Looking for advice on my Eve spaceplane before I send it off

[InfernoSD]

Eve has been a tremendous challenge for me. I've built at least ten unique Eve ascenders but only succeeded in reaching orbit with one of them. That was a 400 ton monster that relied on ISRU, so I wasn't very proud of it. All of those attempts were rockets, and heavy ones, so I thought it would be worth trying to use propellers for the first stage of my next attempt. Well, after a week of trying and countless revisions, I've finally made it to orbit once in testing.

Video synopsis:
First couple minutes of the video are just to show how it flies.
Then I skip to Eve atmosphere and climb slowly to 18.1 km.
From there I drop the propellers and activate six Dart engines, reaching an altitude of 47 km.
Final stage with two remaining Dart engines continues from there.
Ends at orbit with 48 m/s left in the tank. That's less than 1% of starting dV.

So I'm happy that I've made it this far, but since planes are so complicated, I thought I would ask for advice before I call it final and take the next step. I'm sure there's room for improvement. Anything obvious stand out?

I find the drag system entirely confusing and frustrating. I checked some numbers in-flight and was surprised at how much drag I'm getting on parts along the front of the plane, even the nosecones. The MK2 service bay is also showing enormous drag. Can any of these be replaced with something better?

I was having constant problems with stabilization. It was either flipping over, going into a flat spin, or targeting retrograde. Now I finally have it flying pretty smoothly in all three stages. Frankly I'm not sure how I did that, but it must have to do with the exact angle of the wings. Hopefully it sticks.

[Vanamonde]

This is an ambitious project. Good luck. 

[Guest]

Wow this is ambitious. Eve is hard.

Since you asked for suggestions, here's one: minimise your dry mass. That Mk2 cockpit you're using is pretty heavy, and you're also hauling up a lot of wing. The Eve ascender I'm happiest with is a two-stage fully recoverable affair where the lifter stage glides back for recovery, but the orbiter stage's payload is basically a command seat in a service bay, so as light as it can be, effectively.  I have tooled around with attempts at prop-driven SSTOs but man they are difficult, I haven't yet had the patience to finish one, or even get close enough to finishing one that I'd even test-fly it on Eve.

[InfernoSD]

I spent some time looking at Eve SSTO threads and I learned some things.

57 minutes ago, Brikoleur said:

Since you asked for suggestions, here's one: minimise your dry mass. That Mk2 cockpit you're using is pretty heavy, and you're also hauling up a lot of wing.

Thanks, you're not wrong. Dry mass is 20.07t after staging off the propellers. Wet is 62.07t. Unfortunately I'm not great at designing planes in general, so I'm not sure I can do much better with the wings. I experimented in the past with detachable wings, but I lost control instantly without them; the whole plane flips over backward. I think I could perhaps detach the forward wing section alone. I might also be able to remove or simplify the yaw stabilizers and the x-shaped wing strakes.

I also have a big task ahead of me to figure out atmospheric entry. I've never created a single vessel of any shape or size that could enter Eve's atmosphere from orbital velocity without catastrophic damage. I chose the MK2 parts initially thinking their heat tolerance would make reentry possible. Now I'm not sure that's going to happen.

Additional questions:

Should I ascend at a shallower angle? I've never known what the right way to figure out a gravity turn is. It seems like successful Eve SSTOs burn at a shallow angle of between 2 and 20 degrees. I've been starting my burn at 45 degrees to try to get away from the drag sooner.

Is there a good place I can stash fuel cell arrays? I need two for this design, and I'd like to put them somewhere they won't cause drag. Even better would be if I could stage them with the propellers since they're dead weight after that point. Maybe I can put them inside a fairing with the propellers during reentry.

I'm also considering replacing the Darts with a smaller number of more powerful engines. I think having good in-atmosphere isp is less important at the altitudes I'm using them.

If I do get around to making an SSTO some day, I will have to use two different engine types so I can make the most of my fuel weight. And I will have to try a helicopter design so I can skip wings completely.

[Guest]

1. Atmospheric entries for spaceplanes aren't that hard as long as you enter empty. Bigger ones are much more fragile of course, something as big as yours could be a problem even empty. Otherwise, heat shields work. I usually set the Pe somewhere around 60k and hope. The issue isn't really the heat tolerance of individual parts, but the "aerodynamic density" of the entire craft -- if you can present a lot of surface area for relatively little mass, you'll survive. Key is to make the plane as aerodynamically neutral as humanly possible so it can maintain a radial out attitude all the way until it hits the lower atmosphere.
2. If you're making a multi-stage affair where the initial ascent is as a plane, think of your second stage as a rocket: ditch the entire plane, wings and all, and fly to orbit with that. Main difficulty with that design is that you'll have to find a way to point yourself up just when the plane is running out of air, otherwise you'll waste too much fuel turning.
3. The point of using a prop plane like you are is to get out of the thick air with it. It should be able to loft you high enough that you can then burn at a relatively shallow angle.
4. Fairings are your friend, yes.
5. Darts do work on Eve but for something your size, just call it a day and go with Vectors. I've almost always ended up with them whatever I start with.

In my aborted experiments making a prop SSTO for Eve, the main difficulty has been sheer mass. The plane part is a lot of dry mass, and even if it gets you to 20 km or wherever, you still have a long way up, and it takes a lot of rocket to haul all that up. And a very big, heavy plane isn't easy to operate anywhere, and definitely not on Eve.

Another suggestion -- and please feel free to disregard it because everybody plays differently: if you're new to Eve, you might want to start with a bit less ambitious designs, and work your way up from there. The simplest Eve ascent vehicle is an asparagus-staged rocket powered with Vectors, with a Dart-powered orbiter module at the nose: crew capsule, fuel, Dart, decoupler, more fuel, Vector; around that, 4 or 6 fuel tanks with sufficient Vectors (or Darts) on them to get you into the air, with flow priority set so that you can discard them one pair at a time. Build that right and it's not even terrifyingly big. Once you've got that down, go with more complex/ambitious designs.

Eve is a harsh mistress and she will slap you down hard if you disrespect her.

[Cavscout74]

I haven't played much in a while, so I'm not sure if this has changed, but the Mk2 fuselage was very draggy and you basically have 3 complete Mk2 fuselages.  They (used to?) hold the same fuel as the 1.25m fuel tanks, with higher dry mass & much more drag

[InfernoSD]

1 hour ago, Cavscout74 said:

I haven't played much in a while, so I'm not sure if this has changed, but the Mk2 fuselage was very draggy and you basically have 3 complete Mk2 fuselages.  They (used to?) hold the same fuel as the 1.25m fuel tanks, with higher dry mass & much more drag

Hmm, that's good info. I suspect the MK2 is giving me a decent chunk of lift, which shouldn't be overlooked. But I will have to give this a try.

2 hours ago, Brikoleur said:

If you're making a multi-stage affair where the initial ascent is as a plane, think of your second stage as a rocket: ditch the entire plane, wings and all, and fly to orbit with that. Main difficulty with that design is that you'll have to find a way to point yourself up just when the plane is running out of air, otherwise you'll waste too much fuel turning.

It is multistage. I like this idea in theory, but the rocket really likes to flip backward as soon as the wings detach. I'm thinking maybe it's a combination of the tail rudders messing up CoM and the MK2 cockpit which adds significant lift at the nose. Currently my third stage is pretty close to being a regular rocket already. I'll try to shave off as much of the second stage as I can, or maybe I can drop some parts inbetween stages.

2 hours ago, Brikoleur said:

Atmospheric entries for spaceplanes aren't that hard as long as you enter empty. Bigger ones are much more fragile of course, something as big as yours could be a problem even empty. Otherwise, heat shields work. I usually set the Pe somewhere around 60k and hope. The issue isn't really the heat tolerance of individual parts, but the "aerodynamic density" of the entire craft -- if you can present a lot of surface area for relatively little mass, you'll survive. Key is to make the plane as aerodynamically neutral as humanly possible so it can maintain a radial out attitude all the way until it hits the lower atmosphere.

Atmospheric entry on Eve has been the most baffling thing for me in this game. Everyone on this forum makes Eve reentry sound so casual compared to my experiences. Observe: here I am entering with the fuel tanks empty. The props begin blowing up at 89km, so they would have to be kept in a protective shell.

By 85km I have to pitch down hard to keep my wings cool. I can hold 5 to 15 degrees from prograde without overheating them.

Somewhere around 67km, drag finally starts to overtake gravity. Velocity has reached its maximum of 3247 m/s and is now beginning to drop.

Once all the wings are gone the plane presents its belly and falls below 3000 m/s. Also missing are various rotors, adapters, decouplers, and one of the landing gear. But the MK2 parts survive.

I'm inclined to believe the console version uses different physics for atmospheric drag and heat than the PC version, but I've been too lazy to prove it. Surely the 2400K wings are meant to be able to survive entry, right? In any case, I'm sure I will be fine if I add a stage to slow down before entering. The inflatable heat shield blew up last time I tried it.

2 hours ago, Brikoleur said:

Another suggestion -- and please feel free to disregard it because everybody plays differently: if you're new to Eve, you might want to start with a bit less ambitious designs, and work your way up from there. The simplest Eve ascent vehicle is an asparagus-staged rocket powered with Vectors, with a Dart-powered orbiter module at the nose: crew capsule, fuel, Dart, decoupler, more fuel, Vector; around that, 4 or 6 fuel tanks with sufficient Vectors (or Darts) on them to get you into the air, with flow priority set so that you can discard them one pair at a time. Build that right and it's not even terrifyingly big. Once you've got that down, go with more complex/ambitious designs.

Nah, like I said at the top, I've tried lots of different designs. The one successful trip I made to Eve actually was an asparagus design, but it was also set inside a fairing on top of three Mammoth boosters that got it off the ground. This plane was my attempt to make something lightweight, if you can believe it.

Ok, thanks for your responses. Back to the SPH to test things out.

[InfernoSD]

Well, my finalized plane barely resembles what I started with. After an almost complete rebuild, I got as high as 19.9km with the propellers which is probably the best I can hope for with this much mass. The new wings are quite large and practically mirrored on the vertical axis. My final run made orbit with 1200 dV in reserve.

The fuel cell arrays ended up on the front decouplers where the propellers are. It simply doesn't seem worth it to hide them in anything. The kerbals got moved to command seats in the cargo bay -- not exactly travelling in style. The six Darts were replaced with three Skiffs. Admittedly that's less thrust, but it's also less weight and drag, and it has acceptable isp for that altitude.

Lesson learned on wings: If you have weaker engines with low TWR, you're better off keeping the wings for the lift and control they provide. Doubly so if you're lacking in gimbal and reaction wheels. Whereas if you have powerful engines like the Vector, you can power into a bullet trajectory earlier and drop all lifting surfaces.

Lesson learned on performing a gravity turn: Same thing. Stronger engines can power through the atmosphere at a shallow angle. Weaker engines benefit from a high angle to get out of the atmosphere sooner.

If I one day embark on making an Eve SSTO, the question is this: Nerv or Dawn? Nerv makes sense because spaceplane wings and fuselages are some of the few fuel tanks that can carry pure liquid fuel. Dawn makes sense because the use of rotors already requires significant electricity generation.

[Lt_Duckweed]

For an Eve ssto, ions don't make sense unless you are trying to get back to Kerbin.  The total impulse you can extract from an ion engine on an ssto ascent is less than the impulse you can get by just replacing the ion engine with liquid fuel for your nervs.

[Guest]

On 2/1/2021 at 10:32 PM, InfernoSD said:

If I one day embark on making an Eve SSTO, the question is this: Nerv or Dawn? Nerv makes sense because spaceplane wings and fuselages are some of the few fuel tanks that can carry pure liquid fuel. Dawn makes sense because the use of rotors already requires significant electricity generation.

Neither. You'll still need a ton of TWR to make orbit, and neither of them has enough. Vectors and Darts are the way to go.

[Lt_Duckweed]

1 hour ago, Brikoleur said:

Neither. You'll still need a ton of TWR to make orbit, and neither of them has enough. Vectors and Darts are the way to go.

Nerv does help though.  On my Eve sstos, I use a ratio of 2 nervs for every 1 vector.  I start up the nervs and vectors at the same time at 16.5km,  the vectors then run out of fuel at a surface speed of about 2500m/s and the rest of the trip to orbit is done by the nervs.

 

With these ratios I have been able to achieve 5% payload fraction to orbit.

Ideal twr in terms of efficiency balance between engine dry mass and total fuel load seems to be ~78 tons per 1 vector and 2 nerv.

So for a 176 ton ssto this translates out to 2 vectors, 4 nervs, and ~8.8 tons of payload.  I found in my testing that taking away a pair of nervs and replacing with more fuel for the vectors had a negative impact on payload.

Edited February 3, 2021 by Lt_Duckweed

[Guest]

Informative!