SpaceX Updates (Thread 7)

By DocM
in Science News & Discussion

 Share

Recommended Posts

Grand Master

DocM

Posted
  • Author
Posted

@elonmusk
Fourth rocket arrives in the hangar. Aiming for first reflight in Sept/Oct. https://t.co/TqW8d6Cc3U

 

September/October would seem to indicate AMOS-6, SES-10, Echostar 23 or SES-11. SES has been campaigning to be first, and we know 2 customers are bidding for a re-flown booster, so who's on second?

  • BetaguyGZT and Jim K
  • Like 2
Grand Master

DocM

Posted
  • Author
Posted

@jeff_foust
[NASA's]Green: Elon Musk’s Mars plans “tremendously exciting”; discussions will include coordinating launch plans of Red Dragon and Insight in 2018.
>
Green adds they are watching other commercial ventures (like Moon Express) and willing to enter into partnerships “at the right time.”


 

Grand Master

DocM

Posted
  • Author
Posted (edited)

Another possibility,

 

DoD through DARPA has for ages wanted a rapid response launch capability to put birds over less covered hot zones or fill in gaps that limit the number of taskers/day. The current XS-1 program at DARPA is the latest such effort, involving the usual suspects plus Masten Space and others.

 

SpaceX is now accumulating a warehouse full of S1's, and once refurbed they will only require an S2 and payload to launch. This hints at a rapid response capability via COTS. 

 

SpaceX's now advertised intent is to automate S1/S2/payload integration with a 1 hour rollout/fueling/launch. They've already demonstrated the 1 hour fuelling and launch part several times with F9 FT. 

 

Soon they will be eliminating the static wet dress rehersal/static fire, which will allow rollout/erect/fuel/launch on need. After that all they need is the automated integration.

 

DoD has seen the handwriting on the wall.

Grand Master

DocM

Posted
  • Author
Posted

No idea and they haven't given a timetable.

 

Perhaps this won't be the Merlin-powered Falcon but a Raptor-powered follow-on with a recoverable upper stage and interfaces designed for automation. 

 

We already know they can run several development programs at once, and with 5,000+ people, almost 50% more than ULA, I wouldn't put it past them to be working on it.

Grand Master

Draggendrop Veteran

Posted
  • Veteran
Posted

With reference to this article....

 

Carter, Musk to discuss innovation in upcoming meeting 

 

Quote

WASHINGTON – Ash Carter, the U.S. Secretary of Defense, and Elon Musk, SpaceX’s founder and chief executive, will discuss innovation June 8 in a private meeting, the Pentagon’s top spokesman said.

 

“Elon Musk is one of the most innovative minds in this country and the secretary, as you know, has been reaching out to a number of members of the technology community to get their ideas, their feedback, find out what’s going on in the world of innovation,” Peter Cook, the Pentagon’s press secretary said during a June 6 briefing. “The secretary’s had a number of meetings with business leaders and innovation leaders in particular out in Silicon Valley, other parts of the country, and I think that’s his goal here: to hear directly from Elon Musk on some of these issues.”

 

The meeting is private, which means an agenda or discussion items are generally not released. More details were not immediately available.

Musk has had a somewhat strained relationship with the Defense Department in recent years. He believed the Air Force moved too slowly in 2014 in certifying SpaceX’s Falcon 9 rocket to launch national security satellites. SpaceX sued the Air Force that year over its $11 billion block buy contract with United Launch Alliance, arguing that the Defense Department should have put some of those launches up for competition.

 

But since then Air Force and SpaceX officials have talked repeatedly about their close working relationship and earlier this year the Hawthorne, California-based rocketmaker won the Air Force’s first competitive launch contract in more than a decade, an $82.7 million deal to launch the second GPS-3 contract in 2018.

 

Carter has made reaching out to Silicon Valley one of the cornerstones of his tenure this year. During a March trip to California’s high-tech hub, Carter gave some of his most extensive comments to date on space and said one of his core jobs was to “rebuild bridges” between the Pentagon and the tech community.

http://spacenews.com/carter-musk-to-discuss-innovation-in-upcoming-meeting/

 

My take on this, is just as the article hints to. What the average person has witnessed with the recent launches and landing, is incredible....but...this meeting, IMHO, is exactly about innovation, what Elon is working on, it's timeline and particularly, the incredible engineering that is taking place and where it will be in the next few years as well as Elon's view of emerging technologies, transportation, solar power, AI, etc. Carter is mending bridges, getting better intelligence on capabilities in the tech sector for today and in the future, a ring of advisers from the leading companies in their fields. This will allow DOD to make informed decisions instead of listening to "oldspace"....they know how expensive and inefficient that can be.

 

:)

  • BetaguyGZT and DocM
  • Like 2
Grand Master

Draggendrop Veteran

Posted
  • Veteran
Posted

Disclaimer:   What is below, in it's entirety, is my personal opinion, nothing more, nothing less. I am trying to present this post, in a way, to explain the core of this matter, and to do so, for those without a background in the field, so that they may grasp the essentials. 

//

Doc made a post, a few days ago...


"Lars Blackmore's page at MIT, he being SpaceX's landing guy. Some interesting tidbits at his links."
http://web.mit.edu/larsb/www/

 

Seems like a routine info post, and a university link, which is "mouse trap clickbait" to me, just can't resist.

 

That was 4 days ago...and 4 days of realizing, that to me, this is the single most important post I have read on Neowin and maybe, I should say something.

 

From Lars Blackmore's page at the link...

 

"About me: I am at Space Exploration Technologies (SpaceX), where I am responsible for Entry, Descent and Landing of the Falcon 9 rocket, as part of our reusable launch vehicle program. My team developed the precision landing technology for the Grasshopper rocket, the F9R-Dev rocket, and the F9R reusable booster stage. After placing our customer's satellites in orbit, F9R recently completed the world's first landing and recovery of an orbital-class booster stage:"

 

and....

 

"Previously I was in the Guidance and Control Analysis Group at the NASA Jet Propulsion Lab, part of the California Institute of Technology, where I developed control and estimation algorithms for NASA's future space missions. I co-invented the G-FOLD algorithm for precision landing on Mars, and was part of the SMAP (Soil Moisture Active Passive) mission, which launched in January 2015.

In 2007 I finished my PhD at the Massachusetts Institute of Technology in the Model-based Embedded and Robotics Systems group, under the supervision of Brian C. Williams. My thesis was on control and estimation of stochastic systems, especially hybrid discrete-continuous systems, and I am currently continuing research in this area. I am particularly interested in chance-constrained optimal planning, that is, finding the best plans such that the probability of failure is below a given threshold. I am interested in applications in autonomous air and space systems.

Previous research has been in control and estimation for Formula One racing. My MEng thesis was with the McLaren team, and in my first year at MIT I carried out a project with the Jaguar team (now Red Bull Racing)."

//

//

 

If I were to hazard a guess to the racing involvement, it would be control system optimization for fuel conservation through powertrain control and route planning. Sounds familiar, does it not.

 

and Lars has a list of papers, of which, in content summation, are astounding.

//

Elon Musk started SpaceX with a vision for the colonization of Mars, and to do this, reusability is a cornerstone.

 

Reusability entails getting back expensive parts of your launcher, reliably, and getting them back in action with minimal cost while maintaining operational requirements. This very lofty goal was set early, practically at the formation of SpaceX. By recruiting the best available talent and an efficient/flexible company structure, as of recent successes, they are well on their way. Many departments are pushing the bounderies of conventional practices and proving, beyond a doubt, that "newspace" is here to stay and that others are now realizing that they must adapt or relinquish market share.

 

Some may have noticed that I have emphasized control systems in several of my posts, and this post is why.

 

Control system engineering is an engineering niche unto it's own, yet is so powerful, that many aspects are used in everyday engineering design. Conventional and Modern control theory are the delineations, if you will, yet they both are actively used.

 

A control system, at a primitive level, as an example, is a stage (discrete RF amplifier) where one input, example would be a leg current, is monitored for an undesired change. Once a change is detected, a weighted or proportional response is fed back to neutralize this change, allowing the stage to perform it's intended design.

 

The stage (amplifier) is represented by a "model", a mathematical statement, the inputs and outputs are also  mathematical statements. To achieve this model, at varying degrees of complexities ( even down to the atomic scale, an example could be a SPICE model, my personal favorite), requires varying degrees of knowledge to the physical properties that must be described mathematically.

 

If we don't have all the data necessary, we do what we can experimentally, till we have enough relevant data to refine the chosen control systems mathematical requirements.
 
Depending on the system representation used, these properties can relate to many facets such as time invariant or time varying, linear or non linear activities, freqency domain (Conventional control theory) or time domain approach (Modern control theory) as well as many other peculiarities.

The real issue is the proper, indepth mathematical description of the model/system/states, which includes indepth mathematical description of all variables to the system, encompassing all dependencies such as physical limitations, so that a desired controlled output (even range) can be achieved using a particular control regime.


 Control engineering relies on a body of equations and methodology, some of which took either weeks, months, years or generations to develope by many fields. Still, some equations are so difficult, that one must manipulate an equation to use a transform (change mathematical planes) to enable an easier answer using another plane equation or manipulate vector-matrix differential equations for a state space representation. We even have to develope  new approachs to enable eventual conformation to an existing strategy. Control schemes will have specific benefits and liabilities, and again, manipulations will be required to solve an issue in one realm and be able to reintroduce it in the present method of choice. 


The number of inputs/outputs and an indepth understanding/mathematical representation of their dynamic behaviors, increases the complexity of the system, increases calculation and decision times,and hence, system reaction times.


One can easily have several inputs (example, physical position, gravity, atmospheric pressure, temperature, acceleration and velocity at various points in a flight path) and that data can overwhelm a system if not properly designed.

 

// Going back in time, Doc made a post, with video, of the engine design team's use of GPU's for model refinements. The results are cutting edge and have helped produce outstanding durable products such as the Merlin's and Draco's and soon to be seen Raptor. The issue and use of the GPU's is poignant. The mathematical models are being refined, but the issue here, has been the immense data bank. In order to make it cost effective, as related to computational power, a method was derived, using GPU's, to retrieve specific amounts of data, and display this in a model simulation to determine properties of the design for various combustion phases. Again, we see another advanced modeling regime in place, where the model description and it's inputs are causing a computational bottle neck which had to be remedied by cost effective use of available hardware. We can't have a super computer everywhere, and in some instances, it would be overwhelmed anyway, hence a different refined and localized approach. The engine team has done outstanding work, and it shows every launch/landing.
//

The only present way, to get anything substantial, to land on Mars, is by retropulsion. This is where SpaceX has made their mark. As we can see, Lars and his team have been busy for quite some time, and have been progressively developing cutting edge control system designs.They have been able to mimic the Martian atmosphere, by using the upper regions of Earth's atmosphere during reentries and have been refining control systems with every landing attempt.

 

But, they have done much more. Not only are they refining the landing techniques, here on Earth, they have been working on propulsive/guidance control systems, using SuperDraco's for Mars. 

 

Lars and his team are using extremely complicated mathematical control system designs, and are developing new solutions which are required to make an extremely complicated system semi autonomous. The landing on Mars will have an optimized flight path (fuel saving and soft landing)  to a predetermined landing spot. It will be able to detect obstructions and modify to a continued optimal flight path. Systems will be checked for health, not just failure. All this will be done in real time and in autonomous mode, since it can't wait for Earth to call back with instructions, while landing. 

The use of GPS is commonplace today, but there are a variety of "accuracy" standards in use, ranging from consumer use up to high precision military requirements. 

 

The F9FT first stage, on re-entry, is handled by an extremely complex control system, that follows an optimized flight path to get the stage to a preset GPS waypoint. The ASDS is also programmed to stabilize itself around that same waypoint. From what I have read, the first stage and the ASDS do not communicate with each other, and there is no local GPS differential/correction station present. With this information, I would assume low double digit cm accuracy, which would allow safe landing in the case of a directly opposed cumulative error.


The repeated accuracy of the landings/attempts demonstrates the maturity, to date, of this complex control system. It will be continually improved for an optimal flight path that both conserves fuel and presents the least exposure to damaging heat.

 

There is enough video available so that one can see, in real time, how this extremely complex control system can direct and guide the first stage, through severe conditions, to incredible accuracy, placing the stage where required.

 

SpaceX has, through advanced engineering, systematically developed the required blocks to get the launcher we see today, in a cost effective and timely manner. While the development continues in various area's, the "baton" has been given to Lars and his team. They are the ones that, through experimentation and extremely complex mathematical control systems, will put SpaceX on Mars. When one considers that there is no GPS constellation around Mars (for now), no local GPS differential/correction station present (for now) ,and communications will be by DSN (old but reliable) and Mars payload relays, with data rate limitations (NASA is heavily involved in newer optical communication replacements), they will be relying on their own systems as well as pertinent data from Mar's present orbiters. 

 

But all this comes with caution since these are still pioneering days. SpaceX has amassed a wealth of data and proprietary development, which is why it does not use the "broken" patent system. This is extremely important, more so now than before, since SpaceX has demonstrated, several times, that it is in possession of a system that can guide an extremely large, high velocity object, from space, to a position on land/at sea, with incredible accuracy, and low cost, a trait that could be used for less than ideal purposes. 

 

As complex mathematical models of the operating environment include more definition, the causes and effects generate an exponential increase in the data to be manipulated...theoretically more data than is possible to do operations on. This is why advanced control systems must choose methodologies and place limits on descriptor depths to enable a system to be controlled in an expedient fashion and yet have tight control on relevant processes...avoiding mathematically induced technology bottlenecks.

 

Engineering abilities have come a long way since the 60's/70's. I have tried, above, to present a simplified basic  outline, without creating a small paperback. How mathematically advanced are some of SpaceX's teams? A blunt answer is that  Lars Blackmore's team is developing a complex control system that is akin to a low level AI that has a set of "blinders" on and will be able to land a craft within designated mission "success parameters" for the environment, all the while conserving fuel. This makes what you thought were complex equations look like someones grocery list. That is the reality of todays advancements in many fields, and advances will continue, some at torrid paces.

 

Welcome to the future and "buckle up", we're going for a ride! 

Grand Master

DocM

Posted
  • Author
Posted (edited)

Don't remember if I said it here or elsewhere, but at one time I used the term "wizards" to describe both Lars team at SpaceX and the guys running MSNW; fusion rocket, ELF thruster and plasmoid fusion reactor, which is <5 miles from SpaceX's Seattle comms satellite factory

 

We just have to lock these guys and the BFS team in a room and order 6 months worth of pizzas and buffalo wings. 

 

and

 

:rofl:

 

 

Edited by DocM
  • Draggendrop and BetaguyGZT
  • Like 2
Grand Master

BetaguyGZT

Posted
Posted

It's been said before, and I'll repeat it here. Elon Musk is the real-life Tony Stark, minus the "Egomaniac Playboy Party Animal Alcoholic with a penchant for one-night stands with Supermodels". He's all of the work ethic and smarts with none of the bad habits, ego or snark.

 

He's a Developers' Developer. Mad Scientist minus the Mad. He lets his people innovate without interference. He wants them to think outside the box, but he also wants results that are workable, repeatable and reliable. And above all, he wants those results for the right reasons. He's patient, he's not afraid to fail, and he's not afraid to dust himself off and try again when he does fail.

 

That's why SpaceX is successful. :yes: 

  • Draggendrop
  • Like 1
Grand Master

DocM

Posted
  • Author
Posted

NASA up for more SpaceX Mars mission collaborations,

 

http://spacenews.com/nasa-exploring-additional-cooperation-with-spacexs-red-dragon-mission/

 

SpaceX has always said there'll be numerous Red Dragon missions, multiple precursor missions before BFR/BFS starts making big noises, and NASA wants to go along.

  • BetaguyGZT and Draggendrop
  • Like 2
Grand Master

BetaguyGZT

Posted
Posted

Oh man that's gonna be hugely interesting. Gonna test ISRU, Base/Hab and EDL systems & procedures for BFR/MCT along with other new and refined technologies ...

 

I can't wait. Coming to Mars on the 2018 & 2020 Launch Windows ... yeah. I can dig it. NASA and NewSpace are gonna need all of the brainpower, funding and able bodies they can get to make all of this happen across the board. It's like the big Lunar push in the 60's multiplied by a factor of two, but with a tenfold reduction in funding.

 

Are we all up for it?? :yes: Darn it, I really gotta get my Engineering Degree and all my Certifications sorted ... if I work really, really hard I can have it done by mid-2018. That'd be soon enough to get me into the mix enough to get developing for the 2020 Launch Window ... if I choose the Internship Program I can start applying as early as my Sophomore year this coming January.

  • geertd, Jim K and Draggendrop
  • Like 3
Grand Master

DocM

Posted
  • Author
Posted

El9n Musk's IAC 2016 conference schedule in Guadalajara.

 

Date: September 30 (Friday)
Time: 0830 to 1030
Title: Mars (awaiting confirmation)

 

2 hours. Sounds like a rather detailed presentation.

 

There is also talk of info releases ahead of the conference. I hope Twitter and Instagram are ready for this ;)

  • BetaguyGZT, Jim K and Draggendrop
  • Like 3
Grand Master

chrisj1968

Posted
Posted

I'm watching event horizon.. of course a movie. But in theory, would astronauts have to hibernate in some sort of stasis to make a trip from Earth to Mars or would it be From Earth to the moon, resupply with fuel and then catapult to Mars in some sort of sleep stasis?

 

this intrigues me

Grand Master

chrisj1968

Posted
Posted
14 minutes ago, Beittil said:

A trip to Mars with crew under ideal circumstances wouldn't require hibernation or similar tech. We're generally talking (one way trip) about time frames which astronauts currently already spend on board ISS or less.

ok, thanks a lot for the reply. we're losing somewhat the boundaries of what we call fiction and reality. fascinating.

This topic is now closed to further replies.
 Share
Go to topic listing