SSTO Frustration

[Xyphos]

I'm trying to design a Mk3 SSTO rocketplane that can LAND on Minmus in a single leg, no refueling.
the trouble is, getting the dV requirements; leaving Kerbin requires 4000+ due to drag and I have no problems getting the vessel to orbit,
but the Hohmann Transfer eats up the remaining fuel and I have nothing left to land with and crashes into Minmus.

I've seen videos of people doing SSTO's and landing on Minmus but I can't figure out how you guys are doing it
 

[Starhawk]

You will need something like 1300 m/s to get from LKO to Minmus surface.

As far as the videos go, the version with which the video was made matters a lot.  In the pre-1.0 days turbojets were extremely overpowered and one could get to orbit without using chemical rockets for anything except a tiny circularization burn.  I made SSTO's to Duna and Laythe back in the day.

Since 1.0 and the attendant aerodynamic changes SSTO's are much more difficult and the margins are much thinner.  It takes quite a bit of efficiency to have any significant dv left over once you make orbit.

Having said that, it is possible.  This one can land on Mun or Minmus and then come home.

Happy landings!

[Streetwind]

The question of single stage vehicle dV post-ascent is one that can pretty much be directly replaced by the question of single stage to orbit payload.

Since dV is directly computed out of the ratio of fuel mass to dry mass of the vehicle, then that means that a vehicle that can lift more fuel into orbit can achieve a better dV once there. The trick here is that you're not looking for absolute payload mass, but rather for payload fraction. In other words, it will not help you if you can lift twice as much fuel to orbit in a plane that weighs twice as much itself; only lifting twice as much fuel with a plane of the same weight will help you.

Now, since you've not provided a screenshot of your craft, we can't make suggestions to improve it. Instead, may I direct you to the Stock Payload Fraction Challenge thread? If you're not afraid of ridiculous designs, there are some planes there that make it to orbit with more than half their entire mass being dead payload. Perhaps you will be able to draw some insights from them that help you out.

Failing that, I'm paging resident spaceplane junkie @GoSlash27

[Nich]

Are you getting 1500+ dv off air breathing?

are you using a shielded docking port or some other blunt heat protection?

are you using struts or fuel lines? These add drag and weight that kill you.

how close are you to 15 ton per rappier?

how close do you come to crashing into the ocean?

are you using a five degree aoi on your wings?

Edited March 21, 2016 by Nich

[Empress Neptune]

Increasing you range will come down to two factors: how efficiently you fly and how much fuel you're carrying. Assuming your flight profiles are efficient, all you need to do is carry enough fuel fraction to orbit. If you're current design isn't making it try to increase the ratio of fuel to payload. Keep the propulsion fraction the same unless the craft is overpowered. If this still doesn't work you might need higher ISP engines, but I think RAPIER to Minmus should be possible. I haven't tried flying a SSTO there, but my 150 ton empty SSTO has been to the Mun with an extra fuel tank in the cargo bay. Initially I had similar problems to yours in that I couldn't get the range I wanted. I needed to iterate until I got the right ratios of fuel to engine to payload.

[GoSlash27]

 *Starts, looks around groggily*

Actually, I think you've got it covered.

Best,
-Slashy

[GeneralVeers]

Starhawk, is that a nuclear engine at rear center on your space plane? Looks like it, but hard to tell.

[Xyphos]

11 hours ago, Nich said:

Are you getting 1500+ dv off air breathing?

are you using a shielded docking port or some other blunt heat protection?

are you using struts or fuel lines? These add drag and weight that kill you.

how close are you to 15 ton per rappier?

how close do you come to crashing into the ocean?

are you using a five degree aoi on your wings?

1550 dV @ 21000 kM

yes

yes, but the fuel lines and struts are clipped and hidden inside the fuel tanks, so they shouldn't pose a threat?

10 ton per rapier

I don't dive or crash, it's a 20* climb all the way up to the statosphere, level out to get 1550 dV; the vessel is so big, it'd break apart if I tried that.

no, the wings are clipped over & under the vessel

Edited March 22, 2016 by Xyphos

[Starhawk]

 

24 minutes ago, Xyphos said:

yes, but the fuel lines and struts are clipped and hidden inside the fuel tanks, so they shouldn't pose a threat?

Clipping doesn't matter to the drag algorithm.  They will still generate significant drag unless completely enclosed in a fairing or cargo bay.

4 hours ago, GeneralVeers said:

Starhawk, is that a nuclear engine at rear center on your space plane? Looks like it, but hard to tell.

Yup.  That's how it can go to Mun or Minmus.  It certainly couldn't do it using the Rapiers. 

The nuke is clipped way into the empty rear fuel tank.

Happy landings!

Edited March 22, 2016 by Starhawk

[Nich]

1550 should be good enough to get to Minmus just remember 1600 is possible

Based off your answers it sounds like your SSTO is a little over powered and over winged.  Add fuel to get up to at least 15t per rappier.  Most of mine come in closer to 20-25t per rappier.  This should also increase the loading on your wings.  When you do this you will need to change your assent path.  Hold 50m until you get 350 m/s  then pull up to 8-10 degrees.  Adding AOI to your wings will add 200-300 dv in orbit.  Anything from 2-7 degrees.  5 is the aerodynamic optimal for super sonic flight (shift and wasd key) but slightly more allows more lift with less wing giving more dv once your above 30km but you pay for it with more drag and lower airbreathing top speed.  Also @Starhawk is right on the fuel lines and struts

 

[Foxster]

I'm certainly far from an expert on SSTOs, I use brute force and ignorance to make them work. However, I did just slap together one with 4600dV left by orbit. This used mk3 body parts, 5 Rapiers, 3 nukes, 9810 LF and 2613 O. 

I flew it at about 15° until 26km then fired the nukes too and then when the rapiers were out of air switched them to closed cycle and finished getting to orbit with the last of the O. Running on nuke-only after that. 

4600 dV is more than enough for a trip to the Mun or to Minmus and back. 

[Wanderfound]

Why a Mk3? Are you planning on carrying something bulky?

And why not refuel it in orbit? Just send up an SSTO tanker in advance; the cost after recovery is trivial.

It can be done without refuelling if necessary, although it's easier in a Mk1 or 2 than a 3. Either way, you're going to want nukes.

 

For inspiration:

Long range Mk3

 

 

Long range Mk2

 

Edited March 22, 2016 by Wanderfound

[AeroGav]

The game's modelling of aerodynamics and jet engine performance is heavily simplified, save the most egregious errors, there is only so much performance to be had.   Chasing the perfect fuel mass ratio, engine combo, wing loading, and ascent profile has been an exercise in diminishing returns.

That said, it's not that difficult to fly single stage to Minmus with a science payload, as these, whilst sometimes bulky (science jr, survey scanner) tend to be light.       The problem comes with putting an IRSU refinery up there.  This equipment is much heavier and i don't think it's possible to do without a refuel in low kerbin orbit,  which is annoying because i hate flying rendez-vous and docking maneuvers.    It's ironic because SSTOs make great IRSU ships, all those empty tanks they don't discard on the way up are perfect for stockpiling fuel.          And a spaceplane with a scientific payload really comes into its own when you refuel it and fly on to Laythe, Duna etc.

In terms of Delta V,   spaceplanes are very efficient up to low orbit.  Beyond that however, they're carrying a lot of dead weight in terms of air breathing engines, wings, control surfaces, intakes and (mostly empty) fuel tanks.       If you don't plan to visit anything with an atmosphere, how about creating an upper stage for your SSTO?

 Have a nuke at the rear of the fuselage, with a liquid fuel tank, then crew cabin, then docking port in front of that.     Once you're in orbit,  transfer all your fuel to this tail section and detach it,  free to explore the kerbin system with the rest of the airplane safely parked in orbit.

In terms of a true deep space SSTO, this is my current best effort,  first flew on Monday..

https://onedrive.live.com/redir?resid=7003A8806D8A6B2C!713&authkey=!AKArjxC_7Z_h9kk&ithint=file%2ccraft 

 

 

Quick Overview

 

• Two long mark 2 cargo bays either side of CG to minimize loaded/unloaded CG shifting.
• Four pairs of Big S delta wings containing most of the fuel.   Fuel also stored around CG with minimal shifts from full to empty.

•  

  One Rapier engine.  Two pod mounted Panthers serving as boosters at lower altitude – when jettisoned, this uncovers the NERV engines mounted in front of them on the same pods.
• Three Vernier lifting engines in the belly to assist horizontal take-off and landing in low gravity environments.
• Liquid Fuel Capacity 3360 Units. Oxidizer Capacity 440 Units.
• Fuelled but with empty cargo bay it has a take-off weight of 38 tonnes.
• Has very small ailerons which cause sluggish roll response.   While this makes it easier to fly on keyboard, it does make lining up with runway at KSC tricky – recommended to land on the grass.
• Unusually, it’s a taildragger.   Couldn’t get the layout to work any other way, sorry.  Deal with it.  
• Designed to be flown with Kerbal Joint Reinforcement mod and TAC Fuel Balancer.

 

 

Staging Info

 

1. Starts both Panthers in Dry mode and the Rapier.
2. Jettisons the Panthers when they become useless.
3. Starts up the NERV engines.

 

 

Abort Button – Toggles Panther engines between Dry and Wet mode.

The above graph shows my speed/altitude profile and the different power sources in use.  Blue indicates where the Panthers were active in Dry mode.  Red indicates afterburner use.  The Rapier engine has the colour yellow on this chart,  the nukes a nice radioactive green.

What you'll notice - 

• Up to 10km I'm gaining height faster than speed, and not attempting to pass the sound barrier.
• At 10km we nose down and use afterburner to get through the high drag  transonic regime as fast as possible.
• Speed and altitude mainly increase together but comparatively small deviations from this trend, where speed increase is not matched by height gain, cause enormous increase in temperature.

Pictures from Flight & flying tips

Spoiler

 

I’m flying with keyboard controls only so am relying on pitch trim – if you are doing the same, recommend you pay close attention to this picture and set that exact amount of up trim before engine start.  Pitch trim is good because in the longer term , the aircraft will settle into the angle of attack you trimmed for, whereas SAS just holds a particular nose angle  regardless of aerodynamic circumstances.  There will be short term fluctuations however, particularly after a transition – and we’ve just transitioned from the runway to flying.    If the nose starts pitching up too much and speed begins to decay,  use SAS for a couple seconds to quell the deviation.

 

Mach 1.7 is the velocity at which Panthers produce peak power in dry mode , unfortunately they also loose thrust very rapidly with altitude in dry mode whatever the airspeed.

Mach 2.5 is the optimum speed for them in Afterburner.  In Afterburner the loss of thrust is not so rapid with increasing altitude, however thrust falls off a cliff above mach 2.5... by 2.9 it's probably best to jettison these engines, they're just dead weight.

Mach 3.7 is the optimum speed for Rapier engines and thrust tails off only slowly with increasing airspeed.  Above 23km however, thrust halves with each additional 1.5km in altitude.   On one engine we can get up to about 24.5km, two would increase that a little, but not worth the extra 2 tons dead weight that must be carried out the atmosphere.

Nukes join the party..

Mach 5 at 29km, things are getting hot .  Was very relieved when it started climbing again.

On our way up 

 

Final approach

Horizontal landing, using the Verniers to translate upwards as lifting engines.

 

Final result - parked on the greater flats with 900m/s remaining.   Kerbal engineer only takes account of the fuel in the pre-coolers in its delta v, doesn't allow for the fuel in the wings which the nuke engines wouldn't be able to access without TacFuelBalancer or transferring manually.

Design Rationale 

 

 

Spoiler

 

1.       This spaceplane was optimised for deep space missions with less dense scientific payloads.  Hauling full fuel tanks or IRSU equipment to low Kerbin orbit is probably better done by something with more thrust and lots of oxidizer. 

 

2.       For the interplanetary capability I had in mind, I wanted NERV engines and for as much tankage as possible to consist of LF rather than Oxidizer.   That way, Delta V potential when refuelled by an IRSU facility is maximised.

 

3.       In order to be refuelled, it should be capable of reaching a Minmus IRSU facility unaided, with a light load.

 

4.       In case you hadn’t guessed, efficiency trumped aesthetic considerations.

 

5.       To this end, the airplane was designed to accelerate to orbit from the air-breathing ceiling with the meagre thrust of NERV engines alone.  Two NERVs would be provided, which is the maximum a vehicle of this size can reasonably accommodate without the mass of the engines themselves impacting performance.  Nevertheless it is clear the design needs to have very low drag at high altitudes.  

 

6.       Assuming due diligence with regard to parasite drag, the main source of drag in very high altitude flight are the wings.  In turn, the amount of drag a wing makes in supporting a vehicle of a given weight, is dependent on its angle of attack.   Lowest drag, for a certain amount of lift, occurs at 2 degrees AoA at low speeds rising to an optimum of 5 degrees at very high speeds.

 

7.       However, maintaining the most efficient angles of attack is likely to result in the vehicle not gaining altitude quickly enough as speed rises and being destroyed by overheating, unless the aircraft has very large wing area.

 

8.       One obvious objection to adding extra wings is that they become dead weight that reduces delta V out of the atmosphere.  This was addressed by using the Big-S delta wing part exclusively, since it has the same mass: fuel volume ratio as other tank parts, yet also forms a lifting surface.  In fact it’s mass: lift rating ratio is as good as any other wing. 

 

9.       In this design therefore I strove to minimise the amount of fuselage tanks which add weight but no lift, in favour of as many wing tanks as possible, which at least provide lift to offset the mass of the fuel they carry.    

 

10.     The wing layout I came up with fits about as much area as you could plausibly fit on a fuselage of this size without resorting to clipping or biplane/triplane multiplane layouts not suited to high speed flight.

 

11.   In a similar vein, I’ve moved away from using clipping in the propulsion arrangement.  My recent designs all featured Rapiers attached to the rear node of NERVs and offset so that both can be used simultaneously, with tail cones attached and offset last inline to get the benefit of lower drag without actually blocking the nozzle.   Such impossible layouts minimise the frontal area of the craft but are gamey and also force a 1:1 ratio of engine types, which may not be desired.

 

12.   Furthermore, attaching all the engines to the rear of the fuselage like this makes the aircraft excessively tail heavy when empty.

 

13.   So, I decided to return to the trimotor layout as used on my very first spaceplanes, which avoids such CG problems and enjoys perfect thrust symmetry. 

 

14.   However, where these early designs used 2 RAPIER and 1 NERV, I now use 2 Nuke engines and only a single Rapier.  Thrust falls off so rapidly with increasing altitude that adding a second Rapier only increases the air breathing ceiling from 24.5 to 26.5 km for the cost of 2 tonnes engine mass that becomes dead weight once orbit is achieved.  OTOH, dropping down to a single NERV means carrying large quantities of oxidizer, since this engine alone will not have enough power to boost the vehicle to orbit.

 

15.   Structurally, it would have made more sense to mount the engine pods beneath the wings, but that results in a pitch moment that can overwhelm the reaction wheels in the upper atmosphere.  Handling considerations took precedence and the pods were attached to the sides of the fuselage at the exact midpoint of the fuselage height, and the wings mounted to the pods in turn.   Keeps the thrust axis right on the centre of gravity but necessitates the use of Kerbal joint reinforcement.

 

16.   Flight testing revealed the need for the two Panthers.   Due to its large wing area this design generates a lot of drag when trying to fly fast at low altitudes, which meant it was unable to penetrate the sound barrier below 10km.   Unfortunately the Rapier doesn’t produce much power below mach 1, and with only a single engine the climb to 10km could take almost 20 minutes.   Because the available thrust was only just greater than that needed to merely maintain level flight, the design suffered excess “gravity losses” and would use half its fuel in the subsonic phase of flight.

 

17.   In dry mode, the Panthers double the available power and triple the surplus power available for climb.  Furthermore, they have an ISP of 9000 compared to 3800 for the Rapier.   Combined with the reduced “gravity losses”, fuel usage up to 10km is cut by an order of magnitude.

 

 

 

Edited March 23, 2016 by AeroGav

[Vegetal]

There was a challenge a while ago, in 1.0.4, in which I built an SSTO capable of doing the round trip to minmus, and *probably* could make it indefinitely without ever refueling at Kerbin, just on Minmus using an ISRU. The big one mind you, the lighter one only came to be on 1.0.5.

It used 4 rapiers and 2 nukes, Mk3 sized. The real trick is to have JUST enough oxidizer to get close to orbital speed on the rapiers, when you go empty you have to be able to continue just on the nukes. That's the first and most important idea you got to have in mind, the rest is ascent optimization. As Nich said, enough incidence on your wings so you keep your plane mostly aligned with prograde, and the least drag as possible.

Found the album!

Obs: I just noticed I made the plane change after launch. Ha! What a noob!

 

Edited March 23, 2016 by Vegetal

[AeroGav]

8 hours ago, Vegetal said:

 

Obs: I just noticed I made the plane change after launch. Ha! What a noob!

 

I never figured out how to make a plane change before getting to orbit and i know it costs me a fair bit of delta V.   How do you do this?  Do you have to time your launch for the planets being lined up or can you just take off, set target, and fly north/south within the atmosphere in the same way you'd burn normal/anti normal when in orbit?   The problem is , patched conics won't give you an AN or DN because you don't currently have an orbit to compare with the target?

 

[Wanderfound]

2 hours ago, AeroGav said:

I never figured out how to make a plane change before getting to orbit and i know it costs me a fair bit of delta V.   How do you do this?  Do you have to time your launch for the planets being lined up or can you just take off, set target, and fly north/south within the atmosphere in the same way you'd burn normal/anti normal when in orbit?   The problem is , patched conics won't give you an AN or DN because you don't currently have an orbit to compare with the target?

Yes, just fly a bit north/south of east during the initial atmospheric takeoff, and keep that orientation as you burn into space.

There's probably some clever mathematical way to figure out exactly what time to launch to put your initial orbit trajectory in line with your target planet, but personally I just eyeball it. Minor corrections are cheap if you've got nukes.

 

23 hours ago, AeroGav said:

That said, it's not that difficult to fly single stage to Minmus with a science payload, as these, whilst sometimes bulky (science jr, survey scanner) tend to be light.       The problem comes with putting an IRSU refinery up there.  This equipment is much heavier and i don't think it's possible to do without a refuel in low kerbin orbit,

Oh, yes it is.

As well as all the usual tricks, there's one important thing to remember: if you're in LKO with an ISRU rig, you probably also have ore tanks. Pause after the initial boost to LKO, switch on the ISRU and turn the contents of your ore tanks into pure LF to power your nukes. That should give you enough for the boost to Minmus.

In some ways, ore is better than normal fuel, as you can convert it to LF and O in whichever proportions you please. Only lift off with empty ore tanks if you have to.

[Dr. Jet]

https://kerbalx.com/Dr_Jet/Duran-nuclear-SSTO - this one is capable. 20 kerbals, no part clipping, over 2k dV at LKO.

A bit tricky to actually LAND on atmoshpereless planet, since it's not a VTOL, but at Minmus with it's low gravity it's quite possible.

[Vegetal]

@Aerogav: You have to time your launch to be exactly at the AN or DN of the body's plane. KER helps in this regard, as you can keep an eye at the relative inclination as you make your orbit insertion, and so adjust your heading accordingly.

I think the game's map is one of the most lacking parts of UI, it gives you no useful information. So yeah, you have to eyeball it.