.

The economics of re-use say flight rate trumps cost of a few additional cores. To maximize flight rate, you need to have some spares, so that you never miss a launch window if processing is delayed on a vehicle and also to account for both routine and unscheduled maintenance that takes longer than the normal processing time (e.g., an overhaul every 30 flights).

The number of extra cores you need depends on what fraction of the time cores are either undergoing maintenance or waiting to undergo it, just like the number of extra aircraft an airline needs to maintain. It's an interesting unknown number.

A low hangar with as many bays as needed.  Doors  will be on each end of the bays, so stages can towed/driven through.

With a large number of cores in the facility and a hanger the height of the HIF's (how or why would you make it lower?), you're going to get an enormous building that way, too. Sounds plausible, though. I guess it depends on how much land you have to work with.

I've defended this idea more than it really deserves, so I'll stop now.

[AncientU]

The 'strongback' could be the extended sections of a hydraulic crane, with a clamp like on the TEL upper end.  The stage is strong enough to support its own weight (stages always seen with support on each end, nothing in the middle), so the 'strongback' merely needs to lower and guide the upper end of the stage into a horizontal cradle.  Bulk of stage weight is at octoweb end,  so the upper end of the stage wouldn't need massive support.  The octoweb end, which is strong enough to support a fully-fueled vehicle, would easily handle the full stage weight when the stage is lifted for leg retraction.

I don't see this as a difficult rigging problem to solve.  Predict that the cranes and slo-mo landed vehicle handling will be replaced this year by a custom retriever.

[darkenfast]

Or maybe...just maybe....you stick with the simpler, tried-and-true method of hiring a local specialty crane outfit to safely lower your stage and put it on the transporter.  The thing is long and awkward and can only be handled at certain points.  They have this method down.  They use it in Texas all the time.  I would guess that they took longer than usual this time because it is the first time, and they are being very careful.   

[guckyfan]

Or maybe...just maybe....you stick with the simpler, tried-and-true method of hiring a local specialty crane outfit to safely lower your stage and put it on the transporter.  The thing is long and awkward and can only be handled at certain points.  They have this method down.  They use it in Texas all the time.  I would guess that they took longer than usual this time because it is the first time, and they are being very careful.

In McGregor they had no choice. It was put on top of the tripod until very recently. The handling is tricky. As you said it can be handled only at certain points in a certain way. I am not sure they will let a local crane crew handle it, I would expect they flew in the crew from McGregor. A special design for the purpose is at least a possibility IMO.

[AncientU]

Or maybe...just maybe....you stick with the simpler, tried-and-true method of hiring a local specialty crane outfit to safely lower your stage and put it on the transporter.  The thing is long and awkward and can only be handled at certain points.  They have this method down.  They use it in Texas all the time.  I would guess that they took longer than usual this time because it is the first time, and they are being very careful.

It took 48 hours to get the first landed stage to the hanger -- the operation was cumbersome and slow.  I suspect that the division of time in the target 48 hours for processing will be a small allocation to retrieval (maybe 10%/4.8hrs), and most time to ground checks and reintegration of the stage in the HIF(90%).

[DaveH62]

Moving to 48 hour turn around will be a journey, identifying and solving hundreds of critical path issues one at a time. A crane may be ok for now, but eventually they will need a built to suit system. A modular strongback with the louvered base section a permanent part of the launch and landing pad and a upper strongback, stage connection section, able to disconnect from the base. Once the strongback is raised at the landing pad, it connects to the rocket, is lowered to a transport and moved to a hanger for inspection. Once complete, the same strongback stage connection section could be moved to the launch pad.

[AncientU]

Moving to 48 hour turn around will be a journey, identifying and solving hundreds of critical path issues one at a time. A crane may be ok for now, but eventually they will need a built to suit system. A modular strongback with the louvered base section a permanent part of the launch and landing pad and a upper strongback, stage connection section, able to disconnect from the base. Once the strongback is raised at the landing pad, it connects to the rocket, is lowered to a transport and moved to a hanger for inspection. Once complete, the same strongback stage connection section could be moved to the launch pad.

Why use the same strongback?   The TEL is special purpose tooling and would be impractical to run it down to the landing pad.  The LC-39A version runs on railroad tracks as well as having umbilicals, etc. which would be useless at landing pad. 

Also, if the strongback is run down to the landing pad, it could not be integrating the next vehicle, could not be refurbished from the just completed launch, would have to have core offloaded in hanger for integration, etc.  Too many issues... if you want fast turn-around, make special tooling that does each job efficiently.

Nope, there will be special apparatus built for retrieving.

Edit: The TEL strongback is also too long to retrieve first stage plus interstage.  It is designed for a mated pair of stages.

[rcoppola]

I look at how they customized the Marmac's to become ASDS's. (Including modifications as they gained experience.)

I suspect the same will apply here and they may have already started a customized landed stage Vertical to horizontal/transporter at some undisclosed sub-contractor location. Or at the least, are working on final designs to facilitate such. With the potential of as many returned cores as they may have for 2016, they'll want to streamline this process as efficiently as possible heading into the 3rd & 4th quarters of 2016.

"LZ-1, the Falcon has landed. Landing operators move to procedure 11.100 Section 3 on LZ 1 B-Net..." I suspect they'll be amending their post landing procedures constantly through to the stage being returned to HIF.

They have a lot of TE experience at this point having designed and built multiple variations. As well as a lot of crane and customized flat bed experience in moving cores around. So I don't think it's a stretch to think they can take all that experience and create a customized and efficient hybrid de-erector/transporter.

[john smith 19]

I just discovered this thread and have now read through all of it. I have a few points.

1) People have said where think the time is going but the first real task would be plot on a time line where it actually goes.

2)Special purpose stage (not complete rocket) handling machines (probably adapted versions of commercial vehicles or the existing TEL) will be needed at the implied flight rates that need such fast turnaround.

3)To maintain this flight rate you are going to need "buffer stocks" of stages in a storage facility to cover the delay from ocean recovery --> transport to McGregor and McGregor --> Refurb site and Refurb  site --> launch site. So IRL the next booster comes from the refurb store while the last is just landing on the ASDS for transport back to port. That "Buffer store" will be pretty big.
Faster refurb --> smaller store.
Faster return to port --> smaller store.
Faster test firing --> smaller store.

The poster boy for this is the "Single digit press tool changeover" in the auto industry, pioneered by (IIRC) Toyota.  Key fact was that in principal it could have been done any time since about the 1930's if the incentive had existed for it to happen.

[Paul_G]

3)To maintain this flight rate you are going to need "buffer stocks" of stages in a storage facility to cover the delay from ocean recovery --> transport to McGregor and McGregor --> Refurb site and Refurb  site --> launch site.

I know that the FH boosters were taken back to McGregor for a test fire, but am not sure if the other reuse cores have been taken back - I *thought* the East Coast cores were refurbished at KSC / CCAFS and were not hauled back and forth to McGregor.

[stcks]

The only reused F9 (not FH) to go back through McGregor was 1021

[john smith 19]

The only reused F9 (not FH) to go back through McGregor was 1021

Then that buffer stock holding area just got smaller. 

[Paul_G]

The only reused F9 (not FH) to go back through McGregor was 1021

Then that buffer stock holding area just got smaller. 

I thought SpaceX had acquired the rights to use the old SpaceHab facility at Port Canaveral to refurbish, and store cores. That could be the buffer you are looking for.