How to fly a spaceplane

[Spricigo]

Ok,   know that is very craft-dependent but I'm interested more on general advice to help figure out the rest by myself than a in deep guide.  (BTW, also know there is lots of usefull info around but kinda overwhelming for a spaceplane newbie like me) 

 

Question is: what kind of flight profile get more from the airbreathers engines in spaceplanes?  And what kind of kind of common mistakes is expected from inexperienced spaceplanes pilot? 

 

Thanks in advance for all the good input I'm sure will come. 

 

 

Edited December 27, 2016 by Spricigo
Missclik post

[Aegolius13]

First off, as you say, it's VERY craft dependent.  But I will assume your plane is relatively low on the power side, because that's generally the way to make it most efficient (i.e., so you don't haul around a ton of unneeded engines).  I'll also assume you're running Rapiers as your jets, since they're the best for most spacebound applications.  

Since Rapiers don't hit full thrust until you're already going quite fast, spaceplanes often have the most trouble getting from takeoff to around 450 m/s or so, when thrust and speed starts to shoot up.  In particular, transonic drag can be a bear.  There are a couple options to get around this.  One is to do very little climbing (less than 5 degrees) until you've gotten to around 450m/s or so, then pitch up.  Another is to sacrifice acceleration at first, climb at subsonic speeds to where the air is a bit thinner (maybe 5km or so), and then level out to break the sound barrier.  Really underpowered and/or draggy planes may actually need to dive at this point; obviously if you go the dive route you'll want to have enough initial altitude that you won't crash.  

Once you've hit the sweet spot, you can afford to pitch up a little.  But from here until the jets give out due to altitude, your top priority is usually to get horizontal speed as high as possible.  Thus, you don't want to climb too steeply.  I generally do most of my climb at a 10-15 degree inclination, then level out to more like 5-10 around 15km or so, to eke out all possible speed from the jets.  Depending on your TWR, your plane may get pretty hot during this stretch.  But if possible, you should design your craft using high temp-resistant parts, or protecting sensitive parts in a cargo bay.  If you have too much power, I'd suggest pitching up rather than turning your throttle down. 

I generally manually switch my Rapiers to closed-cycle mode when my speed starts dropping, rather than waiting for the jets to completely cut out.  I like to ascend at about a 10-15 degree inclination at this point.  Then slowly point down to horizontal, as you would with a conventional rocket launch.  

If you're packing nukes for vacuum use, it's a good idea to use them during the last half of the ascent, since they're much more efficient than closed cycle Rapiers.  I like to turn them on around 20km, when the jets start losing thrust, and leave them on until my apoapsis hits the desired mark.  Since their thrust is low, it's usually necessary to supplement with a burst from the closed cycle Rapiers, but I try to keep this as short as possible.  It's ultimately possible to carry almost no oxidizer, which helps keep things simple since both your nukes and jets can share LF.  

As far as common mistakes: I'd say newbies tend to put a lot more engine on their planes than is necessary, either because their designs are draggy, or because they want to climb fast, more like a rocket.  Note there's nothing wrong with having an overpowered beast of a plane, but as mentioned above, this tends to limit your total delta-v, which is particularly important if your plane is going somewhere beyond LKO.  So it's often useful to iterate your design, removing engines, simplifying/streamlining the design, and working on your ascent profile to do more with less.  

[tseitsei89]

The different "stages" of "normal" spaceplane ascent go as following (assuming rapiers or rapiers + nukes):

1. Subsonic stage. Get the speed above 400m/s. The point is to get to supersonic speeds where rapiers start working properly and producing significantly more thrust. If your plane doesn't have huge amounts of (unnecessary) thrust you probably have to go pretty much horizontal here and not try to gather altitude yet.

2. After you have sufficient speed (~400-450m/s) you start the climbing stage where you slowly pull the nose up and fly up to thinner atmosphere to minimize drag. The ascent angle here is very craft dependent but can be something like 10-20 degrees.

3. Acceleration stage. Once in the thinner atmosphere you start to lower your ascent angle to keep in the region where jet engines still work to gather as much speed as possible using airbreathing jets. You should be flying almost level flight (just slightly ascending slowly) at around 20-23km here and try to get speed as high as possible before switching to rockets.

4. Rockets and/or nukes stage. Once you stop accelerating with jets it is time to switch rapiers to rocket mode (if I'm using nukes I usually fire them up a little earlier and then switch rapiers to rocket mode after they flame out completely). Depending on TWR and your initial ascent you probably have to burn little over the prograde marker on the navball at first and then lower the nose to prograde when time is right, Also this is the stage where you will have to watch the overheating issues. Just keep burning (as close to prograde as you can without blowing up) until your Ap is out of atmosphere. If executed correctly your Pe should also already be pretty close to appearing above kerbins surface once your Ap reaches 70+km.

5. Circularization stage. Obvious and easy. Keep pointing prograde. Coast to apoapsis and circularize.

[bewing]

My advice is going to be different than that above in that they talk repeatedly about pitching up and down and whatever. Don't do that. Whenever you make an attitude adjustment on your plane by hand, you lose immense amounts of speed. How to avoid it:

When you have SAS in Stability mode, your nose naturally rises over time. Use this fact. If you want to "pitch up", then just wait for 30 seconds and don't do anything, and it will happen by itself.

If your plane is properly designed, SAS will need to keep a small amount of upward pitch control. If you momentarily turn off SAS, then your nose will drop a bit. Use this fact. You can flash SAS off for an instant by holding down F. Using F makes your plane go faster, instead of slowing it down.

Some additional points that differ a bit from what's been said before:

Your spaceplane has wings for a reason. They are there to provide lift, to raise your altitude. The more you pitch up, the more work your engines have to do, and the less work your wings can do. So you are defeating the purpose.

OTOH, the more horizontally you fly, the longer it will take to get to orbit, and the more drag you will encounter. So there is a hard-to-find intermediate point between climbing too slow and too fast.

The biggest mistake newbie SSTO pilots make is in climbing too fast, because they don't expect a launch to take 15 minutes.

Your airbreathing engines burn 10 times as much fuel going fast at low altitudes than at higher altitudes. So wait and climb for a bit at 200 m/s before you go supersonic.

So, a post-launch flight profile:

Climb at 200 m/s to 5km. Whenever your nose goes above 5 degrees AoA, flash F to get it back down. At 5km, go to full throttle, flash F to bring the nose down to 2 or 3 degrees.

Sit and watch as the plane accelerates and climbs to at least 10km. Let the nose rise during this time. At a little over 10km, hold F to bring the nose back down to around 5 degrees AoA. Let it climb and accelerate again.

At around 14km, hold F to bring the nose down to maybe 2 or 3 degrees to finalize your speed run. If you have to level the wings, do it now. Let it accelerate and climb and don't touch the controls again until you need to stage some engines. Adding nukes at around 20km can be a very good thing. Going closed cycle on rapiers when the speed maxes out can be a very good thing.

Once you are above 29km, it depends on what tech you are using. If you are using MK1 tech, you need to climb as fast as you can to 42km to avoid heating. With MK2 or MK3, start bringing the nose back down to 10 degree AoA. Then switch to map mode, and watch your Ap. When it gets to maybe 55km, go pure prograde (not horizontal). When the Ap gets to 65km, kill your engines. Then circularize when you get near the Ap. Note that circularizing can take another 15 minutes.

 

Edited December 28, 2016 by bewing

[GoSlash27]

Everything Tseitsei said.

 There are a ton of design tips to help you build a better space plane, but these are a few design tips that will help you *fly* better.

1: Don't design your plane to be maneuverable. Your plane should be a jet- propelled lumberwagon into space. Maneuvering is bad. Every change in attitude in direction or attitude makes drag. You want your control inputs to be minimal and have very little effect. To that end, you should have very little control authority.

2: Your vertical rudder should not respond to any input. There is no reason to coordinate using active rudder, so turn it off. Otherwise you'll engage SAS and it'll cross control with bank, adding yaw. That makes drag and drag is bad.

3: Turn off your reaction wheels during climbout. Same deal; they hold heading funny and put you into draggy attitudes.

4: Don't use 1 set of controls to respond to 2 inputs. That makes your handling twitchy and imprecise, which makes unnecessary drag and forces you to make more inputs.

Good luck,
-Slashy

 

[Spricigo]

 

Thanks for the answers,  however I noticed the need to clarify a few points:

1. No problem in staging as long as most of my craft is recoverable . I'm using FMRS here and experimenting with the concept of a rocket plane launched from a jet,  both recoverable. 

2. Since my plan it's to use it in a career game I would like to consider panther and whiplash also since they are available earlier.  (in fact most of my experiments were with whiplash and none with rapier) 

3.  Going further away than LKO with a spaceplanes is not in my plans at the moment,  maybe when I get experience I try it.  

 

Anyway,  advice for SSTO is still welcome since,  I suspect,  there are other players interested. And I will try it at some point also. 

 

[foamyesque]

Bah. Me, I design and fly my planes very differently than most of the above.

 

1. Thrust is life. I wouldn't drop below more than one RAPIER per 25 tons of craft; you can get away with even less, but unless you're hyperoptimized for drag, its really, really painful. More thrust means you have more acceleration low down, so you spend less time getting to speed, and it means you have more speed higher up, which means you need to run your less efficient rockets much less. Better TWR also means you waste less dV against drag and gravity when you punch it for orbit on the rockets. Adding more engines can and often does increase your payload capacity, provided you can handle the thermal loading.

2. Contra @GoSlash27, I design my spaceplanes to be agile, with either relaxed stability (CoL in CoM) or sometimes even dynamic instability (CoL ahead of CoM). I do this not for ascent -- though it means almost no control deflection is needed to steer, which helps drag-wise -- but for re-entry and landing, where being able to generate large AoAs quickly is very useful.

3. Balance is crucial. You want to eliminate longitudinal CoM shift if at all possible, so you get consistent flight dynamics regardless of if your fuel tanks are full or empty or you have anything in your cargo bay.

4. You don't need many air intakes. One shock cone can easily feed six RAPIERs throughout flight; if maxing your throttle on the runway doesn't cause flameouts you've got more intakes than you need.

5. Nothing fancy is required for ascents. Point your nose to 5-20 degrees, depending on your TWR, drag, and thermal balances, and just maintain that until your apoapsis hits the orbit you want. Swap to closed-cycle once your airbreathers stop accelerating you, usually around 22km for RAPIERs. I like to aim for an apoapsis of 105km; drag will usually round that off to 100km.

Edited December 28, 2016 by foamyesque

[Aegolius13]

With Whiplash or Panther, the idea is basically the same, though you may be able to avoid the crawl up to the sound barrier since they have better low-end power.  As with a Rapier, your goal is to build up all possible horizontal speed before the engines cut out (which will happen lower and slower).  Of course, since these engines have no jet mode, you'll need a separate rocket component.  Even if you only get up to 15km or so, you still want vacuum-optimized rocket engines - sea level stats can be ignored.  Terriers and Aerospikes are generally my favorite, though with the 'spikes you may need some more reaction wheel torque to hold attitude.  Note that you can turn on your rocket engines before you cut out your jets in this case; I generally do it whenever I start to lose speed, and just let the jets burn out naturally.   

 

One other general thing to add (apropos of the control authority debate): before you finalize a design, make sure you can fly and land it on the way down.  This is a pretty different beast than going up, since you may need to maneuver more; you won't have vectored thrust keeping you steady the whole time; and your center of mass may shift after losing fuel and cargo. 

This entails two main things.  You want to make sure that, on the high/fast/hot part of your descent, you can hold your plane at the desired attitude and have it not explode or flip.  Keys to this are designing for heat resistance, having enough reaction wheel torque, and having your center of mass in front of your aerodynamic center.  Then, once you get into the slower, lower, thicker-air part of the descent, you need to keep your stability, have enough control authority to do the necessary maneuvers, and have a reasonable stall speed so you can land.  You can certainly turn your jet back on for this part if you need to, and if you saved some fuel, but it's good to get a handle on landing mostly or totally unpowered.  

Take particular note of how your ship handles with the fuel tanks empty.  Depending on your design, your center of mass may shift dramatically (most likely towards the rear, if that's where your engines are).  This can lead to a flip-happy plane.  It's easy to do some testing by launching your ship with just enough fuel in the tanks to get you going.  

Cautionary tale: a couple months ago I sent an SSTO to Laythe and back.  Everything went fine up to reentry, but it took many, many, hours and many, many saves to land it because I forgot to optimize the final design for landing, and it could barely leave prograde at all without flipping.  

[Mikki]

I tend to make my designs as they need no input to reach orbit after takeoff, sounds crazy but works very good.
After take-off i aim for 10m/s climbrate, acceleration and the curvature of Kerbin make the rest.
This with Mk3 fuselages and moderate payload capacity, around 20 tons.
Drag and lots of control input is your enemy while ascend.
TWR at launch like 1:1.15, and increasing.

[Blaarkies]

8 hours ago, foamyesque said:

so you spend less time getting to speed, and it means you have more speed higher up, which means you need to run your less efficient rockets much less.

But there are more inefficient engines running simultaneously, which at least uses more fuel per second of use. Your method definitely wins the reward per effort ratio to get stuff in orbit i guess...

[foamyesque]

Just now, Blaarkies said:

But there are more inefficient engines running simultaneously, which at least uses more fuel per second of use. Your method definitely wins the reward per effort ratio to get stuff in orbit i guess...

You use more fuel per second, yes, but:

 

1. Drag losses (both aero and gravity) accumulate with time. Shorter burns lessen the time they have to work; so if for example you have aero drag losses of 1000kN and a thrust of 2000kN, you only get 1000kN of force going to acceleration; if you have a thrust of 3000kN, you get twice as much acceleration for only 50% additional engine mass. Since planes and SSTOs generally have high dry:wet ratios and much higher aero drag losses (compared to staged rockets) this is much less of a penalty than it is with rocketry for a much larger benefit.

2. The additional speed above 1500m/s on the airbreathers, if you can get it, saves you a giant pile of oxidizer (or, for nukes, a long, slow, draggy acceleration), and hence saves weight overall.

 

[Blaarkies]

19 minutes ago, foamyesque said:

You use more fuel per second, yes, but:

 

1. Drag losses (both aero and gravity) accumulate with time. Shorter burns lessen the time they have to work; so if for example you have aero drag losses of 1000kN and a thrust of 2000kN, you only get 1000kN of force going to acceleration; if you have a thrust of 3000kN, you get twice as much acceleration for only 50% additional engine mass. Since planes and SSTOs generally have high dry:wet ratios and much higher aero drag losses (compared to staged rockets) this is much less of a penalty than it is with rocketry for a much larger benefit.

2. The additional speed above 1500m/s on the airbreathers, if you can get it, saves you a giant pile of oxidizer (or, for nukes, a long, slow, draggy acceleration), and hence saves weight overall.

 

The trick is to find a nice balance between power and weight saving to ultimately increase DV reserves. The drag losses are exaggerated(100 tons of drag force? for 200 tons of thrust? thats a brick with some engines on it). Planes generally have high dry:wet ratios, that is mostly true, but I believe @bewing has the real solution to OP's question:

10 hours ago, bewing said:

Some additional points that differ a bit from what's been said before:

Your spaceplane has wings for a reason. They are there to provide lift, to raise your altitude. The more you pitch up, the more work your engines have to do, and the less work your wings can do. So you are defeating the purpose.

The biggest mistake newbie SSTO pilots make is in climbing too fast, because they don't expect a launch to take 15 minutes.

 

Edited December 28, 2016 by Blaarkies

[tseitsei89]

Yeah... There seems to be a lot of different opinions here but I'm just gonna weigh in here again and say just this:

When dealing with spaceplanes LISTEN TO @GoSlash27!!!! He is one of the very best spaceplane guys in these forums. Really. Honestly.

[GoSlash27]

*embarrassed look*...

 I appreciate that, @tseitsei89, but there are quite a few folks on this forum that I'd consider spaceplane gurus including (but not limited to) @Starhawk, @Right, and @foamyesque. I'm sure I'm forgetting people, but you know... senility...
 We don't always agree on everything, so it's wise to pay heed to their advice as well.

Best,
-Slashy

 

[tseitsei89]

21 minutes ago, GoSlash27 said:

*embarrassed look*...

 I appreciate that, @tseitsei89, but there are quite a few folks on this forum that I'd consider spaceplane gurus including (but not limited to) @Starhawk, @Right, and @foamyesque. I'm sure I'm forgetting people, but you know... senility...
 We don't always agree on everything, so it's wise to pay heed to their advice as well.

Best,
-Slashy

 

Yeah I know that at least Starhawk is one of these gurus as well. Havent been that active in the forums to know everyone. So sorry if I'm missing some names here.

But your forum posts were the ones that taught me how to build and fly a spaceplane. So thanks for that

[Chrysoprace]

Also a spaceplane newbie here. Just started trying to mess around with them after I unlocked the Airspike and Whiplash. Like you, I want to figure it out on my own but still need some pointers. I get far more satisfaction from the "try stuff and see what works good" method than I do from looking things up. I have found this thread to be very helpful. 

In looking about for answers to my questions I have found that a lot of the information available is out of date, probably because it is from older versions. I thought it would be good to contribute to this thread with my findings and a few pointers that I found on my own. Maybe I can help some other spaceplane newbies. 

Firstly, a lot of older guides talked about cramming as many intakes as possible onto a spaceplane to increase operational ceiling, but my experiments showed that this is no longer the case - the increased drag appeared to be the main factor, making the extra intakes counter productive. In fact, in all of my tests drag seemed to be the single most important variable. The older advice for ascent profiles also seems to be obsolete now, presumably due to the new aerodynamics, although it is somewhat similar to the advice given above, just with steeper angles.  

Secondly, when staging your rocket engines, the most efficient point seems to be when the thrust from your air breathing engines peaks. This was true for my tests, where I just pointed the plane at 20 degrees inclination due east, got to a 100km orbit, and checked how much oxidizer and liquid fuel was left in my tanks. It may not be true for a more complicated ascent. I have not played with any Rapiers yet, so this is maybe only applicable if you are using a combination similar to my Whiplash and Airspike setup. 

Thirdly, I found it very helpful to put a waypoint at the spawn point on the runway. Just go to the SPH, click on a Hex or something, launch it, right click it, go to Kerbnet access, make a waypoint called “Runway” or “Land Here” or something, then recover it. This makes it far, far easier to get lined up with the runway after reentry. If you can get the waypoint aligned with your navball’s 90 or 270 degree line you are coming straight in.

Lastly, a couple of my own questions. 

I have read about cutting throttle to keep the air breathing engines running longer at high altitude. Is this true and/or helpful?

Are Rapiers and maybe Nukes actually better than a combination of Airspike and Whiplash? I understand that the total weight is less, but the Rapier is less efficient in both modes, and although the Nukes are super efficient they also seem heavy and weak.

Edited December 30, 2016 by Chrysoprace

[foamyesque]

On 2016-12-28 at 5:02 PM, GoSlash27 said:

*embarrassed look*...

 I appreciate that, @tseitsei89, but there are quite a few folks on this forum that I'd consider spaceplane gurus including (but not limited to) @Starhawk, @Right, and @foamyesque. I'm sure I'm forgetting people, but you know... senility...
 We don't always agree on everything, so it's wise to pay heed to their advice as well.

Best,
-Slashy

 

Hey, that's nice of you to say @GoSlash27. For what it's worth, even when we disagree on things, you're one of the people whose thoughts I pay attention to.

 

[GoSlash27]

1 hour ago, Chrysoprace said:

I have read about cutting throttle to keep the air breathing engines running longer at high altitude. Is this true and/or helpful?

Chrysoprace,

 Yes it is true, but no it is not helpful. Beyond a certain altitude and Mach number, air breathing engines will only create *negative* thrust if they're still running. Better to shut them down and switch to closed cycle.

1 hour ago, Chrysoprace said:

Are Rapiers and maybe Nukes actually better than a combination of Airspike and Whiplash? I understand that the total weight is less, but the Rapier is less efficient in both modes, and although the Nukes are super efficient they also seem heavy and weak.

Well... that depends on what you're doing. If you're going interplanetary, than carrying the nukes is worth it. Otherwise, no; the RAPIER is King and the only reason to not use it is because you haven't unlocked it yet. The main reason isn't the additional mass, but the additional parallel nodes. They come with a big drag penalty. You need more air breathers to overcome the drag of the rockets and more rockets to overcome the mass of the air breathers. These combine to create a larger and less efficient spaceplane to accomplish the same job.

Best,
-Slashy

[Raideur Ng]

Drag-optimized SSTO to carry your space-optimized vehicle into orbit. Release, explore. Mixing the two gets ugly.

[GoSlash27]

2 hours ago, foamyesque said:

Hey, that's nice of you to say @GoSlash27. For what it's worth, even when we disagree on things, you're one of the people whose thoughts I pay attention to.

foamyesque,

 Likewise, and I never take offense when you disagree with me. There are a lot of different ways to approach spaceplanes and a lot of different ways to use them.

Best,
-Slashy

[Whisky Tango Foxtrot]

20 hours ago, Raideur Ng said:

Drag-optimized SSTO to carry your space-optimized vehicle into orbit. Release, explore. Mixing the two gets ugly.

Or put a docking port in-line with the SSTO's centre of mass (a shielded port on the nose is good for this) and rendezvous an already-in-orbit Nuke tug with it for any long-range work. I have several of these patrolling the Kerbin system (refuelling at stations orbiting the Mun and Minmus using fuel mined and converted on the surface) and consider them to be an essential part of my space infrastructure. I only ever plan my launches to get their payloads into a stable orbit outside of Kerbin's atmosphere; the tugs always take things the rest of the way.

[AeroGav]

Space plane flight has 3 critical phases

1. getting supersonic

2. the speed run (reaching highest possible speed on air breathing power)

3. flight to orbit. (raising AP over 70km with rocket power)

This list is in chronological order but it also works as a "most likely to screw up on" ordering system as well, since most players seem to do 3) well (experienced in rocket flight) but struggle with 1).

Getting Supersonic

The advice you see most frequently is "fly level straight after takeoff , don't try to climb till over 400  m/s"   or sometimes people will say,  "climb at 20 degrees to 8km then lower nose to 10".

The first piece of advice doesn't work with every design,  and the problem with naming specific climb angles and acceleration altitudes is that they are specific to one aircraft only.   Optimum climbing angles will change depending on the thrust:weight ratio, and speeds and altitudes with wing loading.

I prefer to use AoA (angle of attack - the difference between where the nose is pointing and where the plane is going) as a more universal indicator.    My "universal " rules that should be adaptable to any SSTO :

1.  Don't raise your nose more than 5 degrees above prograde or you'll have excessive drag.  

2.  Between 240 m/s and 440 m/s is the high drag transonic region.  After 440 m/s drag is worse than below 240, but not as bad as being transonic.  Don't try to climb in the transonic region.   Go into a shallow dive or  if down on the deck, just fly level.

3.  As you get higher, the air will get thinner, reducing both drag and thrust.   Up to 14km (Panther)  17km (whiplash) or 22km (R.A.P.I.E.R) thrust falls more slowly than drag, so it is easier to achieve high speed in level flight at higher altitudes - provided you can do so without violating rule 1 or 2.

 

Case in point -   someone downloaded one of my career mode planes and complained they couldn't get over 330m/s.     They were trying to do it at sea level,  because everyone told him that's how you fly an SSTO.

https://kerbalx.com/AeroGav/Basic-Tourist-SSTO

I had to add some pictures to explain how to get supersonic on this aircraft -

The exact altitude you should transition to supersonic flight depends on the airplane.    Some SSTOs have much smaller wings relative to their weight, but more power.   For example,  I had a play with this from @Val's back catalog -

https://kerbalx.com/Val/SSTO-XS-160-Olympus-Mk3

With a 290Ton mass and only 2 pairs of big S wings to support it,  this thing thunders off the end of the runway at 120m/s and almost stalls.   By the time you've gotten fast enough to not have the nose more than 5 degrees above prograde, you're into the transonic region.  The airplane is using AoA to communicate with you - and this one is telling me it wants to go supersonic right on the deck, loud and clear !

At the other end of the spectrum is my first seaplane.  

https://kerbalx.com/AeroGav/K133--Curlew

Like the Olympus, it has two pairs of big S wings, but it only weighs 37 tons - and they aren't the only lifting surfaces.

It can get up to 13km before being forced to raise the nose above 5 AoA or break the sound barrier.

 

The Speed Run

The next task is to get to the highest possible speed on air breathing engines.   In level flight, the optimum altitude for this is 22km on a Rapier ship, because up to 22km, thrust declines more slowly than drag.  After that, power tails off really quickly.  However, the "keep AoA below 5" rule still trumps all, so if you are one something heavy with small wings,  you may not reach 22km till the end of your speed run.

Note that RAPIER engines benefit from a ramjet boost as you go faster, but only up to a point.    Peak power is at about 1150m/s.  After that , the power declines the faster you go.  However it's still managing 80% of it's peak at 1350 m/s.    After that, thrust falls away more quickly.  By 1650 m/s,  it's producing no more power than it does at mach 1.   By 1800 m/s, thrust has fallen to zero.

 

Flight to Orbit

During the speed run you were probably pitching the nose down to stop the airplane climbing above 22km. But once we engage the rocket engines,  we no longer have to stay low enough to keep our jets fed.   The only thing that matters is getting the best lift:drag ratio we can to make the most of our rocket fuel.  At supersonic speeds, this is achieved at an AoA of 5 degrees.   On a plane where the wings were attached to the fuselage with an incidence angle , you 'll probably be better off with a lower body AoA , depending on how much incidence the wings have.   With 5 degrees or more, you probably want a body AoA of zero, prograde lock.

see the seaplane  video for depiction of the speedrun and flight to orbit

 

[foamyesque]

@AeroGav: It's been my experience that absolute maximum speed level flight is easier to achieve on the deck with RAPIERs: the Mach limit at which they hit 0 thrust is much higher in absolute speed because of the faster speed of sound. I also have tended to find that the best spot to pile up speed is the 10-15km region, where your wings can support you as you racket up from 1000m/s to 1500m/s. As I get up to 20km+, unless carrying way more wings than needed anywhere else, there's not enough lift+engine thrust to keep a steady (or accelerating) climb rate whilst simultaneously accelerating. I tend to have 95% of my maximum speed by 15km.

This also aids in a very simple flight pattern of finding an angle that works, and just holding it. Faffing about with squeezing every last possible erg on the airbreathers can take forever, too. You can get the vast majority of a spaceplane's efficiency savings without needing an hour-long flight to orbit; with a non-minimalist TWR and good drag characteristics, powered flight can be five minutes or less, which is right there with rockets.

 

[AeroGav]

2 minutes ago, foamyesque said:

@AeroGav: It's been my experience that absolute maximum speed level flight is easier to achieve on the deck with RAPIERs: the Mach limit at which they hit 0 thrust is much higher in absolute speed because of the faster speed of sound.

 

Thing is though, i think the speed of sound declines pretty rapidly above 5km and at 10km it's much the same as at 22km.Drag and heat are fierce below 17km !

Quote

I also have tended to find that the best spot to pile up speed is the 10-15km region, where your wings can support you as you racket up from 1000m/s to 1500m/s. As I get up to 20km+, unless carrying way more wings than needed anywhere else, there's not enough lift+engine thrust to keep a steady (or accelerating) climb rate whilst simultaneously accelerating. I tend to have 95% of my maximum speed by 15km.

 

You will accelerate quicker because you have more thrust, for the same reason you'll reach a higher speed in a climb.  

However, in level flight, when drag starts to become a limiting factor you're probably better off a little higher up.  It does depend on the craft design, you need enough wing to fly at 22km at these speeds without excessive AoA, otherwise stay lower.   If you're building a pure chemical SSTO, you probably don't fit as much wing,  as you'll have TWR > 1 in rocket mode.

My LF-only seaplane ssto only has TWR of about 0.3 so it needs those wings to hold it up there while the nukes do their thing.  Plus, all those strakes and big S wings give me somewhere to keep the liquid fuel.

 

 

Quote

This also aids in a very simple flight pattern of finding an angle that works, and just holding it.

If you are saying, hold a constant AoA, that is fine because  i often do that and it's transferrable from one design to another. AoA doesn't care about wing loading or TWR.   A lot of my aircraft have angled wings and i just set prograde from engine start to orbit and leave it alone - does the job well enough for most uses.

If you're saying, hold X climb angle, well that's going to change from one plane to the next and it's a lot of trial and error to find that simplistic pitch angle you can hold all the way to orbit, if you can only use it once. 

Quote

Faffing about with squeezing every last possible erg on the airbreathers can take forever, too.

 

During the speed run i'm looking at total drag and total thrust.   Rapier in Air breathing mode has 4x the ISP of a nuke, but when thrust has fallen so low, 75% of it is just going to overcome drag, it's time to "punch it".

Quote

You can get the vast majority of a spaceplane's efficiency savings without needing an hour-long flight to orbit; with a non-minimalist TWR and good drag characteristics, powered flight can be five minutes or less, which is right there with rockets.

 

The Curlew is an extreme example, with a single rapier pushing >37tons of seaplane with no oxidizer. Looking at the time stamps on the video - 

Lift off from the water to mach 1 at 13km   - 10 minutes 25 seconds

Mach 1 to LV-N activation at 1415m/s, 22km  - 9 minutes 19 seconds

LV-N activation to MECO   2200m/s , 42km -  4 minutes

So about 25 minutes powered flight.       Coasting, re-entry, docking etc. sink more time imho,  it just depends how you like to spend your time in KSP.  Also I think you can see from the video, the Curlew is pretty undemanding.  It flies mostly "hands off", i just add or remove a notch of pitch trim every now and again in my attempt to chase max delta V !  I'm flying with keyboard btw and SAS off, so am relying on it's natural stability to hold course.

[foamyesque]

I meant find a specific navball pitch angle, not AoA. AoA varies, as it should; you want higher AoAs for max lift on takeoff and, as speed builds and you start generating enough lift to actively climb, your prograde vector will pull up to your noses' pitch, at which point you proceed to go supersonic. Then your nose stays pitched under your prograde heading to keep your climb rate under control so that your airbreathers have enough time to build speed, until the air thins out enough that your prograde marker starts dropping back down to match your nose again, which is about when you're in your final climb-with-minimal drag phase of the speed buildup. Somewhere around 21-22km the RAPIERs will cease to accelerate the craft, which is my cue to swap to rockets.

Now, granted, this is how I fly the spaceplanes I design, and I design them to fly the way I like to, but when I've borrowed planes from other people's posts on here, I've generally found the same approach works. The precise angle I choose depends on the thrust, lift, drag, and thermal balances of a craft, but 5-15 degrees is generally the band, with high TWR crafts hitting the upper end and lower TWR ones the lower. A non-trivial problem with low-TWR spaceplanes IME is spending too long in the 20-25km area, where drag is still significant. A 5 degree climb rate pushes you through that area with 130m/s+ (depending on your horizontal velocity), which is about the minimum I consider acceptable. My powered flight (excepting circularization) generally ends with 105x5km (or so-- the periapsis varies by a couple of kilometers) as the orbital parameters, at an altitude of 30km or so. There's enough vertical speed that drag will only pull it down to 100x0km - 100x-5km, which is a trivial circularization burn (I have in fact completed it with TWRs at or under 0.05 :v).