Kerbodyne SSTO Division: Omnibus Thread

[eddiew]

Oh nice, I didn't know that was back up. Clearly I shall have to start making more colourful builds ^^ *And* I've just realised that the KER hud is actually movable...

(Side note; argh! First time in 10 years I've had more than one game I wanted to play, and I don't have time for all of them! KSP, Cities Skylines, TerraFirmaCraft and Pixelmon are all calling me...)

Edited May 27, 2015 by eddiew

[lordcirth]

Just a tip for anyone else who's been trying to ages to make an SSTO with a crew cabin for tourists: Forget the 2100kg cockpit, put a probe core & reaction wheels in the 1/2 cargo bay you'll want anyway. Same size as an in-line cockpit, but 250+100+100=450kg total. It's the only way I've found so far to make good passenger SSTO's.

[Wanderfound]

Passenger SSTOs aren't all that tricky:

But a probe core in the cargo bay is a good idea in general, anyway. As well as providing options in "ran out of jetpack fuel on EVA" type emergencies, they also allow you to dispense with specialist pilots. In my career game, Jeb's been sitting at home ever since I unlocked the first stability-equipped core. Scientists are just too useful not to bring one along whenever possible.

OTOH, my usual way of building a small passenger spaceplane is to just swap out the cargo bay, replace it with a passenger cabin and find somewhere else to stick the SAS/batteries etc. Most of the small/medium Mk2 ships here (e.g. http://forum.kerbalspaceprogram.com/threads/90747-Kerbodyne-SSTO-Division-Omnibus-Thread?p=1931901&viewfull=1#post1931901) will work as tourist taxis with minimal modification.

Edited May 27, 2015 by Wanderfound

[lordcirth]

I guess I wasn't too clear about what I meant. I'm playing in Career mode, so Rapier's aren't an option for a while. I'm mostly using Turbos, and Thuds or LV-45 Swivels for now. And I was looking for 4-person or so tourist shuttles that are fast and easy to fly to Kerbin orbit, farther can wait for now. I have a backlog of early tourist contracts that I took while I didn't have turbos, and I figured it would be a good intro to nuFAR spaceplanes.

I'm working on one right now that's shaping up, but still having my usual problems with unstable reentry. Wave drag is pretty low, but it's barely stable in pitch. It seems nuFAR pitch stability is no longer simply a matter of CoL behind CoM. On one draft, adding canards apparently helped stability! I still don't understand exactly how that works now. Also, I ran into a bug, an interaction between FAR and Trajectories - hit map and your wings fall off, even if you're in orbit or aero failures are off. So I have to reenter without Trajectories. I came in on Hold Prograde, it looked like I was still in space, then my fins got heat bars. Pulling up 1 deg flipped me.

The spaceplane:

http://steamcommunity.com/sharedfiles/filedetails/?id=450412999

http://steamcommunity.com/sharedfiles/filedetails/?id=450414313

https://www.dropbox.com/s/jfhp1waygftlg0e/probe%20orbiter.craft?dl=0

[blowfish]

I'm working on one right now that's shaping up, but still having my usual problems with unstable reentry. Wave drag is pretty low, but it's barely stable in pitch. It seems nuFAR pitch stability is no longer simply a matter of CoL behind CoM. On one draft, adding canards apparently helped stability! I still don't understand exactly how that works now.

Use static analysis rather than CoL (which is virtually meaningless in FAR). Sweep AoA at a few different mach values, the largest you need should be about mach 7 (Kerbin re-entry). Make sure the Cm line is sloping down or at least flat at all mach numbers, otherwise you will flip. Give canards negative and elevators positive AoA deflection - that will help with stability a lot.

[lordcirth]

Thanks for the tips, I've been primarily using Stability Derivatives and making the numbers green in common scenarios. I'm trying static analysis now, Cm is great at Mach 1 and steadily flattens out until it's almost flat at Mach 4. Setting elevators to AoA +10 makes a very small difference. Moving the wings back one offset notch as well gives a barely stable graph:

The plane is a bit short on pitch authority, I guess that's the problem. It does not have canards, and I can't really add them because the wings are as far back as they can go, as the screenshots show.

I replaced the AV-T1 Winglets at the tail with AV-R8 Winglets for more control surface, and added vertical AV-T1's on the wingtips to add vertical stabilizer so I could rotate the AV-R8's flatter:

Mach 7, as you suggested:

This didn't help the area ruling though, it went from 1.04 to 1.58 m sq. Will that make a big difference? the turbo has a lot of thrust between mach 0.7 - mach 2, so it might be well worth the drag.

[blowfish]

Thanks for the tips, I've been primarily using Stability Derivatives and making the numbers green in common scenarios. I'm trying static analysis now, Cm is great at Mach 1 and steadily flattens out until it's almost flat at Mach 4. Setting elevators to AoA +10 makes a very small difference. Moving the wings back one offset notch as well gives a barely stable graph:

http://images.akamai.steamusercontent.com/ugc/47631932794754148/76D5253F02F63EE25B819288FF37536A50604599/

The plane is a bit short on pitch authority, I guess that's the problem. It does not have canards, and I can't really add them because the wings are as far back as they can go, as the screenshots show.

I replaced the AV-T1 Winglets at the tail with AV-R8 Winglets for more control surface, and added vertical AV-T1's on the wingtips to add vertical stabilizer so I could rotate the AV-R8's flatter:

http://images.akamai.steamusercontent.com/ugc/47631932794791465/6B3E7978C05C522629D01A47558A2386AE34D0B1/

http://images.akamai.steamusercontent.com/ugc/47631932794804160/249DFF68A8CE265FFFDE6DF270685C70F396A76B/

Mach 7, as you suggested:

http://images.akamai.steamusercontent.com/ugc/47631932794839234/00066D34F0C33D674C0CDDFDB84753F48EF60C04/

This didn't help the area ruling though, it went from 1.04 to 1.58 m sq. Will that make a big difference? the turbo has a lot of thrust between mach 0.7 - mach 2, so it might be well worth the drag.

Okay, let's see if we can sort this out.

One of the first things I notice is that your CoM is shifting very far back as the tanks drain. I'm not sure there's much to be done about it given the design, but this will make the craft much less stable when it is empty (i.e. on re-entry), and AFAIK FAR uses the wet configuration to calculate everything (there used to be a toggle for this but I guess it went away).

The control surfaces on the main wings are quite small (probably why you don't have that much pitch authority). Do you have any bigger ones unlocked at this point in the game?

+10 isn't very much on the pitching surfaces. I usually go with +/- 100.

Hope some of that helps...

[lordcirth]

Okay, let's see if we can sort this out.

One of the first things I notice is that your CoM is shifting very far back as the tanks drain. I'm not sure there's much to be done about it given the design, but this will make the craft much less stable when it is empty (i.e. on re-entry), and AFAIK FAR uses the wet configuration to calculate everything (there used to be a toggle for this but I guess it went away).

The control surfaces on the main wings are quite small (probably why you don't have that much pitch authority). Do you have any bigger ones unlocked at this point in the game?

+10 isn't very much on the pitching surfaces. I usually go with +/- 100.

Hope some of that helps...

Not sure what to do about the CoM/dCoM, the whole point of the design was to have as little dry mass as possible, which makes it hard to balance. I tried +100 AoA, helped a little. I replaced the Elevon 1's with 3's for more surface, it looks a little odd but whatever. I set them to +100 AoA as well. Btw, through all revisions the inner elevons have been pitch, and the outer pitch+roll. Is that correct?

The graph is much more stable now at mach 7:

The other problem I'm having is during ascent, as the turbo loses power I lose control. It seems to be on all axes at once, but it could be that one is causing the others. I'm not sure if it's the wrong ascent profile, kicking the rockets in too early/late, or bad airframe. I get a temperature bar on the nose intake just before it starts to founder.

EDIT: It's always between 25km-30km, with speed anywhere from 900m/s to 1300m/s. The faster I go the higher I generally get before losing control - do I just need more intakes to power through it? I'm aiming fairly steep because I'm on rockets by that time and my nose intake is on fire. Stability Derivatives at 30km don't go green before mach 5. TWR on the twin Thuds is only 1.02 at 25km - this plane is heavier than I'd like. The wings are already tuned down to 0.7, which can survive 80kPa last I tested.

Edited May 28, 2015 by lordcirth

[blowfish]

Your control surface setup is correct. Figure out what the flight condition is when you loose control (mach and altitude). Calculate stability derivatives, and chances are you will see some red. Most likely dynamic pressure is just dropping to the point where your stabilizing surfaces are no longer effective. If this is the case, there are a couple of things you can do:

1) Maintain higher dynamic pressure. This means flying lower, which means more heat. You might be able to mitigate that by replacing the intake at the front with a nose cone and placing several structural intakes on the top of the fuselage - intakes are usually the first things to overheat.

2) Increase your vertical stabilization - replace your three vertical stabilizing surfaces with a single large one, as far back as you can put it. The ones at the ends of the wings probably aren't doing you much good anyway - they're too close to the CoM.

[lordcirth]

Your control surface setup is correct. Figure out what the flight condition is when you loose control (mach and altitude). Calculate stability derivatives, and chances are you will see some red. Most likely dynamic pressure is just dropping to the point where your stabilizing surfaces are no longer effective. If this is the case, there are a couple of things you can do:

1) Maintain higher dynamic pressure. This means flying lower, which means more heat. You might be able to mitigate that by replacing the intake at the front with a nose cone and placing several structural intakes on the top of the fuselage - intakes are usually the first things to overheat.

2) Increase your vertical stabilization - replace your three vertical stabilizing surfaces with a single large one, as far back as you can put it. The ones at the ends of the wings probably aren't doing you much good anyway - they're too close to the CoM.

I had to put a ridiculously large tailfin to get green NB at 25km/mach 3-4, which is where the problem is. I'm not sure this is practical - it actually meshes with the engines and doesn't increase wave drag, but it must have a lot of normal drag.

Removing the wingtips and replacing the fins with a Structural Wing Type A did help, but it's not enough. It could be that the tail just isn't far enough from CoM for this design to work. If I had more Dv to work with, I'd add a docking port or something at the front, but 2km/s is low enough already with a low TWR. What's the lowest rocket TWR that is efficient? Most of my designs in 0.90 aimed for 1.2-1.4 like rockets.

Edited May 28, 2015 by lordcirth

[Wanderfound]

I guess I wasn't too clear about what I meant. I'm playing in Career mode, so Rapier's aren't an option for a while. I'm mostly using Turbos, and Thuds or LV-45 Swivels for now. And I was looking for 4-person or so tourist shuttles that are fast and easy to fly to Kerbin orbit, farther can wait for now. I have a backlog of early tourist contracts that I took while I didn't have turbos, and I figured it would be a good intro to nuFAR spaceplanes.

I'm working on one right now that's shaping up, but still having my usual problems with unstable reentry. Wave drag is pretty low, but it's barely stable in pitch. It seems nuFAR pitch stability is no longer simply a matter of CoL behind CoM. On one draft, adding canards apparently helped stability! I still don't understand exactly how that works now. Also, I ran into a bug, an interaction between FAR and Trajectories - hit map and your wings fall off, even if you're in orbit or aero failures are off. So I have to reenter without Trajectories. I came in on Hold Prograde, it looked like I was still in space, then my fins got heat bars. Pulling up 1 deg flipped me.

The spaceplane:

http://steamcommunity.com/sharedfiles/filedetails/?id=450412999

http://steamcommunity.com/sharedfiles/filedetails/?id=450414313

https://www.dropbox.com/s/jfhp1waygftlg0e/probe%20orbiter.craft?dl=0

I've been finding that while single-fuselage ships are easy to streamline and will get to orbit, its difficult to get one there with more than a couple of hundred m/s ÃŽâ€V in the tanks (which is a thin enough margin that an imperfect ascent profile might put it in danger of failing to reach orbit). With fuselage plus outrigger tanks (e.g. like the Tradizione), there's a lot more gas in reserve.

NuFAR has made deltas a bit tricky; the delta wings can do nasty things to stability that aren't always picked up by the static analysis screens. The rearwards weight bias of a delta design gives extra leverage to the forces trying to flip you and handicaps the ability of your rear-mounted surfaces to compensate. Conventional mid-wing aircraft are much easier to stabilise.

Getting stability right is a higher priority than minimising drag. A turbojet has enough TWR to punch a brick through transonic if necessary. I'll try to throw together a low-tech passenger ship as a demonstrator later today.

[Captain Sierra]

A turbojet has enough TWR to punch a brick through transonic if necessary.

I might sig quote this.

[eddiew]

A turbojet has enough TWR to punch a brick through transonic if necessary.

I might sig quote this.

Sometimes you don't even need a turbojet to make a brick fly...

(Admittedly this was in stock aero, but if anything, that's more punishing than nuFAR!)

...I'm a bit tempted to see if I can make it work with FAR now...

[blowfish]

The turbojet and RAPIER have about the same thrust at Mach 1 EDIT: not quite: The turbojet has about 20% more thrust at Mach 1. The RAPIER has lower static thrust, but it grows much faster and only starts to fall off at higher mach numbers.

Edited May 28, 2015 by blowfish

[SHiftER2O]

Sometimes you don't even need a turbojet to make a brick fly...

...I'm a bit tempted to see if I can make it work with FAR now...

http://i.imgur.com/uZnWxUA.jpg

(Admittedly this was in stock aero, but if anything, that's more punishing than nuFAR!)

That is.... interesting.

[eddiew]

The turbojet and RAPIER have about the same thrust at Mach 1. The RAPIER has lower static thrust, but it grows much faster and only starts to fall off at higher mach numbers.

That... just doesn't match what I see in the right-click menu though... (unless something has changed recently?)

At mach 0.9 and 10km, turbojets are usually producing about 170-180kN, where a rapier will be around 125-135kN (I've done this flight pattern a lot ). Rapiers aren't really hitting their power band until about mach 1.5 - although I freely accept they overtake turbos at high mach and have a higher top speed.

- - - Updated - - -

That is.... interesting.

I like to call out people who moan that new stock aero is too realistic and not allowing any creativity

[blowfish]

That... just doesn't match what I see in the right-click menu though... (unless something has changed recently?)

At mach 0.9 and 10km, turbojets are usually producing about 170-180kN, where a rapier will be around 125-135kN (I've done this flight pattern a lot ). Rapiers aren't really hitting their power band until about mach 1.5 - although I freely accept they overtake turbos at high mach and have a higher top speed.

- - - Updated - - -

I like to call out people who moan that new stock aero is too realistic and not allowing any creativity

Actually, you're right. At sea level and mach 1, the Turbojet should produce about 432 kN, with the RAPIER producing 364 kN (about 20% more for the turbojet). The density falloff is about the same in the lower atmosphere so just scale those numbers...

[SHiftER2O]

I like to call out people who moan that new stock aero is too realistic and not allowing any creativity

bu-but I'm not moaning D:

[Wanderfound]

Here y'go:

Kerbodyne Astrotaxi. High tech not required.

Javascript is disabled. View full album

Craft file at https://www.dropbox.com/s/y7725e0gwgbbvu0/Kerbodyne%20Astrotaxi.craft?dl=0

All of the parts involved are from fairly early in the tech tree, the control authority is mild enough that it isn't too hard to handle (drag chutes for short-runway landings, too), and there's room in the service bay for any minor gadgetry you want to bring along. I deliberately left the streamlining a bit rough, to demonstrate that hyper-slickness isn't absolutely necessary. If you've got 'em unlocked, a pair of airbrakes at the back near the solar and chutes would be a worthwhile optional addition, too.

Edited May 28, 2015 by Wanderfound

[Wanderfound]

Of relevance to all airplane fans: http://forum.kerbalspaceprogram.com/threads/82785-1-0-2-Kerbin-Side-v0-41-2-Kerbin-SideJobs-v1-2?p=1978065&viewfull=1#post1978065

- - - Updated - - -

I had to put a ridiculously large tailfin to get green NB at 25km/mach 3-4, which is where the problem is. I'm not sure this is practical - it actually meshes with the engines and doesn't increase wave drag, but it must have a lot of normal drag.

http://images.akamai.steamusercontent.com/ugc/47631932795414975/A59D53A59411687DC432FAB2994247AA11D5417F/

That's actually not that big a tailfin for a delta, BTW. The size of the fin required is proportional to the distance between it and CoM, so ships with a rearward weight bias tend to need either paired fins or a single jumbo one.

- - - Updated - - -

Actually, you're right. At sea level and mach 1, the Turbojet should produce about 432 kN, with the RAPIER producing 364 kN (about 20% more for the turbojet). The density falloff is about the same in the lower atmosphere so just scale those numbers...

The Turbo is a clear winner if you're staying in the atmosphere, but RAPIERs are the choice for space; as it should be.

The "use a turbo to get over the RAPIER flat spot" works in FAR as well as stock, but a decently streamlined multi-engine FAR ship normally doesn't have that much trouble busting through transonic, anyway.

Apart from low-tech builds, the only time recently that I've found a need for turbojets on an SSTO was for Mk3 monsters that just couldn't get up to speed before the end of the runway without the extra thrust. Thanks to the more realistic atmospheric thrust scaling for rockets, my old trick of "use the LV-Ns as takeoff boosters" isn't really practical anymore.

OTOH, a few RT-5/decoupler combinations bolted onto the back of your engines will get just about anything up to speed in a hurry.

[eddiew]

Actually, you're right. At sea level and mach 1, the Turbojet should produce about 432 kN, with the RAPIER producing 364 kN (about 20% more for the turbojet). The density falloff is about the same in the lower atmosphere so just scale those numbers...

I'd

, but yeah, basically at subsonic speeds, rapiers are a bit sucky. It's why we had a few weeks of people struggling with spaceplanes that couldn't break mach 1 - and why a lot of planes now being posted to these forums have more engines than we're used to seeing.

And as Wanderfound has mentioned elsewhere - stock aero hugely punishes any change of direction. Attempting a dive to gain speed generally loses it all again when you start to pull up. I won't risk any kind of manoeuvre until I'm seeing rapiers delivering over 180kN.

Constantly torn between the soft landings of stock, and the more sensible ascents of nuFAR...

bu-but I'm not moaning D:

Lol, nor was that ship targeted at you - t'was a fairly early 1.0.2 build while people were still complaining that 'nothing' would fly anymore

[Wanderfound]

Whee!

[lordcirth]

Here y'go:

Kerbodyne Astrotaxi. High tech not required.

All of the parts involved are from fairly early in the tech tree, the control authority is mild enough that it isn't too hard to handle (drag chutes for short-runway landings, too), and there's room in the service bay for any minor gadgetry you want to bring along. I deliberately left the streamlining a bit rough, to demonstrate that hyper-slickness isn't absolutely necessary. If you've got 'em unlocked, a pair of airbrakes at the back near the solar and chutes would be a worthwhile optional addition, too.

Thanks so much to both of you, I've learned a lot. It seems that outriggers are indeed the way to go for anything with a payload, even a 2t crew cabin. Transonic drag with the Astrotaxi maxes out around 450kN, which is huge, but each turbo is doing 440kN at that point, so like you said, not a problem. I tried some outrigger designs, but I was using 2x Lv45 & 1 turbo, partly because that lets you ride turbo till flameout and partly because the turbo seems so powerful now. But with the transonic drag of outriggers, it seems that 2 turbos works better. I notice the Astrotaxi only has ~0.8 TWR on the LV-45, which I thought would be way too low, but apparently I'm wrong. My 1-jet 2-rocket craft would struggle to 25km then burn out all their fuel getting 70km Ap, since I thought you needed >1 TWR for the rocket phase. More rocket TWR does increase efficiency to some degree, right?

Also, I'm still not good at reentry - what's the rule of thumb for nuFAR? Still retro opposite from KSC, 30km Ap?

EDIT: I think that's working ok, but I can't keep it steady on reentry - if I Hold Prograde, I dive too low, get overheat bars, and lose control. But I'm not sure how to keep it on course at a given AoA with just SAS. I have Pilot Assistant to tweak SAS, using the numbers you posted a while ago that worked great in 0.90. I had similar problems in 0.90 so it's probably just piloting.

Edited May 28, 2015 by lordcirth

[FleshJeb]

I like to call out people who moan that new stock aero is too realistic and not allowing any creativity

http://i.imgur.com/uZnWxUA.jpg

You may have just changed my mind. After a few hours of fiddling, I had abandoned planes altogether.

[Wanderfound]

Reentry: keep the nose 5 to 15° above Surface Prograde, and hold it there until you've levelled off. From then on, slowly descend further while keeping an eye on your heat; if it gets too hot, halt the descent (and maybe climb a little) until you've cooled off / slowed down a bit. You can get away with a higher AoA during the early stages of reentry, but pull it down as soon as you start to feel the wind.

A 20km periapsis is a good conservative height, but once you get the hang of it you can come in steeper. I often set my periapsis to intersect the ground on the western edge of the KSC continent.

Airbrakes, flaps, spoilers and S-turns will all help to ditch speed, but none of them are absolutely necessary.

Demonstrations:

Javascript is disabled. View full album

Javascript is disabled. View full album

BTW: Astrotaxi would also work well with the forward section replaced by a Mk2 cockpit, docking port and short LFO tank. Adds slightly nicer aesthetics and refuellability, plus your tourists would no longer need to spacewalk to the orbiting resort.

Edited May 29, 2015 by Wanderfound