WI: Carter lets the Space Shuttle die

...Will point out again that a 'lighter' Orbiter in fact does not need the tiles which was actually a known thing but as designed they couldn't make it lighter as long as the engines were attached. Then finding out the engines made the Orbiter fly even worse...
First of all, to get this out of the way, I am very much in favor of an Orbiter that does not have the main boost to orbit engines attached, but that idea was rejected out of hand for STS, was it not? How late was it seriously considered as an option?

A suitable set of J-2S engines would mass less than the 3 SSMEs I believe, by a few tens of percent, but not make a tremendous difference. Putting them in a separate package would probably wind up with a net higher weight, though maybe not with synergistic reductions in Orbiter weight.

Now--much more important to me (though a case for STS being designed with a separate engine package that might have been accepted in early 70s is very very interesting too)
....what's all this about the tiles not being needed? I am fascinated, and must have missed any detailed or extensive mention you made of the tradeoffs available.

Recently rereading Wikipedia on the Shuttle TPS, if I read it correctly the whole TPS system--all of it, all layers, top and bottom--masses a total of 8 metric tonnes. If we consider downmass as the limit, the TPS mass could be raised as much as 10 more tonnes at the cost of totally eliminating downmass without making the returning Shuttle any heavier. If transpiration is in the mix, and the mass difference is the supply of fluid (water, I would think) to be boiled away, getting rid of all of it high up could have the Shuttle heavier on entry (at the cost of an equivalent amount of up-mass payload) but no heavier with a complete downmass payload in subsonic glide and landing, which might work if NASA could face the payload cost.

All the discussion of alternative TPS I ever saw implied that the mass of a more conservative (and robust!) metal shingle alternative would be far greater, enough so to eliminate the payload, so I have gone with the assumption this was the case.

For the shuttle being lighter allowing elimination of tiles, I suppose that means the effective high speed surface area remains the same while the mass goes down, so it is not a matter of a dimensionally smaller shuttle, which is where "synergistic" weight savings come in. Of course a smaller shuttle in cross section would also mass less so TPS just as dense per square meter would weigh less overall, and thus perhaps a heavier per square meter alternative might be considered.

Can you point me to where the tradeoffs in total mass required are discussed in more detail?

Because meanwhile I've had another idea I've wanted discussed critically, for some time, relating to alternate materials and other improvements in TPS logistics while making them safer from damage.
...
The key assumption is the only way to get the engines back is they have to be IN the Orbiter which was an "original" decision but as was pointed out by Rockwell et-al wasn't as much a given as had been assumed. Putting a pod under the ET was considered and studied
How late? it seems like you are talking about fasquardon's time frame, the Carter Administration. That late?
...We can 'fix' it later since they gave us the money this time... But that never really was an option again.
But no technical reason, just a matter of political will? That's what ATLs are for!
A lot like the Shuttle/Orbiter we know but with an external recoverable engine pod and tail covering assembly, possibly with a couple of jet engines in it and some kerosene. Lighter Orbiter without tiles, (but would still have the carbon/carbon panels) with maybe a titanium or metal composite TPS. Extended on-orbit time due to more on board consumables since the Orbiter itself is lighter by a lot. Still have most of the same abort issues/options though the jets would be a benefit.
That sounds absolutely wonderful! With Orbiter already separate from the main engine set, both being recoverable, replacing the orbiter with a faired one-way disposable payload bus to lift cargoes comparable to Skylab to orbit in one shot, that is a Shuttle C, is pretty simple; the major work to be done OTL is already done--developing the separate recoverable engine module that is.

Now the problem is that we have greatly raised the payload mass--yay!--and hence the cost of a tonne to LEO is slashed by the ratio of orbiter payload to Shuttle C payload--huge, factor of 2 at least, more like 3 or 4 probably--so yay!

But--boo! "no one wants such a big payload." Well, shucks, every now and then someone surely does. And maybe commercial customers can be persuaded to batch a whole bunch of competing payloads into one big bus for the sake of the cost discount? Unfortunately customers are liable to want their payloads to go to different inclination orbits...sigh. And batching is a hard sell.

So once again I say--STS is oversized, once you sidestep the inefficient Orbiter. Solution--downsize it! Assemble 3 segment, even 2 segment, short SRBs from the same modules used to make full sized Orbiter 4 segment units. The fifth top section, the nose cap, is identical in function though if it contains a parachute that can be lighter for the short rounds. If we eliminate the farce of "recycling" the SRBs and just let them splash and sink, we can save the parachute mass for all launches by reducing it to zero! The sixth section (on a standard 4 seg booster) is a specialized nozzle, to handle 3/4 or 1/2 the standard mass flow; the nozzle and gimballing actuators are scaled down hence lighter. Perhaps the nozzle section alone should be recovered and reused? Even sacrificing it, by anecdote it is claimed the cost of refurbishing was equivalent to the cost of fresh manufacture.

I fear a single section booster is far too small, but I think 2 section is about right for a single SSME, and 3 for a two-SSME module. So we have to design those smaller engine modules too.

But the upshot is a mini-Shuttle C, that uses fewer of the same elements as needed for standard STS, and if the pace of refurbishing returned components and integrating in one-shot components can be fast and inexpensive enough, turnaround time fast enough and cheap enough, then the smaller payloads (on a Shuttle-C basis--if you want downmass, send up an Orbiter) would be the right size for the existing market, costs might be significantly lower than ELV--this gives the infrastructure making SRB parts and tanks (oh, yes, mini-Shuttle C requires a smaller tank design too, but once designed and tested should be no problem to assemble at Michoud, and to ship with existing full size tank infrastructure) and doing engine refurbishment a steady flow of revenue work, to lower unit costs for each element and maintain the infrastructure needed for the big payload missions--Orbiter launches and heavy payloads in the 60-100 tonne range.

Mini Shuttle evolved from STS Orbiter capable components is not optimal, but it might be plenty good enough and save money on an overall national launch system that can monopolize American launches, undercut all but the best and most visionary foreign designs, lower the cost of payload tonnes to orbit (not as much as promised but far better than delivered OTL) and free up NASA budget for actual space missions of all kinds, robot and crewed.
Plus sides are greatly decreased refurbishment time though you have a greater recovery and processing cost and path due to the engine pod.
I believe you, but perhaps people will be skeptical?
...Upside is you have a modular "Transportation" system rather than just a manned LV....
This excites me the most, Shuttle C, and if necessary min-Shuttle C, for the win!

Now I am going to post on my wacky TPS suggestion.
 
OK, as we know OTL the Shuttle TPS proved disappointing in that it proved dangerously fragile. This meant a lot of refurbishment time and cost in replacing blown off tiles, inspecting those that remained for damage and replacing them, and withal once they were so damaged, the reentering ship was lost with all hands. And in fact there had been another mission--the first or second after the Challenger stand-down IIRC--where the damage to the tiles was done not by foam from the ET but by a piece of one of the SRBs (!) and was just about as bad, in terms of smashing holes in the TPS layer, as what did in Columbia, but fortunately the burn through happened at a location in the aluminum frame with doubled gauge thickness, due to a conduit or hatch or something there, and the thicker metal survived. Though it did warp! Imagine if the damage had been shifted a few feet and that Shuttle had been lost just after Challenger!

Metal TPS would seem an obvious answer, but I have the impression that it would have to be considerably greater in mass per square meter. From RanulfC's post this might be far less true than I have been supposing, but grant that every kilogram of weight it adds, is a deduction from payload, both upmass and downmass.

But here's my idea. If we can't replace the ceramic tiles with solid metal, can we protect them with a very thin layer of such high temperature metals? Can suitable metals that are strong and tough at low temperatures, so falling chunks from the tank or SRBs that hit them don't dislodge or shatter the fragile tiles, be heated to typical tile outer temperatures and remain solid--here they don't have to be tough, because there are no solid projectiles impacting them, just plasma-heated sheets of gas; they just have to keep their shape and hold together. Inside the metal outer layer, which is so thin there is little temperature gradient between its outside and inside, the same sorts of ceramic tiles we used OTL are packed. I believe it would not be possible to bond them to the outer layer, because with being heated, their outer parts will expand, and so they cannot be rigidly attached to the outer sheet of metal. Though each might have a dimple with a pin on the inner layer anchoring it, or even 4 or 5 pinned dimples so they are held in place that way. The tiles, as OTL, get extremely hot on the outside but remain cool on the inside, which surface then can be spaced away from the inner hull by high temperature honeycomb (in case of heat leakage due to damaged tiles) that serves to press the tiles against their outer retaining pins, holding them in place and allowing pressure on the expanded tiles touching the outer metal to be transmitted as aerodynamic force to the inner airframe. The tiles then are not bonded to the hull at all, but sandwiched between inner hull and outer metal layer. The outer metal layer then is constructed in large segments, each containing a great many tiles, and secured to the inner hull with bolts. At launch, and when the tiles have cooled during subsonic descent after high speed entry, the cooled outer metal is a firm, rigid slab that transmits air pressure on it to the outer frame of each metal slab, thence to the attachment bolts and thus via this grid to the inner air frame. During entry, the outer metal is heated up, very quickly reaching temperatures inside and out near or at the peak temperatures the tiles must handle; the heat radiating on the interior to the slightly separated tile surface heats the tiles surfaces up and causes them to expand into contact with the outer metal skin, and transmit air drag pressure more smoothly through the tiles to the airframe. The tiles cannot radiate away heat, but the outer metal layer can and does, so if it can survive the peak entry heating, it starts to cool, allowing the tiles to also cool. As this happens the tiles shrink back away but now the metal is somewhat stronger.

Upon landing and refurbishment, perhaps, if there is some way of verifying that each tile within an outer slab came through undamaged, we can forego visual inspection and just leave the slab in place for the next flight. If there is some indication of damage, or there is the possibility with no way to be sure mandating total inspection, instead of quick visual inspection of each tile followed by very slow painstaking individual tile replacement, there is an order of magnitude fewer metal slabs to unbolt, lower down and look at. Any slabs with bad tiles in them can be swapped out for an alternate set made in advance and warehoused for the purpose; the new slab gets bolted back in place and the next panel that needs to come off (hopefully not every one) can be turned to. Slabs with damaged tiles are inspected for refurbishment candidacy--if they are OK themselves the bad tiles get removed and replaced, and the slab is put into storage as replacement for a future refurbishment, while it is known that the Orbiter will go up with good tiles in good slabs. The tiles are protected from most damage, and when perhaps shattered anyway, the fragments are still more or less in place to serve as damaged but still largely functional insulation on reentry, to be replaced after the mission, on the ground. Since there are far fewer slabs than tiles, even if it is necessary to remove and inspect every one, and that the removal process is comparably long to individual tile replacement, overall the labor and time involved should both be much reduced, with much of it deferred to gradual work on the removed slabs and preparation of new replacement sets if the old ones are too damaged.

Will this system necessarily be heavier overall? We are replacing naked tiles and some glue with a layer of metal, auxiliary pins, flanges and bolts on the metal, and an inner layer of honeycomb to assist in keeping the tiles from rattling around and transfer any aerodynamic pressure from them to the inner hull. All that is additional, the epoxy we have eliminated probably massed very little, so it ought to be heavier. But maybe we can make the tiles a bit thinner, with their outer layer being a bit cooler due to heat gradient across the outer metal, and the inner layer being separated from the aluminum by a vacuum gap maintained by the thin honeycomb layer, so we might be able to offset some of this added mass.

Say we add 50 percent, or 4 tonnes; this lowers downmass limits from 10 to 6 tonnes, and pares 4 tonnes off upmass too. This is bad for Shuttle economics, but it was a failure on that front anyway OTL. It puts limits on the mass of things like Spacelab, but the crew capacity remains as OTL.

With only one mission lost OTL out of a hundred and more due to TPS failure, it might seem not so cost effective even with foresight of that one loss throwing cold water on the glad-handing optimism of NASA "Success Oriented Management." Knowing that a second Orbiter came literally within inches of a similar disaster might ruffle the bean-counting feathers a bit.

The biggest selling point to NASA management would be if the slab-based multiple tray of tiles system of managing tiles is a big savings of manpower and time toward turnaround or not. If it is, then the money and time saved on each cycle of refurbishment brings the Orbiter closer to its (unattainable) program goals, and this lowers the cost of each launch.

Bear in mind that I see this as only a partial solution; the real keys are to go over to Shuttle C and probably to develop mini-Shuttle C, with this sort of slab-based, metal covered TPS serving on engine recovery modules as well as on Orbiters.
 
Last edited:

Archibald

Banned
Not sure about that as Nixon wasn't convinced that Titan/Gemini would address the aerospace program issues. Frankly he had aerospace executives visiting him telling him anything less than a 'new' development program and most of the aerospace industry is going out of business. Having said that I suspect Bush would have been forceful enough to MAKE NASA consider ALL the options from the start rather than making the pre-assumptions they did. AND he might have already been 'in-the-know' enough to realize that the head of the NRO was telling NASA they didn't actually NEED to follow the Air Force requirement as strictly as they did since the Air Force wasn't the "real" customer!

Nixon didn't wanted to handle NASA by himself so he instead got both OMB and PSAC reviewing Shuttle designs. OMB worked on costs, PSAC - on engineering (PSAC: President Science Advisory Committee).
What happened was that PSAC and OMB had conflicting opinions about what shuttle was the best - or no shuttle at all. In my TL Explorers it didn't took much for PSAC to finally convince OMB that the large shuttle was too expensive and that more generally the space shuttle as a whole wasn't needed. My source was Tom Heppenheimer SP-4421 "the space shuttle decision". Since then John Logdson "After Apollo" shed a new light on the process and after reading it on Google Books it more or less confirmed my analyzis. That is, the space shuttle got an extremely close brush with death on late October 1971.

OTL the shuttle fate hanged to what I call the "fat DynaSoar" launched by a Titan III-L. The "fat DynaSoar" saved the shuttle because it had wings and a payload bay. Big Gemini had no such things.

A panick stricken NASA finally listened Mathematica and, frantically working from the "fat dynasoar" they managed to salvage a full scale orbiter (15*60 ft payload bay) out of it, and after three months Nixon screwed the OMB and funded it. Of course the full scale orbiter was saved at the expense of the booster and a large external tank.

In my TL I just had NASA (George Low) screw the "fat dynasoar" in September 1971. Hence when the shuttle got it close brush with death in October the "fat dynasoar" is nowhere to be seen and Big Gemini is picked by both OMB and PSAC, and then to Nixon.
 
Last edited:
Nixon didn't wanted to handle NASA by himself so he instead got both OMB and PSAC reviewing Shuttle designs. OMB worked on costs, PSAC - on engineering (PSAC: President Science Advisory Committee).
What happened was that PSAC and OMB had conflicting opinions about what shuttle was the best - or no shuttle at all. In my TL Explorers it didn't took much for PSAC to finally convince OMB that the large shuttle was too expensive and that more generally the space shuttle as a whole wasn't needed. My source was Tom Heppenheimer SP-4421 "the space shuttle decision". Since then John Logdson "After Apollo" shed a new light on the process and after reading it on Google Books it more or less confirmed my analyzis. That is, the space shuttle got an extremely close brush with death on late October 1971.

OTL the shuttle fate hanged to what I call the "fat DynaSoar" launched by a Titan III-L. The "fat DynaSoar" saved the shuttle because it had wings and a payload bay. Big Gemini had no such things.

A panick stricken NASA finally listened Mathematica and, frantically working from the "fat dynasoar" they managed to salvage a full scale orbiter (15*60 ft payload bay) out of it, and after three months Nixon screwed the OMB and funded it. Of course the full scale orbiter was saved at the expense of the booster and a large external tank.

In my TL I just had NASA (George Low) screw the "fat dynasoar" in September 1971. Hence when the shuttle got it close brush with death in October the "fat dynasoar" is nowhere to be seen and Big Gemini is picked by both OMB and PSAC, and then to Nixon.

Now I am confused.

It is hard for me to see exactly why the concept of a winged reusable manned vehicle with a payload bay depended on one USAF backed concept that was, OTL, still going on paper in October 1971 (if I read you right) and would be sunk if NASA (not DoD or the White House, who could "kill" an Air Force project) if someone cancelled it just one month before a crucial decision in October of that year.

To begin with, I thought Robert McNamara killed Dynasoar Mark 1 long long before 1971.

And as DoD chief, I think he did right, given that the Air Force could not describe an acceptable military mission for it. What should have happened I think is for Dynasoar to be transferred to NASA, since it was not of particular military use (if the Air Force had defined even one distinctly military mission for it not better accomplished by other means, I believe McNamara would have signed off on continuing it in the Air Force, and of course it would then require Titan II to launch it.

As a research vehicle investigating "airplane like" low G (1-3 G versus 5-12) hence long duration atmospheric reentry from orbital speeds with high lift to drag ratio, I suppose even if Von Braun et al regarded it as a distraction from their crewed space projects, with a White House mandate and additional funding attached transferred from DoD budget to NASA, it would find a home somewhere in one of NASAs fiefdom centers, and with NASA "stuck" with it from Von Braun et al's views, they'd come around to launching it with Saturn 1 or Saturn 1B, whichever was required--thus justifying acquiring extra Saturn 1/1B orders and perhaps doubling or tripling the total number of Saturn 1 type rockets acquired and launched versus OTL.

Increasing Saturn 1B orders might in turn tip the balance toward NASA working with EELVs instead of taking on the ambition of developing something reusable--alternatively perhaps someone would look in to the idea of reusing the Chrysler made first stage.

Anyway OTL it was the Air Force's baby and when McNamara took it from them it just died. So--at what point did the Air Force reacquire a mandate to do any level of research, even just paper studies, on a new upgrade of the Dynasoar, such that there was something for Low to kill or let continue at NASA? Was this a case of someone at NASA, one of those other centers not in line with VB and company, adopting Dynasoar as a cool idea they wish someone had done, and doing paper studies (presumably with Titan launchers, not Saturn, since all this is happening under the Huntsville mafia's radar I guess)?

Anyway, what difference does it make to the Shuttle Decision whether Fat Dynasoar existed at all?

I am assuming it was a paper project, an investigation into "what if" by a bunch of engineers more preoccupied with hypersonic aerodynamics than with NASA's main focus on craft designed mainly for vacuum operations with minimally simple ballistic capsules of increasingly known performance. Was it more than that OTL? If it were escalated to the level of authorizing building hardware, presumably scale model entry vehicles launched on off-the shelf available rockets, I would think it would have been a big big deal administratively. IIRC there were indeed tests in the later 60s of various non-capsule lifting bodies and/or winged forms for entry characteristics, done with rocket-launched scale models, from this series the eventual bi-national evolutionary course of the HL-20 lifting body form evolved. (The Soviets stole one of the American designs that was missile tested for their MiG "Sabot" spaceplane orbiter, then Americans observing Soviet tests used the photos and data we got from spying on the Soviet test for defining HL-20 in the late 80s). Was "fat Dynasoar" one of the forms tested in that series?

In any event, what is one more paper study, more or less, especially in the context of actual testing being done on various forms?

Everybody and their brother in the aerospace biz had their own Shuttle ideas before the OTL interagency showdown around NASA's high budget expectations versus Congress's "been there, done that, now shut down so we can say we saved the taxpayer some money in this economically stagnating, costly Vietnam war year" mood. As I understand it, Nixon himself had no great desire to slash NASA back--though I bet he did have a preference for NASA to shift over to doing something quite different from the Kennedy-Johnson Apollo/Saturn legacy. He saw himself as caught between any President's desire to have the public associate him with cool stuff versus Congress's budget hawk mood--with his own OMB tending to side with Congress on the matter to be sure, and NASA with its hyperbolic expectations of continuing peak Apollo funding forever forcing him to downgrade his esteem of their managerial responsibility. But had Congress been willing to fund NASA mad money with no questions asked, I suppose Nixon might have wanted to go with Von Braun's Mars scheme or some such; why not back a legacy that means quite different hardware than the Kennedy legacy Apollo, on an even grander scale? Not because Nixon didn't want it--because Congress was not going to fund it and he had higher priorities to get funded so he wasn't going to bat against Congress on this.

So--all the aerospace firms had plans for some sort of winged spaceplane with a cargo bay. (Except Chrysler, which made the launcher the focus of their reusable concept, seeking as I would to decouple the orbital payload into a generic package of given mass--but even they offered a winged spaceplane as one such possible package)! Archibald, you say that if one particular such plan--Fat Dynasoar--had not been an active program in October 1971, Low et al would be forced to fall back on Big Gemini. I just don't see that. Sure, the FG might have been the closest single design to the bare bones minimal spaceplane NASA dared to consider downgrading to. But the psychology at work here is the notion that some sort of spaceplane, specifically one in which both crew and cargo ride on, in analogy to cargo airplanes, must be made reusable and be reused, and if Fat Dynasoar was convenient to point to OTL, its absence would just prompt Low to have the appropriate center (or perhaps an inappropriate center!) brainstorm a quick sketch of something suitable--able to hold both a crew and a cargo of useful size, keep the crew on orbit for a week or two, aerodynamically reenter and land as a glider. Given that Fat Dynasoar or no Fat Dynasoar, in addition "real shuttle" designs I believe every aerospace firm anyone could name off the tops of their heads (and quite a few they couldn't) had at some point between 1960 and 1970 sketched out some suitable super-Dynasoar type thing, the NASA team mainly needs to ransack the dusty file drawers full of proposed projects to pull out five or ten of these sketches, and pick the one or two that look closest, and splice them together on the back of some envelopes to have something quite as good, once NASA artists are done with their part of the job, as Fat Dynasoar to put before OMB and Congressional delegates.

Now as it happens I think this is unfortunately silly; in retrospective hindsight, this is not what the USA needed done at all.

But let's remind ourselves of the context of conventional wisdom at the time about future generation cheaper access to space, the absolute ideal would be something that would look, from the outside anyway, either like Skylon or like Venture Star. That is, one single hull that contains within itself a) all fuel b) all engines needed to go from the ground to LEO; c) all cargo d) all crew desired to be put into space; upon reaching orbit, can then e)putter around on OMS (or ambitiously a subset of the main engine thrust) to orbital destinations or serve as a platform for launching deep space vehicles (to GTO or translunar or beyond to solar system destinations)--note it is doing so with all tankage and engine mass needed to reach orbit in the first place--then f) retrofire for g) aerobraking and some kind of precision controlled landing of the whole thing, including crew and downmass cargo, to a base suitable for minimal inspection, refurbishing (on the order of checking the oil and refueling and loading in the next batch of payload and passengers/crew to be launched again--and on the assumption that this ground servicing and general turnaround is comparable to an airplane's typical in service turn around time (h). With assumptions like that, it is quite OK if the total upmass and downmass is something in the range of 5-20 percent of the dry mass of the whole thing, thus 0.5-2 percent of the total fueled mass.

Now, today we recognize that doing this on chemical fuels (barring insane risks like using FOOF or other such foolishness) was not only insanely ambitious for 1970 state of the art tech but probably nothing we can accomplish even today, or in the foreseeable future. But in the 1960s designers like Philip Bono churned out a dozen or more systems meant to more or less meet this target, and with million pound payloads too. He may have been nuts but most everyone assumed doing it would be a matter of will; if they agreed it was too much for NASA to commit to in 1971 it was because it would cost too much at that time, given a parsimonious Congress.

OK, so the designers took one step back, and dropped criterion a&b and compromised with a two-stage design. Now engines and fuel would be split between two (sometimes more) hulls where all but the Orbiter would not carry any cargo and would not achieve orbital speeds, but would after boosting the Orbiter partway, separate and return to launch site, certainly by aerodynamic flight, landing like an airplane and jumping to h), being prepared in airplane like turnaround times to boost another Orbiter. Meanwhile the Orbiter would otherwise meet all the criteria, because dang it, affordable access to orbit would look like airplane operations and those specs above are designed around airplane operations.

And except for a few mavericks like Chrysler, this is what everyone meant going into 1971 by a "next generation Shuttle."

Then OTL, the still skeptical and tight fisted Congress and OMB balked at the estimated cost of such a two-piece TSTO Shuttle. No one doubted it could be done at all though in retrospect we now figure this too would have cost orders of magnitude more than estimated to accomplish and probably would have failed at that. But, in the desperate place NASA management was backed into, they had to postpone this ambition too, and look around for approaches with lower up front investment to get some approximation of what they wanted.

OTL, TAOS brought in two more compromises--1) eliminate the integral fly back single booster stage, replace it with SRBs that would allegedly be refurbishable with savings but turned out not to save much if anything over the cost of just making and expending new ones for every launch. 2) separate the reusable spaceplane Orbiter that otherwise stubbornly retained all aspects of the desired SSTO ideal super shuttle from its orbital boost fuel supply, which would go into an expendable tank.

Having thus sacrificed a) from the SSTO wish list completely, retaining only OMS fuel for the spaceplane, and with the b) compromise shifting the vast majority of surface launch thrust to the SRBs, and simplifying the "flyback" of the boosters to parachute braked splashdown in the ocean and recovery by boat downrange, NASA then dug in its heels and refused to sacrifice c)-h). Or rather, as you allude to, briefly it looked like they'd sacrifice the rest of b) separating the launch to orbit sustainer engines--in the Flax Shuttle, or your Fat Dynasoar, these would be disposable. But fortunately (from the point of view of their traumatized entitled mindset, focused on taking some if not all steps toward the SSTO dream, if not from the POV of a really economical and sustainable national launch system) they were able to recover that lost ground by insisting that the boost to orbit engines would be kept on the spaceplane. Thus, they convinced themselves that what they were doing with TAOS-Shuttle was developing an interim Shuttle that could meet most of the checklist, looking forward to the day in the future when they could step by step, building on experience in shaking down partial b and c-g to an economical turnaround, go back first to the TSTO (essentially replacing the SRBs with a reusable flyback booster set of some kind, and hopefully reincorporating the upper stage propellant tank back into the Orbiter) and perhaps someday even the SSTO.

Now even at this point, the two OTL pacing items delaying things in 1978, the SSMEs and the tile based TPS, might have been sidestepped, and in hindsight dispensing with the first would save a lot of time and money and avoid much schedule slip and uncertainty--at the cost to be sure of having to design around significantly lower Isp and accomplishing necessary thrust on the ground with more reliance on booster thrust, and putting a substantial number greater than 3 of developed, lower pressure than SSME 1970-mid-70s state of the art engines-I assume J-2S would be best. Note that since all off the shelf hydrogen engines were not much good at sea level, the boosters alone would have to be relied on until they burnt out--which saves some propellant mass from the prop tank, but this is offset by the lower storage density of 5.5:1 LOX/fuel ratio versus 6:1. J engines could surely be tweaked to use the denser combo I suppose.

Turning to TPS, I await @RanulfC's clarification of just what alternatives were available; a certain sacrifice in mass might be a good tradeoff for a more robust, cheaply maintainable system. I suspect though that without some synergistic composite approach as I suggested in my previous post, the mass penalty would be considerable--but this might be overcome by sheer brute force in the SRBs and J engines, and designers would have considerably more certainty and confidence that a solution within realistic, attainable targets would emerge on schedule. But I am relying on RanulfC's word here; the impression I get from Wikipedia is that on one hand a heat sink TPS strategy would require both a heavy shield layer and a craft hot structure to deepen the heat sink further, meaning the main body could not be a straightforward aeronautical aluminum design, and on the other that while the knowledge base was more developed than the leap in the dark ceramics, no designer could actually say they had ever already designed a complete hot-structure metal TPS spaceplane to fly on any spaceplane. Missile warheads yes, spaceplanes, no. They had a lot of data from Dynasoar investigations, but Dynasoar never flew full scale full speed, crewed or otherwise. It is a shorter leap into murk instead of a long leap in the dark, but there is still some risk and cost develop, and the heavier the overall spaceplane the bigger the booster and engine set and tank must be.

Now you, Archibald, join with RanulfC in suggesting that, beaten back from a, most of b too, that NASA take another step back and compromise on the reusable engines b--either as with Flax Shuttle/Fat Dynasoar per Archibald, dispose of the orbital engines so that the entire thing is tantamount to an EELV, or develop a second recoverable unit in addition to the spaceplane still ambitiously doing c-h, to also be recovered from orbit, and reused.

In fact Archibald your goal is to regress from the ambition of SSTO like an airplane even further, having given up on reusing boost engines completely to also forget the spaceplane aspect of the Orbiter, and simplify it into Big Gemini, which violates most of g as well--it brings the crew back, but since you say it had "no cargo bay" (the versions I know of certainly did have upmass storage, so I presume you mean no downmass cargo capacity); it also, even if the optimistic Rogallo wing could be developed in reliable form for such a large capsule has less precise ability to target a landing site, and maneuvers in the upper atmospheres at higher G's than the goal for spaceplanes. It also uses an ablative, nonreusable heat shield--this might be designed to bolt on and be quickly replaced with a new one to be sure. IIRC, in throwing away the lower tier of the structure before reentry, where the cargo and extended habitat are discarded, it also discards the bulk of its OMS capacity as well, engines along with tanks. This raises the question of how much money if any is saved by reusing the core crew volume.

In being backed into this extreme regression from Shuttle development as the next Big Thing, NASA is giving up on having anything moving them toward the visualized El Dorado of a reusable space plane. In this case, there had better be agreement to budget some operational program using a minimal cost Big Gemini on expendable launcher, almost certainly Titan III--which in 1971 is just barely becoming operational in non-man rated form! Or of course revival of Saturn 1B, but that ship alas has sailed already, and the Air Force hates it anyway.

Now all of us agree that in hindsight, the goal of the SSTO space shuttle, with airplane like turnarounds and costs higher than airplanes chiefly due to higher propellant costs (which are small compared to other real world rocket costs) was a pipe dream, and that what NASA might have done better to focus on, in the face of the refusal of Congress to fund Mars missions, ongoing Moon exploration, or any other gloriously ambitious deep space program, the choices boiled down to either taking a very clean sheet, critically thought out development of a hardheaded cheapening of space launch freed of irrelevant analogies, or else used the money saved from no dramatic cost to orbit reduction initiate as generous a LEO program as Congress would fund (Eyes Turned Skyward approach) and hope that the politicians would be more generous in the future, and that with gradual commercial development of incrementally cheaper launches, NASA would slowly, with building experience and perhaps orbital infrastructure, be able to do more with less someday. My own notions for the rational way to quickly and effectively develop cheaper tonnes to LEO as an investment that might pay off would be to develop a spaceplane that just does e to h, with downmass perhaps reserved for special high cost, high mass later developed models, and do this either concurrently with or (with a hiatus in American manned space ability, or a costly continuance of Apollo or revival of Gemini derivatives) focus on a launch system that first of all cuts costs drastically in the boost phase--I would say, by developing a single standard LRB design that like OTL SRB for Shuttle splashes down in the downrange, but thus recovers tankage and engine for many reuses. This standard booster would be strapped on to central LH/LOX tanks of various sizes, with variable numbers of J engines on the bottom, and the boosters used in variable numbers to raise these tanks of variable mass, each tank/engine set design being developed successively for increasingly ambitious mass to LEO targets, with payload on the nose. This is a far cry from the vision of the self-turnaround single stage Shuttle, but I believe the LRB could be developed very soon after deciding to do it in 1971, and the first tank/engine combo to design two of them to boost would be essentially carried over from the top stage of the Saturn program--about 100+ tonnes of LH/LOX propellant with a single J-2S; the combination should just about match Saturn 1B, but the boosters are recovered and reused, and using a single modified booster and a smaller hydrogen stage one has a competitive booster for the early 70s developed commercial market. The two booster design is sufficient to boost a reasonable sized crew only spaceplane, small enough to use an emergency escape system in case of launch mishap and perhaps its smallness favors the use of simple heat sink metal TPS design too--and it could be a capsule instead of spaceplane. Thus even with the foundation of the launch system postponing a new American crewed vehicle program, Apollo and Gemini derivatives both being scrapped, and the new crewed vehicle having to wait until say 1976 or '77, when its launcher in basic form is undergoing tests, American crewed launch capability should be back no later than OTL, but this assumes even lower NASA budgets in the 70's. With the money and commitment promised to STS by Nixon OTL, and the crewed vehicle program progressing on the back burner until the launcher is ready and then brought to full speed, American astronauts could return to space perhaps half a decade earlier than OTL, with a hiatus perhaps eliminated by ordering a few more Saturn 1B and Apollos to close the gap. After developing a man-rated two booster, 1 J-2S launcher/orbiter combo by '77 or so, the next thing in the lift rocket track is to develop a space station module lifter, with 3 or 4 boosters and maybe going over to 2 J engines, for payloads in the 30-60 tonne range, which might be as far as NASA needs to go for decades. Using the two-booster version, and maybe bigger ones if the space launch market really booms, this design could lift every commercial, scientific and military payload the USA generates as well as catering to foreign customers too, and cut the costs substantially below EELV OTL.

BUT--doing it that way involves contradicting everyone's conventional wisdom, in every age. It falls between the stools of the EELV advocates, who can afford existing expendables and see little purpose in spending NASA dollars on cheapening launch (in competition with their private partners!) and the Shuttle dream, which sneers at these unambitious seeming, pathetic half measures. The latter had the endorsement of every visionary in the aerospace world. My proposal is basically a moderate tweak on say Delta, the variations seeming too trivial to invest public money in, the private sector seeing no guaranteed and expanding market to justify a speculative venture to shareholders.

Going to Big Gemini violates the dream of the Shuttle and is a complete capitulation by NASA to the Air Force.

Developing the Flax shuttle is about as far back into the corner as NASA is likely to be pushed and still survive. And even the Flax shuttle does not conceive of the crucial decoupling I think would be part of any truly economical system--which is, to separate the payload from any recoverable stuff.

Would a Shuttle Orbiter stripped of the Main Engines, whatever engine that may be, really have much cheaper and faster turnaround than with them? The overall system, if we are reusing main engines, still requires that they be refurbished and reused eventually, so the system still waits on that process, wherever it happens. Meanwhile a crew capable Orbiter that also carries all the cargo can only deliver a fraction of its total mass as cargo. Instead, a modest mid-range launch system can alternate between putting up an HL-20 sized spaceplane with minimal to no cargo, and pure cargo payloads with no crew; if it is ever essential to do both in one launch, we can make a bigger launcher, and a still bigger one to enable downmass.
 

Archibald

Banned
You fail at understanding George Low personality and role in the entire story. Fletcher has no experience with NASA and was hired to get the space shuttle. He sticks to his guns until it is too late.
George Low considered the fat dynasoar early on (May 1971, during his time as acting administrator). A personal note of him show he understood that a less expensive shuttle or even a crewed capsule allowed NASA to get a space station funded by 1972 and not AFTER the crewed vehicle development and early flights (OTL 1984).

Look at the situation according to NASA annual budget and cost development vehicles. Most shuttle designs (except Mathematica) cost $8 billion dollar or more, eating more than $1 billion dollar out of NASA budget every year until 1980. Mathematica managed to cut shuttle development costs to $5.5 billion. The capsules (Apollo or big G) were $1.5 billion to $3 billion.
Now if you want to get a space station funding from 1972 (not 1984, AFTER the shuttle) there is no way in hell the crewed vehicle development costs get past $4 billion, since NASA annual budgets are drying faster and faster, bottoming out in 1974.
Bluntly, NASA annual budget was dropping to a paltry $3 billion annually... and the shuttle would eat half of that paltry sum every year until 1978... which become 1981.

Also Low was already there in 1969 when Paine Mars plans failed miserably and Nixon OMB threatened to shut down manned spaceflight entirely.

As for the Air Force, their position on the shuttle program was sickening and somewhat skizophrenic. On one hand, by 1970 they rammed the 15*60 ft payload bay AND crossrange into NASA. On the other hand, they didn't really helped NASA fighting Congress, PSAC and OMB to get the shuttle approved. They didn't really liked the shuttle, in fact they pushed NASA to get all the risks out of the shuttle before comitting themselves to replace their beloved Titan III.
NASA really hoped that the Air Force clout would help the shuttle to get funding with Congress. It never really happened. Air Force was lukewarm at best.

At the end of the day, Low and Fletcher are faced with two options,both $6 billion dollar. Either a cheap capsule plus a small space station, or a space shuttle to nowhere.

Finally, what is important to understand is that NASA had lost control of the space shuttle and manned spaceflight fate. It belonged to OMB and PSAC. That was really the major roadblock and a massive brick wall standing between NASA and Nixon. Nixon deliberately build that wall to keep NASA at arm length.

By october the massive brickwall that were OMB and PSAC had decided that

a) the full size shuttle was not necessery. Does this meant "screw the Air Force ?" hell yes, but it doesn't matter, because Air Force REAL support for the shuttle was next to zero.
They liked their Titan III.

Most importantly, hidden behind the Air Force was the NRO: the NRO hated the shuttle and never flew any KH-9 or KH-11 on it. Yet it is now proven the KH-9 drove the length of the shuttle payload bay !

Can you believe that ? the NRO rams a 60 ft length payload bay into NASA to launch its KH-9 spysats, yet soon thereafter, they loath the shuttle and sticks to their Titan III.It really happened like this IOTL.
since 2006 The Space Review has detailed, declassified accounts about the NRO love-hate relationship with the space shuttle.

b) the entire space shuttle, notably the planned flight rates, was an idiocy.

And the PSAC / OMB were not the only ones: the entire aerospace world (including the Soviets) was scratching their heads and asking " 65000 pounds * 50 flights annually, total 1200 metric tons of payloads launched every year when the entire mass of satellites launched every year since 1962 totalled barely 100 metric tons, TEN TIMES LESS". And we aren't going to Mars anytime soon."

At this point, George Low early thinking and the OMB / PSAC thinking are INDEPENDANTLY - CONVERGING toward a similar concept: that is, a package consisting of cheap crew vehicle plus a space station by 1972.
 
Last edited:
Recently rereading Wikipedia on the Shuttle TPS, if I read it correctly the whole TPS system--all of it, all layers, top and bottom--masses a total of 8 metric tonnes. If we consider downmass as the limit, the TPS mass could be raised as much as 10 more tonnes at the cost of totally eliminating downmass without making the returning Shuttle any heavier. If transpiration is in the mix, and the mass difference is the supply of fluid (water, I would think) to be boiled away, getting rid of all of it high up could have the Shuttle heavier on entry (at the cost of an equivalent amount of up-mass payload) but no heavier with a complete downmass payload in subsonic glide and landing, which might work if NASA could face the payload cost.

As I remember, in the Space Shuttle Decision, it says that NASA thought they could get an ablative TPS that massed the same as the tile TPS.

Not sure how much it would cost to develop an ablative TPS for the shuttle tho.

One option was actually de-rating the SSME but as per anything else that would 'degrade' the Shuttle even more NASA wasn't going to consider it unless forced to do so. They weren't, so they didn't. A LOT of alternatives were simply not given consideration because NASA had hit a point where they were not going to compromise anymore but the problem was they hadn't compromised (really) to begin with and frankly didn't know how to. It would take an outsider with enough authority to be listened to to get it through their heads but by the time of the proposed POD it's mostly too late for anything but quick fixes and kludges.

I hadn't thought of de-rating the engine. I suppose that might shave some money off the development costs.

A lot like the Shuttle/Orbiter we know but with an external recoverable engine pod and tail covering assembly, possibly with a couple of jet engines in it and some kerosene. Lighter Orbiter without tiles, (but would still have the carbon/carbon panels) with maybe a titanium or metal composite TPS. Extended on-orbit time due to more on board consumables since the Orbiter itself is lighter by a lot. Still have most of the same abort issues/options though the jets would be a benefit.

Finding a way to get NASA to plausibly accept an engine pod (and thus make the shuttle-c more plausible) would sure be a neat place to end up.

Though it really does seem the shuttle was just doomed to be a dog once the decision to put large SRBs was taken...

Actually "more money" was always the given answer to fix ANY problem at NASA but I question its validity as they HAD money they just didn't spend it well. I suspect that NASA might in fact not be capable of fiscal responsibility since the "best" administrator I think they had was a "bean-counter" and HE couldn't get them to work within a budget. (Not that his boss helped by suggesting they could in fact have everything they wanted and then not supporting them when they tried to get it)

I have a suspicion, (which is why I'm exploring the concept) that given a high profile enough examination and a frank but balanced appraisal NASA could be convinced to not only live within it's budget but prosper as said budget actually increases again, (it did) over time. IMHO the problem was and is NASA has never been held to that standard and has been 'raised' if you will to not have to. Every time there is progress in learning that lesson it's undercut by a new program and budget re-arrangement which never lives up to the promise.

All reforms take more money since you need new money to do new things even as you are still spending money on the things you are rolling up.

And while you are right that NASA often made poor choices that were expensive, I am not convinced that they do so more than other comparable government departments or corporations.

US aerospace does seem notably borked from what I read. But then that may have more to do with the people complaining being loud enough that I've read their articles, rather than the complaints being correct.

fasquardon
 
As I remember, in the Space Shuttle Decision, it says that NASA thought they could get an ablative TPS that massed the same as the tile TPS.

Not sure how much it would cost to develop an ablative TPS for the shuttle tho.
I have two categories of doubt about this:
1) I used to assume that ablative would be lighter than a conserved, reusable TPS, but after a Vintage Space web video discussed the Chinese use of wooden heat shields, I looked up Apollo figures for comparison, then the Shuttle's OTL system for a third check, and discovered that no, if I did the math correctly (Apollo versus Shuttle are somewhat different problems of course) that actually the Shuttle system (of which the infamous tiles are just a portion of the whole solution) works out to be lighter. That is, if one downgraded the Apollo spec to only have to reenter with a top atmospheric speed of say 9 km/sec instead of 12 (for entry only from LEO versus entry from a translunar trajectory) which would cut the necessary thickness of the ablative layer almost in half--it still works out heavier than if one substituted tiles. I also factored in weight per square meter to equalize the problem in that dimension. Perhaps I did it wrong, but it did seem that the tiles were more efficient in terms of mass, by quite a lot. This does not mean that tiles could have been substituted in Apollo, because presumably in addition to the net integrated heat flux being more per square meter bearing a given mass with translunar entry, also peak temperatures are higher--with ablative, one just has a more rapid rate of shield decomposition and removal at higher temperature, but with fixed systems such as metal heat sink or heat-resistant ceramics, above a critical combination of heat flux and temperature, the material breaks down and ceases to function as it should--it melts and flows away, or cracks, or undergoes chemical transitions that ruin its thermal or mechanical properties, or something. Or just plain vaporizes. Ablatives are supposed to vaporize, so that's all OK. But they mass more, and must be replaced after each entry

2) in addition to needing to be replaced each time, their erosion pattern is never 100 percent predictable. One can make them thick enough so that the deepest burn won't burn through, but each entry will be a somewhat different history and the upshot is, once the vehicle has penetrated deep into the atmosphere and slowed down enough that the layer is no longer ablating away, its exact shape and texture will vary. This is OK for simple, not very aerodynamic (at subsonic final approach speeds anyway-ablative capsules have lively aerodynamic characteristics at hypersonic and supersonic speeds of course) capsules, but the idea here is to have a spaceplane that can manage a decent glide ratio (brick like compared to most subsonic airplanes of course, but glider-like compared to an Apollo capsule) and have good aerodynamic control characteristics for a pilot to learn to manage well. Ablative coated Shuttle Orbiters would each enter subsonic glide phase with unpredictable lift-drag polars; no amount of hours in a simulator or even flying Orbiter practice vehicles released from high altitude carrier planes will prepare the pilots for the particular characteristics they will find randomly imposed on their reentering Orbiter; they will literally have to learn to manage this particular flight on the fly, and everyone's life aboard not to mention the expense of an Orbiter depends on mastering it well enough to get it to the landing field and then land it, with no opportunities for practice in advance. (One could learn the statistical pattern of variation of aerodynamic variables by some means in advance, then send the pilots to the simulators with realistic instances of random variables thrown in to the simulation works). Since the aerodynamics of the Shuttle as a glider on final approach are mediocre at best, throwing random changes in the parameters will presumably only make it worse, never better, and the task of landing one becomes that much more difficult.

Then of course one has to strip off all the remnant outer skin, at least on those sections requiring ablative, and reattach a new complete ablative layer. Considering how much work had to be done with the tiles that would not be so bad as long as the patches of ablative material are much bigger in area than individual tiles were; what matters the most is how many steps workers have to take.

Going with ablative skin would also be yet another blow against the reality of reusability versus the dream, and the whole premise of the basic system design is that reuse will save money every launch and thus cheapen the launches.
I hadn't thought of de-rating the engine. I suppose that might shave some money off the development costs.
The main thing that accomplishes is shortening the development cycle--if the engine was originally designed for a nominal 100 percent, and we never take it over 90, presumably it will past the mandatory battery of tests at 90 with flying colors, if it would have passed at 100 percent even only marginally.So it can be certified man-rated at 90 if not at 100, and missions constrained to 90 planned and launched. Thus the extra time (hence also money) needed to prove it out at 100 can be postponed and the program start operating with crewed launches.

But you will still want to eventually certify for 100 percent spec, and since this is where they had problems, rework the design and re-test it--and this means the improved design is not even rated at 90 until you test it and it passes at least that benchmark. But if you have 90 percent SSMEs in hand already, there is no sense in certifying the next iteration at that low level; you want to press on now to 100 percent.

Overall, the program costs more, not less, if you won't settle for derating permanently. It is not an overall cost saver, it is a more expensive way overall, but it might allow a version of Shuttle to fly with degraded performance versus target on schedule, if there aren't any other holdups.

And it matters how much X percentage degrading costs in Y tonnes reduction of payload to orbit. The cargo share is at the margin, and a single tonne reduction is a big percentage. Over ten tonnes eliminates up mass capability completely, and renders the entire design pointlessly stupid. One now has a vehicle that can take three astronauts to orbit, and support them there for a while--but much of it is sheer useless dead weight, for the cavernous cargo bay that now accomplishes nothing. (It could still be used for down mass, but what system is launching stuff into orbit we'd later wish to take down?) Clearly if one can't get the engines to spec someday soon, an additional program cost now, one ought to do a clean sheet smaller Orbiter that only serves as a crew vehicle, and vehicle to bring the main engines back down again.
Finding a way to get NASA to plausibly accept an engine pod (and thus make the shuttle-c more plausible) would sure be a neat place to end up.
That's El Dorado and Cibola IMHO
Though it really does seem the shuttle was just doomed to be a dog once the decision to put large SRBs was taken...
I strongly dislike the SRBs and wish they'd gone a liquid route, provided they recovered and reused them--no need for fancy flyback, initially, just parachute them to a splashdown downrange and go get them and bring them back as they did with the SRBs.

But, especially if the farce of pretending to save money by recycling the SRBs had been recognized and abandoned, the SRBs are not the show stoppers. A bad design cost us Challenger, and NASA had plenty of operational warning that was a big risk, but the improved design never failed in the same catastrophic way ever again. And the hellish thing is, the final double-ring design that worked well enough existed on paper before the first launch, along with remarks that the somewhat lighter single-ring design was risky. A sufficiently realistic and safety-conscious management could have insisted on the safer design from the beginning and avoided the Challenger disaster. To be sure, perhaps with sufficiently cold launch conditions even the two-ring design might fail, for all I know, and part of the lesson of Challenger was to take the temperature limits Thiokol reported seriously. Also the safer design did mass more and this did cost a bit of payload.

As I've suggested, if Shuttle C were developed, it might turn out that the payload mass was too great to be useful except for a handful of missions, and then STS is still of little use--unless a downgraded version, with lighter, less powerful boosters, a smaller tank and using fewer SSMEs were something that could be easily achieved. All of these elements are technically new equipment requiring new investments in development and testing. But each of them should be relatively cheap to attain, given the bigger versions are already proven and provide operational data guiding the downgrades. The segment design of the SRBs means that making a 3/4 thrust version is a matter of leaving out one of four segments, and redesigning the nozzle section. For that reason I am much more reconciled to the SRBs and Thiokol getting an honest cut than I was before.

Doing it right from scratch would be better, but I am also interested in TLs where we waste less and get more value out of the perhaps misguided initial investment.

I still await commentary on the feasibility of jacketing the ceramic tiles in an outer layer of metal to protect them from their vulnerabilities. Is it even possible--will all known metals melt or soften to an unacceptable degree at the maximum temperatures the tiles reach? Wasn't Columbia lost due to damage not to the tiles but to the solid Carbon-carbon leading edge spar, and can a metal outer layer take the even higher temperatures those areas protected by carbon-carbon had to endure?

And for RanulfC to either point to the specific prior posts here or elsewhere that detail the tradeoffs in going to an all metal heatsink TPS versus ceramic, or point to the data itself.

And to remind everyone--this thread has become something of a forum on what alternate Shuttle Decisions might have been made in 1971, but we are supposed to be looking at 1978, assuming '71 went as OTL.

I just can't see how a last-minute set of fallback kludges can get a workable Orbiter operational any sooner than the to-spec Columbia flew OTL, with a decision as late as 1978 or even the year before, so strictly speaking we should drop this thread and start a clean-sheet alternate 1971 thread instead.

But the discussion has been fertile, and so I for one will probably welcome its continuation anyway.
 
In the spirit of revising 1971, I've been thinking more about Saturn Shuttle, which is the particular approach Nixon was given a model of for a photo op in the Oval Office IIRC. In particular there was that scheme to make a launch system out of turning the Saturn V first stage into something more like an Atlas--dropping the outer ring of 4 of 5 F engines, with the central engine remaining as a sustainer burning a small fraction of the total propellant load after the outer 4 engines drop. The plan was to recover the outer 4 engines, mounted on a ring that would parachute down for a splash and retrieval by ship. For Saturn Shuttle, I'd propose developing that in upgraded form for all 5 engines, burning them all the same standard short interval normal for boost, recovering all 5 while sacrificing the tank mass. This is the same philosophy adopted OTL, and I suspect reusing just the F-1A (or later iteration) engines alone might work out to be more economical than the OTL notion of reusing SRBs. Now we have two discarded tanks to make per launch instead of one.

For the engine recovery section, I am visualizing mounting the 5 F engines to a large inverted dish of high temperature steel, such as Boeing was investigating for the SST. The central engine would be mounted rigidly, and the outer 4 would each have a one degree of freedom simple gimbal allowing them to swing tangent to the disk circumference; this gives adequate 3 axis thrust vectoring with a simple and relatively light structure. The thrust of all 5 would be conveyed to the steel dish, that would transpose the complex thrust structure into a simple combined force and torque the tank would convey on to the craft as a whole. The dish is concave side down, meaning that plumbing carrying in ker-lox propellant either comes around the outside or comes in through a penetration in the dish, and with the engine guts and gimbaling mechanism sits well in from the rim. Upon first stage cutoff, the dish blows loose of the tank stage, which will burn up and sink in the ocean. Perhaps we have some delay first, in which the whole thing coasts parabolically to high reentry heat (much lower than from orbit, but still pretty hot for aluminum based structures like the tank to take). The attached tank helps slow down the whole thing until the air heating is getting too hot for the protected parts of the engines/gimbal/fuel line system, at which point it separates, and some sort of drag structure is deployed to guarantee the engine dish flips around to convex-side down, the side that used to face up to the tank. This is smooth, and is basically the heat shield as well as mount of the engines. The engine bells are already designed to take high heat so they need minimal protection. Eventually parachutes are deployed to slow its final approach to the sea, and then the dish splashes down in the ocean. Being made of a stainless steel, or coated with one, the salt water does little harm, and the engines are floating on a steel raft, protected from sea water intrusion. A ship comes and picks it up.

Now RanulfC has suggested that the Shuttle concept might have taken the suggestion of a separate main engine module from the Orbiter to heart. Here they have an example of one type of design for recovery of engines alone to extrapolate. It needs a lot of extrapolation to be sure! Somewhere between 5 and 7 J-2S engines might be worth designing to recover. Note that Saturn-Shuttle disallows the possibility of ground launch of the hydrogen engines, since the exhaust must be on the bottom of the stack but the hydrogen stage must be high up on the stack. (Well, doing something more like the Soviet MAKS design, where the spaceplane trails behind the hydrogen tank, might allow an Orbiter bearing ground-lit Main Engines to be mounted with its tail even with the F engines exhaust, while the nose, connected to the hydrogen/oxygen tank riding on the first stage with a suitable thrust structure, being the point of propellant intake--but this requires a goofy and probably heavy connection and an Orbiter that is as long as the first ker-lox stage. But here we assume a hydrogen engine module separate from the Orbiter anyway). Therefore there is no call to develop SSMEs whatsoever. The J-2S had inferior Isp to the SSMEs, 436 sec versus 450, but that might be remedied in a much more incremental, much less risky way than the heroic SSMEs, because we are waiving the requirement of striving for decent sea level thrust. J engines were always designed for vacuum thrust, so just improving the nozzle won't help close that 14 second Isp gap much, but perhaps modest chamber pressure increases--the J engines only ran at 30 atmospheres, versus say the F engines at 70, so there is some room for improvement by modest means. Or for the first generation system designers can just work around the lower Isp.

If we have separate stages, no parallel burning, of course the hydrogen engines can be stacked in line with the ker-lox booster system. Here too let us assume 5 J-2S engines, each mounted as the F engines are on a similar steel dome section--or perhaps not steel. In terms of thrust structure, the modest thrust J engines pose a far lesser challenge, by a factor of 8, than the same number of F engines. But in terms of reentry and recovery from orbit versus booster burnout speeds, the heat flux is an order of magnitude worse. If we are not going to rely on developing the OTL ceramic tiles but go with heat sink metal structure, it is that function much more than thrust structure function that will govern the mass of the dome. We get thrust structure for free as it were! Because we must pay a high cost for the heat shield of course.

Now at this point the major reason to put the Orbiter on the side of the stack instead of on the nose is eliminated too. There might be other reasons for a sidesaddle mount, but we at any rate have the option of putting the Orbiter on the nose of the stack instead. This has advantages and disadvantages.

With no SSME project (only an optional plan to improve J engines for reuse first of all, along with parallel improvements in the F engines, and perhaps to improve Isp and/or thrust in either, most of this can wait until after the basic system is proven and flying useful missions, although reuse better become a reality to justify the whole project in the first place, the major focus is the clean sheet redesign of the first and second stages of Saturn V, mainly developing the engine mount/heat shield domes and designing the associated disposable tanks for cheap construction while keeping weight under control and maintaining decent levels of reliability. In parallel, a spaceplane must also be developed. With no ceramic tile program, the TPS must perforce be heat sink metal. There is no need to house 3 SSMEs, which between them mass around 9.5 tonnes and require more mass for gimbaling and merging into the general Orbiter thrust structure. Here the Orbiter is cargo on the rocket, dead weight payload all the way to orbit.

But we won't take another step "backward" that I think is the most important of all--separating launch system from payload completely. The philosophy is still shared that it is OK and indeed an advance to have the cargo/payload of each mission be a mere fraction of the total Orbiter mass, and to have therefore a tonnage to orbit mass requirement half an order of magnitude greater than the payload itself. This makes sense if we can reduce costs per launch by a greater magnitude than that per launch. We also grandfather in the assumption that the Orbiter must be crewed, which is also in hindsight most unfortunate. If it were otherwise--if someone had the brain wave that rather than building an approximation to an SSTO spaceplane they are actually trying to make a cost-effective National Launch System, with a crewed vehicle being just one optional cargo among many, they would scale down the project considerably. It might be marginal to hope that something like the Eyes Turned Skyward Saturn 1C, that used only one each (non reusable) F and J engine for two stages capable of putting some 20 tonnes into LEO on an expendable basis, can maintain a large fraction of that capability with provision for returning both engines back to Earth for reuse. But maybe with 2 F engines it can be done? At the time of this Saturn Shuttle Decision ATL, that possibility is overlooked because of the obsession with putting up a 100+ tonne crewed ship that carries 20-30 tonnes of cargo every time.

On the subject of TPS, I have another suggestion to consider. Suppose we are mainly relying on metal heat sink instead of ceramic. The straightforward thing is to calculate how thick the metal needs to be for its inner surface peak temperature to be in an acceptable range and make shingles of that thickness--like ceramic, the heated metal will expand so there need to be expansion cracks to allow for this, though metal being more elastic perhaps the scoring can be coarser. But suppose we put three layers over it, similar to my suggestion of metal-tile synergy. The scored inner metal main TPS has loosely draped over it a layer of copper, aluminum or suitably conductive alloy, which does not have to be very strong mechanically, it can reach a point of nearly melting, because it is bonded to another layer of high temperature metal, this one thin, as with the ceramic concept, so that it is heated to near uniform temperature. And bonded to that, forming the outer layer, is ablative material. Ablatives though not as it turns out light, do tend to be rather tough, so the whole sandwich is protected during launch and the ablative also acts as a sort of half-assed Whipple shield on orbit against micrometeors and so forth.

Upon entry, the thin ablative layer is nowhere near thick enough to survive the entire heat dose of entry; it is only a quarter or fifth as thick as we'd want for that. But it does bear the brunt of the most intense heating, the initial blast at maximum temperature and heat flux. Furiously ablating away, it delays the moment at which the outer metal layer of high temperature metal starts to get seriously heated, and by the time the ablative is burning through in patches, the peak flux has been passed and the remaining integrated heat dose is significantly reduced. The intermediate layer of highly conductive though not high temperature strong metal evens out the patchy heating due to random burn-through patterns, it heats neighboring more persistent ablative from below as well as above, accelerating further the burn-off of the ablative--but note that this burn off still absorbs substantial heat, which is carried away with the evaporation of the ablative. Soon the outer layer is hot and smooth, with the ablative all burned off, and heat soaks in through the outer two metal layers, now essentially at uniform high temperature, the soft conductive layer wrapped in the stronger outer one. The inner soft layer heats the main block of TPS, which expands differentially, closing the scoring cracks, forming a smooth layer for the copper layer to rest on. It delays the heat pouring in with a fairly linear temperature pattern from outside to inside. But with the delay and partial absorption by the ablative, and small extra heat sink effect from the outer two layers, we might save some weight in this layer, perhaps more than the outer two layers add. With the ablative, the total weight of the TPS at launch and throughout the orbital mission might be higher than with a pure and simple shingle layer, and that is yet more than with the OTL ceramic choice--but the ablative layer will burn off, completely and smoothly due to the fact that it was far too thin for the total job and the conductive metal layer undermining the bonding epoxy from below. Thus, the weight of the ablative is absent when the Shuttle is gliding subsonically, and so should be its aerodynamic fouling effect. It is now necessary to reapply another ablative layer before the next launch, but given that OTL tile inspection and repair was a time-consuming part of every refurbishment cycle, we may come out ahead here too.

So what does anyone think of this approach to STS? It should not take long to realize that the weight of the Orbiter can be replaced by any desired payload, should a very large load be wanted. Turnaround time should be comparable to OTL if not accelerated, despite the fact that another recovery event has to be managed--for the recovery of the orbital booster stage and the J engines will happen within a few orbits of launch, thus the refurbishment of the engines and their mounting/heat shield system will be started before the Orbiter returns to Earth, generally speaking, and run concurrently. The Orbiter itself will avoid two refurbishment issues, which may or may not accelerate turn around time.

Instead of cycling SRB elements back and forth between Utah and Cape Canaveral, we need to make 2 tanks at Michoud instead of one. However, the upper hydrogen/oxygen tank will tend to be smaller, despite the hit in Isp efficiency due to choosing the simpler engine, because during the initial burn fro pad to first stage burnout, the hydrogen J engines will be shut down. The lower first stage tankage is much more massive, but also far more dense due to using kerosene and oxygen. Thus, each tank will be smaller and it might be possible, depending on Michoud's available space, to make both concurrently.

If it turns out that Shuttle element turnaround times are not significantly faster, and that costs are not lower than for STS OTL, the option of going "Shuttle C" is very simply present. If that is not an attractive approach to lowering costs per kg to launch, another round of development may make a smaller booster/upper stack set, going down perhaps to a single F engine with a single J engine for roughly a factor of 5 reduction, and with combinations of 2,3, or 4 engine designs, a sweet spot for cargo up-mass might be hit that with turnaround costs scaled down appropriately, makes for substantial reductions in cost to orbit on an attractive scale of tonnage. Hitting upon the right size, a scaled down crew-only Orbiter could also be made that routinely rides alone on the smaller scale system, while the full 5 and 5 engine design, and perhaps something else intermediate, can launch a combination of cargo and small Orbiter.
 
i went similar way in 2001: A Space-Time Odyssey

Here NASA STS is de facto a two stage Saturn V with Wings
Winged First stage use 5xF-1A engine and bunch of jet engine to get back to KSC
Flight back is rough do Aerodynamic braking from Supersonic speed.

The Orbiter is large, huge because internal propellant tanks and using 4xJ-2S
Do large size the Heat shield is made from Metal instead of ceramic or silicone bricks.
All system are modular, so easy replace during maintenance and allow to STS launch every 45 days.

On launch price
The F-1A / J-2S are around 10 reusable then junk.
NASA not wanted to use them and went for RS-25 in naive beliefe, it can be reuse one hundert times.
in fact the RS-25 was similar: around 10 reuse...
The Saturn Shuttle is cheaper in maintenance, but launch cost roughly same like two stage Saturn V.

but since in TL, the USA in Hot Space Race against Soviets, NASA And Capitol Hill are not interested in Money for moment.
In fact they work and to get Saturn Shuttle first flight and build the third US Space Station.
 

Archibald

Banned
Ablative isn't that much difficult: just pile up enough of the stuff on the exposed bottom and let it burn during reentry. It is a one-shot solution, but it might actually weight less than a reusable TPS.
The Soyuz famously has far more ablative TPS than it actually needs.
 
Shevek,

A few papers that turned up in research lately that may address some of your questions:
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19830014016.pdf --Page 26 has a weight breakdown for a 3xSSME engine pod reference design use on sidemount SDHLV with recoverable engines. Total mass is about 31 metric tons, of which about 20% is bookable to recovery-related hardware--retros, TPS, parachutes, aeroshell, etc.

https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19750024432.pdf
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19730022148.pdf
Two studies on ablative TPS alternatives for the space shuttle, with detailed weight breakdowns. The former is sort of an executive summary two years after the detailed one 1973 one. Minimum system weight for a directly-bonded ablative system may be similar to the ceramics or just slightly less, but with the issues of scraping and re-applying it every flight, which the X-15 proved wasn't exactly easy. Systems with removable panels tended to be much heavier, up to twice the weight of the tiles.

https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19820015388.pdf --A detailed study on the non-ablative alternatives. Many are a bit heavier than the tiles, and several have some effects in regards to panels bowing under loads and heating. The biggest problem for the subject of this thread is that even in '82, the concepts studied were all in fairly low states of technical readiness--not much more advanced than tiles were when selected in the mid-70s. Thus, there's risk associated with picking any single one (such as the fact that the report's preferred solution would have spread carbon-carbon over much of the vehicle's hottest portions, placing it more at risk of foam strike) as well as significant schedule slip: the report estimates no earlier than 5 years from approval to proceed.
 
As I remember, in the Space Shuttle Decision, it says that NASA thought they could get an ablative TPS that massed the same as the tile TPS.

Not sure how much it would cost to develop an ablative TPS for the shuttle tho.

That's what I was recalling as well but those reports seem to suggest it was never that simple, or considered. Given the experience with the X-15 "spray-on" that's understandable but technology was actually progressing pretty fast. Having said that though it was never going to be "simple" and "fast" which the tiles promised initially.

I hadn't thought of de-rating the engine. I suppose that might shave some money off the development costs.

As I understood it if they blew up once more they were going to consider it JUST to get them working but that would have had knock-ons for other developments. Like needing to 'up-rate' the SRBs to carry the initial loading. Thiokol of course said it was 'easy' to do but so far as I can tell no one really bought that.

Finding a way to get NASA to plausibly accept an engine pod (and thus make the shuttle-c more plausible) would sure be a neat place to end up.

Though it really does seem the shuttle was just doomed to be a dog once the decision to put large SRBs was taken...

Arguably a lot longer before that but NASA could be myopic that way. :)

All reforms take more money since you need new money to do new things even as you are still spending money on the things you are rolling up.

And while you are right that NASA often made poor choices that were expensive, I am not convinced that they do so more than other comparable government departments or corporations.

US aerospace does seem notably borked from what I read. But then that may have more to do with the people complaining being loud enough that I've read their articles, rather than the complaints being correct.

NASA consistently undershoots on cost estimates. It's historic, or maybe contractual I'm not sure :) The thing is when they get the cost even 'close' it usually turns against them, ("90-day-study") and then they get blamed for low-balling some rather obvious costs and deny doing so.

Folks,

Let's not forget...

What matters is building the space shuttle with the...

RIGHT SIDE UUUUUUP !!!!

Har, har :) We're a bit past that point at THIS POD though. Still going through the various citations but wow the alternatives weren't all that great. At least officially. And folks are right it's not so straight forward to make major changes this deep in. E of Pi on additional C/C; that might not have been exactly a bad thing as it would have brought the problem forward rather than the continuing assumption of how tough C/C really was. But the delays was what NASA was trying to avoid so I'm under the impression they would have been forced to accept anything BUT additional funding.

Randy
 
Top