Kerbodyne SSTO Division: Omnibus Thread

[eddiew]

Just by looking at your crafts, here is how I would expect them to fly with inexperienced piloting: they should reach 400-500m/s at around 24km and then the nose will pull down. If you attempt to pull the nose up, the nose will probably drift to the side. The plane might even flip.

Actually I design using FAR's stability window, aiming for stable flight at 10km at mach 1, 20km at mach 2, and 25km at mach 3. While I've definitely been having problems getting the nose up with this low TWR model, it doesn't start suffering yaw or roll instability until that awkward 25-30 zone where everything seems to waver

My problem with the earlier Boomerang revision isn't so much that the nose 'drops' or gets unstable, but more that it ends up bouncing along at its service ceiling. It doesn't have the lift to ascend into thinner air at its best speed, and it doesn't seem to have the thrust to go any faster at the current drag, so it goes up 2km then down 2km then up 2km then down 2km... all the while pointing in the same direction. Kicking the rockets in at this point just burns the fuel out before it makes it to orbit :/

You've largely described my flight pattern though - mach 1+ by 10km, mach 2+ by 20km, level out, gain speed until it doesn't want to go any faster, then pull up. I tend to kick the nuke in before turning the rapiers to rockets, since it has ~795 ISP by that point and the extra bit of thrust can get me another km or two of altitude on air. Seems no reason not to use it early, and absolutely, soon as the AP is within striking range of vacuum the rapiers shut off.

Basics to orbit:

* Get as high and as fast as you can before lighting the rockets. Ideally, you want to be over 30,000m and Mach 4 before you touch the oxidiser supplies.

...

* Less is more. Often the way to fix a design is to strip it back rather than bulk it up. See De Minimus, Karnifex and Alkahest (and Skylon...) for examples.

From a quick glance, the Boomerang could afford to lose some torque wheels (one is plenty) and probably has heavier wings than necessary: there's plenty of wing surface, so they aren't too heavily loaded, so they don't need to be super strong.

I did try cutting back on the wings, but it led to plummeting into the ocean at the end of the runway. Unless I beefed up the canards, which then destabilised it at higher altitudes and speeds :/ Overall this plane has been a real pain to work with. 5.1B does pretty well; it can hit mach 4 at 25km, but I do struggle to get anything beyond that point on air, even nursing the throttle. My best so far is with 2 rapiers on LFO and 2 on air at 33km... maybe I should have just turned 2 off instead of toggling to rocket mode?

I've got the mass/strength at 0.9 on the canards, 0.7 in the inner delta area, 0.6 for the outer deltas, and 0.5 on the tailfins - will take them all down low and see how it works out Will also narrow them in a bit once they're lighter, hopefully the gain in TWR and reduced drag will compensate for the loss of lift.

I definitely did overkill torque wheels on the Shortbow, but it was my first venture with vernors and I wasn't really sure how much torque I could lose. There's 1 reaction wheel inside the Boomerang, I'll strip them off the nacelles and cross my fingers.

Got a feeling I should get my design patterns away from delta wings too... they look awesome (IMHO), but it forces the CoM to be a long way back and I think sometimes I'm tripping up on lack of leverage and pitch authority. Not entirely sure those sharp swept edges are as effective as a wider, thinner wing, either... It would be nice to one day get a thing to orbit due to good engineering rather than massive numbers of engines xD

[Wanderfound]

Got a feeling I should get my design patterns away from delta wings too... they look awesome (IMHO), but it forces the CoM to be a long way back and I think sometimes I'm tripping up on lack of leverage and pitch authority. Not entirely sure those sharp swept edges are as effective as a wider, thinner wing, either... It would be nice to one day get a thing to orbit due to good engineering rather than massive numbers of engines xD

When I read this, I was on my way to add an addendum to my previous post, consisting of this:

Get away from deltas if you can. They look cool, and they're superficially uncomplicated, but the rearwards weight bias makes a delta design much trickier to get right than a conventional airframe. The mid-wing/rear-tailplane design is common for a reason: it works.

The only real difficulty with mid-wing designs is the tailstrike hazard, and that's easily dealt with via raising gear on hardpoints and adding tailstrike guards when necessary.

- - - Updated - - -

It might be useful to check how much of the altitude problem is design vs piloting, BTW. I had the Dolphin over 37,000m on jets alone, without a speed-killing zoom climb (it did lose a bit of speed after 30,000m, but not a lot); how high can you get it while still maintaining a reasonable amount of speed?

[eddiew]

I'll give the Dolphin a go after work tonight and find out

It'll give me a chance to see what magic you work that keeps turbojets working at altitude, too! Even with a ton of excess air intake, mine always seem to fizzle by 20km, reducing down to zero thrust by 22-23 :S That said, judging by the flight test, you're at mach 4 about 17km higher than my planes are at the same speed... I think I'm falling foul of the whole 'max thrust at 1000m/s' thing and need to be more aggressive on my ascent so's to have some punch left by 30km...

/me goes away to RTFM

[WhiteKnuckle]

It might be useful to check how much of the altitude problem is design vs piloting, BTW. I had the Dolphin over 37,000m on jets alone, without a speed-killing zoom climb (it did lose a bit of speed after 30,000m, but not a lot); how high can you get it while still maintaining a reasonable amount of speed?

By "jets alone" do you mean only your rear turbojet, or with the RAPIERS in airbreathing mode? Because shouldn't the turbojet be useless over mach 3.2?

[eddiew]

Ok, yeah, I have no idea how you get to mach 4 on a turbojet Mine conks out by 3.2, whatever my altitude, whatever my air intake...

That said, my best try got an apoapsis around 42km, which the rockets appreciated when I kicked them in. Certainly required a more aggressive hand than I usually take; mostly I'm aiming at a 20 degree ascent, where the Dolphin responded best around 30-35, and even steeper for the first 10km.

Sadly I can't really use this ascent profile on my 2 engine Boomerang, it just doesn't have the power to point at 30 degrees and go for gold. Still, I might see if I can turn the 4x rapier version into a 2x2 and get a bit more kick out of the deep atmosphere...

[blowfish]

Are you not using FAR? I don't see an icon anywhere on your screen. FAR changes how intakes work to an extent so that could be it.

[CrisK]

eddiew, You're barely going fast enough there. You need to angle downwards at 25-30km and build up more speed. Ideally you would be going twice as fast before you reach 40km. If you're moving that slow then you will have a very hard time.

[Wanderfound]

By "jets alone" do you mean only your rear turbojet, or with the RAPIERS in airbreathing mode? Because shouldn't the turbojet be useless over mach 3.2?

The standard flight pattern on something like the Dolphin is:

1) All jets on until the air runs low at somewhere over 20,000m.

2) Turbojet off, RAPIERs stay on in jet mode, until the air supply runs low again at about 25,000m.

3) Turbojet on, RAPIERs off, keep it like this until you've maxed speed and altitude.

4) RAPIERs back on in rocket mode, keep the Turbojet running until it's producing zero thrust.

5) Close intakes and shut down the Turbo while the RAPIERs finish off alone.

Turbojets (in FAR/KSP, anyway) keep producing thrust up to around Mach 5.

[eddiew]

Are you not using FAR? I don't see an icon anywhere on your screen. FAR changes how intakes work to an extent so that could be it.

Cropped the screenshot so's not to waste space. Been using NEAR/FAR since my first week of KSP, wouldn't fly without it

eddiew, You're barely going fast enough there. You need to angle downwards at 25-30km and build up more speed. Ideally you would be going twice as fast before you reach 40km. If you're moving that slow then you will have a very hard time.

But the turbojet dies around mach 3.2-3.5, regardless of altitude or intake air... I can get a 4x rapier plane up to a hefty pace, but mastery of turbojets is a mystery to me :/ If there's a magic button I don't know about, please do enlighten me, cos right now I'm flummoxed!

I will however start expanding my stability window and design for 30km altitudes as well, because I've obviously been underestimating how far a plane should be able to go on air. Most of mine kick to rockets at 25-27, which is costing rather a lot of fuel.

[Wanderfound]

eddiew, You're barely going fast enough there. You need to angle downwards at 25-30km and build up more speed. Ideally you would be going twice as fast before you reach 40km. If you're moving that slow then you will have a very hard time.

Yup.

Wings work better the faster you go. So do intakes.

My normal flight pattern is to race to the thin, low-drag air above 20,000m, but then not worry about climbing any higher until I'm approaching Mach 4. There'll normally be a bit of a descent while I build speed; I don't actually dive, but I normally keep the nose to about a 10° pitch relative to the horizon while accelerating. As you gain speed, the AoA required for level flight decreases, and the descent will turn back into a climb.

It might be educational to take one of the single-turbojet trainers up and see how high and fast you can get it.

- - - Updated - - -

But the turbojet dies around mach 3.2-3.5, regardless of altitude or intake air... I can get a 4x rapier plane up to a hefty pace, but mastery of turbojets is a mystery to me :/ If there's a magic button I don't know about, please do enlighten me, cos right now I'm flummoxed!

I will however start expanding my stability window and design for 30km altitudes as well, because I've obviously been underestimating how far a plane should be able to go on air. Most of mine kick to rockets at 25-27, which is costing rather a lot of fuel.

By "Turbojet dies", do you mean "the engine produces zero thrust" or "the plane stops accelerating"? The first shouldn't be the case, but the second is to be expected, especially on heavier/draggier ships.

And yeah, I tend to think of 30,000m as "edge of rocketry". Plenty of tailfin required; the air's thin up there.

Edited February 11, 2015 by Wanderfound

[WhiteKnuckle]

By "Turbojet dies", do you mean "the engine produces zero thrust" or "the plane stops accelerating"? The first shouldn't be the case, but the second is to be expected, especially on heavier/draggier ships.

Zero thrust. (at least that's what I'm talking about)

Regardless of altitude, intake air, or other aspects of ship design, a turbojet will start to loose power over 1km/s, and will be producing zero thrust by mach 3.2

This is listed as the max speed in the discription of the turbojets once FAR is loaded as far as I know (it's been a while since I've played without FAR so I'm not 100% sure)

Edit: Might be due to B9, I don't know why a parts mod pack would screw with stock parts but that was the first thing I came up with when doing a search... reading more now...

Edited February 11, 2015 by WhiteKnuckle

[blowfish]

Right - B9 adjusts the thrust curves of the stock jets to be ... slightly more inline with real life.

[WhiteKnuckle]

Right - B9 adjusts the thrust curves of the stock jets to be ... slightly more inline with real life.

Yeah just found that out.

Goodbye B9, we had some fun times, but I want my turbojets back.

[blowfish]

The changes are isolated in a single CFG, so you can just remove it if need be.

[CrisK]

For what it's worth, I just flew the Dolphin to orbit with the latest version of FAR installed. I took the quickest, least efficient path available: angle 20% up and hold. It made it to orbit with fuel to spare and my speed was 2241m/s. It would have been significantly faster had I taken a proper 5-15% shallow flight path. This is a very, very easy SSTO to fly (nice work Wander!) so there's something strange going on if you run out of thrust.

[eddiew]

Right - B9 adjusts the thrust curves of the stock jets to be ... slightly more inline with real life.

Seriously??

...I've been playing the prat version of Rimmer for 4 years

Right. I'm guessing that's the B9_Aerospace-Squad_Jet_Balance.cfg file? Deleted. With extreme prejudice, malice, and satisfaction. Thank you for letting us know, blowfish! That's quite annoying to discover, but it's better to be aware than not. If B9 is going to do that, it should also alter the description to append "nerfed by B9" so it's clear that stock+FAR spaceplanes are no longer valid.

Under the circumstances I would like to

Not as much horizontal speed as Wanderfound, I grant, but the altitude and upward momentum are still pretty respectable I think

And now I must redesign every spaceplane I ever made that has more than 2 rapiers :/

[Hodo]

Right - B9 adjusts the thrust curves of the stock jets to be ... slightly more inline with real life.

That is actually FAR.

B9 only adjusts the engines in its mod.

[Wanderfound]

That is actually FAR.

B9 only adjusts the engines in its mod.

To clarify for the crowd: FAR nerfs jets to about half their stock thrust, and adjusts the speed curve so that jets cut out around Mach 5. B9 does some other stuff, the details of which I'm not sure of. But if your jets are cutting out at Mach 3, it ain't FAR that's doing it.

[Hodo]

To clarify for the crowd: FAR nerfs jets to about half their stock thrust, and adjusts the speed curve so that jets cut out around Mach 5. B9 does some other stuff, the details of which I'm not sure of. But if your jets are cutting out at Mach 3, it ain't FAR that's doing it.

Yeah B9 has a line in its cfg....

Nerf Squad jets for great justice. That is the line. It reduces the power curve down from dying at mach 5 to losing power by mach 3.2-3.4.

It isn't a bad cfg addition, it brings the turbojets inline with the B9 jet engines. And honestly it is pretty easy to still get to orbit with Turbojets and a small rocket engine.

[eddiew]

It isn't a bad cfg addition, it brings the turbojets inline with the B9 jet engines. And honestly it is pretty easy to still get to orbit with Turbojets and a small rocket engine.

It's bad if you aren't aware of it and are wondering why stock+FAR craft simply don't work properly for you

In the light of this irksome revelation, and now having turbojets that are on a par with everyone else's, I'm pleased to say that the Boomerang MK6 passed her test flight with flying colours. The R&D boys took a serious line about improving the 5.1B and tore the wings clean off to start again. Thank you guys for all your advice; this is now arguably the most stable thing in my fleet. Piloted badly and switching to rockets at 26km and mach 3, this bird left me with 700 delta-v in LKO, which is more than enough to rendezvous with a fuel tanker or space station. She has more lift than I expected and I'll try to make a better job of her next ascent, because I reckon she could reach Munar orbit with a gentler hand.

And check out that wCoM and dCoM! Less than 2 kNm of torque on RCS when vernors are turned off

Also for reasons unknown, removing the B9 config made my clock green... this design was previously down to 30% time speed, now it flies at 96-99% O_o

Edited February 16, 2015 by eddiew

[CrisK]

Seriously??

...I've been playing the prat version of Rimmer for 4 years

Right. I'm guessing that's the B9_Aerospace-Squad_Jet_Balance.cfg file? Deleted. With extreme prejudice, malice, and satisfaction. Thank you for letting us know, blowfish! That's quite annoying to discover, but it's better to be aware than not. If B9 is going to do that, it should also alter the description to append "nerfed by B9" so it's clear that stock+FAR spaceplanes are no longer valid.

Under the circumstances I would like to

http://foxwoodstudios.co.uk/kerbal/spaceplanes/kerbodyne-dolphin-2.png

Not as much horizontal speed as Wanderfound, I grant, but the altitude and upward momentum are still pretty respectable I think

And now I must redesign every spaceplane I ever made that has more than 2 rapiers :/

Eddie, I don't want to come across as constantly negative or critical, but this is not a good flight. The reason why I (and Wander, and Hodo) maintain a horizontal profile for so long (aka: we target the horizon or slightly above it) is so that we build up a great deal of speed, AND while doing so we raise our periapsis to 200,000 or less.

Once your speed is sufficient and your periapsis is around 200,000 then you can gain altitude, but only gradually. As you burn at a 10-15% angle you raise both your apoapsis AND your periapsis so that by the time you're over 80,000 your periapsis is over 75,000. You've achieved orbit without having to stop, aim sideways, and waste fuel on an orbital burn.

In your case you may still be able to get to orbit, but it won't be an efficient path.

Sorry for the negativity, and good work on getting up past the 40k barrier!

Edit: The revised ship looks far better. That plane should be able to take you interplanetary.

[Wanderfound]

Eddie, I don't want to come across as constantly negative or critical, but this is not a good flight. The reason why I (and Wander, and Hodo) maintain a horizontal profile for so long (aka: we target the horizon or slightly above it) is so that we build up a great deal of speed, AND while doing so we raise our periapsis to 200,000 or less.

Once your speed is sufficient and your periapsis is around 200,000 then you can gain altitude, but only gradually. As you burn at a 10-15% angle you raise both your apoapsis AND your periapsis so that by the time you're over 80,000 your periapsis is over 75,000. You've achieved orbit without having to stop, aim sideways, and waste fuel on an orbital burn.

In your case you may still be able to get to orbit, but it won't be an efficient path.

Sorry for the negativity, and good work on getting up past the 40k barrier!

Edit: The revised ship looks far better. That plane should be able to take you interplanetary.

In fairness, that shot was in response to a max altitude "challenge", rather than a presentation of the ideal way to orbit. And the enhanced skin drag of the latest version of FAR does tip the balance towards steeper profiles. The ultra-flat ascents that I used to use (crank it up to Mach 6 at 30,000m with the nose on the horizon) don't work anymore: if you shut off your engines at 30,000m these days, you'll lose all of your speed to drag long before you reach orbit.

You still want the flat jet climb between 20,000 and 30,000m to maximise speed, but you also want to pull the nose up as hard as you can as soon as you light the rockets so as to be out of the drag zone when you shut them down. A bit of a zoom climb shortly pre-rocketry can make the transition easier and ensure that you get high enough to dodge the drag.

[eddiew]

Eddie, I don't want to come across as constantly negative or critical

...

Edit: The revised ship looks far better. That plane should be able to take you interplanetary.

No no, that's completely fine and fair. I was aware it wasn't the best ascent profile, but I was feeling exuberant that I'd found out about that B9 nerf to turbojets and that removing it immediately gave me 5km more altitude even after a couple of strong English beers Now that I'm back on an equal playing field I have the opportunity to reassess the build of all my 3+ engine planes and learn to pilot them again under the new rules. Quite exciting really!

And cheers I'm not sure what the range on it is, if I top it up in orbit, though I'm sure it'll do for ferrying out to Mun/Minmus for those 5-kerbal station contracts

...

You still want the flat jet climb between 20,000 and 30,000m to maximise speed, but you also want to pull the nose up as hard as you can as soon as you light the rockets so as to be out of the drag zone when you shut them down. A bit of a zoom climb shortly pre-rocketry can make the transition easier and ensure that you get high enough to dodge the drag.

Good advice, thank you Biggest problem I'm finding is that making a design that's stable all the way to 30km can be time consuming. 20km @ mach 2 is a doddle compared to 30km @ mach 3, or even 4. Only takes a small tweak like an extra intake and suddenly a working design gets all kinds of squirrelly :S

That and neck-bend at 4x timewarp, but I suspect that's a symptom of overly robust canards xD I do love those procedural wings/control surfaces, but there's probably a reason stock canards are relatively timid.

Edited February 12, 2015 by eddiew

[Wanderfound]

Kerbodyne Gunbalanya. Express crew delivery and retrieval.

Flight test at http://s1378.photobucket.com/user/craigmotbey/Kerbal/Beta/Kerbodyne%20Showroom/Gunbalanya/story

Alternate format at http://s1378.photobucket.com/user/craigmotbey/slideshow/Kerbal/Beta/Kerbodyne%20Showroom/Gunbalanya

Craft file at https://www.dropbox.com/s/kzvxb2ge8yetrmx/Kerbodyne%20Gunbalanya.craft?dl=0

[Wanderfound]

Good advice, thank you Biggest problem I'm finding is that making a design that's stable all the way to 30km can be time consuming. 20km @ mach 2 is a doddle compared to 30km @ mach 3, or even 4. Only takes a small tweak like an extra intake and suddenly a working design gets all kinds of squirrelly :S

That and neck-bend at 4x timewarp, but I suspect that's a symptom of overly robust canards xD I do love those procedural wings/control surfaces, but there's probably a reason stock canards are relatively timid.

Most things will develop some yaw instability over 30,000m unless they're doing Mach 5 or better. This is why I'm so fond of sticking a Vernor each side of the cockpit...

You may notice that a lot of my recent stuff has fairly low powered canards (either by lowering the max deflection directly, or cutting the pitch influence if I want large deflection for the AoA settings). The FAR visualisation thingie is good for tuning control surfaces; set it to highlight stalls, take it for a test flight and then detune whichever surfaces are stalling. Multiple low-powered pitch surfaces work better than a single high-powered surface; they're less prone to stalls.