To Grasp the Heavens

[Shevek23]

To go back to an earlier post in the light of this latest one, on my general theme of "idiot balls," part of the decline of the British position in aerospace despite Selene's success was said to be that the giant satellites Silver Star launcher could put up were going out of style, due to miniaturization enabling lighter satellites to get the job done.

But I doubt that trend would totally wipe out the big payload market. Certainly as miniaturization progresses, some enterprises seek to launch smaller and smaller payloads, which eventually opens up niches for air launch and the like, along with the option of hitching a ride as secondary payloads in a launch initially commissioned and mostly paid for by some bigger payload. We see both approaches today, and satellites and deep space probes being brought down to tens of grams or even less.

But this has hardly destroyed the market for big payloads. Over the decades since Apollo we haven't seen the upper size of payloads rise much, partially because making a bigger launch system for them is a major step. A version of Titan claimed a 30 ton capability but I gather it has had few or no takers. Nevertheless, big payloads to LEO in the 20 tonne range continue to go up.

So, in this ATL, I find it a bit odd to claim that while the massive capability of Silver Star was desired for the massive British domestic TV geosynchs, now that those TVs can presumably be reached with more compact and lighter higher tech 1980s generation satellites, Silver Star launchers are sitting idle.

But wait, I think. Won't at least some commercial entrepreneurs consider what they could do with satellites that are more sophisticated per kilogram, and also have the sheer power and mass of the late 60's-early 70's big British telecom sats? Satellite phones for instance--modern OTL sat phones rely on highly sophisticated equipment in the phone itself to extract useful signal from a weak one from a small satellite, but wouldn't it be possible to use brute force in orbit, combined with mid-80s state of the art microtech, to enable a cruder, simpler handset to get the job done instead? The big satellite has big "ears" to pick up a weak signal from the ground and process it out of background noise, and then a powerful beam to enable a simpler ground handset to pick it out? Modern phones depend on digital processing in the phone itself, might not a powerful enough satellite work effectively with analog signals instead?

I think it is very strange then for the British industry to be left completely high and dry by advancing microtech. The advance over OTL in sheer mass to orbit, and the relative economy that the long shaken down and tested Silver Star launch system offers in the form of reliability and its development cost being largely buffered with large Black Anvil and heavy usage in Selene as well as the commercial branch of the business being something they have a decade of experience in ought to attract entrepreneurs who see possibilities in new satellites that are both big and smart.

BAC should have been enjoying a steady if perhaps limited demand for their launch services, and some of the customers desiring SS launches would be American I would think.

 

[PMN1]

What is the payload weight and fairing size for the Silver Star?

 

[sts-200]

Funnily enough the word cavalier also exists in French with the same meaning.

Interesting, I didn't know that.
Although cavalier is obviously originally a French word (English words ending "ier" almost always are), I had thought the expression was derived from the English civil war (the "Roundheads" versus the "Cavaliers").
If its the same in French as well though, we probably got the whole expression off you.

SDI craziness and hype, exposed. I wonder how will the Soviets react.

In typical Soviet fashion.

Lowell Wood initial plan - "smart rocks" - were to be as light as some grammes, and there would have been ten of thousands of them. It makes breakthrough starshot looks realistic when you think about it.

https://www.washingtonpost.com/arch...08d-a5d0-19167462228a/?utm_term=.80fc6d0e7966

In my opinion, these were by far the most practical of the SDI ideas, and directly or indirectly they helped to drive forward a lot of concepts around smaller, simpler spacecraft.

 

[sts-200]

So BAC basically will fund it's development off of the back of Star Wars? Hahaha that is v cute

The US funding a foreign launch vehicle out of DoD funds? Not too sure about that one.
There might be other ways and means though...

 

[sts-200]

What is the payload weight and fairing size for the Silver Star?

19' diameter external (so about 17' max internal), with the fairing length designed for the Hermes satellite plus the upper stage.
They were capable of putting just over 3.4t into GEO (not just GTO) with a margin.
Without the second stage, the core could put just over 16t into an ultra-low Earth orbit.

 

[sts-200]

To go back to an earlier post in the light of this latest one, on my general theme of "idiot balls," part of the decline of the British position in aerospace despite Selene's success was said to be that the giant satellites Silver Star launcher could put up were going out of style, due to miniaturization enabling lighter satellites to get the job done.

But I doubt that trend would totally wipe out the big payload market. Certainly as miniaturization progresses, some enterprises seek to launch smaller and smaller payloads, which eventually opens up niches for air launch and the like, along with the option of hitching a ride as secondary payloads in a launch initially commissioned and mostly paid for by some bigger payload. We see both approaches today, and satellites and deep space probes being brought down to tens of grams or even less.

But this has hardly destroyed the market for big payloads. Over the decades since Apollo we haven't seen the upper size of payloads rise much, partially because making a bigger launch system for them is a major step. A version of Titan claimed a 30 ton capability but I gather it has had few or no takers. Nevertheless, big payloads to LEO in the 20 tonne range continue to go up.

So, in this ATL, I find it a bit odd to claim that while the massive capability of Silver Star was desired for the massive British domestic TV geosynchs, now that those TVs can presumably be reached with more compact and lighter higher tech 1980s generation satellites, Silver Star launchers are sitting idle.

The payload market is healthy (if anything more so than in reality).
With Silver Star its just as much a lack of self-confidence, a perceived lack of export markets and an unwillingness to invest anything in the industry in the mid-late '70s. Nevertheless the program staggered on until axed early in the Thatcher government.
Doesn't mean the technology is entirely dead through.

But wait, I think. Won't at least some commercial entrepreneurs consider what they could do with satellites that are more sophisticated per kilogram, and also have the sheer power and mass of the late 60's-early 70's big British telecom sats? Satellite phones for instance--modern OTL sat phones rely on highly sophisticated equipment in the phone itself to extract useful signal from a weak one from a small satellite, but wouldn't it be possible to use brute force in orbit, combined with mid-80s state of the art microtech, to enable a cruder, simpler handset to get the job done instead? The big satellite has big "ears" to pick up a weak signal from the ground and process it out of background noise, and then a powerful beam to enable a simpler ground handset to pick it out? Modern phones depend on digital processing in the phone itself, might not a powerful enough satellite work effectively with analog signals instead?

I couldn't possibly comment.
Yet.

I think it is very strange then for the British industry to be left completely high and dry by advancing microtech. The advance over OTL in sheer mass to orbit, and the relative economy that the long shaken down and tested Silver Star launch system offers in the form of reliability and its development cost being largely buffered with large Black Anvil and heavy usage in Selene as well as the commercial branch of the business being something they have a decade of experience in ought to attract entrepreneurs who see possibilities in new satellites that are both big and smart.

BAC should have been enjoying a steady if perhaps limited demand for their launch services, and some of the customers desiring SS launches would be American I would think.

Silver Star itself is out of production, but a lot of the background stuff is still there.
There will be a few surprises along the way, but I don't think I'm giving much away when I say there is an installment coming entitled "Phoenix".

 

[sts-200]

The Shed at the Bottom of the Garden​

Despite being the third nation to launch a satellite and, with France, the first to put a man on the Moon, no orbital rocket had ever been launched from the British Isles. Up until the 16th October 1985, all British satellites had been launched from Australia, French Guiana or the USA.

The "Silver Arrow" that lifts off from the Benbecula Missile Range on that morning was neither large nor particularly advanced. Its first stage consisted of a surplus 1960s Black Knight Mk.3 booster that had been refitted and upgraded on a shoestring budget. The design of the stage's eight HTP-Kerosene fuelled engines dated from the 1950s, but they were simple and reliable, and they successfully burn for 119 seconds, taking the rocket clear of the atmosphere and steering it North towards the pole. Tiny explosive squibs push the stage clear once it had burned out, and second stage flight began as four small solid motors ignited to help settle the liquid propellant.

Stages two, three and the payload were different and increasingly sophisticated. Student engineers at Cranfield and Brunel Universities had developed the second stage from a dubious concept of the 1960s, the "Satellite Upper Stage", or SUS. Originally built to carry and control payloads in orbit, the SUS proved to be hopelessly unreliable and soon went out of service. The unit launched on Silver Arrow had been built in 1962, but its mechanisms had been stripped down, rebuilt and improved in the labs at Bristol. The new electronics and controls owed nothing to the 1960s, and have been designed and built from scratch using systems that were undreamed of when the SUS was originally designed.

For obvious reasons, navigation systems, inertial platforms and other rocket control mechanisms are the top-secret preserve of governments and their defence contractors. The technology needed to pilot a payload, whether it was a research satellite or a nuclear warhead, was closely-guarded ultra-precision engineering. Because of that, it is also mind-bogglingly expensive. The system that steered Silver Arrow into orbit was none of those things; although in some ways it could be considered more advanced than almost any other guidance system. Without massive government backing, and being a collection of junior engineers and university students, the team behind Silver Arrow had to come up with an alternative. Instead of the more usual military-grade systems such as the highly classified fluidic gyros used on Black Anvil-Cavalier, they used off-the-shelf components which were begged, borrowed or bought from commercial suppliers.

There were no space-rated guidance components on Silver Arrow; instead the team used ordinary systems as found on any light aircraft. Several of these were carefully tested, adapted and linked to mimic the function of a 3-axis inertial platform. Light aircraft gyros are not very accurate, and when used on board a rocket they would have to withstand significant, sustained acceleration forces that are not normally found in commercial aviation. However, with the resources of some of the best Aerospace Engineering and Physics departments and some of the UK's brightest young engineers, many of the deficiencies could be dealt with. The effects of acceleration forces could be overcome with changes to the design of the bearing systems, and extensive testing mapped the performance of the gyros under a variety of loads. The next step was to allow for these known errors, and Silver Arrow had one of the most powerful flight computers yet fitted to a rocket, using a set of Zilog Z80 microprocessors. Better known for their use in home computers, these chips had the processing power to allow for the known behaviour of the gyros and accelerometers, and used a fuzzy logic program to produce a synthetically accurate inertial reference from several none-too-accurate sources. In this respect, Silver Arrow was the world's second most sophisticated rocket; only NASA's Shuttle had more processing power on board.

After a 237 second burn, the single Gamma engine on the second stage shuts down and the control system on board starts to orientate the stage for the final boost into orbit. 181 seconds later, the innovative stage has done its job. Its final act is to spin itself up and fire the squibs to separate the unguided, solid-fuelled third stage. Originally, the team behind Silver Arrow had hoped to build a liquid-fuelled third stage, but the cost would have been high and when the RAE offered them a surplus “Moorhen” spin-stabilised motor left over from a 1970s test programme, the design was quickly changed to use it. Ten minutes after liftoff, Silver Arrow has done its job and the payload has been deployed. UoSat-2C is injected into a 585x599km orbit at 97.8deg, close enough to the planned-for 590x590x97.75d.

If anything, the satellite is even more of a marvel of non-traditional space engineering than the rocket that launched it. The fourth in a series of small university-led satellites, it doesn't contain a single traditional “space rated” component. The spacecraft was built at the University of Surrey and masses just 50.2kg. The concept of “off-the-shelf” was taken to a new level, with some items such as the satellite's batteries actually being bought from a local hardware store (although they then went through a rigorous testing process). Spacecraft engineering tests and magnetic field experiments make up the primary payload, and the satellite is controlled by another Z80 processor, a module used with great success on an earlier UoSat spacecraft. The secondary payload carries another commercial microprocessor, a powerful Intel 80286 which can be used to run programs from a small memory, accessible via an amateur radio transponder. It is hoped that future spacecraft might make use of this more capable computer, although on UoSat-2C it plays no part in the control of the satellite.

Besides the interest within the space science community, it came as a surprise that the mission evoked such widespread public interest and enthusiasm, in part due to the way it was portrayed in the popular media. Silver Arrow was not a traditional space mission. It didn't require millions of pounds to be spent on "science stuff" that hardly anyone understood, or need gigantic machines that could only be built by colossal corporations with government backing. Instead, it was small and much more accessible; it was done by students on a shoestring budget in little workshops. The limited scale of the enterprise and the distinctly "non-aerospace" details surrounding it gave it an appeal far beyond the scientific community; stories such as the Hebridean fisherman who lent the launch crew a tarpaulin to cover the rocket's equipment bay on the ferry trip over to Benbecula helped to bring the project to the attention of the public, and to make it somehow seem more applicable to them.
In inspiring people to think about what was possible, it no doubt ranks as one of the most cost-effective space flights of all time. Certainly, the amateur radio experiment and its associated computer were well used by everyone from school science clubs to undergraduate projects between 1986 and 1992, when the payload suddenly failed. More fundamentally, it appealed to something in the British nature; a charming, naive (but sometimes accurate) belief that most great inventions and discoveries began with chaps tinkering about in places such as the shed at the bottom of the garden.

The reality was that these homespun tales made for great publicity, and were emphasised and exaggerated for that reason. However, the project was certainly not some sort of below-the-radar shock to the establishment; it had the backing of the likes of Rolls-Royce, the Royal Aerospace Establishment and the RAF, all of which had helped by provided money, apprenticeships and facilities. Although certainly not the old guard of British aerospace engineering, Silver Arrow and UoSat-2C were built by some of the brightest scientists and engineers in the UK, young men (and increasingly women) who could go on to build tomorrow's spacecraft and aircraft.

 

[Archibald]

I've started reading "Selene" again (and it is massive). What strikes me so far is the carnage of perfectly unuseful OTL British aerospace projects.
So far the death toll includes
- Concorde
- TSR-2 and what followed (AFVG, F-111... Tornado ?)
- CVA-01 (not sure, I'll check the story further, but the Falklands update still has a cranky HMS Eagle)
- The Polaris submarines
- The Europa rocket boondoggle

All these projects surely cost a crapload of money OTL.

I wonder about ITTL RAF. With perfect hindsight, the Buccanner lasted until 1994, without a S.3 variant that could have flown in the 70's and last until the 2000s.

 

[Archibald]

It is fascinating to read Selene and then to compare with OTL ESA history
http://www.esa.int/esapub/sp/sp1235/sp1235v1web.pdf
https://www.esa.int/esapub/sp/sp1235/sp1235v2web.pdf

Make no mistakes, the two documents are dull and boring. Selene is far more exciting.

STS-200 - I don't know if you ever heard of that proposal, but there was an italian scientist, Amaldi, that in 1959-60 proposed the "Euroluna" concept.

Check "Amaldi's dream of a Euroluna before 1965" in the first document I linked.

 

[sts-200]

It is fascinating to read Selene and then to compare with OTL ESA history
http://www.esa.int/esapub/sp/sp1235/sp1235v1web.pdf
https://www.esa.int/esapub/sp/sp1235/sp1235v2web.pdf

Make no mistakes, the two documents are dull and boring. Selene is far more exciting.

STS-200 - I don't know if you ever heard of that proposal, but there was an italian scientist, Amaldi, that in 1959-60 proposed the "Euroluna" concept.

Check "Amaldi's dream of a Euroluna before 1965" in the first document I linked.

Amaldi was a good man, and his input into CERN was of great value.
They were never short of ideas, but as those documents point out though, there were too many people who wanted the early effects to be civilian-dominated, and therefore denied themselves an obvious source of funds and facilities.
Contrast that with the early USAF and US Army space initiatives, and even the civilian NASA never hesitated to benefit from the DoD programs.

Lots of missed opportunities around that time, although even I would admit that Selene takes them to the extreme!

 

[sts-200]

I've started reading "Selene" again (and it is massive). What strikes me so far is the carnage of perfectly unuseful OTL British aerospace projects.
So far the death toll includes
- Concorde
- TSR-2 and what followed (AFVG, F-111... Tornado ?)
- CVA-01 (not sure, I'll check the story further, but the Falklands update still has a cranky HMS Eagle)
- The Polaris submarines
- The Europa rocket boondoggle

All these projects surely cost a crapload of money OTL.

I wonder about ITTL RAF. With perfect hindsight, the Buccanner lasted until 1994, without a S.3 variant that could have flown in the 70's and last until the 2000s.

Glad you are enjoying the old thread once again. Now that I have set the scene and we are reaching the real story with this new thread, I'll probably be slowing down a bit, so you have time to go back.

Some of those cutbacks come out of my attempts to make Selene seem even vaguely economically plausible.
The story of Selene owes a lot to the story of Concorde, suitably changed for a space project.
TSR2 never even made it off the drawing board in the story, and there would have been the same farce with AVFG and F-111. Tornado still happens (I had to give BAC something to do in the '70s), and we may still see Typhoon.

With TSR2 and Concorde, there was also the consideration of freeing up people (not just money) for the space program. Even Britain+France's technical base was much smaller than the USA, so I couldn't have too many people working on other things.

CVA-01 never made it off the drawing board in Selene, although as you see, some of the RN's older ships have longer active careers. Without an agreement with the US, there was no point in Polaris (although the RN still has SSNs).

ELDO/Europa never happened, but the rocket was effectively still built once Blue Streak was deployed. There are a few versions, but "Blue Star Mk.3" bears a strong resemblance to the real Europa (there's a drawing somewhere in about mid 1965).

 

[sts-200]

Power to the People​

Like so much technical infrastructure development, the story of electricity generation in the UK is a patchwork of commercial and public, successes and failures, which in this case only began to be rationalised in the 1940s. The creation of the Central Electricity Generation Board (CEGB) in 1957 provided a strong, centralised, engineering-led management structure to complete the construction of the National Grid and to cope with the booming demand for power. In those days, nuclear generation seemed to offer the promise of an unlimited, reliable and cheap source of energy, and all-electric homes were built in huge numbers in anticipation that electricity might become "too cheap to meter". Engineers at the CEGB knew better, although that did nothing to slow the trend as clean electricity replaced dirty coal and smelly oil as a source of domestic heat. New power stations and new technologies were needed to keep up with demand, and it is to the organisation's credit that throughout the 1960s, the lights always stayed on.

In 1974, that changed, and a whole generation of British politicians learned a fundamental lesson: Power is measured in Watts, not Votes.

Only months after the Oil Crisis was inflicted on Europe and America, coal miners and transport workers across the UK went on strike. Weak and ineffective conciliation by the Conservative government delayed some of the effects, but it also ensured that when the blow fell, it would hit even harder. Even as the nation had a spacecraft on its way back from the Moon, power rationing was announced for all homes and businesses. There was little coal being mined, and few trains were still running to distribute what stocks there were. Within a month, the government fell. Attempts to portray the flight of Selene 5 as an example of the nation's technological prowess backfired dramatically as lights went out in homes across the country, and "space" quickly became a dirty word. The election of a minority Labour administration didn't solve much; Labour or Tory, the unions could see they had the government on the run, and were in a position to demand wage hikes and job guarantees; demands which they ultimately obtained in full. Thanks to these agreements and the necessity of avoiding another bitter dispute, a great deal of money was invested in mining; over the next ten years, the long-term decline in Britain's coal output was halted. Meanwhile, union militancy only increased.

Having avoided a crisis for some years, in 1984, the Conservative government realised that now was the time to strike - or perhaps rather, to provoke one. The proposed closure of a dozen unprofitable mines was the excuse union leaders had been waiting for. They brought out the miners in Yorkshire and Lancashire and urged the public to support them in rolling back the tide of Thatcherism. The refrain "Switch on at Six" was hammered home; urging the public to boost their electricity use at peak times to bring down the grid, and with it, the government.
From the other side, it was personal. The Prime Minister was a member of Ted Heath's government when it was brought down by the 1974 strikes, and she had never forgotten or forgiven. It would be a class struggle, a battle of old technology versus new, socialism versus the market. In this industrial war, and nothing short of total victory would do.
Unlike during those dreadful months in 1974, this time the establishment was prepared, and the country also had other options. Stocks of coal at power stations could be boosted by the use of oil, now readily available from the North Sea. In addition, a new generation of nuclear plants were just coming on line after years of delays, and miners in the Midlands were much more reluctant to go on strike than their Northern colleagues; their pits weren't being threatened directly, and there was a guaranteed market for their coal in the form of the power stations sitting directly above their heads. The strike was therefore patchy, but still, no quarter was given in either direction. Throughout the summer of 1984, it seemed touch-and-go. By the autumn, the strike was still “firm”, but that wasn't good enough; thanks to the efforts of the CEGB, the government could announce with total confidence that there would be no power cuts into 1985. The strike didn't collapse, it just faded away, as first a trickle, then a flood of men went back to work, defeated and dispirited. The pit closure programmed recommenced, and this time there would be no compromise.

To complete the victory, and to forever remove the union's knife from the nation's jugular, would take time and planning. New sources of power would be needed, as well as new ways of funding and managing their construction. If nuclear energy was to play a part, the mistakes of the dreadfully slow AGR programme of the 1960s and ‘70s could not be repeated. Although most of these stations were now online, it was clear that they would not meet their original design goals in respect of on-load refuelling or total power output. There was also a new consideration, that of radioactive waste. In the early days of both military and civil nuclear programmes, this had never really been considered, but it was now clear that large amounts of high-grade waste would need to be stored for millennia to come. The natural and lightly-enriched, low burnup Uranium used in current reactors meant that most of the material within the fuel elements played almost no part in the release of energy, but when extracted it was all still highly radioactive.

Led by the Energy Research Laboratory in Leatherhead, Britain's engineers were challenged to come up with alternative solutions. The obvious short-term candidates were oil and gas, now readily available from the North Sea. Some existing stations could be converted to use these fuels, while older plant could be replaced by more efficient combined-cycle gas turbine systems. An expanded network of small hydro stations was suggested for parts of Scotland and Wales, and technical exchanges with the US provided a great deal of information on the latest designs of pressurised water reactors, which America's extensive nuclear power program had developed over the years. Spearheaded by President Kennedy in his 1961 “Frontiers of Technology”** speech to Congress, the USA had built nuclear power plants in large numbers. Although the image of the industry had been tarnished by two accidents in the late 1970s, the latest GE-X335 design promised to be safer than anything that had come before. France had also invested heavily in nuclear power, and the idea of a common European reactor design had support from across the continent. Beyond those ideas, there were more exotic gas-fuel reactor concepts, and the HTLCR, the High Temperature Liquid Core Reactor, which took the theory behind the AGR design to the next stage by using a continuously-replenished liquid fuel loop, and by raising the reactor temperature to match it to the latest combined-cycle generator technology.

Teams were also tasked to study more unusual concepts, and time and money was set aside to seriously investigate geothermal energy, solar-concentrator plants and wind turbines, and even ideas such as wave generators and space-based solar power were looked at. First investigated seriously by NASA in the 1970s, a space-based solar system would require huge satellites in high orbits to convert sunlight to electricity, then beam it down to Earth using high power radio beams. The advantage was that power would be available all the time, irrespective of day, night or cloud. The disadvantage was equally obvious: any meaningful solar power satellite would mass tens of thousands of tons, while the largest single satellite yet put into geostationary orbit had a mass of just 3.4 tons.
Most of these concepts were quickly rejected. Wave generators were too vulnerable and geothermal was very expensive. Wind turbines showed some promise, but at a high cost and they would never be a reliable source of base load power. The solar concepts were just laughed out of contention; with one spoof report circulated round the CEGB saying "The site near Manchester shows great promise as the world's first combined solar-hydro plant, and there would be no difficulty in operating all year round; one day of solar power, 364 days of rain collection."

Whether their perspective was humorous or imaginative, capable scientists and engineers had the task of analysing all options, and they were determined to present a fair set of reports into realistic technologies.



** This was mentioned in Quaere Caelis, but I didn't emphasise it sufficiently at the time. So, a little clarification:
In 1961, Kennedy didn't give his "landing a man on the Moon" speech. Instead he gave a more generic one, which pushed for a variety of high-tech development, including both space technology and nuclear power. As Alan Shepard had orbited the Earth first, the American public could see that their country was ahead in the space field and so some of the money that was spent on the real-world NASA instead went to other programs.

 

[Simurgh]

A greater shift to nuclear power likely does the US much more good than the expenditure from the Apollo program, so once again they're further ahead. It should also mean in general that the energy situation as the decades pass should be greater than it is today, with consequent benefits. Although given the eco-movement and the nonsense that it's permeated around nuclear energy, perhaps there would be a greater counter reaction than OTL.

 

[sts-200]

A greater shift to nuclear power likely does the US much more good than the expenditure from the Apollo program, so once again they're further ahead. It should also mean in general that the energy situation as the decades pass should be greater than it is today, with consequent benefits. Although given the eco-movement and the nonsense that it's permeated around nuclear energy, perhaps there would be a greater counter reaction than OTL.

It will certainly have had some effects on the trade deficit over the years, less imported power and the probability of US reactor exports goes up. The oil crises will still have bitten though.
There's also been less of a boom-and-bust for NASA, for good or ill. They didn't land on the Moon, but conversely they haven't had their budgets cut by 50% from the peak. Much more to come there.
Both sides of the nuclear lobby will certainly figure in the story, although it's fair to say that we have a slightly more "nuclear aware" world than in reality.

 

[sts-200]

Flightless Bird ​

At 10:15 on the 23rd March 1985, fifty-one seconds after she leapt off her launch pad at Cape Canaveral, a huge burst of flame and smoke exploded out of the side of the Space Shuttle “Falcon”. The flight was routine, the ninety-seventh Shuttle flight and the sixteenth to be wholly dedicated to carrying a DoD payload. It wasn’t a highly publicised or anticipated NASA flight, but there was still a sizable crowd of mission technicians, dignitaries and casual sightseers watching the launch, many of whom hoped that the winged orbiter would fly out of the ball of smoke, before making a quick landing, or perhaps attempting to ditch in the sea. To universal dismay, there was no sign of the orbiter, just a single solid rocket booster, still firing and gently arcing away off course. Seconds later, this too exploded into a ball of white smoke and flame.

The shock at the loss of a Shuttle, her five USAF astronauts and an unspecified payload threw the programme into chaos. Despite a high-profile public image and a record of 96 largely successful flights, the Space Shuttle was not all that it was supposed to be. The entire system was late, overbudget and operated too slowly to cope with the requirements of all of the missions for which it was intended. Behind the scenes, questions were being asked about its future even before the accident.

Investigations into the loss of the Falcon started immediately, while attempts to salvage the wreckage were still underway. The Navy cordoned off the sea area surrounding the crash, supposedly for "public safety and to facilitate the investigation". In practice, the primary objectives were to locate any remains of the crew, and to secure components of the orbiter's highly-classified payload.
Working out what had happened was not especially difficult. Ground-based cameras showed a burst of flame from the centre of the port SRB, starting at T+50.92 seconds. The explosion shattered the external fuel tank, first rupturing the lower LH2 tank, ripping the orbiter off the back of the stack and then rupturing the upper LO2 tank, leading to a violent Hydrogen-Oxygen burn barely a second after the initial event. Film and telemetry show the that shuttle itself started to roll and yaw as it broke free of the tank, and explosive modelling showed that the underside of the orbiter would have been broken and buckled by the force of the initial blast. A combination of this damage and the interplay of thrust and aerodynamic forces then caused the orbiter to break up less than a second later. Further fires and explosions can be attributed to Helium and RCS propellant tanks and lines disintegrating as the shuttle broke up. The starboard SRB broke free of the external tank and continued, unguided, for a further 16 seconds until it was destroyed by the RSO.

Working out why all of this happened would take longer, and the investigation would lead to the discovery of numerous unsettling problems within the Shuttle program. Clearly, there had been a problem with the port SRB and suspicion fell, in succession, on the N2O4 thrust-vector system feed tank, the silicon-putty joints between segments of the motors, material flaws in the steel casing and finally on the solid fuel itself. Painstaking analysis of fragments of telemetry and fragments of booster eventually concluded that there was a problem in the fuel grain. The issue lay with the presence of a void in the plasticised solid fuel. A new manufacturing technique had been introduced that saved time and money, but new tests showed that this technique tended to produce a greater number of voids and discontinuities within the fuel grain.

On OF-97, there was probably a void close to the middle of the port SRB, out near the wall of the casing. As the solid fuel burned away during the flight, this void was brought closer and closer to the combustion surface, allowing it to become hotter and hotter. As it warmed, gases trapped in the void would have risen in pressure until eventually they ignited, setting fire to the walls of the void and causing the pressure inside to spike. The relatively weak fuel matrix would have burst into the central combustion chamber of the SRB. A chunk of this unburnt fuel probably blocked the passage of burning gasses down the centre of the motor for a fraction of a second - long enough for the still-burning SRB to produce more hot gas, raising the pressure inside the motor and bursting the steel casing.

At NASA and within the USAF, the conclusion of this internal investigation was well-received. It should never have happened, but it was a straightforward manufacturing fault, primarily attributable to the use of a new process. The manufacturer would receive a rap on the knuckles, the process would be changed and everything would be just fine. They were able to announce that Shuttle flights would resume in the summer, with relatively minimal interruption to the assembly sequence for Space Station Freedom or in the launch of national security payloads.
To say that this attitude raised a few eyebrows would be an understatement. The space agency was accused of everything from incompetence to tactlessness by journalists and politicians who were hostile to the program. It seemed as if they had blown up five astronauts, and were now saying "Ooops ... carry on". The agency was obliged to announce some of the results of its investigation, while Thiokol (the SRB's manufacturer) came in for heavy public criticism, and was later sued by the families of the crew.

What did not come out at the time was the full extent of the political and technical battles within NASA, the USAF and the Shuttle program itself. From a technical standpoint, the Shuttle had been in difficulty from the very beginning. When the first flight of the “Eagle” was made in 1977, she was over a year late and close to a billion dollars over budget. There had been problems in the fabrication of the orbiter’s Titanium structure and delays in the development of the Silica tiles that protected the underside of the craft. The successful XS-20 spaceplane program in the mid-1960s had given false confidence in the ease with which the Shuttle could be developed. Many of the orbiter’s systems had proven to be more complex to build and more expensive to maintain than had been thought in the early ‘70s.

When Commander Armstrong radioed that “our Eagle has landed” as his shuttle touched down at Edwards Air Force Base, he sounded calm and confident, and the flight was lauded as a “highly successful test”. Nevertheless, both pilots and controllers knew that he and his co-pilot Bob Crippen had suffered several close calls. There had been issues with the hypersonic re-entry glide, damage to heat tiles and higher than expected stresses on the orbiter’s structure during liftoff.
These problems took time to solve, and it would be seven months before there was a second flight. In 1978, they managed to make four flights, with the Space Shuttle declared "operational" after the completion of OF-5 in November. When designed and approved, the Shuttle system was supposed to be capable of completing 25 flights per year. In 1979, there were eight flights, then 13, 15, 17, 15 and 17 in the years out to 1984. Everything took a bit longer than was planned for, and by 1982 it became clear that this situation was about to become a lot worse. The re-usable orbiters were supposed to receive a thorough refit after completing 15 flights - a refit that would see each one out of service for about 8 months. Five orbiters had been built, of which the first was of less use than the others (she was a heavier, highly instrumented vehicle, effectively a semi-prototype). All this meant that in 1984 and 85, there would be only 2 or 3 orbiters available at any one time, meaning no more than 8-12 flights would be possible during those years.

A streamlining and operational improvement plan was put in place, to gradually refit the orbiters "in service", and to cut down the time and cost of preparing the Shuttle for launch. The revised manufacturing process for the SRBs was one of the results of this program, while other changes seemed equally well-meaning but ultimately led to an even worse margin of safety than before. Systems inspections were cut back, orbiter processing and integration was sped up and worst of all, numerous short-cuts were unofficially tolerated - that is, they were not official operating procedure, but "everyone knew" that's what you did to get the job done. In addition, numerous variations in expected behaviour were treated as being normal, without proper investigation as to why the system was not functioning as it should do. If it had worked and wasn't damaged afterwards, that was deemed to be OK.

After the accident, a new risk analysis done from the ground-up by the USAF concluded that a Shuttle mission had a 1:80 chance of "total failure", and a 1:25 chance of a mission-threatening malfunction. When the Shuttle was designed in the early 1970s, those numbers were supposed to be 1:1000 and 1:200. Equally importantly, the vehicle had always struggled to meet the Air Force's needs, and with the potentially enormous number of launches needed under the nascent SDI program, it was clear that it never would. Other solutions were needed, and in the spring of 1985, the Air Force's Future Launch Vehicle program received a high priority and generous funding. Its aim would be to replace the capabilities of the Shuttle and the USAF's aging “Titan” with a range of new, more responsive, more capable rockets.

In the near future, the Shuttle was still needed and despite the controversy and internal disagreements, work to put the vehicle back into service continued apace. A record of 96 safe flights (with just one mission failure, when a cooling fault forced an early return from orbit) was an impressive achievement, and the Shuttle still had many supporters from both within the industry and in the public at large. Some changes to operational practices could be put in place relatively quickly, enough to provide an acceptable standard of safety assurance. Improvements to quality control, management structure and review processes didn’t necessarily involve complex and time-consuming hardware changes. There was a genuine desire and an urgent need to put the Shuttle back into service, but despite the criticisms of complacency that were levelled at NASA, the agency did not just “paper over the cracks”. By late summer there was sufficient technical confidence to restart the program.
In September 1985, the Shuttle returned to flight, and the orbiter “Condor” safely reached the part-constructed SS Freedom two days after liftoff. Nevertheless, without Air Force support, and with Titan's days numbered too, NASA were faced with some difficult choices.

 

[Shevek23]

I've learned a few things I did not know about solid rocket systems tonight. One I still don't know is whether the sort of failure that happens to Falcon here happens with any frequency to other solid rockets. I've never heard of this sort of failure mode described before.

You may have picked up from comments of mine on your threads or elsewhere--I really don't like hypergolic rockets much. Not when they use hundreds of tons of dragon's blood poison reactants for launch from Earth anyway. Well, I don't like solids much either. I was going to post something a long time ago about various rocket systems, but this stuff about how the Falcon's booster failed underscores that what I would really like to see is the Shuttle program, and hopefully then Pegasus, develop a reusable liquid fueled alternative to the solids.

In the course of trying to write an earlier reply, I computed the approximate size and thrust I suppose the four solids boosting your ATL Shuttle, which I recall you saying is about half the size developed OTL, and to my amazement, if you are using four solids each one is about half the size of the OTL UA-1205 which was the workhorse of the Titan III launches OTL. Since Titan III was set in motion before the 60s began, initially with the hope of launching DynaSoar from it, I would think that ATL boosters for it would be pretty similar to OTL. So, for your half-sized ATL Shuttle two standard UA-1205 should be about right sized--maybe a little too small. Why NASA and the Air Force agreed to develop a half-sized version is a bit of a mystery to me. I might speculate it was to increase the range of the Pegasus launch system, so that the ultimate reduction would be to have a small hydrogen stage atop a single standard small booster, which would be right for something about 1/4-1/3, maybe 1/2 considering the upper stage is hydrogen burning, the sort of payloads that a Titan III could lift OTL, then a two booster version would come in at say 2/3 OTL Titan III, and so on.

Anyway it seems that a single booster would average during an approximately two minute boost some 300 tonnes-force of thrust, that is 3 MegaNewtons. But solids, at any rate the OTL Shuttle SRB, would not generally be designed to give a constant boost; taking advantage of their nature the boost thrust can be tailored over the burn to an extent, by designing the grain and core plenum cut right, and I believe the OTL SRB would actually start at 16 MN (if it were in vacuum--being at sea level it only thrusts at I estimate 13 MN or so) and falls pretty linearly down to about 8 MN before burning out completely. So to match that sort of launch profile, a liquid fueled booster would need to start at around 4 MN (vacuum, down to 3.5 or less at sea level) and fall to 2 MN before burnout. To get this, one could take 4 H-1 engines from the old Saturn 1, fire them for half the time (about 60 seconds) then shut down two, leaving two to burn up the last third of the propellant over another minute. Of course in this TL by now Saturn 1 and 1B are museum pieces. OTL the leftover H engines NASA released went to Delta rockets. But IIRC, in your ATL the Thor missile was never developed...

And once again I find I misremember your TL! I probably got this mixed up with Kolyma's Shadow I guess. In your TL, Britain does accept the Thors, which are developed, and has them removed again by 1963, and that's all you said about it, and nothing one way or the other about Delta missiles. We can still presume then that spare and later surplus Thors were repurposed as OTL, though perhaps the resulting family of launchers is not called Delta? That's good news for the option of H-1 derived engines for alternative recoverable boosters, since the Delta program probably ordered upgraded engines as OTL when their NASA stash ran out.

Apparently this was the RS-27, which assuming Delta development happens more or less as OTL would be the version available around the time of the Falcon disaster. Eventually, at the end of the Eighties it was upgraded to RS-27A and shortly after that RS-27C. These engines retained vernier additions for roll control.

So, this ought to be doable. The RS-27 engines appear to have had their thrusts downgraded versus the legacy H engines, whereas they will not need verniers for the application I have in mind.

Sadly, I fear no one in the TL will take this up. Too bad, in your TL the interesting option of using hydrogen peroxide for the oxidant thus eliminating cryogenics from the boosters would be more likely to get a hearing.

 

[sts-200]

I've learned a few things I did not know about solid rocket systems tonight. One I still don't know is whether the sort of failure that happens to Falcon here happens with any frequency to other solid rockets. I've never heard of this sort of failure mode described before.

-Happily, no, not with any frequency, but it does happen. Mixing solid fuel is a fairly crude mechanical process (not unlike making a large cake) that has to be carefully controlled. Poor mixing (i.e. an area with too much oxidiser/fuel/binder) can happen, and that can also lead to problems developing as the fuel sets to a solid.

Most notably, a few months after Challenger, a Titan launch failed due to a type of void fault. If you want to see the horrible footage, this is it:

The accident I’ve described isn’t identical, but it’s related to this and a similar incident with a smaller boost motor.

You may have picked up from comments of mine on your threads or elsewhere--I really don't like hypergolic rockets much. Not when they use hundreds of tons of dragon's blood poison reactants for launch from Earth anyway. Well, I don't like solids much either. I was going to post something a long time ago about various rocket systems, but this stuff about how the Falcon's booster failed underscores that what I would really like to see is the Shuttle program, and hopefully then Pegasus, develop a reusable liquid fueled alternative to the solids.

-They have their place, but I would agree that is preferably in orbit. The problem then is all the launch preparation. HTP and N2O aren’t too bad to work with, although N2O performance is low.
Solids have their place – apart from missiles, they can be made small and still maintain a high prop mass fraction and/or thrust - but there is no getting away from their nature, they work perfectly or not at all.
I once worked on a hybrid rocket project (liquid oxidiser, solid fuel), and as far as I am concerned they are the worst of both worlds.

In the story, Shuttle development is certainly one option, if they can find the money and enthusiasm.

In the course of trying to write an earlier reply, I computed the approximate size and thrust I suppose the four solids boosting your ATL Shuttle, which I recall you saying is about half the size developed OTL, and to my amazement, if you are using four solids each one is about half the size of the OTL UA-1205 which was the workhorse of the Titan III launches OTL. Since Titan III was set in motion before the 60s began, initially with the hope of launching DynaSoar from it, I would think that ATL boosters for it would be pretty similar to OTL. So, for your half-sized ATL Shuttle two standard UA-1205 should be about right sized--maybe a little too small. Why NASA and the Air Force agreed to develop a half-sized version is a bit of a mystery to me. I might speculate it was to increase the range of the Pegasus launch system, so that the ultimate reduction would be to have a small hydrogen stage atop a single standard small booster, which would be right for something about 1/4-1/3, maybe 1/2 considering the upper stage is hydrogen burning, the sort of payloads that a Titan III could lift OTL, then a two booster version would come in at say 2/3 OTL Titan III, and so on.

Anyway it seems that a single booster would average during an approximately two minute boost some 300 tonnes-force of thrust, that is 3 MegaNewtons. But solids, at any rate the OTL Shuttle SRB, would not generally be designed to give a constant boost; taking advantage of their nature the boost thrust can be tailored over the burn to an extent, by designing the grain and core plenum cut right, and I believe the OTL SRB would actually start at 16 MN (if it were in vacuum--being at sea level it only thrusts at I estimate 13 MN or so) and falls pretty linearly down to about 8 MN before burning out completely. So to match that sort of launch profile, a liquid fueled booster would need to start at around 4 MN (vacuum, down to 3.5 or less at sea level) and fall to 2 MN before burnout. To get this, one could take 4 H-1 engines from the old Saturn 1, fire them for half the time (about 60 seconds) then shut down two, leaving two to burn up the last third of the propellant over another minute. Of course in this TL by now Saturn 1 and 1B are museum pieces. OTL the leftover H engines NASA released went to Delta rockets. But IIRC, in your ATL the Thor missile was never developed...

-The story’s shuttle uses a pair of “UA-1207B”, a hypothetical 7-segment motor based on the real-world 1207. The Shuttle was supposed to lift a payload of 30,000lbs into orbit, but they haven’t quite got there.
Quite right on the thrust profile, hence my made-up “1207B”, as a Shuttle would need a different profile, even if only to help minimise dynamic pressure loads on the stack.

If they’re going to be thinking about a liquid-booster shuttle, the design constraints would be load-dominated (unless they changed the whole design). However, a “softer” liftoff could be countered by a slightly longer burn time for the boosters, staging higher and faster in return for lower acceleration at key times.

And once again I find I misremember your TL! I probably got this mixed up with Kolyma's Shadow I guess. In your TL, Britain does accept the Thors, which are developed, and has them removed again by 1963, and that's all you said about it, and nothing one way or the other about Delta missiles. We can still presume then that spare and later surplus Thors were repurposed as OTL, though perhaps the resulting family of launchers is not called Delta? That's good news for the option of H-1 derived engines for alternative recoverable boosters, since the Delta program probably ordered upgraded engines as OTL when their NASA stash ran out.

-Delta still exists in the story, very much as OTL. We certainly haven’t heard the last of it, however don’t forget in the story the Shuttle is more successful (not better, just more successful) than the real one, It also failed on the 97th flight, not the 26th, launching a top-secret mystery payload, not a heavily publicised one.

Apparently this was the RS-27, which assuming Delta development happens more or less as OTL would be the version available around the time of the Falcon disaster. Eventually, at the end of the Eighties it was upgraded to RS-27A and shortly after that RS-27C. These engines retained vernier additions for roll control.

So, this ought to be doable. The RS-27 engines appear to have had their thrusts downgraded versus the legacy H engines, whereas they will not need verniers for the application I have in mind.

Sadly, I fear no one in the TL will take this up. Too bad, in your TL the interesting option of using hydrogen peroxide for the oxidant thus eliminating cryogenics from the boosters would be more likely to get a hearing.

-IIRC, RS27 had longer nozzles than the old H-1s, so it wasn’t really a downgrade, just that they produced less thrust at sea level in return for more performance at high altitude.

Much as I like HTP, I wouldn’t recommend it as a booster fuel. LOX is so much better, and it isn’t actually that difficult to handle (much easier than LH2, and even the nastier hypergolics).

 

[Shevek23]

...
I once worked on a hybrid rocket project (liquid oxidiser, solid fuel), and as far as I am concerned they are the worst of both worlds.

Good thing I shelved an earlier post attempt then! I was going to go on about hybrids. But really, I don't see much advantage if any in substituting a solid fuel grain for liquid fuel as well as oxidant. It might seem less complicated, and for some applications such as small handlaunched tactical missiles there might be an advantage, especially if the liquid oxidant can be just to launch and the fuel grain is later burned by hot ram air. For something like a launch system booster though, I suppose with a decision to switch to one liquid component it is on the whole better to pump the fuel as well. As noted, one good trick you can do with solids is to vary the thrust, more simply than attempts to throttle liquid fuel engines--but a throttled hybrid would introduce the difficulties of stable combustion with varying flows anyway I suppose.

In the story, Shuttle development is certainly one option, if they can find the money and enthusiasm.

-The story’s shuttle uses a pair of “UA-1207B”, a hypothetical 7-segment motor based on the real-world 1207. The Shuttle was supposed to lift a payload of 30,000lbs into orbit, but they haven’t quite got there.
Quite right on the thrust profile, hence my made-up “1207B”, as a Shuttle would need a different profile, even if only to help minimise dynamic pressure loads on the stack.

I started writing this very confused. From this post in the Selene thread and a few other responses you made to others above it, I got the impression that on the whole we are looking at about 1/2 the mass for the components. That is,

Spoiler: OTL Shuttle statistics and stuff

OTL Shuttle launches clustered very closely around 2050 tonnes on the pad, of which 1180 was the SRBs, between 35 and 27 the dry tank, typically 726 the LH2/LOX propellant, and this consistently left 117-125 tonnes for Orbiter itself, of which about 7-8 would be consumed after MECO and separation of the Orbiter from the tank for 200 m/sec final delta V to achieve initial parking orbit, typically. Thus 109-118 tonnes would arrive in LEO, though more OMS fuel might be needed to maneuver to higher orbits, and some had to be reserved to deorbit for return of course. I do not believe they ever actually achieved a 65,000 pound cargo, which would be just under 29 tonnes. Obviously that would be to the easiest orbit, 28 degrees inclination at 185 km altitude from Cape Canaveral, whereas they often aimed for higher orbits especially during construction of ISS and later crew/cargo shuttle missions there, which could only be accomplished by Discovery and Endeavour. For those missions the cargo up mass would be half this or less, and the Orbiter would launch heavy on OMS fuel instead. Now, supposing another 8 tonnes is the minimum amount of reserve propellant needed for deorbit and contingencies from a low orbit, and mass to LEO is maximized at 118, we have 110 left over, leaving 82, or 90 with the reentry fuel reserve. I think 82 tonnes dry is pushing the structural dry mass by a few tonnes at most, and the crew and their supplies and equipment slated for recovery do mass something. I've never had good luck pinning down a detailed mass breakdown for a mission that doesn't leave some variables open to speculation!

At any rate for years I've been claiming STS put "125" tonnes into orbit and clearly that has always been wrong--125 at MECO is more like it, but the plan was to fail to achieve a sustainable orbit on the main engines in order to guarantee quick disposal of the fuel tank.

Note that pushing for the 65,000 pound goal requires use of the advanced ultralight ET at 28 tonnes, versus STS-1 where the tank--with a heavy white paint job--massed 35 tonnes. There have been slight improvements in the effectiveness of the SSME and also some lightening of basic structure in the last two Shuttles built (every model built was lighter than the ones before actually) but the main thing was lowering the tank dry mass, and that was offset by the need to add some weight to the SRB dry structure after Challenger--which is to say that before Challenger, they were running on a design that was over-light since it was known in advance that the single-ring joints between segments was not fully reliable whereas the two-ring design belatedly adopted met specifications. I forget if the second launch already dispensed with the tank paint job which IIRC saved some 5 tonnes right out the gate. Tank ultra-lightening then compensated for the heavier boosters pretty much, and was necessary to meet the full spec.
s

Now ITTL, in the post I referenced you said that "not everything scaled" and that sometimes, in order to meet the 30,000 lb payload goal, which is just 13 1/3 metric tons, they'd have to use 4 SRB, while lighter launches could get away with two? That seemed very strange because with OTL STS and also apparently here, despite the fact that the main engines do light, the lion's share of sea level thrust is borne by the solids, and it seems to me varying thrust by a factor of two ought to be far too much. Granted it is only boost thrust, which only lasts 2 minutes or so, so it won't be nearly a factor of two difference in payload, but I think it works out to be overkill even so.

Anyway, if two SRBs OTL lifted one OTL STS stack, and the ATL stack masses half as much, then four ATL solids should equal one OTL SRB, and that pointed to individual solids with thrusts around 3 MN, whereas the Titan 7 segment solids you actually use have twice that thrust, leading me to wonder why you wanted half-sized boosters. Now I see you don't though.

Spoiler: An inferred model of the ATL Shuttle

In my head I worked out that the ATL Orbiter without any payload but otherwise all up for a crewed orbital mission would mass some 55 tonnes on the launch pad and at MECO, and then have to consume some 4 tonnes of hypergolic OMS propellant to arrive in orbit just over 50 tonnes. It occurred to me that if the overall thrust of the Orbiter main engines at launch were larger in proportion than OTL shuttle, that might explain why two additional solids are needed, but really that can't be because the higher the thrust, the larger the propellant mass must be and that raises the tank mass out of proportion, so the low-mass launch would not work. The J derived main engines must max out at something like 250 tonnes force of lift, 2.5 MN, and since we are told there are 4 of them, that's 625 kN for each. J-2S already got some 1140 kN in vacuum, so it would not require much improvement to meet the sea level launch spec--but it would mean that the hydrogen-oxygen mix is being consumed at a prodigal rate. Perhaps if the improvements were aimed at the opposite goals of making the engine 10 or more times reusable in one direction, and raising sea level Isp at the cost of vacuum thrust hence Isp, we find that the ratio of main engine thrust for takeoff is indeed superior? But this comes at the cost of having a tank sized for a hundred or more tonnes more than is available for the higher efficiency push to orbit, and that at a lower Isp than the J-2S 436 sec, already lower than the SSME's 453. I still think they'd do better to launch a smaller tank with the smaller thrust of just the solids alone, and air light standard, already developed J-2S engines, perhaps with their vacuum Isp nudged up a bit more. But trying to improve thrust at sea level or in vacuum will probably involve raising chamber pressures and temperatures, and this works against reuse. Starting from a J engine standard and aiming for marginal improvements, I think it can still be done more easily by far than developing the SSME, but the biggest step to easy early development is to forget ground lighting the hydrogen engines I think. Why exactly did they adopt that goal then?

Anyway with everything scaled down by 1/2 or less for the no-cargo launch, I'd think just two of the heavier Titan solids you've described would be more than plenty. After all, the solids I was looking at from OTL were made with just 5 segments--the more or less standard UA-1205. But you are starting with the seven-segment upgrade designed in the mid-60s for Dynasoar and MOL, and not deployed OTL until the late 80s in response to Challenger's loss, where the Air Force suddenly needed a heavier Titan to lift heavy NRO satellites meant for launch on the Shuttle.

Teasing inferences out of the data given by EA, I conclude that the vacuum thrust given is, as with the Shuttle SRBs and for the same reason, a surge some 20 percent above the average thrust of 5,961,027 Newton, a figure I got by noting a 120 second burn time, inferring the propellant mass, determining average mass flow from that and multiplying by the 272 sec vacuum Isp given. Comparing the sea level thrust given, dividing by the vacuum thrust and noting that the ratio is the same as the ratio of SL Isp of 245 to 272 (90 percent) I infer that in vacuum this booster would start at the given 7116.90 kN and fall down to 4805.154 over the burn then burn out. On Earth it starts at 6410.4 and presumably puts out most of the vacuum thrust at burnout; a linear interpolation would probably underestimate total impulse but gives an average of 5608 kN or 95 percent the vacuum average. These estimates are probably close enough for approximate work. The dry mass of one is 51.23 tonnes and the propellant load is 268.08 for all up mass of 319.33 tonnes; this gives an overall ratio of dry to full mass some 10 percent lower than the OTL SRBs. I estimate the average vacuum thrust of an SRB to have been something like 10,880 kN on average so the UA-1207s give 54.8 percent of the thrust of a Shuttle SRB on average and thus, just two of them should be quite adequate for an overall system 1123 tonnes on the launch pad. Two of them mass 638.66 tonnes, with the minimum mass Orbiter massing 55 tonnes, this leaves 429.5 tonnes for propellant and tank. Taking the less favorable OTL ratio of 35 tonnes out of total 761 tonnes the tank would mass 20, leaving 409.5 tonnes of propellant.

Plugging all of this into Silverbird Launch calculator, using OTL vacuum thrust of the J-2S of 1140 for each and Isp of 436, and 25 kN for the OMS, and 4 tonnes out of 55 for the OMS burn, the calculator gives 603 kg for payload, which is pretty on the nose for the stipulation that 2 solids are adequate for crew launch but not for cargo. This launch was assumed to go to minimum LEO--185 km circular orbit at 28 degrees inclination from Cape Canaveral. What this means is that aiming for any more ambitious orbit, such as 55 inclination or to a 450 km altitude, would probably be impossible without scanting some important supply, and completely impossible even pushing every reserve and tossing aside every removable weight the ship can operate without overboard before launch beyond very narrow limits. The two booster version then can barely accomplish a test flight and those only to limited orbits. Without further lightening of the ship or improved engine Isp, it simply is not practical with so few boosters.

Spoiler: Variations 1) Messing with O:F ratio and propellant mass

Of course we are told that improvements of some kind with the J engines have been made. If I kick the Isp up by 8 seconds, roughly half the difference between J-2S and SSME in vacuum, and change nothing else, payload goes up nearly 3 tonnes. A 450 kg orbit at 28 degrees with practically no payload (48 kg) is possible then, as is a 200 km orbit to 37 degrees--we cannot launch to 27 degrees though.

There is another neat trick I can think of too. I use some software designed to simulate aspects of rocket engines called RPA 1.2.8 Lite, and I noticed yesterday that although the SSME's parameters indicate a peak in ISP at an oxygen/fuel ratio of 5, of course the real engine uses 6. The Isp difference is very slight though, a matter of 2 seconds, falling in roughly a parabolic pattern. Kludging around with variable O:F ratios, I found that since the mass of propellant that stores in the tanks increases with higher O ratio (since oxygen is far more dense than hydrogen, and a higher ratio means burning hence storing and pumping more LOX and less hydrogen, that by raising the ratio we could put a given mass into a faster trajectory, since the mass ratio would improve, or the improved mass ratio could benefit us in the form of more mass, hence more payload, to a given speed. The catch is that the higher the ratio of oxygen to hydrogen (up to a point anyway, which is far beyond the cargo mass optimum) we go toward a stoichiometric mix at 8:1, the hotter the combustion chamber and other engine parts get, by hundreds of degrees. The SSME is already pushing the red line in that respect so it would be unwise to mess with it. Maybe we can ease it up on other hydrogen designs though, such as the J-2S. This is the kind of change that works against reuse of course!

Looking at OTL figures for J-2S, I believe you are mistaken, author @sts-200, to believe the S version improved on the J by raising chamber pressure--it is only 30 atmospheres, so you can see why I think improving it for ground lighting is a dubious project. I just checked, the chamber pressures were identical. What the J-2S did was raise the nozzle expansion a bit (another counterproductive thing to do for SL use) and replace the gas generator turbopump with a tap-off gas source, making the chamber its own gas generator in effect, and lower the overall engine weight.

An RPA model predicts flow separation at 8.3 km altitude; to fix that by a lower expansion ratio (J-2S is at 40) would of course lower the vacuum Isp. Raising the chamber pressure would help but probably raise the temperature too, and of course make pumping harder, perhaps to the point the simple tap-off would be no good. I again say, just air light the engines and use a lighter tank! Anyway, my goal here is to mess with the O:F ratio, and a variable to watch is the chamber temperature, which the model gives at 3318 degrees K. Another interesting pair of variables is the mass flow at the nozzle throat, times the velocity at nozzle exit, which gives a thrust index. Here those are
1289.637 kg/m^2, 4250.1146 m/sec for an index of 5481105, suggesting the throat area is 0.208 square meters
Also of interest is the molar mass in the chamber and at exhaust, 12.5623 and 13.1032 respectively.

I get a peak theoretical Isp of 436.5944 at O:f of 4.6; at 5.5 the same plot shows 434.4374. This can be modeled with a parabola of form 436.5944 - 2.663*(x-4.6)^2; this predicts the plot very well out to 6.5 and pretty well, off by just 3, at stoichiometric 8. Using a math program I can set the higher mass ratios we get with higher tank densities versus the lowered Isp using this function to get an estimate of the ratio that gives the highest payload--unfortunately the complication of two stage operation messes it up so it is just suggestive. The curve I get suggests we get maximum payload over and above the Orbiter mass at 7.3, which means that a tank that could hold 409.5 tonnes at a 5.5 ratio can hold 482.677 or over 73 tonnes surplus, an increase of nearly 18 percent. Raising the oxidizer ratio to that value in RPA and leaving everything else unchanged drops delivered vacuum Isp down to 418 as predicted, raises the chamber temperature to 3472 K, an increase of 154 degrees or a bit under 5 percent; the thrust index is 5604050 which is a 2.24 percent increase, suggesting vacuum thrust would be 1164. The molar masses are 15.1 and 16.71, meaning I believe that the molar density in the chamber is only 88.6 percent that of the 5.5 ratio nominal engine, while the exhaust product is made of denser molecules moving more slowly as we would expect.

With these masses and Isp and slightly raised thrust due to greater mass flow rates, Silverbird calculates a payload of 2888 kg versus a bit over 600. Earlier kludging like this with OTL Shuttle figures gave me more disappointing results until I scaled the thrust of the SRBs up in proportion to the increased mass of the stack thanks to the extra propellant. If we do that too, the weight is up by 73 added to 1123 tonnes for a 6.5 percent increase. Total thrust was 16482 kN, we need 1071.4 more (about 100 tonnes weight force) and we get 100 more out of the main engines due to the denser flow, so we need 485 more from each solid. In proportion, that is eight percent more, and presumably we need to kick the mass up by that much--which is about equal to adding 3/5 of another segment to make most of an 8 segment rocket. Now doing that would also raise the mass of each by 26 tonnes, nearly doubling the mass increment and sending us round for another iteration and probably another too.

Spoiler: Variations 2) making an 8 segment booster

Instead of doing that, I propose to just add yet another segment to the 7. We might guess this raises mass by 8/7 but actually the "seven" segment solid has 9 segments--it is capped by a nose segment that provides aerodynamic shaping and retains the plenum burning chamber pressure of the segments, and a tail nozzle section that shapes the flow and includes the mechanism--in this case, guided flows of N2O4 to divert the main flow, so it has tanks of that stuff--to steer. By adding another segment, we do not change the burning time from nominal 120 sec, because each segment burns independently of the others and they are all the same, and we do not change the plenum pressure--all we do is increase the volume hence mass flow of the reaction, and hence the thrust while Isp stays the same. The nose segment has no different job to do, it stays the same shape (pretty much anyway) and has the same pressure and heat to contain, so it ought to mass just the same. The additional fuel segment masses the same as all the others. The nozzle has got a bigger job to do to be sure--it needs to have 8/7 the throat area to pass more gas flow at the same temperature and pressure, and thus its proportionally shaped nozzle would be (8/7)^1/2 times longer, unless we can get away with squashing it, but we probably can't as the nozzle was carefully shaped for the specific gas flow. Thus the nozzle must mass (8/7)^3/2 as much, and we need to make the vectoring fluid tanks 8/7 as big too since the volume flow is that much greater--or maybe worse. We need to make the nozzle section 1.22 times more massive then, call it 25 percent. Well, that averages with the unchanged nose to about 1.125, depending on their relative weights--probably the nose masses less on the standard 7 segment job, and so the nozzle increase dominates. Overall the thing is probably going to increase by more than 8/7, but not much more. Let's add 50 tonnes to each, and a bit more to make then mass 370 each exactly, and then note that propellant is exactly 8/7 what it was, or 306.4 now, so each has empty mass 63.6. Assume that is really 64 exactly and we have 370.4 all up, and a thrust of 6812.5 kN each.

Putting that in for the 8 segment booster, we get a total of 4887 kg payload, which is 11,000 lb or a bit over one third the design target payload.

--------------
Now after doing all that, I was much surprised to find that doubling the number of boosters had a less dramatic effect than I thought it would--with 4 upgraded 8 segment boosters and a souped up, hot running J-2S burning a 7.3:1 ratio mix massing over 480 tonnes and getting modestly more thrust, the outcome is 20.262 tonnes of payload, which is 15.37 over the 2 booster version--it is also 50 percent beyond target! Earlier I tried the same thing, that is simply doubling the number of standard 7 segment boosters with standard J-2S and 409.5 tonnes of 5.5:1 propellant mix, and also got a 15 tonne payload increment.

But recall that then the 2 booster version was extremely marginal; only by either cutting weight off somewhere or tweaking up some parameter above nominal figures could that version achieve a decent margin for error in its design.

But indeed I have verified that in principle a Shuttle of this type can operate with extremely low cargo masses using just 2 boosters, and using 4 achieve the nominal target.

Now then, I suppose we should consider carefully which parameters the program would want to tweak to get the target.

Spoiler: Tweaking 1) bigger boosters with J-2S groundlit

Note that using 3 boosters instead of 4, with all stops pulled out--8 segment boosters, souped up K engines using more propellant mass--the payload is 13.598 tonnes, or just over the target! We can then either change the canon to say that there is a 2 solid and a 3 solid version, and perhaps an option to add 15,000 lb surplus weight somewhere in the Orbiter to achieve super-nominal missions of some kind with 4 boosters. But I suppose the author chose to ignore odd numbers of boosters for a reason, such as it requires mounting the odd one on the far side of the tank from the Orbiter, which is a big deviation from OTL designs.

The innovation of shifting the propellant ratio up to 7.3 strikes me as rather extreme and risky, since it raises the engine temperature and is the opposite of trying to get superior ISP. Perhaps we should discard this and consider what the augmented 8 segment boosters accomplish with standard J-2S engine parameters, and also perhaps consider 9 segment boosters if the 8 segments don't accomplish enough.

And I think we should look at what happens when we forego the project of a sea level burn of the J engines, and air light them instead.

First, the bigger solids with the standard J engine. By the way, the RPA program predicts sea level Isp of around 170 sec for the J engines, and correspondingly slashed thrusts too, so you can see why I want to ditch surface lighting of them. In fact I took the author's specific mention of raising the oxidant ratio to mean a modest raising of it to just 6, same as OTL SSME, and that might help a bit. However we are going to lose vacuum Isp if we do something like shorten the expansion ratio below 40, which might be vital.

OK--assuming for the moment that we can just input into Silverbird figures including standard parameters for J-2S engines that don't account for poor sea level performance, and that the calculator still gives reasonably true results--possible because the thrust contribution of the J engines at launch is relatively small (which is why I want those engines shut down completely of course)--as long as nozzle flow instabilities don't wreck the nozzles outright anyway, which is a concern considering separation happens at 8.3 kilometers-- it actually looks good. Two boosters gives us 2351 kg to LEO minimum, which suggests to me a little bit of margin for higher altitude or inclination. As a crew bus, we can get by with just 2 8 segment boosters. With 4 boosters payload to 200 km, 28 degrees rises to 16.322 tonnes, comfortably above our target of 30 k lbs. Raising the orbit to 500 km, it falls to 10.115 but that's still quite a lot, and higher than the ISS. 55 degree inclination at the same altitude can be reached with a bit over 9 tonnes, which is well over half of 30 k lbs. From Vandenburg we still have over 3 tonnes of payload to a 500 km 90 degree orbit!

Spoiler: Tweaking 2) 6:1 O:F J engine

To comply with author mentions of raising the engine propellant ratio to a modest 6, first of all this would raise the tank capacity somewhat.

Based on a 31 tonne tank for OTL Shuttles--I believe this was accomplished by removing the paint--a 22 tonne tank should hold 429 tonnes.

The Isp of a 6:1 engine will fall to 432 sec, thrust will probably rise a bit but I will just assume it stays the same. With these assumptions, the bigger boosters, using just two, get to minimal 200 km, 28 degree inclination with a payload of 1469 kg. Again this is just barely adequate but good enough for test flights. Can it reach 500 km? No. Can it reach 29 or 27 degrees inclination? No. With 4 boosters we reach 200 km at 28 with 16 tonnes; we reach 500 at that inclination with nearly 10, and 500 at 55 degrees with 8742 or almost 2/3 nominal. From Vandenberg a polar orbit at 500 can be reached with almost 2800 kg, and a 95 degree inclination with over 1900.

Spoiler: Tweaking 3) Serial Burning and why we should

Now, let us look at what can be accomplished with serial burns, first the solids to bring it up to altitude and then air lighting the J engines. We keep high density 6:1 ratio with 432 Isp but might seek to increase that with a higher expansion ratio. Since the launch is lacking a percentage of thrust due to the J engines sitting things out we compute a mass penalty first and reduce the tank mass accordingly. For a two solid burn, the penalty fraction is a quarter; for a four solid launch however it is only 14.3 percent. Also bear in mind that really the nominal thrust of the J engines would plummet at sea level, so the real penalty would be far less. But starting with 14.3 percent, that is out of an all up mass of around 2000, some 287 tonnes. Now that is pretty appalling, it would slash our tank down more than in half, so let's rethink. RPA says sea level specific impulse of the J engine is down to 117 sec or so. To be sure as the rocket ascends it will climb back to around 430, say to 400 at the time of separation, so it behaves like an average of 260, and thus average thrust during the solid burn from the J engines is down to 2740 from all 4 engines. Thus, when we look at the 2 solid configuration, the real penalty of not using the J engines is only 17 percent, and out of the two engine configuration, that is 208 tonnes. Taking that from the tank, and reducing tank mass by the 2/3 power law (assuming wall thickness does not decline, only surface area) so the tank dry is 14 tonnes and the propellant supply is a mere 221 tonnes, we fall short.

I just threw caution to the winds and iterated with different propellant masses and calculated tank masses in 2/3 power proportion to a 31 tonne OTL tank holding 726 tonnes, and around 450, with a 23 tonne tank, the stack starts to reach orbit with some margin. We can take a small mass up to a 55 inclination orbit at 500 km altitude. We cannot get anywhere from Vandenberg. But by doubling the first stage masses and thrusts, which is to say attaching 2 more solids, we jump the minimum orbit payload up to 26843 kg orover 60 k pounds, double the target! We can reach a 55 degree inclination at 500 km altitude with 18910 kg. well over nominal-obviously such missions would require a bigger payload bed or particularly dense cargo.

What is going on here is that with staged burning, the thrust of the solids is most crucial and slashing it in half with just two solids makes launches very marginal and so only a rather oversized tank can do the job of getting up to LEO at all, but when we double them the extra tank weight pays off in extra payload. Despite the greater difficulty of making a bigger tank to barely achieve a dry run test flight with just two boosters, with four we far exceed specs.

Again the reason I urge this course instead of parallel burning is that the J engines are well suited, coming right out of the box, to high altitude efficient boost, but miserable at sea level, and I think it is overoptimistic to hope to get an engine that is both efficient in vacuum and also reasonably effective--say Isp over 320 sec--at sea level, without going the route of the Shuttle SSME and working with ridiculously high pressures and all the other complex and expensive and hard to maintain nonsense the SSME entailed. I deny that the SSME was expensive because it was made to be reused--it was expensive because it was made to be a ground lit hydrogen burning engine that also delivered high Isp at altitude. Splitting the ranges in which different rocket engines work is the best approach to cheap reusability IMHO, and at low pressures, we have many good hydrogen engines to choose from with many pumping strategies and prospects for being reworked into inexpensively reusable systems. Alternatively these are also the kinds of engines that instead could be made cheap to fire once and then dispense with.

Having specified ground lighting, for reasons that elude me but I suppose the mesmerizing logic of the single stage to orbit goal had much to do with it, I fear that either you must face a much more expensive program to get a single engine type that is good at both takeoff and the long drive to orbit, or sacrifice heavily at one or the other end of the flight regime. Since flow separation involves heavy vibration of the nozzle if we must light an engine at sea level, then we had better design it not to separate on the pad, and without SSME type high pressures, this means low expansion, which costs us Isp in vacuum. If we could avoid separation with minimal sacrifices at sea level, we should concentrate on as good an Isp as we can get consistent with that. To reiterate one more time, the best way I see to do that is to simply wait until we have climbed into thin air to light the high energy vacuum optimized engine.

If they’re going to be thinking about a liquid-booster shuttle, the design constraints would be load-dominated (unless they changed the whole design). However, a “softer” liftoff could be countered by a slightly longer burn time for the boosters, staging higher and faster in return for lower acceleration at key times.

-Delta still exists in the story, very much as OTL. We certainly haven’t heard the last of it, however don’t forget in the story the Shuttle is more successful (not better, just more successful) than the real one, It also failed on the 97th flight, not the 26th, launching a top-secret mystery payload, not a heavily publicised one.

-IIRC, RS27 had longer nozzles than the old H-1s, so it wasn’t really a downgrade, just that they produced less thrust at sea level in return for more performance at high altitude.

All the homework I did above set me straight on the question of the solids, which apparently had to be about twice the thrust I guessed at in order to do their jobs.

Spoiler: Proposed LRB features

It seems that we would want initial sea level thrusts on the order of 6140 kN per booster, with this scheme of 2 boosters for minimal crewed missions and 4 for full payload loading. I would note that if it is possible to save a lot of money with fully reusable liquid boosters designed to nominally survive a boost and recovery cycle with no cumulative damage preventing it from being reused as is, for a number of launches anyway before undergoing refurbishment, and engines designed for say 10 reuse and scrapping with minor cost refurbishment each cycle, then the logic of saving money by economizing on booster use is weakened and every launch goes over to maximal use of boosters for maximum payload, which should bring down overall cost of payload per tonne in orbit.

Recognizing that the solids are designed to start with high boost and taper down, whereas it is not so easy to design rocket engines to throttle and this may make their maintenance more expensive, rather than compromise on the critical initial boost phase I would design each reusable booster to have a multiple number of engines--say four, or perhaps six. Then during the boost, shut down spare engines in groups to mimic stepwise the ramping down that solids accomplish. I was thinking four, but now am thinking six, because it makes for a closer match to the solid boost pattern to step down to 2/3, then later to 1/2, whereas if a single engine of the cluster were to fail, it might be possible to match the overall delta V requirement with fewer engines burning longer. Doing that would change the thrust relationship between separate boosters of course--which is why I was frustrated that the ATL design that began as a serial "Saturn Shuttle" degenerated into a multi-solid parallel burning thing like OTL, which I think was a mistake in both TLs! With a single booster stage, such engine shutdown workarounds redistribute one pool of propellant and don't pose the wasteful and perhaps mission failure inducing balancing acts that trying to manage two or four boosters, one of which has been forced into an off nominal burn schedule by an engine shutdown. It also means recovering one booster stage instead of two or four.

Anyway, living with the fact that a four-booster plan has been chosen and that I think a LRB for each of those had best have six engines installed, and each of those ought to be able to deliver 1023 kN at sea level--you see that once again we land in the ballpark of a 100 tonne sea level engine, right back to the H-1 or its RS-27 derivatives, or of course there are other kerlox engines even in America that can be upgraded or downgraded to that level. The next level of design would be to figure out how much propellant such an array would consume in a suitable load-matched ascent profile of about 120 seconds, which determines propellant mass, and thus minimal tankage. Then I'd recommend recovery by simple splashdown, meaning they free fall toward the ocean, deploy a parachute set, splash as gently as necessary and float, waiting to be fished out of the water--that way we don't sacrifice any performance to stuff like flyback wings, or fuel reserves for vertical return and landing, or any of that nonsense. Instead, of course, we pay some penalties: the things will suffer some aerodynamic heating which will also stress them, and then hit the water at a speed we can only somewhat limit, suffer stresses being batted about by waves and loaded aboard a recovery ship or being towed back by one. And we have to pay to maintain a fleet of recovery ships, one per booster, which will make more sense with more frequent launches, but might be a recurring cost that would be worth offsetting by flyback. I'm not at all sure that cost was ever fairly evaluated against the cost of developing fancy stuff like flyback, because the designers would want to justify doing the latter for the technical fun of it, and disdain something as low tech as hiring what amount to a bunch of fishermen. I'm prepared to be shown figures that prove the fancy flyback stuff is more cost effective, but don't think there is any reason to assume a priori that the simple splash, fish, hose off and relaunch model is going to be costly compared to the fancy stuff. Another cost involved is that deal with all that without posing a high probability of critical damage that costs a lot of money and time to fix and puts the integrity of the booster in doubt, the solution boils down to making the things strong. I envision them being made of some sort of high tech aeronautical grade steel developed for SSTs, which is why it is most interesting that in this TL the USA is heavily invested in those. Of course that also tempts them to design the booster to fly back as an SST, but if that is desired, it is best to group all the booster elements into one spaceplane! In short go back to Saturn Shuttle, but the version with wings on the booster stage.

24 100 tonne lift engines are of course equivalent to 3 F-1A, which since the "Saturn" of this ATL was smaller than our Saturn V, is exactly right sized for a single booster with 3 of those and if they like, wings to fly it back again--going back toward the "Right Side Up" concept here obviously! But it could still be a splasher after all, only now we always only need one big boat, not four smaller ones, to go chase it.

The biggest problem with a good reusable first stage of course is the logistics one. American firms are interested in turning out widgets year after year. If Thiokol had been forced to forego the OTL circling Robin Hood's barn process of "refurbishing" the SRBs they would hardly have suffered; their revenues came in mainly in the process of filling segment casings with propellant grain, and that was a simple process of make it, use it once and buy another one next time you want to launch, for as long as STS was operational--and now, if only the nation could afford frequent SLS launches (which is a laugh of course) they'd be back in business again. But if we hire Boeing to make a set of big Saturn type single reusable boosters, or four times as many smaller side boosters, after ordering enough for planned operations and a small reserve for contingencies, NASA and/or USAF takes delivery, shakes their hands, and that is that...until the projected EOL of the boosters, which considering they don't get used in succession but are staggered with other ones, could be half a decade or more hence. Assuming it was a good system that delivers on its promises and thus it is desired to continue it, then they have to go back to Boeing and order another batch, after a hiatus of years in which the workers have moved on to other things. Obviously the thing to do is design a procurement plan whereby the stages are acquired and put into service gradually, and staggered so that the rate of replacement is pretty steady, and thus Boeing's original production line is geared for this snail's pace and the workforce is kept working steadily at it, and the layout of production gear is justified in a way that building five times as much and letting it sit idle.

This means that if the project is oversold the way Shuttle was OTL, the production line will be oversized and considerable opportunity costs borne by the contractor, which will drive up the operational costs of the system, whereas if a cautious and level-headed management methodically takes a wait and see attitude, they will undersell their project, make it look less impressive, and then when it works out well the contractors will need years to expand to meet the new demand, and there will be a tendency to have a rush of procurement followed by a lull.

Much as I like HTP, I wouldn’t recommend it as a booster fuel. LOX is so much better, and it isn’t actually that difficult to handle (much easier than LH2, and even the nastier hypergolics).

Spoiler: Bottom line: HTHP systems run cooler

My major reason for thinking that HTHP might be superior to oxygen in the context of kerosene fueled reusable booster stages is the notion that if we use the peroxide to drive the turbopumps, the operating temperature of the generator system would be far lower than that of typical ker-lox gas generators, still less other systems like expansion or staged combustion. The relatively cool flow of decomposed HTHP means that the turbines capturing the energy operate at lower temperatures, hence for a given state of the art, they are easier to design, easier to construct, less subject to wear and tear and easier to maintain. We could go with relatively cheap and easy to work with materials that wear down at rates comparable to the more demanding parts of the system such as the combustion chamber and nozzle bell, so it all is ready to scrap at the same time and if turbine breakdowns loom a bit ahead of schedule, it is cheap and easy to repair or replace that part. Or we could make them out of stuff as expensive and difficult to work with as the rest of the engine, but count on these parts sailing through scarcely worn, nearly pristine when the other engine parts have to go, and perhaps swap it into a succession of engines. All of this is down to low operating temperatures.

Less dramatically so, but still significantly cooler would be combustion chambers and nozzles that run on HTHP. They too produce the same thrust, for more mass flow but at lower temperatures, so again we can either use a cheaper and/or more easily inspected and repaired lower grade material because of the lower temperature, or get more uses out of a given high grade material.

Soot of course is far less of a problem too, even using rather mediocre (hence cheap/available) grades of fuel.

It seems to me that HTHP shines when you want your rockets to just work at a low state of the art. This was a missed opportunity IMHO in the Space Race of the 50s, to get something workable cobbled together and pretty much functional before the other guys did, and it is a missed opportunity in the realm of reusable rockets.

Spoiler: Why stick with LOX anyway

Now against that some obvious points:

1) Kerlox is inherently more energetic per kilogram, and thus more efficient. Partially offset by the higher mass ratios the higher density of ker-peroxide offers but fundamentally true. Not only does one use less propellant mass, but a greater part of it is the relatively cheap fuel.

2) LOX is fairly cheap--perhaps after all cheaper even than kerosene, pound for pound, or maybe even so much cheaper that the fact that you use double the mass and more of it versus kerosene is offset by being less than half the price? Anyway, given power and engineering, anyone can have all the LOX they want, that they can pay the power and plant maintenance bills for anyway. It comes from ambient air. You just filter out the dust, compress and cool the air, and LOX comes raining out. The Redstone missile, which still used alcohol for fuel in the late 50s when it was operationally deployed, came with a set of auxiliary service trucks, and one of them was a LOX generating plant. I doubt they used it to completely fill the rocket in the field, I suspect that most of the LOX came from tanks from a central supply, but it would boil off in field conditions and I guess the solution was the LOX truck which would put out a trickle to refill the tanks with, countering boil off. So, technically, even a major user of LOX does not have to actually buy it from a chemical company, although I think most do because they can do it most cost-effectively and make up for their markup with guarantees of quality and so forth. But an operation like NASA's Cape Canaveral can invest in their own LOX extraction/purification plant and given power and maintenance, churn out as much as they like from the free air around. HTHP is not nearly so cheap; the chemical processes to make it in electronic grade (40 percent or so concentration IIRC) are complex, involving many inputs and stages, and then that product is not suitable for propellant; operations interested in peroxide as oxidant for rockets must further refine commercially available grades laboriously to get high grades, and I would aim for as near perfect purity as possible.

3) Therefore the fact that the mass ratio of peroxide to fuels is characteristically very high is a mixed issue, a blessing and a curse.

I like that it is "room temperature" but it isn't quite stable in that condition; I want to keep as near to perfectly pure HTHP as possible chilled to within a couple degrees of freezing. I have thought of nifty ways to do this, such as chilling a bunch of helium, or in a pinch nitrogen, down to that temperature and bubbling it through the storage volume, including loaded into a stage, at whatever rate it takes to maintain the desired cold temperature. This should work well even when summer day temperatures are soaring over 100 F I suppose. Well it is not a huge cost, but it is a cost, and the alternative is to risk the sorts of catastrophic cascading breakdown that deterred people from working with it in the 1950s.

 

[sts-200]

Replying in the thread to Shevek’s post (#58), as he has done a lot of analysis there and it’s well worth a read – don’t get confused though!

This will be just about the Shuttle, I’ll come back to some of Shevek's other points separately.

The closest estimate of the story’s Shuttle is the “Spoiler – Bigger boosters with J-2S groundlit”.

Unfortunately, there is a problem with all of the estimates – they assume a low-pressure J-2S engine.
J-2 ran at about 750psi, the J-2S that was tested in 1969 ran at about 1200psi. There are lots of docs on the J-2, but this is one of the best:

https://archive.org/details/nasa_techdoc_19690072871

I warn you, it’s about 1,400 pages long, but for anyone interested in J-2/Saturn improvements, it’s a good reference point, and there are lots of semi-technical graphs and diagrams alongside the details.

I’ve tweaked my fictional “J-2R” to run at about 1,350psi and accept a O/F of 6. Expansion Ratio is still 40. That will put out about 225klbs at sea level, 295klbs in vacuum. Like the Saturn, the Shuttle will use M/R shift and vacuum thrust will drop at some point in favour of increased efficiency, and to ensure simultaneous depletion of both propellant tanks.

Now for the Shuttle, which I have now looked at in much greater detail thanks to this – it was never really a core part of the Selene story:

In brief, it can put 27,500lbs (plus a small crew) into a 180km Earth orbit – so it hasn't met its design spec.

In that mode, orbiter has a wet mass (ex fuel & payload) of 53.5t, plus 4t OMS fuel. It has four J-2R engines, which are ground-lit.
If they want to go to the space station or any other higher orbit, the payload has to be smaller. That might seem a bit light, but the story’s shuttle has a Titanium structure, which would be slightly lighter and would also save weight on the TPS relative to the real design.

The external tank has a burnout mass of 25t and carries 610t propellant (607t by the time of liftoff). As you will see, that is quite an aggressive design, but the load paths are somewhat more benign than the real one (I think I hinted at that somewhere in Selene, but it probably wasn’t very clear).

The fictional “UA-1207B” boosters are where we run into greater differences. They use a slightly higher performance propellant, operate at a slightly higher pressure than the real 1207 and have a longer nozzle. I modelled c* of 1548, PC=955psi, E.R=11.7 and an HTPB-based propellant (these were cutting edge, but were considered for the real shuttle). SI is 241 at sea level, 269 at altitude.
259t of propellant in each of the two boosters, with a burn action time of 112s. The booster casings are not recovered.
In short, the 1207B is the same as its real-world cousin in name only. However, that’s quite inkeeping with the tricks and fiddles that would be needed to get it funded…
“Oh yes, Mr. Nice Congressman, it is the same broom, all we’ve done is replace the head and the handle…”

The attached sheet is a simulation of a ULEO launch. I suggest you don't bother with it unless you are very deeply interested. It’s probably almost completely incomprehensible as I wrote the sheet for my own use a very long time ago. It's not a fully optimised trajectory, but cell AD510 shows the burnout velocity – 7758m/s at 177km with a small climb rate (that is a slightly sub-orbital trajectory).

So in summary, my Shuttle never quite met its design specs, and they should probably have gone with a four-booster version, but that’s cost saving over-optimisation for you.

As to future upgrades, there are plenty of possibilities, some of which may yet feature in the story…

 

[sts-200]

Bottom line: HTHP systems run cooler"]My major reason for thinking that HTHP might be superior to oxygen in the context of kerosene fueled reusable booster stages is the notion that if we use the peroxide to drive the turbopumps, the operating temperature of the generator system would be far lower than that of typical ker-lox gas generators, still less other systems like expansion or staged combustion. The relatively cool flow of decomposed HTHP means that the turbines capturing the energy operate at lower temperatures, hence for a given state of the art, they are easier to design, easier to construct, less subject to wear and tear and easier to maintain. We could go with relatively cheap and easy to work with materials that wear down at rates comparable to the more demanding parts of the system such as the combustion chamber and nozzle bell, so it all is ready to scrap at the same time and if turbine breakdowns loom a bit ahead of schedule, it is cheap and easy to repair or replace that part. Or we could make them out of stuff as expensive and difficult to work with as the rest of the engine, but count on these parts sailing through scarcely worn, nearly pristine when the other engine parts have to go, and perhaps swap it into a succession of engines. All of this is down to low operating temperatures.

Less dramatically so, but still significantly cooler would be combustion chambers and nozzles that run on HTHP. They too produce the same thrust, for more mass flow but at lower temperatures, so again we can either use a cheaper and/or more easily inspected and repaired lower grade material because of the lower temperature, or get more uses out of a given high grade material.

Soot of course is far less of a problem too, even using rather mediocre (hence cheap/available) grades of fuel.

It seems to me that HTHP shines when you want your rockets to just work at a low state of the art. This was a missed opportunity IMHO in the Space Race of the 50s, to get something workable cobbled together and pretty much functional before the other guys did, and it is a missed opportunity in the realm of reusable rockets.

Its true that HTP turbines can run cooler, but it isn't that much more difficult to run fuel-rich to achieve the same goal using the rocket's own propellants. Whatever temperature you design for, you need a good safety factor on thermal and mechanical stress, so seals and bearings tend to be the limiting factors in turbine life.
I agree that it had some relevance in the 40s and 50s (Redstone, R-7, Black Knight etc.. did use HTP in this way), but if you can build higher temperature turbines, you want to do so - turbine power can be a significant limiting factor in engine design.
HTP is a great coolant, but keeping combustion chambers and nozzles cool isn't a huge problem with Kerolox until you get to very high pressures. "Mediocre" fuel is not something you want in a rocket anyway, it is cheap because it is impure and tends to contain odd Hydrocarbons and Sulphur. Those can cause problems in the cooling channels or chamber.

Why stick with LOX anyway"]Now against that some obvious points:

1) Kerlox is inherently more energetic per kilogram, and thus more efficient. Partially offset by the higher mass ratios the higher density of ker-peroxide offers but fundamentally true. Not only does one use less propellant mass, but a greater part of it is the relatively cheap fuel.

2) LOX is fairly cheap--perhaps after all cheaper even than kerosene, pound for pound, or maybe even so much cheaper that the fact that you use double the mass and more of it versus kerosene is offset by being less than half the price? Anyway, given power and engineering, anyone can have all the LOX they want, that they can pay the power and plant maintenance bills for anyway. It comes from ambient air. You just filter out the dust, compress and cool the air, and LOX comes raining out. The Redstone missile, which still used alcohol for fuel in the late 50s when it was operationally deployed, came with a set of auxiliary service trucks, and one of them was a LOX generating plant. I doubt they used it to completely fill the rocket in the field, I suspect that most of the LOX came from tanks from a central supply, but it would boil off in field conditions and I guess the solution was the LOX truck which would put out a trickle to refill the tanks with, countering boil off. So, technically, even a major user of LOX does not have to actually buy it from a chemical company, although I think most do because they can do it most cost-effectively and make up for their markup with guarantees of quality and so forth. But an operation like NASA's Cape Canaveral can invest in their own LOX extraction/purification plant and given power and maintenance, churn out as much as they like from the free air around. HTHP is not nearly so cheap; the chemical processes to make it in electronic grade (40 percent or so concentration IIRC) are complex, involving many inputs and stages, and then that product is not suitable for propellant; operations interested in peroxide as oxidant for rockets must further refine commercially available grades laboriously to get high grades, and I would aim for as near perfect purity as possible.

3) Therefore the fact that the mass ratio of peroxide to fuels is characteristically very high is a mixed issue, a blessing and a curse.

I like that it is "room temperature" but it isn't quite stable in that condition; I want to keep as near to perfectly pure HTHP as possible chilled to within a couple degrees of freezing. I have thought of nifty ways to do this, such as chilling a bunch of helium, or in a pinch nitrogen, down to that temperature and bubbling it through the storage volume, including loaded into a stage, at whatever rate it takes to maintain the desired cold temperature. This should work well even when summer day temperatures are soaring over 100 F I suppose. Well it is not a huge cost, but it is a cost, and the alternative is to risk the sorts of catastrophic cascading breakdown that deterred people from working with it in the 1950s.

LOX is very cheap (cheaper than bulk fuels like Propane/Gasoline, much cheaper than RP-1) and/or DIY as you say. However, rocket engineers will be happy people when they have to start worrying about the cost of the fuel.
If using chilled HTP, why not just use LOX? Its a better oxidiser and the moment you are into on-pad refrigeration you are dealing with a host of extra problems. Chilled HTP is also a double edged sword, because it doesn't decompose quite as readily once it reaches the catalyst.

As you say HTP does decompose naturally, and that is affected by temperature, but even when warm it isn't a catastrophically high rate unless you confine it. There are also chemical inhibitors which can reduce it.
Most of the problems people had in the early days were due to poor cleaning of HTP storage and lines. If the stuff comes into contact various materials (usually oils and lubricants), it can decompose and burn. Even a thin film of oil on a tank surface, or grease leftover in a joint could trigger that. Unfortunately that left HTP with a reputation it doesn't deserve.
All that is also is true of LOX, and most other oxidisers have similar issues.