Thursday, October 27, 2022

Getting to Low Earth Orbit

I have seen a lot of history flying multi-stage vehicles to low Earth orbit (LEO).  In recent decades,  the trend has stabilized on two-stage liquid rocket vehicles.  In recent years,  a few of those have been made reusable,  with dramatic reductions in the launch price per unit payload mass delivered to LEO.  

I have also quite recently seen a resurgence in interest for using some kind of hypersonic-capable spaceplanes to deliver payload to LEO.  These always presume hypersonic airbreathing propulsion most (or even nearly all) of the way to LEO.  That is NOT going to happen,  for many reasons!  The technology to support such a thing just does not yet exist.   This is true for reasons of some fundamental physics that cannot be ignored,  just because they are inconvenient for marketing or dreaming purposes.

Vertical Launch

Vertical launch of a two-stage vehicle,  whether it is winged or not,  requires rocket propulsion.  Period.  Modern rockets (particularly solids) have a very high frontal thrust density,  about an order of magnitude (or two) higher than any imaginable airbreather.  Thrust to weight ratio greater than 1.5 is required for launch kinematics that do not waste propellants.  Experience confirms that!

There are a very few jet aircraft (gas turbine engines) with thrust to weight greater than 1 subsonically (F-16,  SR-71),  or perhaps even transonically (SR-71),  and only near sea level at that.  Very few indeed!  None of these shows thrust/weight greater than 1 supersonically,  much less hypersonically (no turbine engine can survive at such speeds).   All have service ceilings down in the lower stratosphere (under 30 km).  None have ever exceeded Mach 3.5 speeds,  and for very good reasons of high air temperatures.

The corridor for lifting ascent to orbit is a rather narrow band of altitudes,  and inherently involves a sharp pull-up at very high speeds to enter LEO at any practical altitude.  There is a feasible corridor only if one uses both superalloy metallics and very significant heat shielding,  because it is essentially entry flown in reverse.  The exposure times for ascent are far longer than those for entry,  so the lifting ascent heat protection problem is very much worse than that for entry.  Those high-temperature things are inherently very heavy,  and lifting ascent requires much more of them. 

For a two-stage vehicle of any kind,  there are three variables of crucial importance at the stage point.  Those are staging velocity (most important),  path angle well above horizontal,  and altitude (least important).  The path angle,  if steep,  eliminates the need for the next stage to pull up sharply in the thin air at high altitude,  quite the crucial result!  There are both lift and thrust constraints on that process.  Lift requires huge wings,  which are heavy and draggy.  Thrusted pull-up wastes a lot of propellant.  So,  sharp pull-up is to be avoided,  if at all possible!  It’s just physics!

Two-stage vertical ascent sidesteps all those problems entirely!  For a two-stage vehicle,  the stage point is outside the sensible atmosphere,  and at a very large path angle above local horizontal.  If the staging velocity is not too high,  then stage 1 entry can be had without serious heat shielding requirements,  with (at most) a modest entry burn.  That’s a far lighter inert stage mass.  From there,  the first stage can be flown to a propulsive landing,  or if winged (as recommended here),  can glide back and land horizontally.   See Figure 1.  All figures are at the end of this article.  See also ref. 1 from which the 15% stage inert mass fraction came,  where actually I rough-sized this vehicle.

The “ultimate” version of the two-stage vertically-launched vehicle would be the reusable two-stage winged spaceplane shown in the figure.  If the vehicle stays below hypersonic speeds until out of the sensible atmosphere,  then shock impingement heating is NOT an issue with a second stage mounted parallel to the first!  The same was true with the Space Shuttle cluster.  Be aware that this is a “killer” issue!  It is covered in more detail in ref. 2

The first stage might (or might not) need a modest entry burn for avoiding the need for any heat shield.  It will need significant structure for attaching the second stage,  and for retractable landing gear.  Something like 15% stage inert fraction (counting the second stage as payload) would be reasonable for this winged first stage.  Less is not at all realistic for reusable flight,  and more would prove infeasible with chemical propulsion.  My favorite design approach is mounting them belly-to-belly,  with similar stage shapes,  as indicated on the figure. 

Such a configuration meets all the stagepoint requirements for speed,  path angle,  and altitude,  plus it eliminates any sharp pull-up problems with either stage.  It does so at a “reasonable” inert mass fraction.  All you have to do is size the rocket engines for adequate liftoff thrust/weight ratio,  and perhaps retain a small propellant quantity for go-around or divert at landing.

The second stage flies a non-lifting thrusted gravity turn toward LEO.  There might (or might not) be a circularization burn.  There will be a de-orbit burn.  There might well be modest burn requirements for rendezvous and docking.  A small propellant quantity might be retained for go-around or divert at landing.  This is the sort of thing a second stage can handle at a realistic mass ratio,  with inerts near 15%,  and still shoulder the bulk of the delta-vee (dV) to orbit.  Winged makes a glide landing feasible.

A crucial point here is that every single supporting technology already exists for a design like this.  No new technology development is needed,  only their recombination into this particular vehicle design.  That is a good prescription for a “design-build-fly” project.   Projects requiring technology development,  because one or more supporting technologies do not yet exist in a usable form,  almost never fly.  That’s just an observation from history.  But it is very true.

Horizontal Launch

The horizontal takeoff scenario that is touted by so many enthusiasts,  must use airbreathing propulsion to reach the initial stage point.  For static takeoff,  that implies gas turbine propulsion,  or a turbine-based combined cycle propulsion of some sort.  Turbine has a max speed capability in the Mach 3.0 to 3.5 range,  past which there must be 100% diversion of 100% of the captured airflow around the turbine core,  for it to survive the high captured air temperatures.  Ramjet can take over in the Mach 2 to 2.5 range,  and reach Mach 4 to maybe 6.  Scramjet can take over only in the Mach 4+ range,  and can reach Mach 10+.  Turbine and ramjet are fully operational.  Scramjet might be flying in some prototypes,  but is not fully operational,  not yet!  There are no operational combined-cycle technologies.  There is only parallel burn,  and that can reduce frontal thrust density,  depending upon exactly how it is done.

There is either a stagepoint at high subsonic / low supersonic speed before entering the ascent corridor,  or else a hypersonic stagepoint somewhere within the ascent corridor,  as indicated in Figure 2.  There is more detail given in ref. 3.  The ascent corridor itself is quite narrow in terms of altitude,  as indicated in the figure.  Above it,  the air is simply too thin for there to be sufficient lift at any “reasonable” wing area size.  Below it,  the aeroheating is simply unendurable,  with any schemes that we currently understand,  or materials that we currently possess.  The heat protection technologies required for a lifting ascent simply do not yet exist in a usable form!  Super-ceramics are not yet a usable technology for this application,  because they are high-density,  high-thermal conductivity materials,  and so the backside temperature will not be much below the temperature of the exposed face.  How do you hang onto anything that hot?  See also ref. 4 for more details. 

If transonic to low supersonic at staging,  staging velocity is only about 0.3 to 0.6 km/s,  leaving some 7.3 to 7.6 km/s dV demanded of the upper stage,  and that’s unfactored for gravity and drag losses!  If a single upper stage,  at typical chemical specific impulse,  that’s over 85% propellant in that stage,  leaving little room for structure and no room for payload!  So,  two upper stages are simply required,  for the lower-speed first stage point.  The stage point must be well hypersonic (nearer 2 km/s or more,  Mach 7+) to use a single upper stage,  and the higher the staging speed,  the better.

A sharp pull-up requirement into essentially-vacuum conditions at near-orbital speeds is simply inherent with this type of lifting ascent,  because of the way the corridor bends upward just as speeds near orbital.  That requires a propellant-wasteful thrusted pull-up,  no two ways about it!  Plus,  ascent exposure times to the aeroheating are simply far longer than any imaginable entry exposure times;  making the heat shielding problem very much worse,  and any solutions to it very much heavier.  There is simply no way around that problem currently,  3-D printing capability and super-ceramics notwithstanding. 

Where the air is too thin for feasible values of lift,  it is also too thin for any imaginable airbreather thrust to accelerate or climb the vehicle!  See again ref. 3 for more detail about that.    That simply means chemical rocket propulsion is requiredPeriod.  And chemical rocket specific impulse is lower than airbreather specific impulse!  Period!  There are no other operational technologies to apply for this purpose!  Even nuclear thermal propulsion requires further development for this application,  because of the recent mandates to use low-enriched uranium,  versus the highly-enriched used decades ago.

The problem here is that many enthusiasts are seduced by the higher potential specific impulse values of airbreathing propulsion.  They forget that frontal thrust density in the thin air at high altitude is vanishingly-small with airbreathers!  If it cannot climb and accelerate,  it simply cannot work for ascent to orbit!  Period!  That’s just physics!  Your thrust is proportional to your air mass flow,  and at extremely-low air density,  there really isn’t any!  That is the “service ceiling” effect. 

What this amounts to is that you may have a hypersonic stage point with a two-stage vehicle,  or you may have a subsonic-to-low-supersonic stagepoint with a three-stage vehicle,  as indicated in the figure.  The hypersonic stagepoint is quite infeasible with a parallel-mounted second stage,  due to shock impingement heating effects!   Designing such a flight vehicle with a serial-mounted,  or internally-stowed,  second stage,  is very difficult indeed!  Internal stowage costs inert mass,  and serial mounting introduces fatal center-of-gravity problems for a lifting craft.  Plus,  there are well-known serious-to-fatal high-speed store separation effects.  None of this is likely to meet the needed 15% inert mass fraction for either (or any) stage. 

With the roughly-transonic stagepoint at lower altitudes,  you simply must have a 3 stage vehicle.  Period!  Chemical rocket specific impulses for stages 2 and 3 are simply too low to allow a single second stage,  especially since (1) the first stage dV is so low,  and (2) there is a sharp thrusted pull-up required of the third stage,  or possibly the second stage.  One or the other.  It is NOT avoidable! 

Either way,  you will violate the need for high path angle at staging,  somewhere.  And that also costs you dV,  in one upper stage or the other.  Not to mention the heavier inert masses.  It’s just physics!

Results

With vertical launch,  there are known technology solutions for every issue,  which are not already avoided inherently by that choice.  Winged just makes it more reusable,  if a little bit heavier.  Falcon stages with no wings,  no heat shield,  and minimal landing gear run about 5% inert.  The recommended  vehicle stages must have wings,  heat shields or stage-attach structures,  and fully retractable landing gear.  15% inert fraction will be a demanding design requirement,  but it should be achievable.

With horizontal launch and lifting hypersonic ascent,  assuming chemical propulsion,  there are very serious unresolved technology issues with unready hypersonic airbreathing propulsion technologies,  with unsurvivable hypersonic shock impingement heating,  and with feasible materials of construction and heat shielding for a heating problem far more severe than entry,  due to the extended exposure times.  Unresolved,  these issues make that approach very infeasible!  New technologies must be developed to resolve them,  which pretty much rules out any chance of success as a “design-build-fly” project.  Not to mention the difficulties of achieving realistic 15% stage inert mass fractions with tougher heat shields,  the need to do a propellant-wasteful thrusted pull-up,   and the need for developing brand-new hypersonic airbreathing propulsion technologies for the two-stage option that combine with surface takeoff. 

Conclusions (author’s professional opinions):

I cannot recommend at this time in history the horizontal takeoff approach utilizing airbreathing propulsion in the first stage.  The necessary supporting technologies for heat protection and hypersonic airbreathing propulsion simply do not yet exist!  There is no way to avoid the thrusted pull-up,  and it is unclear how to arrange the stages to avoid fatal hypersonic shock impingement heating. 

I can recommend at this time in history the vertical launch of a two-stage parallel-mounted winged rocket spaceplane.  All the supporting technologies already exist,  thus enabling success at “design-build-fly”.  And the thrusted pull-up,  shock impingement heating,  and extreme hypersonic ascent heating pitfalls are completely sidestepped! 

References:

“Exrocketman” refers to http://exrocketman.blogspot.com,  for which there is a rapid navigation tool on the left of the page.  It is faster and easier than just scrolling,   but you need the title and date of the article.  Click on the year,  then the month,  then the title if need be.  Clicking on any figure in an article lets you see all of them enlarged.  There is an X-out to top right of that screen,  which takes you right back to the article itself.

#1.  G. W. Johnson,  “Two-Stage Reusable Spaceplane Rough-Size”,  7 Sept. 2022,  on “exrocketman”.

#2.  G. W. Johnson,  “Shock Impingement Heating Is Very Dangerous”,  12 June 2017,  “exrocketman”.

#3.  G. W.Johnson,  “About Hypersonic Vehicles”,  1 June 2022,  on “exrocketman”.

#4.  G. W. Johnson,  “Entry Heating Estimates”,  1 April 2020,  on “exrocketman”.  

Figure 1 – The Vertical-Launch Two-Stage Winged Rocket Spaceplane Approach

Figure 2 – The Horizontal-Launch Two or Three-Stage Lifting Ascent-to-Orbit Approach


4 comments:

  1. Looking at figure1 it seems like the only point at which the two staged vertical launch configuration uses its wings is for the lifting glide back to the launch site. Are the propellent savings from gliding rather than preforming a boost back burn substantial enough to justify the dry mass of the wings? Or is there another benefit to them that I have missed?

    ReplyDelete
    Replies
    1. It's less about a tradeoff and more about achieving a desired capability. Gliding to a landing covers a range, which could be cross-range. If you add go-around capability by retaining some propellant, you add a great deal of crew safety.

      Delete
    2. That makes sense, though the first stage presumably is uncrewed, so it could get away with slightly lower safety. Falcon 9 landings have gotten pretty reliable, pure propulsive return and landings may be enough for unmanned stages for the near future.

      In the long run I could see even uncrewed stages paying a lot of dry mass for near airliner like reliability so that they could be in service for decades, but I'm not sure how far out that would be. For near term applications cross range capabilities seems a bit overkill on unmanned stages. But maybe that is old space thinking on my part.

      Delete
    3. You might be right. I was just thinking crewed to ease the automatic controls design, and increase the cross-range capability if something goes wrong.

      Delete