Rescaled Kerbin and efficient SSTOs

[Chris Bolland]

Hey there all!

I've recently been experimenting in my 7.6x rescale play-through trying to build an orbit capable SSTO with some payload mass (like 1 ton is fine, just tiny satellites). Over the course of a few weeks I've been constantly failing to even break the 120km mark, (where space begins with my rescale settings) and have been getting a bit discouraged. I just can't seem to get going fast enough on panther engines to make it up to the point where rockets can take over. I need 9000m/s of deltav for orbit, and carrying enough fuel for the rockets is taking up a lot of mass, which limits my speed.

I've made a few SSTOs  in stock and have watched every video on them religiously, but often people use measuring sticks and rules of thumb that work in stock, but not in rescale. How should I proceed?

If you have any tips or guidance feel free to post below! Here's my rescale settings for anyone interested.

From the 7,6xScale file:

// Base Settings
    
    @Resize = 7.6 - everything is 7.6x bigger than stock
    @Rescale = 7.6
    @Atmosphere = 1 - atmosphere height is 120km changed below.
    @dayLengthMultiplier = 1
    
 
    // Advanced Settings
    
    @landscape = 0.169 - This number I got through trial and error, trying to make MT K2 (Kerbin's highest peak) the same height as Everest roughly. Otherwise the mountains next to KSC are too tall to fly over.

 

From the PlanetSpecificConfigs file:

 @Kopernicus:BEFORE[SigDim2]:NEEDS[SigDim]
    {
        @Body:HAS[#name[Kerbin]]
        {
            @SigmaDimensions
            {
                @Atmosphere = 1.714285714285714 - I've found this value offers the most realistic atmosphere height and falloff.
                    @dayLengthMultiplier = 3.333333333333333
            }
        }
    }

[Vanamonde]

I have no tips seeing as I suck at spaceplanes and only ever made three that could reach orbit in the un-modded game. Good luck.

[miklkit]

I only play the stock system and the Panther engine struggles to reach orbit there too.  It is just too weak and dies off at too low of an altitude.  The Whiplash engines are much better. 

[Leganeski]

Reaching Kerbin orbit at 7.6x scale requires achieving a surface velocity of about 6200 m/s. No matter how much you optimize the ascent profile, there's no way to get around the fact that orbital velocity is fast.

Jet engines are efficient, but their downside (apart from requiring oxygen) is that their thrust starts decreasing beyond a certain speed dependent on the exact engine, and eventually drops to nothing. The Panther engine, for instance, completely loses thrust at Mach 3.5 (≈1100-1200 m/s). Other less efficient jet engines such as the Whiplash or RAPIER are better, continuing to function until Mach 5.5-6 (1800-2000 m/s), but that's it.

At stock scale, where you only need to achieve a surface velocity of 2100 m/s, jet engines can take you almost all the way there, and you can get away with having only a little bit of rocket power that pushes you the rest of the way into orbit. Since you're so close to orbit, centrifugal force reduces the TWR requirements considerably, so there are a lot of options: nuclear engines to avoid the need for oxidizer, small LFO engines for lower dry mass, or really anything with enough Δv.

At 7.6x scale, the situation is completely different: even 2000 m/s is nowhere close to orbit, and you need to accelerate the remaining 4200m/s (or probably more) with rocket power alone, while fighting drag and gravity without much help from the curvature of the larger planet. The Δv requirements for the one stage practically necessitate nuclear engines, but the TWR requirements (something like 1.0-1.5 g) are difficult to achieve with the stock NERV, especially if you're also using it to carry stuff like wings and payload.

There's a reason why SSTOs have never been built in the real world, and the comparatively limited selection of parts in KSP makes designing one even harder. Your 4560-km-radius Kerbin is not quite as difficult to orbit as the 6378-km-radius Earth, but I don't think that this minor decrease in orbital velocity is enough to make the challenge doable without a whole lot of trial and error, and a very small payload fraction.

 

Edit: As @Lt_Duckweed points out, I was neglecting the possibility of continuing to remain at low altitude during most of the ascent, which would allow the plane to use lift to lower the TWR requirement to manageable levels. This requires very precise control over both altitude and angle of attack to maintain just the right amount of lift, and generates a ton of heat at high velocities. However, the somewhat unrealistic wing parts in KSP take zero damage from any temperature below 2400 K, so they might just be able to survive. 

I have never been successful flying an SSTO like this: either I encounter too much drag and quickly run out of fuel, lose control of pitch and spiral into a crash, or make too sharp of a turn and break my wings off. It's definitely possible, though, and I imagine that an autopilot mod would help reduce the risk of all of these problems.

Edited January 21 by Leganeski

[Lt_Duckweed]

On 1/19/2024 at 3:37 PM, Leganeski said:

The Δv requirements for the one stage practically necessitate nuclear engines, but the TWR requirements (something like 1.0-1.5 g) are extraordinarily difficult to achieve with the stock NERV, especially if you're also using it to carry stuff like wings and payload.

You don't need anywhere close to this much twr on your nervs.   When your rapiers stop being able to accelerate you somewhere a bit past Mach 5, you should still have a lift to drag ratio close to or exceeding 4 if you have paid attention to some basic drag optimizations.  You are also high enough that the nervs are at essentially 100% thrust and ISP.  At this point a initial twr of 0.3-0.4 will cut it, and as you burn off fuel and also get closer to orbital speed, the needed twr to overcome drag will drop even as your actual twr will climb.

It's harder than stock scale sstos for sure, but it's a lot easier than stock part RSS sstos, which have been done before by a number of people.

[DeadJohn]

On 1/19/2024 at 11:21 AM, Chris Bolland said:

I need 9000m/s of deltav for orbit, and carrying enough fuel for the rockets is taking up a lot of mass, which limits my speed.

At that scale stockalike jet engines will only get you to perhaps 2000 m/s. You'll still need an extra 7000 m/s of DV from rocket engines. which is very difficult to do in a single stage. Some things you might try:

Hypersonic or scramjet  engines from mods: Some mods have super airbreathing engines that can go much faster than 2000 m/s, and they flame out at higher altitudes than the whiplash or rapier. Try Mk2 Stockalike Expansion, Mk3 Stockalike Expansion, or OPT.

Build the airbreathing engines plus their intakes and fuel tanks as droppable modules. Stage them away when they stop producing thrust. Yes, this violates the "SSTO" definition, but is more realistic in terms of what might be done with current technology. You can put parachutes or probe cores on the drop pods to recover them with the Stage Recovery or FMRS mods.

[Chris Bolland]

Thanks to all who have given their input so far including Leganeski, Lt_Duckweed, and DeadJohn. You guys are masters and very helpful.

2 hours ago, DeadJohn said:

Try Mk2 Stockalike Expansion, Mk3 Stockalike Expansion, or OPT.

Build the airbreathing engines plus their intakes and fuel tanks as droppable modules. Stage them away when they stop producing thrust. Yes, this violates the "SSTO" definition, but is more realistic in terms of what might be done with current technology.

This is actually a great thing to hear, as I thought I was just doing it wrong, but judging by real life, it simply might not be feasible to make an SSTO at 7.6x scale with the ability to bring payloads into orbit. It might be just cheaper and easier to do it with disposable rockets instead. I mean NASA's space shuttle ended up being overly complex and in the end kind of useless, yet it was still able to put massive payloads into orbit. I might try doing something more along those lines, rather than a pure SSTO.

I'm still open to more comments so keep em coming if anyone has anything to add, and I'll be tinkering away as time goes on. Thanks all!