Author Topic: LV Technology  (Read 2507 times)

Offline martin hegedus

  • Full Member
  • ***
  • Posts: 353
  • Liked: 0
  • Likes Given: 0
LV Technology
« on: 11/24/2010 05:20 PM »
Sorry, not really sure what to call this thread, so the title may not adequately describe the topic.

Anyway, I made some off topic posts on the Predicting SLS thread and JohnFornaro asked me to repost here since he would like to comment.

My first post:

...

I'm on of those nerds crunching numbers.  Even worse, I develop CFD codes AND crunch numbers.  And it's so much fun!  Seriously.  It's a blast.  And I know for a fact that companies like SpaceX use engineering computer codes to reduce their costs and make what was not possible 20 years ago possible today.  And the usage of these codes definitely enhance the learning experience.

I'm surprised that I have not seen more computer analysis on this forum.  It's been very minimal.  To me there seems to be more arm waving then necessary.  Disappointing, kind of scary, and somewhat saddening.


My second post:

Here is another metric.  I've been looking for an open source trajectory optimization code.  I think it would be cool.  Much more fun than the rocket equation.  Sorry, for the younger generation the rocket equation is kind of boring.  It doesn't stand a chance against a good interactive experience.  But the codes I've found (POST, OTIS, TAOS, and Astros) are all restricted.  Yes, I am very very very well aware of ITAR.  And I've talked to the points of contact for two of the codes and have been told the inside story.  The foundations for these codes, trajectories and optimization, are clearly in the public domain and are taught at the college level.  It's not rocket science.  And I've done similar programing for an optimization class in grad school.  (However, that was for vehicles with lift and that stay in the atmosphere.  A little more complex and a little more fun!)  Of course a civilian could write a trajectory optimization program for the purpose of entertainment and education, and apply for a Commodity Jurisdiction request with the State Department.  I believe the State Department must return a response within 60 days.  And, here is the thing.  No one has done it.  There are no barriers and no one has done it.  I guess this sort of thing just doesn't excite people.  Yes, it is fun to see a rocket go up.  But, people are not excited by the technology and design aspects of sending one up.


Don't get me wrong.  My statement should not be viewed as black and white as it seems.  For example, I know that there are groups doing high power model rockets.  And, of course, there are the people on this site.  But overall I think the technical interest in the engineering is low.  Again, I love the engineering.  But cutting edge cars and airplanes are also fun.  And maybe what I interpret Ross and others saying holds true, i.e. launch vehicles are not about technology, they are about politics.

Online edkyle99

  • Expert
  • Senior Member
  • *****
  • Posts: 12938
    • Space Launch Report
  • Liked: 4000
  • Likes Given: 761
Re: LV Technology
« Reply #1 on: 11/24/2010 10:31 PM »
I don't see a reason why anyone could not offer a program that does at least a basic simulation of launch and on-orbit maneuvering - to let people play with design ideas.  Nor do I see why any government approval would be required.  As you say, the equations are in the textbooks - and the hardware performance data is either available or can be extrapolated well enough for these purposes. 

The only reason no one has done it is probably that it would not make money!

 - Ed Kyle     

Offline mmeijeri

  • Senior Member
  • *****
  • Posts: 7454
  • Martijn Meijering
  • NL
  • Liked: 76
  • Likes Given: 171
Re: LV Technology
« Reply #2 on: 11/24/2010 10:45 PM »
Actually, there is a very nice freeware simulator that does just that:

Orbiter Space Flight Simulator
We will be vic-toooooo-ri-ous!!!

Offline martin hegedus

  • Full Member
  • ***
  • Posts: 353
  • Liked: 0
  • Likes Given: 0
Re: LV Technology
« Reply #3 on: 11/25/2010 12:14 AM »
Orbiter is definitely a cool tool.  But, my understanding from reading the literature and simcosmos's post is that the tool is open loop and the trajectories are not automatic.  So it takes time and experience to set up a trajectory.  (Granted this is true, to a certain extent, for an optimizer.)  So setting constraints (for example max g), and seeing the results of those constraints, is not a simple task with a pure trajectory code.  Granted, if one is only interested in determining mass to orbit this may not be that important and the rocket equation is probably good enough.  But, if someone is interested in taking engineering brainstorming to the next level, then more detail is necessary.  And, LVs are interesting engineering problems in that they are very "simple" and very "complex" at the same time.  So, should I read anything into the impression, at least to me, that there does not seem much need for this data at the non professional level?

Offline 93143

  • Senior Member
  • *****
  • Posts: 3038
  • Liked: 292
  • Likes Given: 1
Re: LV Technology
« Reply #4 on: 11/25/2010 01:38 AM »
I'd use it.

I'd try to write it, but I'm kinda busy with CFD of a rocket combustor, and I have other stuff eating up my spare time...

My attempts at trajectory design have all both been single-stage and hand-optimized.  Matlab.  A general optimizer would probably be substantially more work.

Offline mmeijeri

  • Senior Member
  • *****
  • Posts: 7454
  • Martijn Meijering
  • NL
  • Liked: 76
  • Likes Given: 171
Re: LV Technology
« Reply #5 on: 11/25/2010 12:02 PM »
Orbiter is definitely a cool tool.  But, my understanding from reading the literature and simcosmos's post is that the tool is open loop and the trajectories are not automatic.  So it takes time and experience to set up a trajectory.  (Granted this is true, to a certain extent, for an optimizer.)

Absolutely true. The visualisation does add a new level of understanding though, at least it did for me.

Quote
  So setting constraints (for example max g), and seeing the results of those constraints, is not a simple task with a pure trajectory code.  Granted, if one is only interested in determining mass to orbit this may not be that important and the rocket equation is probably good enough.  But, if someone is interested in taking engineering brainstorming to the next level, then more detail is necessary.  And, LVs are interesting engineering problems in that they are very "simple" and very "complex" at the same time.  So, should I read anything into the impression, at least to me, that there does not seem much need for this data at the non professional level?

I'd be interested in this, and I've been working on something like this for a while, but in a scattershot sort of way. There's so much to learn and there are so many subtopics which are all very interesting and fun that it's hard to focus.
We will be vic-toooooo-ri-ous!!!

Offline JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 9170
  • Delta-t is the salient metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 620
  • Likes Given: 332
Re: LV Technology
« Reply #6 on: 11/25/2010 04:50 PM »
One of the things which has always bothered me about the explanation of this subject is that there often seem to be two poles.  There's the NPR science guy explanation;  very broad brush, explicitly avoids math, general and not specific.  Then there's the scientific paper; chock full of detailed explanations of part of the subject, assumes you know all the math already, etc.

A middle way, if you will.

Another thing is the software interface.  What's the best way to present?

Another thing is, the rest of the LV. What about the rocket?  The nozzle?  The injectors, turbopumps, RCS?  The EDS?  The Lander?  The only book I've read is Sutton.  How do you put it all together?
Sometimes I just flat out don't get it.

Offline simcosmos

  • Veteran
  • Full Member
  • ****
  • Posts: 484
  • Portugal
    • SIMCOSMOS
  • Liked: 4
  • Likes Given: 0
Re: LV Technology
« Reply #7 on: 11/25/2010 06:02 PM »
Orbiter is definitely a cool tool.  But, my understanding from reading the literature and simcosmos's post is that the tool is open loop and the trajectories are not automatic.  So it takes time and experience to set up a trajectory.  (Granted this is true, to a certain extent, for an optimizer.)  So setting constraints (for example max g), and seeing the results of those constraints, is not a simple task with a pure trajectory code.  Granted, if one is only interested in determining mass to orbit this may not be that important and the rocket equation is probably good enough.  But, if someone is interested in taking engineering brainstorming to the next level, then more detail is necessary.  And, LVs are interesting engineering problems in that they are very "simple" and very "complex" at the same time.  So, should I read anything into the impression, at least to me, that there does not seem much need for this data at the non professional level?


Hi Martin,

What I have shared on older posts at AJAX thread was mostly related with the specific method (Vinka's multistage.dll) that I'm using to implement and assess performances of launch vehicles in Orbiter Simulator:

http://forum.nasaspaceflight.com/index.php?topic=22266.msg651065#msg651065
http://forum.nasaspaceflight.com/index.php?topic=22266.msg653130#msg653130


However, as I have tried to also mention on those posts, such methodology is *not* the only possible method of performance or trajectory implementation and optimization for that simulator: what I have tried to demonstrate was then only one of the methods / one of the tools / possibilities (what I called of "the tip of the iceberg"), which yes, that specific method lacks in some automation but, again, I have also referenced that there exist other methods or tools which can, for example, be used to research a little better things such as, in this case, ascent trajectory optimization for that simulator environment.

Very briefly, there are a number of 'automatic pilot' tools that can be used in a separated way or, as  I have also referenced, even work *together* with the method I have referenced (Vinka's generic dlls).


The above does not change the fact that some caution should be taken in what regards compatibility check between a number of those automatic tools vs the Orbiter Simulator version for which those tools have been made vs other constraints, neither changes the fact that there isn't an equivalent to POST for Orbiter simulator but I would like to use this opportunity to specify that it is possible to have higher degrees of automation than what I have referenced in the posts that linked above.


Such tools are usually available at Orbit Hangar Mods (OHM: http://www.orbithangar.com/ ) or at other Orbiter Community related sites. I will not provide here a complete list because:
 
a) a new Orbiter version is out (2010P1) and there is an high probability that many tools aren't compatible (although older Orbiter versions are still available for download at OHM)... By the way, if eventually downloading Orbiter related stuff I ask NSF forum members to please read with attention the documentation provided on each addon as well to check for eventual compatibility requirements, conditions of use, or to check the context of a given addon version (for example, if it is a development version or a stable release, if the aim is to be realistic or else contain 'playability' liberties, etc)...

b)  each tool is a case and the most correct tool for a given user depends of what he/she wishes to do vs other eventual utilization constraints.

In any case will leave one of many possible examples:
http://www.orbithangar.com/searchid.php?ID=2783
http://www.orbithangar.com/searchid.php?ID=3800
(Powered Explicit Guidance)

In a similar way, there are tools which greatly assist 'Orbinauts' - sometimes even with automatic sub-routines (the user might need to feed a few parameters) - on the optimization of procedures such as ascent, trajectory (main injection burns, rendezvous procedures, etc), landings... Some of such tools are vehicle 'independent' and the user interacts with them via MFD. Other tools or algorithms might be specific of a given addon (or at least might have been made for a specific vehicle but might be used with other vehicles). Other tools have almost hands-free capabilities. Other tools do not have as many automatic features but can be used to assist 'manual' flights for which telemetry can be recorded  with a number of different methods and, such telemetry could then be adapted / tweaked to implement a custom automatic guidance, for example... There are many possibilities, I'm trying to be very generic on this post. 

Of course that everybody is also free to contribute with new tools (for Orbiter or standalone) related with the topic of this thread as long as having not only the skills but also the time and dedication to code and document such tool(s) ;)

Making a summary: as far as I'm aware, there isn't an equivalent to POST for Orbiter Simulator. There are however a number of different and / or complementary methods (beyond the one for which I have made specific references on AJAX threads) that can be used in and with Orbiter as well, given Orbiter simulator expansion possibilities, there is the possibility to make something more complete - in a given area - than what is currently available 'as default' or by using existing 'addons' as long as someone is able to dedicate to it. 

António

PS: yet another example: vehicle (launcher / spacecraft) systems simulation can also be implemented in Orbiter... This does not mean that every addon out there has it! Orbiter should be looked as an environment for which many things are possible to be implemented: much higher complexity and higher degrees of automation of the final products / results depends of each developer's (or team of developers) C++ Jedi skills, free time, etc

« Last Edit: 11/25/2010 06:29 PM by simcosmos »
my pics @ flickr

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 32440
  • Cape Canaveral Spaceport
  • Liked: 11187
  • Likes Given: 331
Re: LV Technology
« Reply #8 on: 11/25/2010 06:16 PM »
One of the things which has always bothered me about the explanation of this subject is that there often seem to be two poles.  There's the NPR science guy explanation;  very broad brush, explicitly avoids math, general and not specific.  Then there's the scientific paper; chock full of detailed explanations of part of the subject, assumes you know all the math already, etc.

A middle way, if you will.

Another thing is the software interface.  What's the best way to present?

Another thing is, the rest of the LV. What about the rocket?  The nozzle?  The injectors, turbopumps, RCS?  The EDS?  The Lander?  The only book I've read is Sutton.  How do you put it all together?

SMAD

http://astrobooks.com/

Offline martin hegedus

  • Full Member
  • ***
  • Posts: 353
  • Liked: 0
  • Likes Given: 0
Re: LV Technology
« Reply #9 on: 11/25/2010 07:00 PM »
Even with all the internet technology (along the lines of this forum, etc.), I find the old school way of talking to people productive.  Of course it helps if you have an engineering school close by.  Granted there are enthusiastic contacts and, well um, those with big egos.

As for software interfaces, I like Java and OpenGL (JOGL).  It is what I program with in regards to user interfaces.  Granted this is such a moving target.

As for putting it all together, my only thought is one piece at a time.  I took a look at my copy of "Rocket Propulsion Elements" (5th Edition) and saw that the first edition copy right is 1949.  1949 was a while ago.  And they definitely didn't have the resources we have today.  Yet it seems like they got things done.

Other references are "Launch Vehicle Design Process, Characterization, Technical Integration, and Lessons Learned" by J.C. Blair etc. and "Information Flow in the Launch Vehicle Design/Analysis Process" by W.R. Humphries Sr. (both free from ntrs) and "Design Methodologies for Space Transportation Systems" by Walter Hammond (available through AIAA).  Granted Ross stated that the information in those references is not state of the art (a statement I disagree with) and Jim mentioned they are not note worthy.  Also, the references are top level but they do show how pieces fit together.  And I like it for that.

Back when I was an undergraduate at U.C. Davis a group of us participated in the AIAA aircraft design competition.  And for that year we had to design a fighter aircraft.  Basically the students follow Roskam's aircraft design books and take it to the next level.  In our effort we supplemented the knowledge in the books with USAF airplane DATCOM, rudimentary FEA (this was back in 1990), and designed our own analog feedback control system.  I hope those sort of things are still begin done.  That was very motivating and awesome.

I don't want to give the impression that building a successful launch vehicle is "easy" or that a design for a "real" one should be attempted at a non professional level since there are legal issues, esoteric things, and a lot of time consuming details that need to be addressed.  And its a job, not necessarily fun for the broader audience.  Furthermore, if one goes to that extent the knowledge you learn probably is not very applicable to outside this field.  But, topics like trajectories (i.e. time integration of state vectors), optimization, structures, CFD, feedback systems, guidance and control are.

In regards to pieces and rocket thrust, here
is something that takes things past "just" Isp and with enough detail to make it fun for those with engineering curiosity but not enough to build something.  Mr. Benson at Glenn seemed to be enthusiastic that it may be used for the AJAX brain storming.

Offline martin hegedus

  • Full Member
  • ***
  • Posts: 353
  • Liked: 0
  • Likes Given: 0
Re: LV Technology
« Reply #10 on: 11/25/2010 08:13 PM »
LOL, (I'm laughing at myself) for all the internet searches, regardless of my experience at NEAR, regardless of my contacts, regardless of the earlier posts regarding state of the art LV design, etc. etc. etc. I never came across astrobooks.  And I just did a search on NASA Spaceflight on astrobooks and other than Jim's recent post, only saw one other reference.

Sometimes this is more like one of those video games were you search and you search for clues to get something.

Offline martin hegedus

  • Full Member
  • ***
  • Posts: 353
  • Liked: 0
  • Likes Given: 0
Re: LV Technology
« Reply #11 on: 11/26/2010 06:01 AM »

Another thing is, the rest of the LV. What about the rocket?  The nozzle?  The injectors, turbopumps, RCS?  The EDS?  The Lander?  The only book I've read is Sutton.  How do you put it all together?

After some some "digestion" (haha on this thanksgiving day) maybe I didn't understand your question or really address it with the "one piece at a time" comment.

So here are four design objectives (drivers) which, in my opinion, tie things together.  This is a sample only, there are more.  One is the structural model.  For example trajectories => aerodynamics => structures and then back to trajectories.  Another is the design of the control system.  An example reference would be "Control of Spacecraft and Aircraft" by Arthur E. Bryson (Same author as "Applied Optimal Control")  And of course there is cost.  But to do costs effectively, in my opinion, you need access to the details of past contracts and that will only happen if you are an insider.  But, I'm far removed from knowing much about costs and the state of the art in determining them.  And there is operations.  What works and what doesn't in the real world.

Offline JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 9170
  • Delta-t is the salient metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 620
  • Likes Given: 332
Re: LV Technology
« Reply #12 on: 11/26/2010 02:50 PM »
Sometimes I just flat out don't get it.

Offline JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 9170
  • Delta-t is the salient metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 620
  • Likes Given: 332
Re: LV Technology
« Reply #13 on: 12/06/2010 04:59 PM »
Well.  SMAD arrived the other day.  Wrapped in plastic, with a flyer included.  The flyer had a little yellow star:

"The best text and reference in astronautics-made better, more complete, easier to use"

Well good.  'Cuz if it weren't going to be E-Z to use, I was going to send it back and ask for a refund.  First laugh.  So I turned to page one.

"Space mission analysis and design begins with one or more broad objectives and constraints..."

Borrrinnnggggg.  Let's try page two, first sentence:

"Space is expensive." 

I just about fell out of my chair laughing.  I really had no idea!  Now I get it!

Seriously, this is an excellent book!  I found the last sentences on page five to be noteworthy:

"Our activity in space depends on what what we can afford to do or what we choose to do.  Therefore, we must carefully analyze why we choose to use or explore space. We must select each space mission, not just to achieve something that could not have been done before, but to achieve something that should be done or is worth doing."

So I look at:

https://docs.google.com/fileview?id=0BzLX4wxxT-QNMmE1NTdjOWItYWE1ZS00NTBmLWJmNzItOWJmNGVlMjkwOTdj&hl=en

and see:

Page 2: Work within budget scenarios.
Page 18: Schedule for Unconstrained Budget

I look at:

http://www.spacenews.com/commentaries/101115-ushering-new-era-human-space-exploration.html

and see:

Quote
This mission would take the crew 12 million kilometers from Earth or about 32 times the distance to the Moon, marking the first time that humans have gone beyond the gravity well of the Earth to venture into truly interplanetary space.

And I realize that I still don't get it.
Sometimes I just flat out don't get it.

Tags: