Jump to content
  • 0

Improving a spaceplane


Question

planes are pretty much the only part of this game that mostly eludes me. unfortunately, i have a penchant for making highly reusable missions that can also explore the ground. Landing an ssto on laythe is trivial, but if i want to also go multiple biomes, a plane is pretty much a necessity. so here i am, stuck with making planes.

So here's the deal: I want a plane that I can send inside Eve, coupled with an ascent stage (possibly one that can be detached on the surface if it is too bulky to explore around freely). then the plane goes back to orbit, shedding the ascent stage. then later the same plane descends on laythe; it gets back from laythe all in one piece, and is later reused at tekto, from outer planet mod. I also want the plane to have a good IVA view, this rules out any crew pod without large windows. I actually have a workable design for laythe and tekto, but it does not work on eve and it has other shortcomings, which i was hoping to fix. so here is the attempt

Ocb7ylp.png

yes, i know what you are thinking. "omg Mk2 parts! Vade retro! vade retro!"

Yes, i know. they are kinda bad. unfortunately, my tried-and-true design with Mk1 parts explodes from heating on Eve, especially the crew capsule. Mk2 parts have better heat resistance, hence this. the in-line cockpit has larger windows, which is why i'm choosing it.

I also have other reasons for this Mk2 design:

- the cargo bay is larger, i can fit more stuff inside. the other version had 3 Mk1 cargo bays, and it had much less room than this one which has a single 250 kg cargo bay. It works a lot better. having the kerbalism mod, i can use it to store more and more redundant life support systems, and i can fit inside reaction wheels in a way that they can be swapped out if they break. something i could not do with the other model

- the crew pod has 2 places instead of 1, and it is pressurized (again, a thing of kerbalism), which is better for the pilots.

So, i really would like it if i could use this version.

 

Unfortunately, it has many problems:

- tends to flip

- difficult to take off, requires 100 m/s to lift

- poor manueverability

- heavier

- huge drag when using rockets

 

"heavier" cannot be avoided. I have 2 pilots instead of 1, in a larger cabin. it's heavier, and it requires more fuel to lift.

For the first three, I already know the problem: the center of mass must be shifted forward! yes, ok. how?

shifting the lateral tanks to the front would then require to also shift the wings, because the center of lift must follow the center of mass. and then i'd aso have moved the control surfaces, which will be the same distance from the center of mass. i could shift the crew cabin backwards of add more fuel tanks in front, i don't want to do that because it would block the view from the crew cabin. also, pushing the front wings forward would look ugly, as they would go over the part where the fuselage starts to thin, leaving a hole between wing and fuselage.

even worse is the drag

AYcysg1.png

i'm only at 350 m/s, but i am losing more for drag than for gravity. on a plane that looks very aerodinamic. i've seen flat top launchers with less drag losses. and I really have no idea what's causing all that drag, as i repeat, the shape is very aerodinamic. what are those two huge red arrows? which parts are causing so much drag, and why?

btw, the purple lines are not real, the propellers are safely stored inside cargo bays, but the game keeps the aerodinamic arrows if you activated them before.

 

i like this design. i'd rather use it. it has many pros, and it looks good. but it's unworkable for two reasons, one the center of mass that i can't figure out how to fix, and the drag, which i can't figure out where it comes from. I am open to suggestions

 

Link to post
Share on other sites

Recommended Posts

  • 1
Posted (edited)

A couple things:

1. It is possible to take off from the runway at 100 m/s at about 1/3 the wing loading you are using, the overwing is also cutting into your l/d ratio by requiring a non ideal AoA.  Move CoL and landing gear closer to CoM and add canards to greatly ease the takeoff flare, and remove one set of big-s and replace with fuel tanks.

2. Something is generating an absolutely horrendous level of drag in those side stacks.  Go into the Alt+f12 menu and go to the aero tab, turn on "aero data in part action windows" and open the action windows for all parts on the craft one by one to find the culprit.  Do this on a run where you do not use the props at all, keep them closed in the bays all the way from launch (there is an issue where closing the bay does not update the aero gui and part action windows, so it makes diagnosing drag issues harder), I suspect the issue is likely something attached to the wrong node by mistake, resulting in a large flat face presented to the airflow.  Double check the attachments around the prop bays to make sure everything is actually attached how you would expect.

3.  Get the mods "CorrectCoL", and "RCSBuildAidContinued", they are vital for efficient spaceplanes, they streamline the build process immensely.  Between the two of them, you can see your wet and dry CoM at the same time, see the TRUE CoL (taking ALL parts into account), and generate a stability graph for a selected speed and altitude.  Between these you will be able to accurately diagnose the flipping issue, and better tune the craft wet/dry CoM excrements and the CoL-CoM relationship.

4.  Once properly tuned, you should expect to see a lift to drag ratio of at least 2 at Mach 1.15, and at least 3.5 at Mach 2 (Maximal values for a non clipped craft are somewhere near 2.9-3 at Mach 1.15, and 4.8-4.9 at Mach 2).

Edited by Lt_Duckweed
Link to post
Share on other sites
  • 0

Couple of things I noticed.

  • Looks (to me) like it has a lot more wing than it needs.  
  • I don’t angle the wings upwards.  Especially not with the MK2 fuselage, which also produces lift.
  • To reduce drag, I try to build planes longer instead of wider.  Two of those engines could be mounted inline with the MK2 fuselage.
  • If the plane is ‘tippy’, try adding small canard elevator surfaces to the nose of the craft.
  • Putting engines inline with the fuselage will displace the shielded docking port you have there.  It can be relocated to the belly or even the nose, it’s actually not that draggy.   Also, there is an inline MK2 docking port..
  • What is powering your craft in the photo going 350m/s?  If that speed is on the propellers alone, that’s about as good as it will get on props alone.
Link to post
Share on other sites
  • 0
16 minutes ago, 18Watt said:

Couple of things I noticed.

  • Looks (to me) like it has a lot more wing than it needs.  

really? if it can't take off, i'd assume the opposite. also, consider it has to fly on laythe too. and swapping big-S wings for other wings make me lose fuel, so 4 big-S are better than 2 big-S and some smaller wings, even though the second may fly better.

Quote
  • I don’t angle the wings upwards.  Especially not with the MK2 fuselage, which also produces lift.

can a plane with zero angle of attack even work?

Quote
  • To reduce drag, I try to build planes longer instead of wider.  Two of those engines could be mounted inline with the MK2 fuselage.
  • Putting engines inline with the fuselage will displace the shielded docking port you have there.  It can be relocated to the belly or even the nose, it’s actually not that draggy.   Also, there is an inline MK2 docking port..

there is another limitation here: i want to leave the docking port behind because for eve i dock there the ascent stage. so it must be backward, aligned with the center of mass.

also, with kerbalims every engine has a small chance to fail on ignition. the plane should be able to reach orbit with 2 engines, at least on laythe (and tekto should be easier, with the propellers clearing most of the atmosphere), and i have 4 in case one breaks (and i also have to shut down its symmetric). so, the four engines are pretty muich a necessity. i could swap out a couple of darts for some cubs placed radially, but they increase drag too, and they are less efficient.

Quote

If the plane is ‘tippy’, try adding small canard elevator surfaces to the nose of the craft.

didn't knew that, i'll try it

Quote

What is powering your craft in the photo going 350m/s?  If that speed is on the propellers alone, that’s about as good as it will get on props alone.

i wish. propellers get me to 170 m/s. afterwards, i closed the propellers bays to reduce drag, and i activated the darts. I have 4 darts pushing, for a TWR of around 1.5, but i can't go any faster than that because of drag.

By the way, if I could optimize the plane to clear up more of the atmosphere on propellers alone, i could also accept more drag, it would be less of the problem.

Link to post
Share on other sites
  • 0
Posted (edited)
45 minutes ago, king of nowhere said:

really? if it can't take off, i'd assume the opposite. also, consider it has to fly on laythe too.

Being unable to takeoff is a different problem than not being fly able.  It’s often a problem of the distance from the landing gear to the pitch-control surfaces.  Try moving the main landing gear forward, and also adding canard elevators to the nose.

45 minutes ago, king of nowhere said:

can a plane with zero angle of attack even work?

Absolutely.  Honestly, I never angle my wings.  They get to space just fine.  My MK3 spaceplane can haul 16 Kerbals to Minmus, has 6 Rapier engines, and only uses two Big-S wings.  They are not angled.

45 minutes ago, king of nowhere said:

there is another limitation here: i want to leave the docking port behind because for eve i dock there the ascent stage.

That is a problem which I don’t have an answer for.  You may be stuck with engines mounted in lateral pods.  In terms of drag, that’s not ideal, but you should be able to work around that.

Edit, to add:  I realize Rapiers won't help you- they're useless on Eve..

Edited by 18Watt
Link to post
Share on other sites
  • 0

I just noticed something else in your screenshot.  You are supersonic (over Mach 1.0) at 7,000m altitude.  That's actually pretty good.  I suspect you will see your speed increase greatly at much higher altitudes.  

At low altitudes going supersonic you end up using a lot of fuel just to overcome the drag.  There's no easy way around that.  I do suggest trying to limit the time you spend at low altitudes going fast.  In other words, with the engines you have I would try to get altitude quicker, and not worry about speed until you get out of the thick atmosphere.

Link to post
Share on other sites
  • 0
18 hours ago, king of nowhere said:

Unfortunately, it has many problems:

- tends to flip

- difficult to take off, requires 100 m/s to lift

- poor manueverability

- heavier

- huge drag when using rockets

This is generic not necessarily a reference to your design:

- issues with CoM / CoL placement. Lack of control surfaces can also cause.

- move rear landing gear to just behind the CoM. I like to raise the rear gear up slightly to create a natural nose up attitude. Add more lifting surfaces.

- add more ailerons. Fill that delta wing at the back. I find “tail wings” help immensely as well.

- check ascent profile. Inefficient climb means more fuel = heavier.

- try lower profile engines like aerospike/rapier etc.

 

Finally, check out some Matt Lowne videos on YouTube (he posts craft files here as well) as he builds lots of spaceplanes to do similar things to your mission.

Link to post
Share on other sites
  • 0
On 5/9/2021 at 11:58 AM, 18Watt said:

Absolutely.  Honestly, I never angle my wings.  They get to space just fine.  My MK3 spaceplane can haul 16 Kerbals to Minmus, has 6 Rapier engines, and only uses two Big-S wings.  They are not angled.

Not angling your wings is a horrifically bad decision.  Proper wing incidence + the consequent reduction in needed wing area can more than double a craft's lift to drag ratio, allowing ascent with half as many engines.

Link to post
Share on other sites
  • 0
4 hours ago, Lt_Duckweed said:

Not angling your wings is a horrifically bad decision.  

I respectfully disagree.  I thought perhaps something had changed in the KSP aero model, so I did some testing.  While there may be some advantages to adding wing incidence at low altitudes and slow speeds, there is a significant penalty at higher altitudes and speeds.  The blanket statement that angled wings is always better is simply not true.

Test Methods:

Spoiler

I built a very basic MKII space-plane.  From front to back, it has a Shock-Cone inlet, short MKII adapter tank, MKII Inline Fuselage, Long MKII rocket fuel Fuselage, short MKII adapter tank, and a RAPIER engine.  I used the enormous Big-S wings (since we're discussing those..), small rudder, small elevators on the tail, and small canard elevators on the nose.  Also has 3 medium landing gears.  I reduced the LF in the wings to half-full, but left the rest of the tanks 100% full of LF and O.

I evaluated the plane at 5Km, 10Km, and 20Km in level flight.  For testing, I activated Infinite Fuel and Infinite Electricity, to ensure the plane would be at exactly the same weight and CG during all testing.

I'm playing 100% stock, with BG and MH expansion packs.  So maintaining level flight was tricky, and absolutely an opportunity for errors.

I used the same plane for both tests.  However, it's worth noting that when I angled the wings up 5 degrees, that affected the COL.  To keep things fair, I did my best to optimize the plane with angled wings, by moving the wings forward/aft.  However, even though all the parts and test mass are identical, angling the wings does change the plane's characteristics.  Again, I tried my best to optimize the version with angled wings, and not penalize it for an altered COL and CG.

Test Results:

Spoiler

Level Flight, 5,000m:

Straight Wing:  760m/s, Fuel Flow = 0.86u.  Max Speed = 1,490m/s.

Angled Wing:  760m/s, Fuel Flow = 2.27u.  Max Speed = 760m/s.

Winner:  Straight wing, by a large margin.  However, for reaching space, I'm not too concerned about performance that low.

Level Flight, 10,000m:

Straight Wing:  1,000m/s, Fuel Flow = 0.65u.  Max Speed = 1,550m/s+.  (I had to back off, it was getting too hot!)

Angled Wing:  1,000m/s, Fuel Flow = 1.44u.  Max Speed = 1,160m/s.

Winner: Straight wing, again by a large margin.

Level Flight, 20,000m:

Straight Wing:  1,000m/s, Fuel Flow = .44u.  Max Speed = 1,605m/s.

Angled Wing:  1,000m/s, Fuel Flow = .31u.  Max Speed = 1,565m/s.

Winner:  Angled wing!  Not by much, but at 20K and 1,000m/s the angled wing was more efficient.  The straight wing was able to go faster, but not by very much.

Why none of this may be relevant:

Spoiler

Your goals may be different than mine, each situation will have an optimal machine which may differ greatly from the optimal machine from another situation.  Pure fuel efficiency may not be your only goal.  It may not be a goal at all!  I do think most of us look for ways to increase efficiency.  In some cases, adding the 5 degrees of wing angle may make for a better vessel.

I like to make spaceplanes which can get to orbit from Kerbin and Laythe.  (If it works on Kerbin, It'll work on Laythe.  Usually..).  The best way to do that efficiently is to get as fast and as high as you can on air-breathing engines.  On Kerbin, my goal is 1,500m/s when the engines flame out, around 24,000m altitude.  

Eve is much different.  I have no doubt that if you can get a plane to orbit from Eve, getting to orbit from Laythe will be easy, that seems to be the goal of the OP here.  My disdain for wing-angling may not be relevant on Eve- I've never built a plane (or anything else) capable of getting to Eve orbit from the surface.  Perhaps an angled wing would be advantageous.

However, the blanket statement that angled wings are always better is simply not true.  In fact, I can't remember ever building a plane where angled wings improved the performance.

A final note about Eve.  While on Kerbin (and Laythe) you want all the speed you can get (to reach orbit)- On Eve Speed is probably your enemy.  My focus would be (as an amateur..) to get as high as you can with wings, to get out of the thick atmosphere, before lighting your rocket engines.  I wouldn't worry about speed until I was quite high up in Eve's atmosphere.

 

Link to post
Share on other sites
  • 0
2 hours ago, Lt_Duckweed said:

A couple things:

1. It is possible to take off from the runway at 100 m/s at about 1/3 the wing loading you are using, the overwing is also cutting into your l/d ratio by requiring a non ideal AoA.  Move CoL and landing gear closer to CoM and add canards to greatly ease the takeoff flare, and remove one set of big-s and replace with fuel tanks.

actually, I don't want to take off from the runway at 100 m/s. I want to take off from Laythe on patches of irregular terrain. I also want to be able to land on irregular terrain, and my piloting skills are meh. This is a propeller plane, not a jet-powered thing. and it's for exploring outer planets, not for using around kerbin.

moving the center of lift and and landing gear and adding canard were good tips. I don't know what did the trick, but now i can take off at 50 m/s. which is more or less the target for something that's comfortable to use on laythe.

 

Quote

2. Something is generating an absolutely horrendous level of drag in those side stacks.  Go into the Alt+f12 menu and go to the aero tab, turn on "aero data in part action windows" and open the action windows for all parts on the craft one by one to find the culprit.  Do this on a run where you do not use the props at all, keep them closed in the bays all the way from launch (there is an issue where closing the bay does not update the aero gui and part action windows, so it makes diagnosing drag issues harder), I suspect the issue is likely something attached to the wrong node by mistake, resulting in a large flat face presented to the airflow.  Double check the attachments around the prop bays to make sure everything is actually attached how you would expect.

I was starting to suspect something like that, indeed it was the case. the darts were attached not to the end of the cargo bays, but to the rotors inside. so both the darts and the cargo bays had some virtual big flat surface exposed to the airflow. i fixed that, and now it's solved the drag problems. Now the plane can obrit kerbin, which means it can also orbit laythe and tekto. as for eve, it all depends on the lift stage.

pYm47PH.png

on the other hand, i wonder how the plane could flip with so much drag on the bottom...

i also discovered that the crew cabin is quite draggy. it would improve the plane to move it backwards, but i don't want to lose the better view. which, so far, is not obstructed too much by the wings.

PyMbbrj.png

finally, the clamp-o-tron is very draggy, despite being shielded. in fact, i tried as an experiment to strap two more normal clamp-o-trons, the drag difference is negligible; the shielded one is maybe 15% less draggy.

jjst1YN.png

both are also much draggier than the aerospikes. i also tried to swap places to see if it was an artifact, but no, those data were consistently reproducible; the shielded clamp-o-tron reduced drag is so small, it may not even be worth the extra 50 kg it adds.

 

Quote

3.  Get the mods "CorrectCoL", and "RCSBuildAidContinued", they are vital for efficient spaceplanes, they streamline the build process immensely.  Between the two of them, you can see your wet and dry CoM at the same time, see the TRUE CoL (taking ALL parts into account), and generate a stability graph for a selected speed and altitude.  Between these you will be able to accurately diagnose the flipping issue, and better tune the craft wet/dry CoM excrements and the CoL-CoM relationship.

i tried to install kerbal wind tunnel, and it was utterly ineffective. i've been very wary of aerodinamic mods ever since.

well, no, not really. i've  simply never tried to make a complex plane ever since. as i said, i don't like planes and i'm just forced to make one by my self-imposed challenge.

Quote

4.  Once properly tuned, you should expect to see a lift to drag ratio of at least 2 at Mach 1.15, and at least 3.5 at Mach 2 (Maximal values for a non clipped craft are somewhere near 2.9-3 at Mach 1.15, and 4.8-4.9 at Mach 2).

yes, i was at 2 at mach 1.1. not the best flyer, but that's often the case when trying to make a vehicle fit for multiple tasks. as long as it flies well enough and it reaches orbit, it'ìs sufficient. it doesn't have to bring any extra payload except for a bit of science

Link to post
Share on other sites
  • 0
Posted (edited)
47 minutes ago, 18Watt said:

I respectfully disagree.  I thought perhaps something had changed in the KSP aero model, so I did some testing.  While there may be some advantages to adding wing incidence at low altitudes and slow speeds, there is a significant penalty at higher altitudes and speeds.  The blanket statement that angled wings is always better is simply not true.

Test Methods:

  Reveal hidden contents

I built a very basic MKII space-plane.  From front to back, it has a Shock-Cone inlet, short MKII adapter tank, MKII Inline Fuselage, Long MKII rocket fuel Fuselage, short MKII adapter tank, and a RAPIER engine.  I used the enormous Big-S wings (since we're discussing those..), small rudder, small elevators on the tail, and small canard elevators on the nose.  Also has 3 medium landing gears.  I reduced the LF in the wings to half-full, but left the rest of the tanks 100% full of LF and O.

I evaluated the plane at 5Km, 10Km, and 20Km in level flight.  For testing, I activated Infinite Fuel and Infinite Electricity, to ensure the plane would be at exactly the same weight and CG during all testing.

I'm playing 100% stock, with BG and MH expansion packs.  So maintaining level flight was tricky, and absolutely an opportunity for errors.

I used the same plane for both tests.  However, it's worth noting that when I angled the wings up 5 degrees, that affected the COL.  To keep things fair, I did my best to optimize the plane with angled wings, by moving the wings forward/aft.  However, even though all the parts and test mass are identical, angling the wings does change the plane's characteristics.  Again, I tried my best to optimize the version with angled wings, and not penalize it for an altered COL and CG.

Test Results:

  Reveal hidden contents

Level Flight, 5,000m:

Straight Wing:  760m/s, Fuel Flow = 0.86u.  Max Speed = 1,490m/s.

Angled Wing:  760m/s, Fuel Flow = 2.27u.  Max Speed = 760m/s.

Winner:  Straight wing, by a large margin.  However, for reaching space, I'm not too concerned about performance that low.

Level Flight, 10,000m:

Straight Wing:  1,000m/s, Fuel Flow = 0.65u.  Max Speed = 1,550m/s+.  (I had to back off, it was getting too hot!)

Angled Wing:  1,000m/s, Fuel Flow = 1.44u.  Max Speed = 1,160m/s.

Winner: Straight wing, again by a large margin.

Level Flight, 20,000m:

Straight Wing:  1,000m/s, Fuel Flow = .44u.  Max Speed = 1,605m/s.

Angled Wing:  1,000m/s, Fuel Flow = .31u.  Max Speed = 1,565m/s.

Winner:  Angled wing!  Not by much, but at 20K and 1,000m/s the angled wing was more efficient.  The straight wing was able to go faster, but not by very much.

Why none of this may be relevant:

  Reveal hidden contents

Your goals may be different than mine, each situation will have an optimal machine which may differ greatly from the optimal machine from another situation.  Pure fuel efficiency may not be your only goal.  It may not be a goal at all!  I do think most of us look for ways to increase efficiency.  In some cases, adding the 5 degrees of wing angle may make for a better vessel.

I like to make spaceplanes which can get to orbit from Kerbin and Laythe.  (If it works on Kerbin, It'll work on Laythe.  Usually..).  The best way to do that efficiently is to get as fast and as high as you can on air-breathing engines.  On Kerbin, my goal is 1,500m/s when the engines flame out, around 24,000m altitude.  

Eve is much different.  I have no doubt that if you can get a plane to orbit from Eve, getting to orbit from Laythe will be easy, that seems to be the goal of the OP here.  My disdain for wing-angling may not be relevant on Eve- I've never built a plane (or anything else) capable of getting to Eve orbit from the surface.  Perhaps an angled wing would be advantageous.

However, the blanket statement that angled wings are always better is simply not true.  In fact, I can't remember ever building a plane where angled wings improved the performance.

A final note about Eve.  While on Kerbin (and Laythe) you want all the speed you can get (to reach orbit)- On Eve Speed is probably your enemy.  My focus would be (as an amateur..) to get as high as you can with wings, to get out of the thick atmosphere, before lighting your rocket engines.  I wouldn't worry about speed until I was quite high up in Eve's atmosphere.

 

you do realize this is a propeller plane built for exploration?

it's a very different concept. most of my flying will indeed be done at low speed and low altitudes. once on a planet, i move around on the various biomes, landing and taking surface and atmospheric science potentially everywhere. the propellers are powered by rtgs, so they can go indefinitely. and I put enough food for one month. i will probably scout around a few hours before getting bored.

when i leave, i go up as high as i can on propeller power. there is no worry about optimal ascent there, because propellers have infinite range. all that matters is that I reach as high and as fast as i can. then i close the propellers bays and i activate the darts, and in a few minutes I reach orbit, where i have my mothership waiting me.

you, on the other hand, seem to be giving advice related to making a kerbin-based, rapier-propelled spaceplane to make a fully fuel-powered ascent and bring as much payload to orbit as possible. you don't have to worry about landing on a mountain in the middle of the sea. or driving it on the ground to align it with enough precision that you can take a scan of a geyser with the robotic arm.

so yes, we have very different design objectives. i am going to pilot this thing for hours at speeds between 100 and 200 m/s, making a lot of take-off and landing on rough terrain during that time. I want a plane that's comfortable to drive at that speed and somewhat forgiving on the landing. easy to take off, because i won't have the luxury of a flat runway. i care about having large windows, it makes driving much better.

on the other hand, i only spend a few minutes orbiting, and even though those may be miserable, as long as the plane does the job, it's ok. heck, my other design was a real pain in the high atmosphere, completely unstable, and it was mediocre even at low speeds. but it could take off from water on propellers alone, and with a mission of exploring laythe and bring back samples from any biome, this far outweights any other concern. I'd keep using that design, but having to add life support, radiation shielding and redundant pieces was too much and i couldn't take off from water anymore anyway.

Edited by king of nowhere
Link to post
Share on other sites
  • 0
Posted (edited)
1 hour ago, 18Watt said:

I respectfully disagree.  I thought perhaps something had changed in the KSP aero model, so I did some testing.  While there may be some advantages to adding wing incidence at low altitudes and slow speeds, there is a significant penalty at higher altitudes and speeds.  The blanket statement that angled wings is always better is simply not true.

Test Methods:

  Hide contents

I built a very basic MKII space-plane.  From front to back, it has a Shock-Cone inlet, short MKII adapter tank, MKII Inline Fuselage, Long MKII rocket fuel Fuselage, short MKII adapter tank, and a RAPIER engine.  I used the enormous Big-S wings (since we're discussing those..), small rudder, small elevators on the tail, and small canard elevators on the nose.  Also has 3 medium landing gears.  I reduced the LF in the wings to half-full, but left the rest of the tanks 100% full of LF and O.

I evaluated the plane at 5Km, 10Km, and 20Km in level flight.  For testing, I activated Infinite Fuel and Infinite Electricity, to ensure the plane would be at exactly the same weight and CG during all testing.

I'm playing 100% stock, with BG and MH expansion packs.  So maintaining level flight was tricky, and absolutely an opportunity for errors.

I used the same plane for both tests.  However, it's worth noting that when I angled the wings up 5 degrees, that affected the COL.  To keep things fair, I did my best to optimize the plane with angled wings, by moving the wings forward/aft.  However, even though all the parts and test mass are identical, angling the wings does change the plane's characteristics.  Again, I tried my best to optimize the version with angled wings, and not penalize it for an altered COL and CG.

Test Results:

  Hide contents

Level Flight, 5,000m:

Straight Wing:  760m/s, Fuel Flow = 0.86u.  Max Speed = 1,490m/s.

Angled Wing:  760m/s, Fuel Flow = 2.27u.  Max Speed = 760m/s.

Winner:  Straight wing, by a large margin.  However, for reaching space, I'm not too concerned about performance that low.

Level Flight, 10,000m:

Straight Wing:  1,000m/s, Fuel Flow = 0.65u.  Max Speed = 1,550m/s+.  (I had to back off, it was getting too hot!)

Angled Wing:  1,000m/s, Fuel Flow = 1.44u.  Max Speed = 1,160m/s.

Winner: Straight wing, again by a large margin.

Level Flight, 20,000m:

Straight Wing:  1,000m/s, Fuel Flow = .44u.  Max Speed = 1,605m/s.

Angled Wing:  1,000m/s, Fuel Flow = .31u.  Max Speed = 1,565m/s.

Winner:  Angled wing!  Not by much, but at 20K and 1,000m/s the angled wing was more efficient.  The straight wing was able to go faster, but not by very much.

Why none of this may be relevant:

  Hide contents

Your goals may be different than mine, each situation will have an optimal machine which may differ greatly from the optimal machine from another situation.  Pure fuel efficiency may not be your only goal.  It may not be a goal at all!  I do think most of us look for ways to increase efficiency.  In some cases, adding the 5 degrees of wing angle may make for a better vessel.

I like to make spaceplanes which can get to orbit from Kerbin and Laythe.  (If it works on Kerbin, It'll work on Laythe.  Usually..).  The best way to do that efficiently is to get as fast and as high as you can on air-breathing engines.  On Kerbin, my goal is 1,500m/s when the engines flame out, around 24,000m altitude.  

Eve is much different.  I have no doubt that if you can get a plane to orbit from Eve, getting to orbit from Laythe will be easy, that seems to be the goal of the OP here.  My disdain for wing-angling may not be relevant on Eve- I've never built a plane (or anything else) capable of getting to Eve orbit from the surface.  Perhaps an angled wing would be advantageous.

However, the blanket statement that angled wings are always better is simply not true.  In fact, I can't remember ever building a plane where angled wings improved the performance.

A final note about Eve.  While on Kerbin (and Laythe) you want all the speed you can get (to reach orbit)- On Eve Speed is probably your enemy.  My focus would be (as an amateur..) to get as high as you can with wings, to get out of the thick atmosphere, before lighting your rocket engines.  I wouldn't worry about speed until I was quite high up in Eve's atmosphere.

 

Ah, I see where the disconnect is.  When you use wing incidence you have to use a different design philosophy.

You do NOT pick an altitude AND speed to fly at.  You fly at 0 Angle of Attack, and put exactly enough wing on the craft so that it achieves 0 angle of attack at the intended altitude and airspeed.  Your test craft has FAR too much wing for it's mass and is flying far too low for it's speed in the first 2 tests.  A single pair of Big-S wings will happily hold up at least 50 tons of plane.  At MINIMUM you should be using a wing loading of 5 tons mass per 1 wing area.  My standard ratio is 1 rapier and 1 nerv per 36 tons of takeoff mass, and 1 wing area per ~5 tons.  This typically results in a flight ceiling of about 20 km at 1650 m/s, after which I kick on the nervs and cruise to orbit.  With this style of craft, 8km/s in LKO, or alternately, 50% payload fraction, is trivial to obtain.

When built and flown correctly, wing incidence will always improve efficiency to orbit, via an enormous reduction to body drag by flying 0 AoA through almost the entire flight.

Also forgot to mention, but Mk2 parts are really, really bad.  They generate both regular body drag, AND wing drag.  So they have extremely high drag at 0 AoA, and it only gets worse as you add AoA

Edited by Lt_Duckweed
Link to post
Share on other sites
  • 0

This craft has 2.49 kn of body drag out of 25.815 kn of drag total.  This is a body drag of 9.6%

Wing area is 6 strakes with Angle of Incidence of 5 degrees, it has a mass at altitude of 30.7 tons and cruise speed of 1700m/s, and an estimated runtime and endurance of something like 20 circumnavigations in ~14 hours of flight.

ojkBh0k.jpg

Link to post
Share on other sites
  • 0
55 minutes ago, king of nowhere said:

when i leave, i go up as high as i can on propeller power. there is no worry about optimal ascent there, because propellers have infinite range.

I'm watching your trials closely.  I need to send a Kerbal to the surface of Eve for a challenge.  While I don't need to bring them back to complete the challenge, I hate to leave them on the surface of Eve indefinitely.  So I'm trying to figure out how to get a Kerbal back to orbit.  Your approach looks like my best shot.  Optimal on Eve is much different than optimal on Kerbin.  Eve may be an opportunity to use wing-angling.

39 minutes ago, Lt_Duckweed said:

A single pair of Big-S wings will happily hold up at least 50 tons of plane.

I agree.  My go-to SSTO weighs 98T on takeoff, and has two Big-S wings.  Can go to Minmus, land, get back to Kerbin, and land at KSC.  Yes, the takeoff speed is very high, around 85m/s.  But two Big-S wings on a 20-ton MKII plane is not even remotely a good choice.

39 minutes ago, Lt_Duckweed said:

Also forgot to mention, but Mk2 parts are really, really bad. 

Yeah.  Sometimes the form-factor is more appropriate.  But I agree.  All things considered, it's almost always better using anything but the MKII.  Too bad, because it looks cooler.  

53 minutes ago, Lt_Duckweed said:

When you use wing incidence you have to use a different design philosophy.

Absolutely.  At speeds in the 100-250m/s range, I'm confident that the ability to maintain lift with almost no fuselage drag would be beneficial with angled wings.  However, the OP has mentioned that fuel economy at those speeds is not important- he is using electric propellers.  The critical part is going to be punching through at least some of Eve's atmosphere, while burning precious rocket fuel.  So while efficiency while low and slow is great, that's not really the problem.  With infinite electric propulsion at those speeds, it doesn't make as much difference how efficient it is down low and slow.  The critical part is how much fuel is it going to take to punch through Eve's upper atmosphere?

Link to post
Share on other sites
  • 0
Posted (edited)
6 minutes ago, 18Watt said:

The critical part is going to be punching through at least some of Eve's atmosphere, while burning precious rocket fuel.  So while efficiency while low and slow is great, that's not really the problem.  With infinite electric propulsion at those speeds, it doesn't make as much difference how efficient it is down low and slow.  The critical part is how much fuel is it going to take to punch through Eve's upper atmosphere?

And wing incidence helps there too.  Both of the sstos in these videos use 3 degrees angle of incidence to greatly reduce body drag

 

Let me emphasize again, if you build and fly a craft correctly, some incidence will always be better than none.  The only question is how much.

Edited by Lt_Duckweed
Link to post
Share on other sites
  • 0
1 minute ago, Lt_Duckweed said:

This craft has 2.49 kn of body drag out of 25.815 kn of drag total.  This is a body drag of 9.6%

Sorry, but it's just a bright yellow blur in the screenshot!

Seriously, I think one of the most common problems with KSP aircraft I see is way too much wing for the aircraft.  I think there is a tendency to say, 'Hmm, it doesn't fly right, so I'll keep adding wings until it does.'  That is seldom the right answer.  I can't think of any KSP plane examples I've seen in the help forums where I can honestly say it needed more wings.

Maybe add a screenshot of your dart in the SPH, to demonstrate that you don't need a lot of wings.  

As a further point, Eve in particular does not require much wing at all.  I'm currently testing a MKII boat, with no wings at all.  Above 50m/s, it is in danger of going flying!  With no wings or control surfaces, that always ends badly.

I agree that the MKII doesn't make great planes.  But they make excellent boat hulls!

Link to post
Share on other sites
  • 0
1 minute ago, 18Watt said:

Sorry, but it's just a bright yellow blur in the screenshot!

Seriously, I think one of the most common problems with KSP aircraft I see is way too much wing for the aircraft.  I think there is a tendency to say, 'Hmm, it doesn't fly right, so I'll keep adding wings until it does.'  That is seldom the right answer.  I can't think of any KSP plane examples I've seen in the help forums where I can honestly say it needed more wings.

Maybe add a screenshot of your dart in the SPH, to demonstrate that you don't need a lot of wings.  

As a further point, Eve in particular does not require much wing at all.  I'm currently testing a MKII boat, with no wings at all.  Above 50m/s, it is in danger of going flying!  With no wings or control surfaces, that always ends badly.

I agree that the MKII doesn't make great planes.  But they make excellent boat hulls!

FnaPJrX.jpg

Link to post
Share on other sites
  • 0
Posted (edited)
11 minutes ago, Lt_Duckweed said:

Both of the sstos in these videos use 3 degrees angle of incidence to greatly reduce body drag

Ok, how do you get 3 degrees?  I'm playing stock, and near as I can tell my choices are either zero or 5 degrees.  The way I like to fly, I think 1 or maybe 2 degrees would be helpful, but 5 is way too much.  Do you need a mod for that, or is that possible in stock KSP?

Edit:  Also, nice videos!

Edited by 18Watt
Link to post
Share on other sites
  • 0
4 hours ago, 18Watt said:

Ok, how do you get 3 degrees?

Editor Extensions mod. I won't play without it. You can set any angle you want.

Link to post
Share on other sites
  • 0
12 hours ago, 18Watt said:

Absolutely.  At speeds in the 100-250m/s range, I'm confident that the ability to maintain lift with almost no fuselage drag would be beneficial with angled wings.  However, the OP has mentioned that fuel economy at those speeds is not important- he is using electric propellers.  The critical part is going to be punching through at least some of Eve's atmosphere, while burning precious rocket fuel.  So while efficiency while low and slow is great, that's not really the problem.  With infinite electric propulsion at those speeds, it doesn't make as much difference how efficient it is down low and slow.  The critical part is how much fuel is it going to take to punch through Eve's upper atmosphere?

 

12 hours ago, 18Watt said:

Seriously, I think one of the most common problems with KSP aircraft I see is way too much wing for the aircraft.  I think there is a tendency to say, 'Hmm, it doesn't fly right, so I'll keep adding wings until it does.'  That is seldom the right answer.  I can't think of any KSP plane examples I've seen in the help forums where I can honestly say it needed more wings.

true. however, there is also the matter that i want to fly on laythe too. air is much thinner, so the plane must have decent flyability in those conditions too. Maybe all those wings are not needed, but if I have to land on the side on an hill, then i want my plane to be very manueverable. which means more wings. I will try reducing wings later, but i won't have the luxury of a flat runway.

i also want to be able to make water landings. and possibly i would also like to take off from water with a small help from the darts without spending too much fuel. else how do i take surface samples from degrasse sea?

 

less wings may certainly be beneficial for a quick trip to orbit, or for a fast rocket plane. but they improve manueverability, especially at low speed. if i have a plane that takes off at 80 m/s on kerbin, it will need over 100 m/s on laythe, and then i won't be able to land it on mountains without breaking something.

As for the Mk2, one of the first things i said in the opening post is that i'm aware they are heavy and draggy and overall bad. but they have high thermal resistance and the right crew size and the right cargo bay size, and that's worth the inconvenience.

Yes, I am trying to juggle a lot of different engineering needs. I like it this way. on the plus side, the plane does not need to be excellent at any of those tasks, merely good enough. the latest design is already there, minus a bit of trimming here and there. Oh, I still have to test if it goes to orbit on two engines alone; if it doesn't, I'll have to figure out something or add a third couple of darts. and then check that nothing else broke in the process

Link to post
Share on other sites
  • 0

@king of nowhere, I have a few questions.  I'm working on a design that meets your criteria, and I'm running into problems.  MKII, with the aft center node free (with a shielded docking port).  Right now I'm using two lateral aerospike engines, although I know you would prefer four.

So far, I can get to orbit on Kerbin.  But just barely.  First problem is the propellers only get me to about 7,500m altitude.  Second problem is the takeoff speed is too high, around 80m/s.  That's fine from the KSC runway, but a little too fast for Laythe, with few good flat runways.  I was able to hit 220m/s, but at the maximum altitude (~7,500m) I was down to ~180m/s.

The flight profile which has worked for me is to get to 7,500m on propellers.  Then, light the aerospikes, and pitch up to about 75 degrees.  The aerospikes do not have enough thrust to push the plane through the air, so I point the nose up, and get out of the atmosphere as fast as I can.  Passing 30km altitude, I let the nose come down to start gaining speed.

My question:  You mentioned that the aft docking port is to possibly join with an ascent stage?  How would that work?  If you mount it to a vertical ascent stage, how do you attach them on Eve, with a crane?  I'm curious because I was only able to make orbit of Kerbin just barely, and am doubting my ability to get to Eve orbit without some extra help.

@Lt_Duckweed, You'll be happy to know I did experiment with angled wings.  Although it didn't help me get any higher on the props, it also did not hurt me any, and reduced my takeoff speed by about 10m/s.  Thank you for the tip on angling wings in tiny increments (turn off angle-snap..).  Actually, at the very slow speeds I can hit with propellers, I found 2-3 degrees incidence to work nicely.  

Link to post
Share on other sites
  • 0
Posted (edited)
1 hour ago, 18Watt said:

@king of nowhere, I have a few questions.  I'm working on a design that meets your criteria, and I'm running into problems.  MKII, with the aft center node free (with a shielded docking port).  Right now I'm using two lateral aerospike engines, although I know you would prefer four.

So far, I can get to orbit on Kerbin.  But just barely.  First problem is the propellers only get me to about 7,500m altitude.  Second problem is the takeoff speed is too high, around 80m/s.  That's fine from the KSC runway, but a little too fast for Laythe, with few good flat runways.  I was able to hit 220m/s, but at the maximum altitude (~7,500m) I was down to ~180m/s.

The flight profile which has worked for me is to get to 7,500m on propellers.  Then, light the aerospikes, and pitch up to about 75 degrees.  The aerospikes do not have enough thrust to push the plane through the air, so I point the nose up, and get out of the atmosphere as fast as I can.  Passing 30km altitude, I let the nose come down to start gaining speed.

My question:  You mentioned that the aft docking port is to possibly join with an ascent stage?  How would that work?  If you mount it to a vertical ascent stage, how do you attach them on Eve, with a crane?  I'm curious because I was only able to make orbit of Kerbin just barely, and am doubting my ability to get to Eve orbit without some extra help.

 

first, i managed to take off from kerbin at 50 m/s, which means around 70 on laythe. good enough.

getting barely to orbit on kerbin is fine, if i can do that i can also get to orbit comfortably on laythe. by the way, since i had a bit of space in the cargo bay, i added a dawn engine and a bit of xenon in it; it's got a few hundred m/s on xenon, which is enough for some orbital manuevers if i need an extra. it can also help me circularize.

regarding the ascent stage, it works twofold; on tylo (and slate, which is similar) the ascent stage would be a normal rocket. the rocket part would have enough deltaV to take care of descent. the plane would land like a rocket, though it would turn 90 degrees just before landing, to hit the ground with the wheels.

then, for take off, i'll just start by accelerating on the ground. I'll start flying the moment i hit a bump at sufficiently high speed.

on eve, the ascent stage would have wings and additional propellers, so it could keep working as a regular plane. I did something similar with the Absolutely NOT Albatross - Jool stage

at 8 minutes the climb starts, shedding empty tanks from behind.

in fact, my new design is directly derived from this old one, adapted a bit for kerbalism and slightly different targets. but i envision the eve ascent stage as something similar, where i reach some altitude with propellers, then i turn on rockets, then i discard some empty tanks along the way. incidentally, i did try the not!Albatross on Eve, and it can't reach orbit. it couldn't even reach space, despite making it to 8 km on propellers before turning on the rockets with 3000 m/s of deltaV. but then, not!Albatross was quite unstable in the high atmosphere, and it had low thrust. its major selling point was that it could take off from water on propellers alone, and it was plenty of fun on Laythe for this reason, but I had to sacrifice a lot of aerodinamics to make that possible. Also, not!Albatross can't land on eve because it explodes for the heat, which is one of the reasons i'm using the heat resistant Mk2 this time.

with this new model i sacrifice water take-off, in exchange for heat and aerodinamic improvements - well, mostly center of mass which was very tilted backwards, it made the whole plane very unstable

the eve ascent stage will need working on, though. if i can't make it work, i can make a dedicated lander just for eve, though recycling landers is more elegant. but all this is a long term project on which i'm working saltuarily, my plate is full at the moment

Edited by king of nowhere
Link to post
Share on other sites
  • 0
59 minutes ago, king of nowhere said:

first, i managed to take off from kerbin at 50 m/s, which means around 70 on laythe. good enough.

Yes, that is workable!  One consideration is that while hopping around biomes you don't need any fuel, so those takeoffs and landings should be very easy.  At some point, you need to takeoff with full tanks to reach orbit.  If you're already down to 50m/s on Kerbin, that's awesome!

As I mentioned, I'm following this closely because I preparing to send a Kerbal to the surface of Eve, and haven't figured out how to get them back to orbit.  A agree that a reusable vehicle would be ideal.  I'm close to giving up on the MKII.  I like the form-factor of the MKII, but I'm having trouble squeezing enough performance out of it.

Link to post
Share on other sites
  • 0
Posted (edited)
8 minutes ago, 18Watt said:

Yes, that is workable!  One consideration is that while hopping around biomes you don't need any fuel, so those takeoffs and landings should be very easy.  At some point, you need to takeoff with full tanks to reach orbit.  If you're already down to 50m/s on Kerbin, that's awesome!

As I mentioned, I'm following this closely because I preparing to send a Kerbal to the surface of Eve, and haven't figured out how to get them back to orbit.  A agree that a reusable vehicle would be ideal.  I'm close to giving up on the MKII.  I like the form-factor of the MKII, but I'm having trouble squeezing enough performance out of it.

heh.

A reusable eve launcher based on the concept of propellers up until 15 km and then rockets is potentially feasible, some people did it, but it's extremely hard (without kraken aerodinamics). if i had to rate the hardest challenge of this game, i'd put that one first, without doubt.

i tried it in the past, i spent weeks working on several projects, but i never came close. i have also seen a pure rocket that can go ssto on eve on rocket power alone, if it lands on the highest plateau.

barring that, you can use a regular big rocket. my first eve ascent vehicle was 400 tons and barely functional, but my second one was 250 tons, had a crew of 3, and its last stage could also be recycled as a laythe lander (rocket, not plane).

if you want to try a plane, without the huge challenge of making it ssto, then propellers + staging is probably the way to go. one of my project was an helicopter rotor strapped on a rocket, it failed to come anywhere close to ssto, but it did reach 15 km, so if you could then eject the rotors and go up as normal two-stage rocket, it would save a lot of weight on the first stage. that, or a staged plane like not!Albatross

if you want to make it a pure eve ssto, good luck with that. at least you know aerodinamics and ascent profiles better than i do, you've got a better chance.

 

Edited by king of nowhere
Link to post
Share on other sites
  • 0

I've been unable to get the plane to have low drag during engine ignition. Probably it's the Mk2 fault, mostly. The big wings certainly don't help, too. Unfortunately, the characteristics that make a good rocket spaceplane and those that make a good explorer propeller plane are often mutually exclusive. add in the need to add radiation shielding, food, redundant life support; of course performance is going to take a hit there.

Anyway, this means that the best way to launch is, after getting as high as possible with propellers, to point mostly upwards. A good spaceplane accelerates mostly in the atmosphere, this one cannot.

The bad news is that I need all four darts for this. as i mentioned, under kerbalism engines can have mechanical failures; I'm not going to risk the mission of the chance of mechanical failure if i can help it, therefore i added two engines. they are unnecessary, redundant, they reduce performance overall. but they are my guarantee if an engine breaks. I considered putting an extra dart into a seq container, but they are too heavy to be handled on a surface. Anyway, they are necessary. So here is the final look of the plane

1uoXTOj.png

B9uCWl1.png

gbUyll9.png

it needs a bit higer speed to land and take off, but i tested it on laythe and it works. it has fairly good manueverability with propellers, though not as much as i'd like. Its predecessor could land obliquely on the side of a hill, i'm not sure this one can. it is very stable, though.

it is pretty poor as spaceplane, but it still reaches laythe orbit easily with 4 engines. i haven't tried it on tekto, i haven't yet installed the outer planet because i'm afrait it will mess up with my current missions, but tekto has lower gravity and thicker atmosphere, should be even easier.

i failed to bring it in kerbin's orbit, but that's not a required mission parameter.

rkL2I5s.png

the cargo bay is quite crowded with science and life support. You can see why i wanted to use the Mk2. the bigger rtg are from near future electrics, they provide adequate power. it has food and air for 100 days.

there is also an ion engine, one of those tanks has 350 kg of xenon. it's worth 700 m/s in orbit. it's too low a thrust to finish circularization, but i can use it for some orbital manuevers to rejoin the mothership. and the whole setup is less than 600 kg

ch8aSih.png

the rover arm is fully functional.

 

i'm not 100% happy about this plane. at best, it's a jack of all trades. but i reached the point where i mostly gave up on improving it. it is adequate on all mission requirements, and touching anything is likely to screw up some other characteristic.

 

As for Eve, i am coming to believe it's best to use a regular rocket. while this plane could work as a last stage, it weights 50 tons for 3000 m/s of vacuum deltaV. To bring this to eve's orbit, i'd need at least 2000 extra m/s, even assuming a good flight capability. and that would require a very heavy ascent stage. a smaller rocket for orbit, together with a small, lightweight, disposable plane for exploring the surface (i recently used one that was just 2.5 tons) is probably going to be cheaper.

It's quite the paradox: i started the project with the mk2 because i needed the heat tolerance for eve, i ended up with a functional project, and then i decided to not use it at eve after all. but i'm still keeping the design, especially the despised Mk2, for reasons of cargo bay and crew comfort.

I want to also use it as rover on tylo and slate (because i'm not carrying around 50 tons of spaceplane without trying to reuse it as much as possible); but that requires rover wheels, which have bad aerodinamics and bad heat tolerance. i can install them with eva construction before the aforementoned landings, and remove them afterwards.

 

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Answer this question...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...