Help me improve my SSTO

[McRoll]

Hello, finally decided to register and ask for help because I'm not getting any progress on my design anymore and am in need for some advanced tipps.

This is the object:

http://imgur.com/a/kYvri

the goal is to create an efficient SSTO with following requirements:

- 2 crew members

- full science equipment storage (except science lab)

- Have some basic utility, like a bit of monopropellant to help with landing on moons and maybe docking ( dont know where to put docking port yet). Would be excellent if it could take some payload like a small probe, though probably not possible.

- only very minimal part clipping or other cheaty stuff (want it to look good and like it could actually fly)

Craft file:https://dropfile.to/93YGtGE

Stats:

8 Rapiers and 2 Nervs | 24K DeltaV on runway | just under 170 tons

Why does it look that way?

I started building around a large liquid fuel tank since I wanted to have a 2 man crew and a bit of cargo load, so going the minimal approach was probably not the best way (?). The first obvious steps were to use as few parts as possible to reduce weight and probably drag, then adding just enough engines to propel it outta the atmosphere. I had to experiment around a lot with various wing forms so it would be able to recover from a stall and reduce some oscillation/wobbling during flight (large delta wings as canards in front didnt work out very well). Had to use struts on the main wings because they would break under load otherwise. Moving wings around results in instability once the fuel is depleted, I built this one with a balanced out center of mass on zero fuel. It now flies kinda well for me, though there is surely room  for improvement.

Right now it has around 5,1k dv after orbit circularization, with around 1k oxidizer left for help with starts and possible vernor engines. I'd like to impove that maybe to 6K or even more if possible, so I can comfortably get a Duna landing and make it back and stuff like that. What would be the optimal ascend profile? Right now I'm accelerating in a very slight angle to about 10 km height, get some speed there and start pulling up in a 20° angle once im at around 1200m/s speed and around 18 km height. Still havent figured out the most optimal way.

Things I tried:

-More fuel adds more mass and the TWR is not enough to accelerate quickly enough from the runway

-More engines mean more mass to drag around in space and delta v goes down. I think I found a pretty good spot between engine count, TWR and mass.

Anything I should know for maximizing efficiency bar the obvious stuff? Like some hidden values, musthave parts? Should I go big or rather as small as possible? Can this design be improved further or should I try a different approach? Also, a rough estimate what I could actually do with 5k dv? I guess its not enough to make it to Duna/Ike and back, right?

Edited August 11, 2016 by McRoll

[Snark]

Moving to Gameplay Questions.

[Wraith977]

If you can give a craft file it would make it a lot easier to try possible modifications on it and report back anything useful. One thing that immediately comes to mind is increasing the wing surface area to give more lift, should get you to orbit with a bit more fuel but of course the downside is that they just add more mass in space, try to find a balance between the two.

[McRoll]

Edited OP with craft file.

[Aegolius13]

A few random thoughts:

Have you experimented with swapping out a couple of the Rapiers with Whiplashes, or even Panthers?  The latter provide a little better thrust right off the bat, and can help get your craft off the runway in a timely fashion, as well as to cross the transonic barrier more quickly.  I generally do 2 Whiplashes on my LKO cargo planes, which are similar to your ship mass-wise.  

You might staying more level after hitting 18k feet, to eke out as much speed as possible from your jets before they give out.  I might be wrong, but I generally prioritize speed over altitude at this stage, then angle up more when I switch mode on the Rapiers.  (As long as you don't overheat and blow up, of course).

When do you start firing your nukes?  You could try at around 20km altitude, or even slightly lower.  They're way more efficient than closed-cycle Rapiers even at this altitude, so every bit of thrust you can get from the nukes will help keep your fuel (and especially oxidizer) needs down.   Though with only two and a plane this big, they may not make a big difference.

It depends on mission, but you might not need much (or any) oxidizer left in the tank once you reach LKO.  For example, Laythe can often be done with just jets and nukes.  

To add a docking port, you could switch your cockpit to an inline, and put a Mk2-Mk1 adapter and a shielded docking port in front.  That might also give you a bit better heat resistance, if that's a problem.  You could also put an inline Clamp-O-Tron behind the cockpit, but it will be a pain to dock if your plane is the moving one of the pair.  

Finally, if you're willing to refuel in LKO, you can get by with a MUCH smaller craft to get similar delta-v, which will be a lot easier to fly at your ultimate destination.  But for some people that compromises the SSTO concept.

[mk1980]

the mk2 cockpit is holding you back. you want to reach at least ~1600 m/s in airbreathing mode before you switch over to closed cycle or fire up the nukes. that's not going to happen with an mk2 cockpit (explodes before you reach that speed).

try using the inline cockpit instead and add a (short) mk2->1.25m adapter tank with a shielded docking port on the nose. alternatively, you can also use the mk2 passenger cabin instead of a cockpit. you'll need a probecore for control, but you get 2 additional seats that way. the mass is about the same (both mk2 cockpits and the mk2 cabin are about 2 tons)

that fixes both the heat problem and the docking port problem. and by going faster in airbreathing mode, you also save quite a bit of deltaV.

 

regarding your mission to duna: 5k deltaV is overkill. from experience, you need about 3.5 km/s for that roundtrip (assuming almost perfect transfers and aerocapture in both systems) some more won't hurt. more than ~4k would be overkill, though.

 

i don't think the wings on that plane are anywhere near enough to land on duna. though. you'll just crash into the ground at several hundred m/s. you'd have to add LOTS of parachutes for the landing. those wings are probably just enough to slow your plane down to safe parachute deployment speed.

 

also, ascent back to duna orbit might be difficult. not sure if 1k oxidiser will give you enough of a push to make it back. the 2 nukes alone definitely won't. i've done a "single stage to duna" a while back. that plane also had 2 nukes, but it was less then half the weight of your design and after landing on duna it was down to ~35 tons or something. the 2 nukes were barely enough to get that thing back to orbit. took me several reloads.

 

is there a specific reason why you have to bring all the fuel all the way? you could cut down the fuel mass dramatically if you add some drills and an ISRU converter and refuel the vessel in the target system.

[Foxster]

Some good stuff above. You can reduce drag a lot and that will leave more dV on orbit... 

You have 5 body stacks and reducing them to one (or three if you just have to) will help with the drag. 

Compound wings tend to be more draggy than a single stock wing but YMMV. Yours look cute and that's fine if that is your priority but not if you want to improve the efficiency. 

Do you really need those big compound canards at the front? Similarly, do you need two big compound tail fins?

That quad adapter at the back will be very draggy. Consider radially mounting a bunch of "Advanced Nose Cone - Type A". Get them aligned forward and then attach your engines to them.  They do transfer fuel OK like this. 

Taper the tail end too. It's as important as the nose in terms of drag. 

Switch to a mk1-sized cockpit.

Put an extended Communotron 16 antenna at the very tip of whatever nosecone you go with. 

Of course, you could go the other way and make your craft smaller. This has pretty much the same stats as yours (5.5kdV on orbit, 2 Kerbals, full science, docking port)...

 

Edited August 12, 2016 by Foxster

[McRoll]

Wow, didnt know some of the stuff, thanks so far. I guess I will try to improve the existing design, particularly with the inline cockpit idea and more lift area and will try another variant where I will try to make it as small as possible.

Basically I want to have a nice allround SSTO which is able to reach and land on most bodies by itself and come back - will probably try a version with and without ISRU.

Will post updates if I reach a significant improvement, of course more ideas are always welcome. Just dont want to copy existing designs, for me its important that I come up with something myself but nothing wrong with getting an idea here and there

[dire]

Good luck! One thing I don't think was mentioned is that the Mk2 bodies are also wings. If you turn your Mk2 stack 90 degrees so it is sticking straight out from the Mk3 stuff, you'll probably be able to eliminate the outer row of your wings -- this'll reduce your drag quite a lot. If you switch some of the inner row of your wings for a Big-S delta-V you'll reduce your drag even more.

If it was me I'd try to swap out the big Mk 3 long fuel unit for a half-size Mk 3 medium fuel unit and maybe switch your outer Mk 1 layer for a Mk 2 liquid fuel layer like I mentioned earlier -- if you have to have the drag anyways it might as well be working for you, and that would let you stick another pair of nukes on there.

Edited August 12, 2016 by dire

[GoSlash27]

 ^ While Mk2 bodies do produce lift, I don't recommend taking advantage of it. They are very inefficient and create a lot more drag than the equivalent wing surface would.

SSTO space planes are all about controlling drag.

The most important piece of wisdom I can impart is this:

 Rotate your wings a few degrees so that they produce enough lift to keep your prograde vector aligned with your nose at Mach 1. Your fuselage is very draggy when it's not perfectly aligned with the airflow.
 This allows you to use less engine mass to get out of the atmosphere, which means more payload in orbit and less cross section drag.

 Also, I recommend the Big S delta wing for space planes. They're basically "free" fuel tankage from a drag perspective. Also more rigid than cobbling together panels, which will allow you to get rid of those draggy struts (struts are the devil).

Best,
-Slashy

 

Here's an example of a very simple Mk. 3 space plane to illustrate:

This one weighs 126 tonnes at launch, yet requires only 4 RAPIERs to achieve orbit with 31t of payload.

You can see where I've rotated the wings for static incidence and paid careful attention to form drag.

Best,
-Slashy

[foamyesque]

I think you're carrying too much tail on that, @GoSlash27; you could probably get away with using standard canards for your control surfaces. Less drag and better responsiveness.

@McRoll: You say you balanced it, but it looks to me like there's going to be significant CoM shifting as you burn fuel; all your engines are at the very back. Unless I've got something small enough for the cockpit to balance out the engine I generally try to put engines as close as I can to the midbody, and balance out the fuel fore-and-aft so that the CoM moves as little as possible. That way you get consistent aerodynamic performance regardless of how much fuel you've burned at any point. Look at the Skylon for an example.

@Foxster: What's this about the quad adapter being draggy?

Edited August 13, 2016 by foamyesque

[GoSlash27]

Another extreme example:

268 tonnes of spaceplane can still get hypersonic, even with just 8 RAPIERs.

Best,
-Slashy

13 minutes ago, foamyesque said:

I think you're carrying too much tail on that, @GoSlash27; you could probably get away with using standard canards for your control surfaces. Less drag and better responsiveness.

Actually, the responsiveness is just about right for a spaceplane. I don't recommend a lot of agility for spaceplanes. They should sled their way out of the atmosphere with minimal deviation to minimize drag.
 I knocked this one together as a quick example, so it's not optimized.
 

Best,
-Slashy

[foamyesque]

On ascent, sure. I tend to lock in at 15 degrees (with Rapiers-- Panthers require more fiddling :v) and just go. Agility is more useful on the entry and landing; particularly, all-moving tails allow for stronger changes to the drag profile when you're in the stall area, and strong responsiveness allows for better flares and approaches.

EDIT:

One problem with the sample designs you're giving him, though, @GoSlash27, is that he wants to land it on Duna. That's going to require a pile of wing area-- he'll be hugely over-winged leaving Kerbin.

Edited August 13, 2016 by foamyesque

[GoSlash27]

1 hour ago, foamyesque said:

 

One problem with the sample designs you're giving him, though, @GoSlash27, is that he wants to land it on Duna. That's going to require a pile of wing area-- he'll be hugely over-winged leaving Kerbin.

True, but these examples aren't intended for that. They just show what can be accomplished by minimizing drag instead of adding engines.

Spaceplanes are very tolerant of extra wing area. Just reduce the incidence and the only penalty is weight.

Best,

-Slashy

[McRoll]

An update from me, I played around with various designs and decided to try a new approach and to implement some of the stuff you guys mentioned, I came up with this one:http://imgur.com/a/03O

This is an improvement over the old one in the way that it features an ISRU converter with a drill and full science suite in the lower cargo bay and carries a small ore survey probe in the frontal bay which can decouple, make a survey and be stored again. It also can dock with space stations now via frontal docking port - and can get faster in the atmosphere without burning up. All this while still maintaining over 4k dv after orbit circularization. It is also lighter, which makes it easier to land. As far as I can say drag is also not much of a problem since it has only a 0.68 TWR but accelerates quite decently.

However, I still have problems with wings, a delta wing brakes up if I pitch too hard at high speeds - these wings are enough for Kerbin but I wouldnt be able to land on Duna with these, so there is still some work to be done. Those vertical wing parts are needed to make it somewhat stable but it still spins out if you dont fly perfectly. Has a little bit drag to the side, but I think it comes from the ore probe in the bay not being perfectly symmetrical or something like that. Also, I'd like to get still more dv out of it - I tried a mission to Gilly and it nearly worked out, however I missed like 500 dv in the end to land on Gilly. My flightpath was awful however, bad launch window and less than optimal maneuvering. I think it can be done with this design. But I'd like to have some safety room if i screw up a maneuver.

Overall I'm quite pleased with this one, just needs some tweaks. Once it reaches 5k dv and some more stability then I will have what I wanted. I can already do that with detachable fuel tanks - but for aesthetical reasons I wanna stay 100% reusable.

Edited August 19, 2016 by McRoll

[luizopiloto]

I have made this one recently...
https://kerbalx.com/luizopiloto/X-98L-BlueJay

[Aegolius13]

Could you use another Big S wing on the front set, instead of the compound wing?  They tend to be more break resistant as well as adding fuel capacity.

[luizopiloto]

 

23 minutes ago, Aegolius13 said:

Could you use another Big S wing on the front set, instead of the compound wing?  They tend to be more break resistant as well as adding fuel capacity.

My designs are partially built for the looks... the Big-S wing strakes wasn't a good fit on that one...

[dire]

Are you flying with fine motor control engaged? Hit the capslock key and your control widgets will turn blue. You should not be needing more vertical wing surface than you have horizontal surface (those mk2's are still wings remember) unless you are flying sideways.

If you are still having control issues maybe switch out all those vertical triangles for tail fins. If it was me I'd angle them 45 degrees so you get an X shape, that gives you control along all three axes (yaw, roll, pitch). If you wind up with eight tailfins and you still are not stable in flight, then something else is wrong.

[mk1980]

i don't think it's worthwhile to add more wings just to be able to land on duna. in my duna SSTO, i had some 1.25m service bays added somewhere inside with the tank stacks and had radial parachutes inside the bays for the actual landing. 

that way, you don't need that much wing mass. my plane had enough wings for kerbin and that wing size was also enough for a direct aerocapture at duna that slowed the plane down just enough to deploy the chutes at "safe" speeds. felt like the easiest workaround. 

you'll have to bring an engineer to repack the chutes if you want to do multiple landings, but you'll want a high level engineer for the ISRU/dill anyway and you have 4 seats, so that shouldn't be a problem.