What would a better STS Have Looked Like?

[Lobo]

I think I heard at some point that each Shuttle was originally envisioned to take about 50 people around a week to turn around and be ready to fly again.  But it turned out to take like 1000 people 2 months to turn around?

Not sure if that is accurate or not though.  Just recall hearing it at one point.

Don't know about the personnel count, but I recall reading old claims of a two-week turn-around.

two-weeks may have been what it was, and I wasn't remembering it correctly.  Either way, obviously it was supposed to take far fewer people than it did, and turn around much faster than it did.  Had it been closer to originally envisioned, it would have been a much more sustainable system.

[93143]

I heard somewhere that the tiles had a tendency to crack, so they just cut them in half both ways, quadrupling the number of tiles...

[Jorge]

I heard somewhere that the tiles had a tendency to crack, so they just cut them in half both ways, quadrupling the number of tiles...

That is obviously disproven by looking at the pattern of arrangement of the tiles. They may have redesigned for smaller tiles at some point but they certainly did not just cut them in the way you describe.

[93143]

Sorry.  Poor choice of words, both on my part and on the part of whoever said that (I do seem to recall something about simply dividing the design).  As you say, it is obviously not designed in quads, so whatever actually happened had to be more involved.

Still, if it's even partly true, it roughly accounts for the two weeks/two months discrepancy...

[Lobo]

I didn't go back and look to see if this link was posted by someone else:

http://www.pmview.com/spaceodysseytwo/spacelvs/sld036.htm

But I think a better STS would have clearly been either the S-IC flyback booster, or the Titan-IIIM booster.  Not sure about the drop tank concept for that though.  Might have been better to have a single belly drop tank, although your forces might have been too out of line.  Guess it depends on how high the Titan booster could have gotten it before staging, and thus how large the drop tank would need to be.  Which is probably why they put the drop tanks where they did.  Still, maybe there could have been a better way to do that. 
The nice thing about the Titan booster is the booster overhead was paid for by the USAF.  NASA didn't need to develop or maintain the line itself.  It could focus on the Orbiter and paylaods. 

[vulture4]

An extensive study of converting the Shuttle to liquid fuel booster was conducted in the late 1980's after the Challenger loss. It examined four booster configurations, pressure fed and pump fed boosters using either kerosene/LOX or MMH/N2O2. The conclusion of the study was that any of these strategies was technically feasible, but would require considerable investment. By that time the redesigned solid rocket motor was in service and the risk of a recurrence of the Challenger loss was relatively low. The refurbishment cost for the liquid-fueled boosters was difficult to predict with any precision, and the flight rate wasn't really high enough for the possible reduction in refurbishment cost to make a real impact on program cost, so the program did not support further work on the liquid fuel boosters. In retrospect the thorniest problem was damage to the TPS from foam loss from the ET, which a change in the boosters would not have helped. 

Nevertheless liquid propellant (and a flyback booster) was the first choice of the shuttle program back in the 70's and would have had real advantages if there have beena n opportunity to test the concept with subscale prototypes a different course might have been followed.

[Lobo]

An extensive study of converting the Shuttle to liquid fuel booster was conducted in the late 1980's after the Challenger loss. It examined four booster configurations, pressure fed and pump fed boosters using either kerosene/LOX or MMH/N2O2. The conclusion of the study was that any of these strategies was technically feasible, but would require considerable investment. By that time the redesigned solid rocket motor was in service and the risk of a recurrence of the Challenger loss was relatively low. The refurbishment cost for the liquid-fueled boosters was difficult to predict with any precision, and the flight rate wasn't really high enough for the possible reduction in refurbishment cost to make a real impact on program cost, so the program did not support further work on the liquid fuel boosters. In retrospect the thorniest problem was damage to the TPS from foam loss from the ET, which a change in the boosters would not have helped. 

Nevertheless liquid propellant (and a flyback booster) was the first choice of the shuttle program back in the 70's and would have had real advantages if there have beena n opportunity to test the concept with subscale prototypes a different course might have been followed.

Vulture,

All good points.  But the purpose of this thread is to go back to the initial development of STS, with a little bit of knowing what we know now, or at least with people then having some realism and pragmatism instead of the unrealistic optimism and hubris that seemed to affect everyone to not seem how the STS they came up with couldn’t be a cheap or useful as they thought it would be.

As I understand LRB’s were in the running during development, as well as using Saturn V hardware, using upgraded Titan hardware (so there were be sharing with DoD), as well as various flyback boosters.
Seems SRB’s were chosen primarily because they were realistically the only “reusable” boosters that they’d be able to get with their budgets and time lines.  And “reusable” was seen as all-important, whether it made real-world economic sense or not.  Flyback LRB’s would be ready in time and for the budgets NASA ultimately had to work with.  However, if they’d gone with LRB’s instead of SRB’s, then flyback boosters could have been an upgrade option down the road.  And would have been much cheaper/easier to do than it was later after SRB’s had been flying.  Because they would have required cancellation of one flying booster (hard to do politically), and development of a brand new booster along with flyback ability.  To many, prior to Challenger especially, that would have seemed like a lot of money to reproduce the same capabilities of the SRB’s in effect.

So, for the purposes of this thread, we don’t want to look at concepts that were considered after the PoR STS had been selected and developed, but before those decisions were made.  If you read back through this thread (not sure if you did), you’ll see many good concepts discussed. 

One of the concepts I liked the best were using Saturn INT-21, with the shuttle mounted on top (no SSME’s obviously) and the whole stack in line.  This would have required minimal modifications to KSC compared to what STS required with SRB’s.  The Moble Launcher UT’s would need to be modified to have the crew access arm relocated.  The Pads wouldn’t need any modification.  The VAB was already set up for stacking the S-IC and S-II stages.  Just would have modified the upper part to stack the shuttle in place of the S-IVB and Apollo CSM.  The Shuttle would have been more Buran-like.

Another concept I liked was to retire all of the Saturn cores, and developed a new 6m CCB.  (Something might have been worked out with Douglas, as their 6.6m S-IVB would have been pretty close.  3X6.6m cores would be too wide to fit through the VAB doors I think, as I think that’s what’s limiting SLS LRB options to 5.5m diameter).  Then Shuttle could have flown on a tri-core booster.  That should have still fit within the existing flame trenches and the VAB doors.  A couple of configurations could be looked at with a tri-core.  First, is two kereolox boosters on a hydrolox core, all 6m diameter.  The kerolox boosters could use a pair of F-1A engine, and the central Hydrolox core would use a cluster of J-2S’s.   Perhaps it would be ground lit, perhaps it would air-light after liftoff like Titan’s central core.  This configuration is similar to Energyia-Buran.  It has a lot of off-center loading, but Energyia-Buran seemed to fly ok.  The kerolox boosters would be upgraded to flyback boosters later. 

Another possibility would be three kerolox cores, with possible crossfeed.  That would allow more commonality.  Not sure if the central core would stage high enough with enough energy though for the Shuttle to make desired orbit with it’s OMS boosters.  So that would be an issue of course.   But the advantages are 3 common cores, and no thick LH2 insulation near the Shuttle’s TPS.  Although if you could stack the Shuttle on the central core instead of side-mount, that would solve that.  Not sure if stacking the Shuttle on a 6m-ish central core would be feasible though.  But if it could be, That’d be interesting.  Maybe a Centaur could be added to the Shuttle to help get to useful orbits.

The advantage of these smaller cores, is they could then be used for stand alone MLV’s, and by adding a true upper stage to it, could be a Heavy lift BLEO LV down the road if/when political will turned back to BLEO missions.  With STS, there was really no way to make it BLEO, as the Shuttle was integral to the LV rather than a payload. 

The concept of using a “Titan III-L” booster for shuttle was interesting too, and would share commonality with DoD.  IT wouldn’t be a reusable booster, but it would be a pretty inexpensive booster so that way it wouldn’t really –need- to be reusable.  Or perhaps there could be an in between concept, Using say 4 Titan SRB’s, along with a central hydrolox core with J2S engines.  If the Shuttle was mounted in-line rather than side mount, then they could probably mount six Titan SRB’s around it.  Make the core wide enough to comfortably mount the Shuttle on top, and then mass-produce the J2S engine and the CCB’s.   Remove the Shuttle and add 2nd stage, and now you have your Heavy lift BLEO LV again down the road when it’s desired.  And it’s a scalable LV that can mount 2, 3, 4, 5, or 6 Titan SRB’s.  So at least the whole booster portion has cost sharing with DoD.  NASA only maintains the Shuttle and the central core overhead.  And that’s only one existing J2S engine, rather than the brand new SSME development.  You still have SRB’s, but NASA didn’t have to development them or pay for their overhead.  In fact, NASA could have probably worked with North American on modifying the S-II production line to produce the new Shuttle core.  Modify it to be a sustainer stage rather than a 2nd stage with a cluster of J2S engines, and with mounts to take multiple Titan III SRB’s, and an adaptor on top for the Shuttle.  And that could possibly have saved NASA a fair bit of costs in developing the new 8.4m ET core.  Again, later, put an upper EDS stage on top and you have a large BLEO LV.  Guess that would be sort of an upgraded “Saturn II”. 

[Lobo]

After some recent discussion on the “What if Satrun/Apollo had never been cancelled” thread, it made me think a bit on this.

I tend to think if you had to fly a reusable spaceplane, as it sounds like the political forces demanded in the 70’s, if NASA had kept the S-IC and S-II stages and put the Shuttle on top of that, that would have been best.

But, perhaps a little better might have been to use the S-II stage only, with strap on Titan boosters.  On this chart, it looks like the performance of such an LV would be between 39K-44K mt to LEO, depending on which Titan boosters are used.
That wouldn’t have been enough to launch the STS orbiter, but perhaps it could have with more Titan boosters, and air starting the J2S engines on the S-II stage? (as they were designed to do anyway). 
Side mounting the Shuttle would have been possible, if the SRB’s were clustered on each side, say 3 to a side, and then putting 5 J2S engines on the orbiter and saved them.  (would have needed better insulation obviously on the S-II stage).
However, I think the better configuration would have been the Shuttle on top, with like 8 Titan boosters around the S-II stage in a Delta II SRB configuration.  Then it could have also launched on it’s own as a stand alone LV with a large LEO capacity.  (for large Space station modules in particular).  Various SRB configuration could be used for the specific payload requirements.  From 2 to 8.  Then just mass produce the J2S, which was designed for cheaper manufacturing anyway.  The Orbiter would have been lighter by not having the SSME’s or MPS.   

That would have saved the whole expensive development program of the Shuttle SRB’s, it would have put the Shuttle on top protecting it from ET debris, and putting the crew farther away from the fuel, and it would have saved the whole expensive SSME development program, as well as the 8.4m ET development.  The S-II stage just needed a bullet dome on it. 

And it could have been sold as having the Titan SRB’s initially expendable, but a side project would have been to make them recoverable later (which could have been maybe studied, but probably never implemented). 
Basically, NASA would claim to be starting with a reusable spacecraft, then phasing in a partially reusable booster over time, so that there wouldn’t need to be too many new technological advances all at one time, as STS ended up needing. 

I found this interesting from the Astronautix entry for Buran:

“Glushko believed that while a Soviet cryogenic engine of 200 metric tons thrust could be developed in the required time, to develop a reusable engine would be impossible due to limited experience with the propellants.
This conclusion led to other important design decisions. If only expendable engines were to be used, there was no need to house them in the re-entry vehicle for recovery. This meant that the orbiter itself could be moved from the lateral mounting of the space shuttle to an on-axis position at the top of the rocket core. The result was the Vulkan - a classic Soviet launch vehicle design: booster stages arranged around a core vehicle, with the payload mounted on top. The elimination of the lateral loads resulted in a lighter booster, and one that was much more flexible. The vehicle could be customized for a wide range of payloads by the use of from two to eight booster stages around a core equipped with from one to four modular main engines. Either a payload container for heavy payloads (Glushko's LEK lunar base) or the military's required spaceplane could be placed on the nose as the payload.”
http://www.astronautix.com/craft/buran.htm
For reasons they explain later, they ended up copying the US shuttle, and the side mount configuration.  But obviously they too saw the advantages of a system that put the payload/orbiter on top, rather than on this side, with boosters arranged around the core.  In our case, those would be Titan SRB’s rather than LRB’s, but the idea is the same. 
I’m liking this idea.

[RocketmanUS]

After some recent discussion on the “What if Satrun/Apollo had never been cancelled” thread, it made me think a bit on this.

I tend to think if you had to fly a reusable spaceplane, as it sounds like the political forces demanded in the 70’s, if NASA had kept the S-IC and S-II stages and put the Shuttle on top of that, that would have been best.   

That would have saved the whole expensive development program of the Shuttle SRB’s, it would have put the Shuttle on top protecting it from ET debris, and putting the crew farther away from the fuel, and it would have saved the whole expensive SSME development program, as well as the 8.4m ET development.  The S-II stage just needed a bullet dome on it. 
{Snipped}

If shuttle was placed on top it most likely would have looked different. Perhaps even more like a soda can shape.

So what type of shapes could a top mount reusable vehicle look like with a large payload bay?

[Ben the Space Brit]

So what type of shapes could a top mount reusable vehicle look like with a large payload bay?

A lot like the Liberty crew vehicle, except with a sealed bay rather than a vacuum-exposed cargo rack.  It probably would have had propulsive landing too.

[Quindar Beep]

During the proposal stage for STS in 1969-70 there was an extremely odd STS suggestion from Chrysler, the SERV, which could have made a better STS. It's so different from anything that's actually been built that it's hard to argue against it or for it, though. My guess is that since it was SSTO it would have been too sensitive to the usual upwards creep in launch mass as it was developed.

https://falsesteps.wordpress.com/2012/08/18/servmurp-chryslers-space-truck/

It could have gone with a small winged re-entry vehicle on top, a crewed capsule, or even unmanned with an aerodynamic spike on top for lots of cargo, even more cargo, and "good lord that's a lot of cargo" to LEO.

The Phase A and Phase B studies Chrysler turned into NASA make fascinating reading:

[Lobo]

So what type of shapes could a top mount reusable vehicle look like with a large payload bay?

A lot like the Liberty crew vehicle, except with a sealed bay rather than a vacuum-exposed cargo rack.  It probably would have had propulsive landing too.

I imagine it could have looked like a lot of things.  It really –could- have looked like the Shuttle, with probably some cosmetic differences around the aft to it would correctly with the interstage from the S-II stage.  Maybe more circular instead of more squarish like Shuttle and Buran. 

Personally, as I have discussed previously in this thread, I would prefer a large reusable capsule rather than a winged orbiter.  Basically, a Command module around the size of the Shuttle cabin, something along the lines of Big Gemini.  So a large crew cabin for fairly long durations in LEO, with all the same facilities as the Shuttle, an airlock, a robotic arm, and even perhaps a small unpressurized cargo bay (for storing things like an MMU or some equipment to be used during EVA’s that you’d want to bring back down with you.).  The capsule could have a pusher LAS system and land propulsively like Dragonrider, with parachute backup for ocean abort.  It would really be a command module and service module all in one. 
Behind it would be an expendable cargo carrier, similar to Direct’s SSPDM.  The Big Capsule CSM would perform RCS and act as a tug, perhaps with some help from aft RCS thrusters on the SSPDM.  Not sure if you’d need aft RCS thrusters or not to wrangle the SSPDM and cargo around.  Be good to keep as much of that as possible on the CSM for reuse, and make the SSPDM as much just a simple static carrier frame as possible. 
If not landing propulsively, it could land with a soft wing and skids like Gemini was intended to.  The whole capsule would be fully reusable, with a metallic TPS or PICA heat shield.  As the heat shield would be so large like on a winged space plane, the heavier metallic could be an option.   Could even do tiles similar to Shuttle, as they’d be covered until reentry so no risk of damage during launch or in orbit. 

You could also do something similar to that, with a small lifting body instead of the large capsule, but would function the same.  Or perhaps even a very large lifting body.  The Soviets seemed to think that was the way to go in that Astronautix article on Buran, and they thought the mass penalties for wings were too great:

“As far as the manned orbital vehicle itself, three different primary configurations were studied extensively, as well as a range of more radical proposals. The obvious choice was a straight aerodynamic copy of the US shuttle. The shuttle's form had been selected by NASA and the US Air Force only after painstaking iterative analysis of over 64 alternate configurations from 1968 to 1972. It would obviously benefit the Soviet engineers to take advantage of this tremendous amount of work.
However the NPO Energia specialists who had developed the Soyuz capsule disapproved of the winged US shuttle design. They knew from the extensive aerodynamic studies undertaken to develop Soyuz that there were large weight penalties and thermal control problems in any winged design. Their studies indicated that a lifting body shape capable of high angles of bank at hypersonic speed could nearly match winged designs in cross range. Therefore their preferred 1974 design was an unwinged spacecraft, consisting of a crew cabin in the forward conical section, a cylindrical payload section, and a final cylindrical section with the engines for maneuvering in orbit. This unwinged MTKVA would glide to the landing zone at low subsonic speed. The final landing maneuver would use parachutes for initial braking, followed by a soft vertical landing on skid gear using retrorockets. After a great deal of detailed analysis the definitive MTKVA design proposed in May 1976 had a refined aerodynamic shape with a rounded triangular cross section. The 200 metric ton vehicle had over twice the shuttle's mass and nearly three times the shuttle's payload.
The third configuration was a smaller spaceplane launched by a Proton-class booster. OKB MiG had been developing the Spiral lifting body spaceplane since 1965, but the project was underfunded and years behind schedule. Spiral was an ambitious concept that was to be launched by a hypersonic air breathing first stage. But the spaceplane itself had been refined in form as a result of years of analysis, wind tunnel, and sub-orbital sub-scale model tests. Chelomei's OKB, whose Raketoplan spaceplane had been cancelled in 1965 in preference to Spiral, also had a contender, the LKS. Evidently owing nothing to earlier Raketoplan designs, this used a shuttle-type wing on a smaller 20 metric ton spacecraft.”
But it sounds like the main reason they went with a copy of the Shuttle, was to maximize all of the research we’d already done, rather than have to do the research themselves.
“The specification of the TTZ set forth payload requirements a bit greater than those set for the US shuttle. It required that the OK orbiter be accomplish the following:
•   Denial of the use of space for military purposes by the enemy
•   Research into questions of interest to the military, science, and the national economy
•   Applied military research and experiments using large space complexes
•   Delivery to orbit and return to earth of spacecraft, cosmonauts, and supplies
•   Delivery of 30 metric ton payload to a 200 km, 50.7 degree inclination orbit, followed by seven days of orbital operations and return of 20 metric tons of payload to earth.
•   Exploit the technology developed for the American space shuttle in order to enhance Soviet space technology capability
The MTKVA and Vulkan were used as a starting point, but modified to meet this requirement. Study of the competing designs indicated that despite the evident advantages of the MTKVA approach, there were serious technical and operational problems with that design. There was considerable technical risk in realizing the vertical landing itself - and considerable operational risk in completing the fast and complex series of operations necessary to achieve the landing. There were also problems in ground handling - how to move the vehicle after it had landed, especially if this occurred outside of the normal landing zone. The final analysis of the problems indicated that the rational solution was an orbiter of the aircraft type. There was severe criticism of the decision to copy the space shuttle configuration. But earlier studies had considered numerous types of aircraft layouts, vertical takeoff designs, and ground- and sea- launched variants. The NPO Energia engineers could not find any configuration that was objectively better. This only validated the tremendous amount of work done in the US in refining the design. There was no point in picking a different inferior solution just because it was original.
Therefore a straight aerodynamic copy of the US space shuttle, was selected as the orbiter configuration on 11 June 1976.”
So, I’m not sure.  Perhaps for a vehicle of it’s size and required payload capacity, something like the Shuttle and Buran was the logical choice.  I think the side mount config on the booster was also copied, but obviously they thought putting it on top of the booster would be better.  And I agree with that.  I believe the reason for Shuttle side mount was because they wanted the hydrolox engines on the shuttle so they could be reused, so you can’t put it on top and do that obviously.  Since the Soviets weren’t trying to reuse the hydrolox engines, there really wasn’t much reason Buran had to be on the side.  Unless the horizontal transportation of the stack to the pad played some role in that.  But Since our Shuttle would be stacked and transported vertically as Satrun V was, that wouldn’t have been an issue for us. 
I wonder how much the cost of five new J2S engines per flight to be expended on the S-II vs. the per flight costs of manufacture and refurbishment of the SSME’s would come out?  But I feel pretty sure cost wise, reusing the SSME’s would be on par or more expensive than mass produced J2S’s, that would have been rolling off the production line at a rate of 15 to 40 of them per year (3-6 launches per year was typical rate for STS, with up to 8 launches per year at times, so 40 J2S engines those years).  We might have seen some Merlin 1 like economics of scale start to manifest after a few years of that rate.  Plus it would have invited streamlining and automation in production for more savings as time went on.  As I understand, the SSME was labor intensive because of their low production rate, and there was never any incentive to streamline and automate their production because of that low rate.  Too hard to recoup the investment.  But at up to 40 engines being produced a year?  Certainly there would have been. 

[Archibald]

(Bumping that to no hijack the other thread)

What I realized recently in Dennis Jenkins landmark shuttle book was that the orbiter shape was essentially frozen late August 1971 - months before Nixon green-lighted the program on January 5, 1972.
It was essentially frozen around the full-length payload bay (15*60ft) and a delta wing.
But that was not the end of the day.
On the basis of delta wing and 60ft long payload bay were different orbiters  named MSC-037, 040A, 040C, and 042.
That basic orbiter was able to cope with either 3*J-2S (037), 4*J-2S (040A), 3*SSMEs (040C, here we are!), or... no engines at all.
Yes, MSC-042 was pretty much an American Buran, with a Titan III-L (four engines and larger core)  in the role of Energia !
Significantly, the basic orbiter was also able to cope with most of the booster / SRB / tank combinations examined - and there was many of them. The tank / booster combination ripped through the orbiter engines, and that's why they could be swapped, or even removed entirely.
There was also another variable - when do you light the engines ? series burn, that is to say, in flight ? or parallel burn, meaning start on the pad ?
Series versus parallel, plus a lot of booster options (Big Dumb Booster or S-IC or Titan or SRMs or SRBs, and on), combined into a large numbers of possible combinations.
The debate started in August 1971, and was only closed on March 15, 1972, with the SRB decision...
The fact that a broad orbiter shape could be adapted easily to all these options is a testimony of NASA state of genius or... despair. It is all a matter of point of view.

[Lobo]

The bit recently about the X-37B got me thinking about this again.

First, I like the idea of NASA retaining INT-21 rather than developing the ET and SRB’s and SSME’s, and scrapping all of that great Saturn hardware.  Launching the orbiter on top of the INT-21 stack, means the changes to KSC would have been much more minor.  The pads could have stayed clean and would only have needed some tower modifications.  That would have eliminated a lot of development and modification costs, as well as eliminating the foam shedding problem with the heat shield.  (Which I think was a concern, even from the beginning)

However, the S-ID stage, i.e. the Saturn V-B got me thinking.  It can’t get the Shuttle to space, but I wonder if the stretched S-ID, which could put about 30mt to LEO, coupled with an X-37B like design could have worked?  That would have been a big X-37B design granted, but integrated the hydrolox 2nd stage into the orbiter sort of like how the X-37B has the propulsion section attached to the rear.  Could a big hydrolox tank be integrated into the aft of the orbiter where the SSME’s and MPS was?  This would use J2S engines that would then be reused.  So the Shuttle would have looked like a really big X-37, and mounted to the top of the S-1D. 
The Shuttle would have been longer than STS was, obviously, with even more tiles, but it’d have some interesting differences.  The wings would probably have been smaller, as the whole shape is more of a lifting body. Just like the X-37B’s wings are proportionally smaller than the Shuttles.  A lot of that heavy hypergolic fuel for the OMS pods would be replaced by lighter hydrolox, which would do the upper ascent and the insertion burn.  The Shuttle was usually only in orbit for about 2 weeks, so that’s all the longer the residual hydrolox would need to be kept for the retro burn.  Not sure if that would be a problem or ok.  That means the only hypergolic fuel needed is for the RCS system.  (unless it were to use a ULA type system where it used GH2 boiloff as a monopropellant for the RCS system? If so, then it makes “safing” the orbiter at touchdown much easier.  GH2 and GOX could also be used form the main tank to generate power and water as the individual fuel cells did for the real orbiters).

Maybe this tank would be something along the lines of the S-IV (or S-II) common bulkhead tank, which would be the diameter of the Shuttle fuselage (which could have been fatter than the real shuttle ended up being, making for a shorter tank, and wider payload bay.) 
The wings with landing gear would be located where necessary to facilitate proper balance for landing.  But the tank at that time would be mainly just an empty tank with a few J2S engines on the rear (which weren’t that heavy).  Most of the weight would be forward of the tank anyway. 

Anyway, it seems like a pretty cool idea.  If the S-1D engine ring were recovered, that’d make for a reusable enough system that the politics of the time should be satisfied.  It’s be reusable “enough” like STS was.  (although as we know that didn’t turn out to be any cost advantage).  I suppose the main question is, could a big enough tank be added to perform enough of the ascent burn, orbital insertion burn, and deorbit burn without making the geometry too untenable?
Another interesting thing would be, if it was feasible to do this, if the S-II stage was added (this would assume the INT-21 stack was retained, and S-1D developed), that could boost the orbiter into high enough orbits that it might have been more useful as a satellite repair, launching and retrieval platform (where it was economical or desirable to do that).