An SSTO as "God and Robert Heinlein intended".

[Urwumpe]

The F9 and F9 1.1 share that same operational components and the same launch infrastructure, so don't see where your complaint is coming from. Are you seriously suggestion that whole concept of incremental development in aerospace is unsound? Because that is what it sounds like.

No, I just learned the concept as: Before you improve something, you should know it - and know what you want to achieve by improving it.

While components of the Falcon 9 1.1 are the same, many major assemblies had been changed, also the flight dynamics. Ground operations have also changed, if you look at the changes to the launch complex to accommody the Falcon Heavy as well.

 

[Hlynkacg]

No, I just learned the concept as: Before you improve something, you should know it - and know what you want to achieve by improving it.

While components of the Falcon 9 1.1 are the same, many major assemblies had been changed, also the flight dynamics. Ground operations have also changed, if you look at the changes to the launch complex to accommody the Falcon Heavy as well.

So how many falcon 9 flights would you consider reasonable before allowing development to continue. 5? 50? 5000? The problem with the attitude that "nothing must ever be flown that has not already been proven" is that nothing new ever gets flown, and that nothing is ever proven. All you do is end up sinking large amounts of time and money into inferior hardware and procedures, or in the case of NASA not doing them at all, while this is perfectly reasonable way to minimize risk when working a "cost plus" style contract it is actively detrimental in all other cases.

 

[Urwumpe]

So how many falcon 9 flights would you consider reasonable before allowing development to continue. 5? 50? 5000? The problem with the attitude that "nothing must ever be flown that has not already been proven" is that nothing new ever gets flown, and that nothing is ever proven. All you do is end up sinking large amounts of time and money into inferior hardware and procedures, or in the case of NASA not doing them at all, while this is perfectly reasonable way to minimize risk when working a "cost plus" style contract it is actively detrimental in all other cases.

A good measure if you look at the bath tub plot, is 8-10. Thats not really
much, but enough to exclude that you are not judging alone by early quality problems, but on real operational problems.

The rest of your post sounds like a big misunderstanding. If you have a fixed cost or managed service contract, you minimize risks. As cost+ subcontractor, risk is not your business, but the business of your customer.

For SpaceX, minimizing risks makes sense and is done, if you look at the rather boring rocket design. But still, it makes no difference to the fact that they are not knowing their own rocket when they replace it.

Also, this has another dimension: Customers have the same problem, when they can't know your offered services.

 

[Hlynkacg]

Except that you seem to be saying that early quality problems should go undressed in the interests of building a sufficient dataset. By that reasoning there is realistic distinction between early quality problems and real operational problems. They are one and the same.

As I stated before, because each rocket is a one-off, each rocket is effectively a prototype of the next. Requiring that a production model be exactly like it's prototype, warts and all, defeats the whole purpose of prototyping in the first place.

as for the latter part.

In a a fixed cost or managed service contract environment your first and last priority is always CYA. As such there is very strong incentive against change or trying anything that hasn't already been tried. Spaceflight is a high risk endevour dominated by such contracts so this is exactly whet we see.

If we assume that Musk is serious when he says that SpaceX's objective is "to open the frontier" It is in SpaceX's best interest to be only as cautious as is required to keep insurance adjusters happy, AKA what they have been doing. Over-caution slows development and comes with severe performance penalties.

 

[Urwumpe]

By that reasoning there is realistic distinction between early quality problems and real operational problems. They are one and the same.

Sorry, but that is wrong - read any publication on quality assurance, and you will see, that there is a big difference between both kinds. Ever heard of the bath tub plot?

---------- Post added at 10:45 PM ---------- Previous post was at 10:42 PM ----------

E It is in SpaceX's best interest to be only as cautious as is required to keep insurance adjusters happy, AKA what they have been doing. Over-caution slows development and comes with severe performance penalties.

And that is wrong - SpaceX is not insured currently. They fail to provide quality assurance documentation to the insurance companies so they can do their own risk assessment.

 

[Sky Captain]

I doubt it will see much GTO duty. The rocket does not appear to be optimized for GTO, the performance at that stage ratio optimized for LEO should be extremely poor. It is a offering to do GTO missions with it, but not a realistic offer, since there are very few customers left building such small satellites.

I think that's why they are developing Falcon Heavy to compete with Proton and Ariane 5 for geostationary orbit launches. Since they lack LH2 powered second stage they have to make up by strapping two additional first stages as boosters. That makes LEO payload capacity far in excess of what's currently needed, but at least they can also launch any satellite to geostationary orbit.

 

[Hlynkacg]

Sorry, but that is wrong - read any publication on quality assurance, and you will see, that there is a big difference between both kinds. Ever heard of the bath tub plot?

Yes I've heard of it, I just don't think it applies to the systems being discussed. "Wear out" failures are a non-issue because each rocket is expendable and refusing to make changes to or apply lessons learned in previous design iterations will result in a flat failure rate rather than a downward curving one.

And that is wrong - SpaceX is not insured currently. They fail to provide quality assurance documentation to the insurance companies so they can do their own risk assessment.

To satisfy their internal actuaries then. It doesn't matter how many failures they have so long their overall rate/cost is low enough that they can still get contracts.

---------- Post added at 14:13 ---------- Previous post was at 14:11 ----------

I think that's why they are developing Falcon Heavy to compete with Proton and Ariane 5 for geostationary orbit launches. Since they lack LH2 powered second stage they have to make up by strapping two additional first stages as boosters. That makes LEO payload capacity far in excess of what's currently needed, but at least they can also launch any satellite to geostationary orbit.

Well so long as the cost is competitive with Proton and Ariane there is no reason not to have the additional capacity.

 

[Urwumpe]

Yes I've heard of it, I just don't think it applies to the systems being discussed. "Wear out" failures are a non-issue because each rocket is expendable and refusing to make changes to or apply lessons learned in previous design iterations will result in a flat failure rate rather than a downward curving one.

Thats wrong: First of all, the overall reliability is the sum of all reliabilities, generally speaking. Every component has a bath tub plot of its own. And there are quite many components in the system, that are reused even in expendable rockets: The ground segment.

Then, you also have a background noise of failures in your operations. A procedure can be failure prone, without appearing like it initially. You can get more failures in a part initially appearing safe, when you get higher flight rates.

Also, it can become harder to produce components, when their parts become obsolete - what happens quickly today. Then you will get again an increased number of failures, even if your parts are produced new for every flight - this usually happens at the end of the life time of a ELV again.

Usually, you freeze the requirements of the new product generation not before your current one has entered the low and stable part of the bath tub. Then you know enough about the hardware and get ideas about the wetware. When the new product is entering test operations, you have enough experience in the wetware from the previous product to improve the operations. And repeat it, when the early operations of the new product are done.

Regardless what the product is. There are less aggressive R&D cycles, and multi-dimensional R&D cycles, but the general best practices don't change.

 

[Hlynkacg]

None of this explains why SpaceX is wrong for pursuing their aggressive R&D schedule if anything issues with obsolescence would seem to encourage it. Nor have you adequately explained why improvements should not be made if the opportunity presents itself.

As I said before, from where I am sitting you appear to be arguing that the whole concept of iterative design is unsound.

 

[Urwumpe]

As I said before, from where I am sitting you appear to be arguing that the whole concept of iterative design is unsound.

Yes, and from where I am sitting, I have the feeling that you never did any iterative design yourself, but instead just start many new expensive projects to fix a few bugs that disturbed you on the first try.

 

[Hlynkacg]

I think the problem here is that I am viewing each successive F9 launch as an experimental prototype. A flight test in the development of the SpaceX Reusable Launch System shown in that video.

As such it just seems silly to me to think that SpaceX would not update their design or their procedure based on the newest available test data.

You on the other hand view F9 development as essentially complete and as such view any changes to the design or procedures as an unnecessary "rocking of the boat".

 

[Urwumpe]

You on the other hand view F9 development as essentially complete and as such view any changes to the design or procedures as an unnecessary "rocking of the boat".

Well, no. I rather have the classic NASA attitude towards prototypes: "Lets fly it until it breaks, and then fly it a bit more." :lol:

I don't see it as unecessary "rocking of the boat" as you claim, I simply think that you need to collect some amount of information before you can improve something. For a small software, one bug to fix can be enough. For a big software system, you apply hotfixes and gather bugs to improve in the next version, so you have the final tests and installation only once for them.

For hardware, you collect improvements until the gains you expect by the improvements exceed the costs for changing the hardware and production lines.

 

[RGClark]

I repeat, show your work.
What are your masses for the stage and fuel. What allowances were made for maneuvering and contingencies. How does your thrust at sea-level compare to thrust in vacuum, what does your dynamic pressure curve look like? How heavy is your recovery equipment and landing reserve and what did you base that estimate? How much do your altitude compensating engines and their associated thrust structure weigh in comparison to the existing Merlins?
Its easy to make a first order estimate and say that something should be possible. Making it actually work is much much harder.
That and the assertion that the F9 first stage would be a more efficient launch vehicle than a complete F9 sounds suspiciously like homeopathic rocketry.

John Schilling is not likely to reveal the calculations that go into his simulator as it is proprietary. I have suggested to him that he might allow users to input also the sea level thrust and Isp to get more accurate estimates, but he chose not to change his calculator.

To see how efficient an SSTO can be, you have to use altitude compensation. I was surprised myself how high the payload got for the F9 v1.1 when you assume the first stage engines got the same vacuum Isp as the Merlin Vacuum. But the point is methods of altitude compensation that can work have been known for decades.

What would be needed is for these altitude compensating additions to the engines be lightweight. However, recent research with high temperature ceramics suggest this is also now possible:

Ceramic matrix composites make inroads in aerospace.
Published on May 14th, 2013 | Edited by: Jim Destefani

Oxide CMC exhaust ground test demonstrator consists of a 1.60-m diameter nozzle and 1.14-m diameter × 2.34-m conical centerbody with titanium end cap inspection portal. Credit: Steyer; IJACT.
http://ceramics.org/ceramic-tech-to...c-matrix-composites-make-inroads-in-aerospace

A More Efficient Jet Engine Is Made from Lighter Parts, Some 3-D Printed
Composite and 3-D-printed components will mean jet engines that use 15 percent less fuel.

In the LEAP engine, the ceramic matrix composites will replace only some of the nickel alloy parts. But in the future, they could be used for more engine parts, further reducing losses from cooling. This change could also allow engines to run at higher temperatures, making it possible to get more thrust from a given amount of fuel. Furthermore, composites could make engines lighter—parts made from these materials weigh one-third as much as the equivalent nickel alloy parts.
http://www.technologyreview.com/new...-is-made-from-lighter-parts-some-3-d-printed/

Bob Clark

 

[Loru]

@RGClark.

Yes. I use John Schilling's calculator as base for my rockets but I noticed that results vary from those in orbiter. I'm not sure what ascent profile is used in calculator but for example Themis-A and HCLV have less payload capacity than in calculator (both are throttling down to avoid more than 4G forces on payload).

So it's nice prototyping tool but not magic ball to answer all your questions.

BTW Have you ever tried putting numbers in for example Velcro add-on and testing propsed ideas?? It's not that hard TBH. There are already some Falcon meshes so you can test it yourself. Velcro already have 2 ISP values (for vacuum and sea level).

Yes. SSTO is possible but if it was economically vialble it would be around for some time already.

 

[RGClark]

@RGClark.
Yes. I use John Schilling's calculator as base for my rockets but I noticed that results vary from those in orbiter. I'm not sure what ascent profile is used in calculator but for example Themis-A and HCLV have less payload capacity than in calculator (both are throttling down to avoid more than 4G forces on payload).
So it's nice prototyping tool but not magic ball to answer all your questions.
BTW Have you ever tried putting numbers in for example Velcro add-on and testing propsed ideas?? It's not that hard TBH. There are already some Falcon meshes so you can test it yourself. Velcro already have 2 ISP values (for vacuum and sea level).
Yes. SSTO is possible but if it was economically vialble it would be around for some time already.

Yes. I will, finally, give Velcro a try.

About the Schilling calculator, I haven't done this with all known liquid fueled rockets, but for the ones I looked at the Schilling calculator was within about 10% of the listed payload value. I noticed though it was much less accurate when I applied it to the Ares I or Ares V or SLS, perhaps because they use large solids with highly variable thrust values. I also haven't tried it for the case of rockets using small strap-on solids.


Bob Clark

 

[TheMineGamer32]

Why don't make a winged spaceplane (fully loaded and fueled) that weights the same as a Titan II first stage (without motors) and we equip it with an LR-87-11 engine? That's a possible SSTO?

 

[orbitingpluto]

Why don't make a winged spaceplane (fully loaded and fueled) that weights the same as a Titan II first stage (without motors) and we equip it with an LR-87-11 engine? That's a possible SSTO?

You wouldn't have a SSTO afterward, and here's why:

Any rocket can have it's total mass be broken down into three categories, the propulsion system, the propellent, and the everything else. For a Titan 2 first stage, the everything else is everything that isn't engine, fuel, or oxydizer: it's the tanks for the propellent, the structure to marry everything together, and importantly, the payload. The Titan 2 1st stage, in it's unmodified form, was able to orbit something in the neighborhood of a few hundred kilograms. Adding structure, like wings, increases the amount of 'everything else" in relation to the working parts(engines and your propellent), either cutting into payload, or demanding performance increases like more propellent and/or engines with more thrust/efficiency.

Since your example idea doesn't allow for more engine or getting heavier, my assumption is your spaceplane body reduces your performance(substituting more structure for less payload and propellent) to something suborbital, meaning your modified Titian 1st stage isn't a SSTO anymore.

 

[TheMineGamer32]

And using composite structure and tanks? Is possible, the Titan II used hypergolic propellants. And i've selected the LR-87-11 engine because -obviously- it's updated and that means more thrust.

 

[orbitingpluto]

And using composite structure and tanks? Is possible, the Titan II used hypergolic propellants. And i've selected the LR-87-11 engine because -obviously- it's updated and that means more thrust.

It would be best do the math on this if your serious- nothing else would give a useful answer to your questions. Before we do anything else, or modify anything, let's first ground ourselves in the data for a stock Titan 2 first stage(all sourced from Wikipedia, by the way):

Masses

First stage dry mass: 4,319 kg
First stage wet mass: 121,200 kg
LR-87 mass: ~850 kg
Propellent mass: 116,881 kg

LR-87-5 specs

Thrust sea level: 956.5 kN 1913 kN
ISP sea level: 259 s
Thrust vacuum: 1096.8 kNs 2193.6 kN
ISP vacuum: 297 s

Now, we could get out the pad and pencil and do the math for finding the Delta-v of a stock Titan 2 first stage the hard, rewarding way, or we could use a online calculator, like this one. Since I chose to wimp out, the calculator gave me this result: 9930.41 m/s. I know from experience and book learning that a ascent to orbit, launching due east from the Cape will take about ~9000 m/s of delta-v, depending on how efficient I am, and how high I put my initial apogee; so there is some excess performance for payload/new structure.

Figure out what wings might mass, and add them to your dry mass(to keep with your original idea), or add the mass to both dry and total masses so you don't lose out on propellent. Play with the numbers, trying to ensure the modified vehicle can still make at least 9,000 m/2, and if you find something that you like, you can try using Spunik's CVEL Titans to fly it in Orbiter; if you encounter trouble setting up the scenario, I might be able to help you work it out.

 

[Loru]

Pure dV derived from rocket equation is not enough. You have to acount for gravity losses, which are determined mostly by thrust/weight ratio in initial phases of flight.

Silverbird rocket calculator says that Titan 1st stage with single LR-87-5 can't achieve orbit on it's own. Adding second engine to the stage gives me astonishing payload of 28kg! And that's for straight eastward launch on equator.