Author Topic: The trend towards ever more capable rockets  (Read 5009 times)

Offline Hyperion5

  • Full Member
  • ****
  • Posts: 1609
  • Liked: 1125
  • Likes Given: 219
The trend towards ever more capable rockets
« on: 06/17/2017 09:52 PM »
The computer world has long known of something called “Moore’s Law”.  This “law” was first proposed by Intel co-founder Gordon Moore in 1965.  It refers to an observation made by Moore, when he noticed that the number of transistors per square inch on integrated circuits had doubled every year since their invention.  Although Moore’s Law today is no longer holding up as we near the limits of physics, computers are still continuing their exponential growth albeit at a slower rate.  Much like with computers, the capabilities of rockets are also growing at an exponential rate, albeit a much slower exponential rate.  It is something so common amongst families of orbital carrier rockets that you might call it “Wernher’s Law”.  Instead of doubling in capability every 12-18 months, many rocket families are doubling in payload capability every 25-30 years.   There are of course some exceptions, like the Proton rocket family.  But even this family is being replaced by an even more potent successor family of rockets that will only grow in capabilities over time. 

I first noticed this trend when reading the ULA announcement of the Vulcan LV.  Out of curiosity, I looked up the history of its predecessor rocket families, Atlas & Delta.  One of the earliest Atlas rockets, the Atlas D, debuted in 1959 and could lift a mere 1400 kg to LEO.  The latest descendant, the Atlas V, debuted in 2002 and was able to lift 10,470 kg sans SRMs, some 7.48x as much as its ancestor.  The rocket family had grown over that time, sometimes in spurts, sometimes slowly, but when a new variant was introduced, it was almost always more capable than its predecessor.  In turn, the latest Vulcan news confirmed the trend would continue, as it would be capable of lifting ~20% more to orbit than its predecessor.  Yet even this version will soon be superseded with a version topped by a larger and more capable upper stage.  Curious, I began looking at other rocket families and saw the same trend appearing, including even in the R-7 family. 

I decided to quantify this trend across several rocket families and/or countries to see how universal it was worldwide.  As it turned out, orbital carrier rockets getting ever more capable and larger over time is so common and reliable, it’s almost the orbital carrier rocket version of Moore’s Law.  Below I’ve catalogued a number of rocket families or rockets from the same country.  The Mass CAGR is a calculation of the compound annual growth rate of gross mass is for the rocket family or rockets between the early version’s first launch and 2017.  The Payload CAGR is the equivalent but for growth in payload to LEO.  Finally, the Improvement Ratio shows how much faster the Payload CAGR is growing relative to the Mass CAGR.  The higher the ratio the more the gains in payload are being driven by gains in rocket efficiency rather than increases in mass. 

 
Rocket Family/CountyFirst RocketFirst flightGross MassLEO PayloadLatest RocketFirst flightGross MassLEO PayloadMass CAGRPayload CAGRImprovement Ratio
ArianeAriane 11979-12-24207,200 kg1,400 kgAriane 5 ES2016-11-17777,000 kg21,000 kg3.64%7.59% 2.08
AtlasAtlas D1959-04-14119,000 kg1,400 kgAtlas V2002-09-21334,500 kg10,470 kg1.80%3.53%1.96
Falcon 9 Falcon 9 v1.02010-06-04333,400 kg10,450 kgFalcon 9 v1.2 Full Thrust Block 42017-06-03549,054 kg22,800 kg7.39%11.79%1.59
IndiaAugmented Satellite Launch Vehicle1987-03-2441,000 kg150 kgGeosynchronous Satellite Launch Vehicle Mark III2017-06-05640,000 kg10,000 kg9.59%15.03%1.56
JapanN-I1975-09-09131,330 kg1,200 kgH-IIB2009-09-10531,000 kg16,500 kg3.38%6.44%1.91
Long MarchLong March 11970-04-2481,570 kg300 kgLong March 5-5002016-11-03879,000 kg25,000 kg5.19%9.87%1.90
R-7Sputnik 8K71PS1957-10-04267,000 kg500 kgSoyuz 2.1b2006-12-27308,000 kg8,200 kg0.24%4.77%19.875

Admittedly Compound Annual Growth Rate is an imperfect way to measure improvement in performance over time, as some families, like the R-7 family, show drastic early improvement in the first 10-20 years and modest improvements thereafter.  Although the compound annual growth rates are unlikely to stay as high for some rocket families, given the push for RLVs, these trends are likely to continue unabated.  Now let’s see what happens when we extrapolate the Mass CAGR & Payload CAGRs from the Atlas V in 2002 to the year 2020, when the Vulcan is expected to debut.  I’ve added in a half rate extrapolation for a more conservative estimate, and it’s close to the estimated specifications of the Vulcan LV.     

Rocket family/CountryYearMass Extrapolation (current trend) Payload Extrapolation (current trend)Mass Extrapolation (half rate)Payload Extrapolation (half rate)
Atlas successor2020462,383 kg19,747 kg393,040 kg 14,382 kg

Although there are limits in physics on how much larger rockets can get, we are nowhere near hitting those limits at the moment.  If there is a Wernher’s Law equivalent to Moore’s, it might read:

1)   Rocket will on average continue doubling in capability every 25-35 years. 
2)   Their growth in capabilities will outpace their growth in mass due to greater efficiency.

The question that continues to bug me is why this is happening.  Are we seeing ever more capable rockets because of military requirements, the requirements of commercial customers, or because rocket makers are trying to gain an edge on their competitors?  More recently however the margin required for reuse seems to suggest the exponential growth will continue for an altogether different reason.  Sound off with your comments, suggestions, and questions below!
« Last Edit: 06/17/2017 10:01 PM by Hyperion5 »

Offline spacenut

  • Senior Member
  • *****
  • Posts: 2120
  • East Alabama
  • Liked: 297
  • Likes Given: 191
Re: The trend towards ever more capable rockets
« Reply #1 on: 06/18/2017 02:08 AM »
10 years ago, on this Forum, many said if you get rockets to get 50 tons to LEO, any type of space exploration is possible.  To the moon, Mars, etc.  In space assembly of 50 ton modules and fuel depots would get us were we wanted to go.  We are now approaching that. 

Vulcan/ACES with solids, FH, and NG coming on line in within a few years.  These three with above mentioned fuel depots and 50 ton payloads could make SLS obsolete. 

NASA needs to take a hard look at the near term future, say 5-10 years out, and using what will be off the shelf reusable and partly reusable rockets with lower costs.  Design a modular in space NautilusX type vehicle for transport people and supplies to the moon and Mars.  Many of these could be built over time for a robust mining and colonization of the solar system. 

Offline IRobot

  • Full Member
  • ****
  • Posts: 1162
  • Portugal & Germany
  • Liked: 206
  • Likes Given: 191
Re: The trend towards ever more capable rockets
« Reply #2 on: 06/18/2017 08:55 AM »
At some point volume becomes the bottleneck.
And modularity is not a solution for all.

Using an oil rig vs containers as a comparison (or even Skylab vs ISS) , let's say that you want to deploy a large refining/mining machine in space.

Option a) you to make it on Earth and ship it in small pieces, analogue to sending parts in containers. You need a huge amount on in-space assembly, including a lot of either robotics or EVA's operations.

Option b) you want to send the whole thing already assembled, like an oil rig being towed to final location. Maybe most of the machine is empty space with lots of piping, so you end up with volume issue well before the weight issue.

Offline Lar

  • Fan boy at large
  • Global Moderator
  • Senior Member
  • *****
  • Posts: 8333
  • Saw Gemini live on TV
  • A large LEGO storage facility ... in Michigan
  • Liked: 5117
  • Likes Given: 3417
Re: The trend towards ever more capable rockets
« Reply #3 on: 06/18/2017 01:30 PM »
Interesting thesis. I hope it's true but wonder if, instead, the "growth" is mostly marketing retaining names of families...
"I think it would be great to be born on Earth and to die on Mars. Just hopefully not at the point of impact." -Elon Musk
"We're a little bit like the dog who caught the bus" - Musk after CRS-8 S1 successfully landed on ASDS OCISLY

Online Darkseraph

  • Full Member
  • ****
  • Posts: 450
  • Liked: 168
  • Likes Given: 86
Re: The trend towards ever more capable rockets
« Reply #4 on: 06/18/2017 01:59 PM »
There's too few data points to extrapolate a trend from with 25-30 year intervals. Orbital rockets have existed for slightly less than 60 years, which is just about two intervals. A trend for smaller rockets also appears to be emerging, indirectly due to Moore's Law.

Although the datapoints are limited and there is little to draw conclusions from, I have a hunch that the launch market is going to fragment between systems capable of supporting human spaceflight and those that need only carry satelites and small robots. We can't shrink people any time soon, but we can still shrink satelites quite a bit more, or pack more capabilities into the same size package. These can be launched on smaller, cheaper vehicles with lower safety margins. Human capable vehicles will probably continue to grow but this will taper off at some point, and instead the flight rate of these vehicles will go up - just like other forms of transport.
« Last Edit: 06/18/2017 01:59 PM by Darkseraph »
"For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled." R.P.Feynman

Offline RedLineTrain

  • Full Member
  • ****
  • Posts: 494
  • Liked: 233
  • Likes Given: 221
Re: The trend towards ever more capable rockets
« Reply #5 on: 06/18/2017 02:49 PM »
It strikes me as possible to confirm this trend by plotting the growth in payloads mass.  Ed Kyle may have the necessary information.

Edit:  It looks like Ed's payload mass information goes back to 2003.
« Last Edit: 06/18/2017 02:51 PM by RedLineTrain »

Offline Hyperion5

  • Full Member
  • ****
  • Posts: 1609
  • Liked: 1125
  • Likes Given: 219
Re: The trend towards ever more capable rockets
« Reply #6 on: 06/18/2017 03:23 PM »
Interesting thesis. I hope it's true but wonder if, instead, the "growth" is mostly marketing retaining names of families...

Thanks Lar.  Growth is mostly marketing retaining names of families?  I understand the debate about a Delta IV or an Atlas V really being "members" of their respective families.  However they come from the same manufacturers and build on the design heritage from their predecessors, even if some of them are new rockets.  If the manufacturers consider them a part of the same family, then I think it's fair to consider them for the hypothesis. 

Offline Hyperion5

  • Full Member
  • ****
  • Posts: 1609
  • Liked: 1125
  • Likes Given: 219
Re: The trend towards ever more capable rockets
« Reply #7 on: 06/18/2017 03:29 PM »
There's too few data points to extrapolate a trend from with 25-30 year intervals. Orbital rockets have existed for slightly less than 60 years, which is just about two intervals. A trend for smaller rockets also appears to be emerging, indirectly due to Moore's Law.

Although the datapoints are limited and there is little to draw conclusions from, I have a hunch that the launch market is going to fragment between systems capable of supporting human spaceflight and those that need only carry satelites and small robots. We can't shrink people any time soon, but we can still shrink satelites quite a bit more, or pack more capabilities into the same size package. These can be launched on smaller, cheaper vehicles with lower safety margins. Human capable vehicles will probably continue to grow but this will taper off at some point, and instead the flight rate of these vehicles will go up - just like other forms of transport.

Well I didn't have time to go through every launch family just yet, but I can add in the Titan family from beginning to end and see if that follows the trend as well.  Any others we should add in that you can think of?  A few families (Proton) don't follow the trend, but again that family is being replaced by the even more potent Angara family. 

Offline butters

  • Full Member
  • ****
  • Posts: 1670
  • Liked: 307
  • Likes Given: 70
Re: The trend towards ever more capable rockets
« Reply #8 on: 06/18/2017 03:46 PM »
I agree with RedLineTrain that payload mass (or perhaps payload energy to correct for injection orbits) would be a more informative measure of launch trends than the evolution of launch vehicle families.

I think that the ubiquity of the 6-ton GTO communications satellite has driven the growth of "medium" LVs over the past few decades, with workhorses of past generations, such as Delta II, coming up short of the comsat sweet spot. Proton, for example, happened to find itself perfectly positioned for the GTO comsat market long before it materialized, requiring little if any performance evolution.

But the GTO market probably won't rule the launch industry forever, with physical limits on the number of GEO satellites and the relatively long lifetime of these satellites. LEO constellations may be the next big thing in satellite communications and a major evolutionary pressure on the launch industry. Will this pressure cause launch vehicles to grow or shrink? Possibly both.

It's not entirely clear whether heavy-lift RLVs with multi-satellite dispensers or dedicated smallsat ELVs will come to define this coming era of LEO constellations, but it seems like those are the two predominant approaches to the problem. Perhaps the biggest surprise is that nobody is known to be pursuing small RLVs (and F9 is scraping the ceiling of "medium" at this point).

I suspect that the number of LEO satellites could very well grow exponentially over the coming decades before running up against physical limits. This is likely to be the most vaguely "Moore's Law-ish" characteristic of the launch industry. How that translates into launch rate or launch vehicle growth is an implementation detail.

Online Hobbes-22

  • Full Member
  • ***
  • Posts: 349
  • Acme Engineering
    • Acme Engineering shop
  • Liked: 98
  • Likes Given: 74
Re: The trend towards ever more capable rockets
« Reply #9 on: 06/18/2017 03:59 PM »
Quote
1)   Rocket will on average continue doubling in capability every 25-35 years.
2)   Their growth in capabilities will outpace their growth in mass due to greater efficiency.

I don't see point 2 happening. Many rockets have mass fractions on the order of 95%, end engine efficiencies in the 95% range. Both leave little scope for improvement: we're near the limits of currently-known materials.
We've gotten close to those limits over the past 30 years thanks to computers that allow us to calculate things like structural strength, so we can build a rocket that has the structural strength it needs, with very little margin. 30 years ago, those margins were much larger.
So we've seen a lot of your #2 happening in the past decades, but this process is running into physical limits. 

Offline envy887

  • Senior Member
  • *****
  • Posts: 2581
  • Liked: 1160
  • Likes Given: 731
Re: The trend towards ever more capable rockets
« Reply #10 on: 06/18/2017 04:51 PM »
Quote
1)   Rocket will on average continue doubling in capability every 25-35 years.
2)   Their growth in capabilities will outpace their growth in mass due to greater efficiency.

I don't see point 2 happening. Many rockets have mass fractions on the order of 95%, end engine efficiencies in the 95% range. Both leave little scope for improvement: we're near the limits of currently-known materials.
We've gotten close to those limits over the past 30 years thanks to computers that allow us to calculate things like structural strength, so we can build a rocket that has the structural strength it needs, with very little margin. 30 years ago, those margins were much larger.
So we've seen a lot of your #2 happening in the past decades, but this process is running into physical limits.

Efficiency is also a function​ of size, however. For example, larger rockets have lower drag losses, and can be flown by avionics of the same mass as a small rocket.

Offline spacenut

  • Senior Member
  • *****
  • Posts: 2120
  • East Alabama
  • Liked: 297
  • Likes Given: 191
Re: The trend towards ever more capable rockets
« Reply #11 on: 06/18/2017 04:59 PM »
When I mentioned modularity, I mean the rockets themselves.  Same attach points for payloads, same or similar fairing sizes.  This would be like container transportation.  All containers are built the same size to fit trucks, trains, and ships. 

Modularity for size and attach points on the rockets.

Modularity of connection points for all docking and attachments.  For instance a Russian piece of space equipment could attach to an American unit.  Or other countries and the various companies. 

Equipment itself could be anything, but where it attaches is another.  SpaceX may be able to get a 50 ton piece of space equipment in orbit, but would an ESA 20 ton piece be able to attach to it?  Or a Russian 20 ton piece?  Could a Blue Origin 40 ton piece attach to any of the others.

Also, modularity in fuel depot connections for all worldwide launchers.  Just like gasoline stations. 

Offline Hyperion5

  • Full Member
  • ****
  • Posts: 1609
  • Liked: 1125
  • Likes Given: 219
Re: The trend towards ever more capable rockets
« Reply #12 on: 06/18/2017 08:18 PM »
Quote
1)   Rocket will on average continue doubling in capability every 25-35 years.
2)   Their growth in capabilities will outpace their growth in mass due to greater efficiency.

I don't see point 2 happening. Many rockets have mass fractions on the order of 95%, end engine efficiencies in the 95% range. Both leave little scope for improvement: we're near the limits of currently-known materials.
We've gotten close to those limits over the past 30 years thanks to computers that allow us to calculate things like structural strength, so we can build a rocket that has the structural strength it needs, with very little margin. 30 years ago, those margins were much larger.
So we've seen a lot of your #2 happening in the past decades, but this process is running into physical limits.

Most rockets do not have mass fractions that high.  The lone exception to that has been SpaceX's Falcon 9 family, though we may see Blue Origin's New Glenn series approach that mark as well.  For the overwhelming majority of rockets, there is plenty of room for improvement.  Take the Atlas V for example.  While its Centaur stage has a good mass fraction for a hydrolox stage, the same cannot be said of its core stage.  While its 93% propellant mass fraction is decent amongst all rockets, it's worth remembering that the Saturn V bested that figure in 1967 with a 93.8% PMF on its first stage.  That's a stage built in 1967 beating one in 2002, and Lockheed Martin had access to better computer modeling, materials, welding and more.  Steven Pietrobon calculated that if you did nothing but replaced the core stage with one built with better alloys and used a common bulkhead, the Atlas V would have lifted 25% more to orbit.  The result of Atlas V's inefficient core stage is that the Falcon 9 v1.2 actually delivers more % of launch mass to GTO than the Atlas V.  If that doesn't suggest plenty of room for improvement still exists even amongst US launchers, I don't know what would. 

Offline ChrisWilson68

  • Senior Member
  • *****
  • Posts: 3519
  • Sunnyvale, CA
  • Liked: 2076
  • Likes Given: 2412
Re: The trend towards ever more capable rockets
« Reply #13 on: 06/18/2017 11:12 PM »
A big difference between number of transistors per unit area and number of kg of payload per rocket is that the former is fundamentally a measure of technology and the latter is fundamentally a measure of economics.

How many transistors can be put in a given area has always been limited by technology.  Every year, we've put as many transistors per cubic centimeter as we knew how.

With rockets, we pretty much know how to build them much bigger if we want to.  Back in the 1960s we built the Saturn V, and then we abandoned it because it wasn't worth the cost to us.  And we haven't built anything as large since.  The size of rockets is determined by economics -- what size rockets are justified for the value of the payloads they can launch?

Another major difference is that you're trying to plot trends within each "family" of rockets.  Moore's Law doesn't talk about trends from a particular organization -- it talks about cross-industry trends.

I think by choosing to look within each family you're defining the problem to suit the result you're looking for -- some kind of analog to Moore's Law.  If you were looking across the whole launch industry, it wouldn't look nearly so Moore's-Law-like.
« Last Edit: 06/18/2017 11:13 PM by ChrisWilson68 »

Offline edkyle99

  • Expert
  • Senior Member
  • *****
  • Posts: 12669
    • Space Launch Report
  • Liked: 3461
  • Likes Given: 578
Re: The trend towards ever more capable rockets
« Reply #14 on: 06/19/2017 02:33 PM »
The question that continues to bug me is why this is happening.  Are we seeing ever more capable rockets because of military requirements, the requirements of commercial customers, or because rocket makers are trying to gain an edge on their competitors?
Payloads have long been the driver for launch vehicle upgrades.  Usually, but not always, it is government payload requirements that drive launch vehicle size.   

 - Ed Kyle
« Last Edit: 06/19/2017 02:33 PM by edkyle99 »

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 31222
  • Cape Canaveral Spaceport
  • Liked: 9494
  • Likes Given: 298
Re: The trend towards ever more capable rockets
« Reply #15 on: 06/19/2017 02:46 PM »

The question that continues to bug me is why this is happening.  Are we seeing ever more capable rockets because of military requirements, the requirements of commercial customers,

Yes and yes.  Payload requirements drive rocket development.  All changes were driven by a customer that had requirements for a larger or heavier spacecraft.  It wasn't "build it and they will come".  It was "they is coming, so build it".

From another thread:

These are grouped by mission

Thor Agena A, B, D, TAT Agena, LTTAT Agena, Titan IIID, Titan 34D, Titan IV   CORONA, HEXAGON, KH-11/12
Atlas LV-3 Agena B,  SLV-3 Agena D, Titan IIIB, Titan 34B   GAMBIT-1/3
Atlas SLV-3A Agena D, Titan IIIC, Titan 34D, Titan IV    GEO SIGINTs

For each new Delta configuration, there would be a spacecraft that drove the need.
« Last Edit: 06/19/2017 03:13 PM by Jim »

Offline Jim

  • Night Gator
  • Senior Member
  • *****
  • Posts: 31222
  • Cape Canaveral Spaceport
  • Liked: 9494
  • Likes Given: 298
Re: The trend towards ever more capable rockets
« Reply #16 on: 06/19/2017 03:19 PM »
The computer world has long known of something called “Moore’s Law”.  This “law” was first proposed by Intel co-founder Gordon Moore in 1965.  It refers to an observation made by Moore, when he noticed that the number of transistors per square inch on integrated circuits had doubled every year since their invention.  Although Moore’s Law today is no longer holding up as we near the limits of physics, computers are still continuing their exponential growth albeit at a slower rate.  Much like with computers, the capabilities of rockets are also growing at an exponential rate, albeit a much slower exponential rate.  It is something so common amongst families of orbital carrier rockets that you might call it “Wernher’s Law”.  Instead of doubling in capability every 12-18 months, many rocket families are doubling in payload capability every 25-30 years.   There are of course some exceptions, like the Proton rocket family.  But even this family is being replaced by an even more potent successor family of rockets that will only grow in capabilities over time. 


The law applies to the spacecraft and the launch vehicles tag along

Offline TrevorMonty

Re: The trend towards ever more capable rockets
« Reply #17 on: 06/19/2017 06:13 PM »
The question that continues to bug me is why this is happening.  Are we seeing ever more capable rockets because of military requirements, the requirements of commercial customers, or because rocket makers are trying to gain an edge on their competitors?
Payloads have long been the driver for launch vehicle upgrades.  Usually, but not always, it is government payload requirements that drive launch vehicle size.   

 - Ed Kyle
All smallsat LVs being developed is a good example.

Offline DarkenedOne

  • Full Member
  • ****
  • Posts: 936
  • Liked: 50
  • Likes Given: 6
Re: The trend towards ever more capable rockets
« Reply #18 on: 06/23/2017 07:28 PM »
There's too few data points to extrapolate a trend from with 25-30 year intervals. Orbital rockets have existed for slightly less than 60 years, which is just about two intervals. A trend for smaller rockets also appears to be emerging, indirectly due to Moore's Law.

Although the datapoints are limited and there is little to draw conclusions from, I have a hunch that the launch market is going to fragment between systems capable of supporting human spaceflight and those that need only carry satelites and small robots. We can't shrink people any time soon, but we can still shrink satelites quite a bit more, or pack more capabilities into the same size package. These can be launched on smaller, cheaper vehicles with lower safety margins. Human capable vehicles will probably continue to grow but this will taper off at some point, and instead the flight rate of these vehicles will go up - just like other forms of transport.

I doubt satellites will shrink much principally for three reasons. 

1.  While it is true that improvements in semiconductors has made it possible for satellite service providers to retain their current capabilities with smaller satellites it also allows them to increase their current abilities with the same size satellite.  Most service providers are interested in increasing their capabilities. 

2.  There are many satellite operations where improved semiconductors will not make much difference.  Astronomy and earth observation satellites are limited by the diffraction limit and the limiting magnitude, which are determined by the telescope size.  For communications you are limited by power and antenna size. 

3.  The cost for large satellites is coming down.  When and if the Falcon 9 Heavy comes online launch costs for launch satellites will drop by more than half in many cases.

Offline Lars-J

  • Senior Member
  • *****
  • Posts: 3397
  • California
  • Liked: 2636
  • Likes Given: 1664
Re: The trend towards ever more capable rockets
« Reply #19 on: 06/23/2017 08:27 PM »
I don't see why this is so bizarre.

Once you create a new launch vehicle, the natural inclination is to tinker with it and improve it. Not downgrade its performance.

Or you do expect to see people who mess around with cars to tweak them to run worse?  ;D

Tags: