Why don't my SSTOs work?

[katateochi]

16 hours ago, Rune said:

Sleek, sleek, sleek. For a winged SSTO, drag is your single worst enemy. Things that create lots of drag: unused nodes, not-pointy parts sticking out of cargo bays, intakes. Have as few intakes as you can get away with, as few nosecones, and absolutely no uncapped nodes.

This is the truth right here! 
As an example, the other month I'd come up with this really cool looking design for a set of engines for an SSTO I was hoping would be able to take an orange tank to orbit. It was a complete fail, such a fail that it couldn't even get up to the magic 400m/s.  But then I was poking around on the post your SSTO pics thread and saw a design that was almost exactly the same (at least visually) to the one I'd just made, and it was merrily hauling an orange tank to orbit no problem (and in fact by doubling up on engine sets he took it to crazy levels and was taking 4 and then 6 orange tanks to orbit! this post here). The engine setup was so similar to mine, but his worked perfectly and mine, well mine was floating in bits on the ocean. So I got his craft and picked it apart and realised that I'd left the rear connection node on the main body of the engine uncapped and I'd used a more draggy method to attach the engines and intakes. That was it. Reduced the draggy attachments and made sure all nodes where capped and now it's happily hauling 40ton payloads to orbit. #dragkills

 

[Rune]

3 hours ago, Kryten 2X4B 523P said:

If you really want to go through the process, start with a single rapier powered craft, keep it minimal and when youre successful, work out the ratios of lift  / mass & fuel per rapier, and go from there.

That is very sciency of you. You'll get far in this game thinking that way!

1 hour ago, Vanamonde said:

I think my issue is more a matter of piloting ascent profiles than craft building, but yes, I only wanted to make planes that can actually do something useful, not just go up and come back down.  

Yup, that is mighty important. But also mighty simple, these days: fly as close to the ground as you dare until you hit the magical 400m/s, the ascend as shallowly as you can without burning up (very low TWRs can't actually burn up, so you just leave SAS on and let it increase pitch on its own as kerbin curves under you). That is a big change from the days of the souposphere and the slow climb to 10,000m, but once you make it flying SSTOs is actually easier than rockets.

57 minutes ago, Jarin said:

No longer true for intakes. Data from testing here shows that shock cones and other intakes are among the least draggy nose components; regardless of open or closed. (for a while the shock cone was apparently better than anything else, but it's now on par with other nosecones)

Not that airhogging does you any good anymore, so there's no reason to spam intakes or anything. But they're not something that needs to be avoided, either.

Yeah, not exactly what I meant there. Engines now need very little in the way of intakes, so minimizing your intake area cuts into your overall cross-sectional area. Take the Spatha, for example (first pic, the one with four RAPIERs, a small bay, and 20mT+4 kerbals capacity). The four RAPIERs are attached to one precooler each, then the precoolers to each of the four nodes of the Mk3 engine mount (then I moved them around to make them look cool, which took a lot of finagling but is besides the point, you could do it better and easier with a quadcoupler). Long story short, I have four engines happily fed, and my cross section is a single Mk3 tank. If I wanted to do the same thing with one shock cone per RAPIER, I would have a cross section of one Mk3 tank, because of the bay... and four 1.25 tanks, one for each intake. So if you do use engine pods, just one shock cone for each two RAPIERs will be enough, and thus you can halve your frontal intake area. Never mind what I did on the Calymore, that was mostly aesthetic reasons.
 

Rune. Precoolers FTW, basically, there's a reason they are in the final tech node.

Edited December 21, 2016 by Rune

[Jarin]

20 minutes ago, Rune said:

Yeah, not exactly what I meant there. Engines now need very little in the way of intakes, so minimizing your intake area cuts into your overall cross-sectional area. Take the Spatha, for example (first pic, the one with four RAPIERs, a small bay, and 20mT+4 kerbals capacity). The four RAPIERs are attached to one precooler each, then the precoolers to each of the four nodes of the Mk3 engine mount (then I moved them around to make them look cool, which took a lot of finagling but is besides the point, you could do it better and easier with a quadcoupler). Long story short, I have four engines happily fed, and my cross section is a single Mk3 tank. If I wanted to do the same thing with one shock cone per RAPIER, I would have a cross section of one Mk3 tank, because of the bay... and four 1.25 tanks, one for each intake. So if you do use engine pods, just one shock cone for each two RAPIERs will be enough, and thus you can halve your frontal intake area. Never mind what I did on the Calymore, that was mostly aesthetic reasons.
 

Rune. Precoolers FTW, basically, there's a reason they are in the final tech node.

Okay, gotcha. That reduces my comment to a mere nitpick. We want to minimize all forward-facing nodes, regardless of cap. The way you talked about it before, I thought you were implying that using nose cones rather than intakes was preferable. I'm totally onboard with minimizing frontal drag profile.

[sardia]

25 minutes ago, Rune said:

That is very sciency of you. You'll get far in this game thinking that way!

Yup, that is mighty important. But also mighty simple, these days: fly as close to the ground as you dare until you hit the magical 400m/s, the ascend as shallowly as you can without burning up (very low TWRs can't actually burn up, so you just leave SAS on and let it increase pitch on its own as kerbin curves under you). That is a big change from the days of the souposphere and the slow climb to 10,000m, but once you make it flying SSTOs is actually easier than rockets.

Yeah, not exactly what I meant there. Engines now need very little in the way of intakes, so minimizing your intake area cuts into your overall cross-sectional area. Take the Spatha, for example (first pic, the one with four RAPIERs, a small bay, and 20mT+4 kerbals capacity). The four RAPIERs are attached to one precooler each, then the precoolers to each of the four nodes of the Mk3 engine mount (then I moved them around to make them look cool, which took a lot of finagling but is besides the point, you could do it better and easier with a quadcoupler). Long story short, I have four engines happily fed, and my cross section is a single Mk3 tank. If I wanted to do the same thing with one shock cone per RAPIER, I would have a cross section of one Mk3 tank, because of the bay... and four 1.25 tanks, one for each intake. So if you do use engine pods, just one shock cone for each two RAPIERs will be enough, and thus you can halve your frontal intake area. Never mind what I did on the Calymore, that was mostly aesthetic reasons.
 

Rune. Precoolers FTW, basically, there's a reason they are in the final tech node.

 Where are these pics? And do you have any liquid fuel only space planes with full complement of all the usual goodies (RCS, docking port, chutes, cargo, and passengers?)

[Rune]

1 hour ago, sardia said:

 Where are these pics? And do you have any liquid fuel only space planes with full complement of all the usual goodies (RCS, docking port, chutes, cargo, and passengers?)

Previous page, in my post just before the one you quote (the one @Jarin was referring to in the paragraph I quoted). And no, I don't have a liquid fuel only bird at the moment, no. The closest thing would be the Espada, my longest-range bird (3.5km/s on LKO with six kerbals for a let's-get-to-lvl3-in-one-flight training mission), but that uses a bit of oxidizer to reach orbit efficiently, and then it uses the nuke for mileage. Much better that way.

 

Rune. Gravity losses are another kind of drag.

Edited December 21, 2016 by Rune

[Jarin]

2 hours ago, sardia said:

 Where are these pics? And do you have any liquid fuel only space planes with full complement of all the usual goodies (RCS, docking port, chutes, cargo, and passengers?)

Pretty sure I've got something in this ballpark in the hangar somewhere. Give me a few to check. Like @Rune, most of my distance planes are hybrids. I mean, I'm using RAPIERs for best atmospheric velocity already, so tossing in a bit of oxidizer for that initial kick is an easy choice. "Pure" LF SSTOs become more of a "because I can" exercise. Might have an old whiplash/nuke plane though.

[fourfa]

4 hours ago, Rune said:

So if you do use engine pods, just one shock cone for each two RAPIERs will be enough

I've only been experimenting with this lately, but I'm successfully running five RAPIERs per shock cone in a cargo SSTO spaceplane that I'm pretty happy with.  They will starve at full throttle and 0m/s.  I put the throttle at 50%, go loud, then full throttle at 40m/s and never run short of air.  The lack of drag could help a lot - I was frankly surprised how easy it was to lift 50T with that plane.

On 12/20/2016 at 9:26 AM, Freds said:

FAR and MK2 Expansion are a must-have for SSTO

I just want to add a 7th or 8th thumbs-down to this idea.  And so far no one has complained about the conflation of "SSTO" and "hybrid/air-breathing spaceplane" so I'll throw this in here:

 

[fourfa]

8 hours ago, Tex_NL said:

Properly placed RCS ports will only add a very small amount of drag.
Again with your craft as an example add a single RV-105 RCS Thruster Block  and a single Place-Anywhere 7 Linear RCS Port on each side of the CoM. ....etc

I've become a huge fan of hiding all docking ports and RCS inside cargo bays.  See the one linked above, or this.  Each little bit hidden from drag may be small, but it does add up.

[Jarin]

4 hours ago, sardia said:

And do you have any liquid fuel only space planes with full complement of all the usual goodies

Okay, meet the Royal Albatross. As ugly to fly as it is to look at, but it gets an orange tank to orbit without burning a drop of Ox. Lumbers off the end of the runway while still losing altitude, but levels out before hitting the water. Run low and fast to get the rapiers up to speed. Your ideal acceleration profile looks like this:

Spoiler

 

(I'm cheating with pilot assist. Do not attempt waveskimming the Royal Albatross unassisted)

Somewhere between mach 2 and 3, pull up to 10 degrees and leave it there. Kick on the nukes passing 20km. It'll probably bounce, but it gets to orbit eventually. 

No parachutes because I never have them, but it does come with airbrakes and very sturdy landing gear. Fully stock, but does take advantage of expanded offset limits, so don't try to move those engine pods yourself unless you have the same.

---

Like every LF-only craft, this thing's biggest issue (aside from just kinda being a terrible design all around) is lack of TWR. Even one of those tanks having some Ox in them would make it 100% easier to reach orbit, and would probably get there with a ton more dV on the nukes since you wouldn't spend so long in middle atmosphere clawing for speed.

Edited December 21, 2016 by Jarin

[n.b.z.]

On 20.12.2016 at 2:04 PM, Jeb&Bob said:

I'm currently about 15 hours into the game

 

On 20.12.2016 at 2:46 PM, Jeb&Bob said:

So far, I've only built planes and they are easily able to land and fly.

Building a plane that just handles reasonably well, after only 15 hours of playtime, is good! You will be doing just fine with KSP!

I don't know about the Mac, but KSP has its own screenshot key, which at least on Win is F1. (And F2 gets the interface out of the way for the pretty ones.) Screenshot files should end up inside the "Screenshots" directory within the game's main directory.
_____

A short remark on drag: it's not only open nodes that can hurt badly, it's also high angle of attack. Having too much wing area obviously causes excessive drag and adds useless mass, but having too little wing area can also hurt badly, as the vehicle may be flying with a too high angle of attack. Building such an angle into the wings can help with that.
_____

A rather long comment on droptanks: true, anything dropping tanks is not an SSTO by definition. So, no argument.

But this discussion may miss, or obscure, the point of making good spacecraft, that get a job done efficiently. A point that Freds seemed to want to get across if I'm not mistaken:

The capability of a spaceplane that can be flown as an SSTO can be greatly enhanced by adding droptanks! Who cares if the plane loses the SSTO sticker, when you can massively improve the capability of the vehicle by making a small fraction of its dry mass expendable? When you fly wreckage retrieval contracts around the Mun or Minmus, conveniently flying retrieved pieces and rescued Kerbonauts right down to the KSC runway, for the price of a few tanks and nose-cones?

There is no need to design something either completely reusable, or completely expendable. Sure, spaceplanes with droptanks are not SSTOs, but can make a lot of sense nevertheless.

I'm in my very first career game (but not new to the game at all), and my workhorse for about half of all contracts has been the "VentureWing", a spaceplane I made as soon as I had Rapiers. It's the best one I ever made. The cargo hold is made from two long MK2 cargo bays in a row, the vehicle's CG is exactly on the boundary between them, and the fuel tanks are almost perfectly balanced (the exception are the two "NCS adapter" tanks which move the CG slightly forward when full).

It carries most of the science gear (except large scanners) in its own equipment bay without cluttering up the cargo hold. It features a medium dock port, a small one on the bottom, RCS for translations, and lots of reaction wheel torque. There are enough seats to be useful for most occasions. Heck, there is even a contingency fuel cell and two Oscar reserve tanks in case someone manages to break both solar panels. (No, why, I'm not looking at Bill at all.) The plane handles marvellous, having a natural "trim speed" suited for making the approach to the runway a joy to fly without SAS if a joystick is used, and extending the flaps will even change the trim to landing speed.

It would be more efficient if it carried only one nuke, but carrying two of them gets me a lot of the nuke's efficiency while still retaining enough fun factor (a term highly correlated to "TWR"; some would even say these are basically synonyms).

This spaceplane is an SSTO for all missions that stay in Kerbin orbit. It can even reach the surface of Minmus or the Mun and return to Kerbin, but not with a sensible payload. The value for my career game is that I can say "to heck with the SSTO label", hang a bunch of big LF droptanks under the wings, and actually go places with it, or enjoy a comfortable delta-V margin to be able to just wing it without looking at any numbers. I have retrieved four stranded Kerbals AND one piece of wreckage from four different Mun orbits - one of them was retrograde - in a single mission ... and could have done a half-hour sightseeing flight in the mountains with the fuel I still had when coming back. (I didn't.)

Optional droptanks were part of the design from the beginning: the four 1.25m service bays on the outer pods - which happen to be arranged around the vehicle's CG perfectly - and their content are just useless ballast for missions in Kerbin orbit. But when going to the Mun, the two downward-pointing Spark engines in each bay make vertical landings on the gear a breeze. And when Kerbonauts ventured to another planet for the first time - Duna, of course - the Sparks were assisted by large parachutes deployed from these bays. This worked so well, that the crew did a visit to Ike's surface on the way back just because everyone was having so much fun. (Okay, I'll admit it: they refueled on Minmus first AND took mining equipment on this pioneering trip for ease of mind. For five Kerbals going to an atmosphere in which the plane had never been tested, having plenty of fuel margin was reassuring. But they would have made it back easily without it, if they had left out Ike.)

How do I get away with so much radially attached stuff? 4 radiators, 4 RCS parts, 1 small dock port, 2 wingtip antennas, 2 ladders, 2 drogue chutes? ....Not to speak of that surplus of intakes that were somehow left over from an earlier, non-Rapier version, and then seemingly just forgotten?

First: I don't. There is a performance penalty, but the four Rapiers are strong enough to deal with it. When using large droptanks, I pitch down to almost level flight to get through transonic drag, while lighting the nukes for a little extra push. In other words, ascent profiles are so important, they can make the difference between "land on the Mun" and "get stuck on the boring side of the sound barrier forever". Once Rapiers hit about 420 to 450 m/s, they basically become unstoppable.

Second: The outer fuselage columns are NOT attached radially. Instead, the outer Rapier engines were attached back-to-back to the inner Rapiers, rotated 180° and shifted out; eliminating the draggy rear nodes of all Rapiers. The outer fuselage columns then were built from there to the front, and are held in place by magic auto-struts.

Third: The main wing has an angle of attack built in that allows lift to be generated while the vehicle is at zero pitch. In other words, when fast enough, the vehicle's attitude marker on the Navball is nicely aligned with the prograde marker. While the wing itself is more draggy this way, everything else is less so!

 

[sardia]

I managed to add some extras to the craft, still no RCS yet. But it's getting there. It's still one of the fastest SSTO launches I've made. They usually take forever to get to orbit. This one got to orbit in 5 minutes. Why is there no cargo bay for 1.25m form factor?

Just found this guide that shows the relative drag values of each nose cone. And the tip that a RCS one way can be used to protect your craft from heat. Still testing that one.

Edited December 22, 2016 by sardia

[fourfa]

53 minutes ago, sardia said:

Why is there no cargo bay for 1.25m form factor?

There's a 1.25m service bay, which can work just fine

[Aerospike]

9 hours ago, sardia said:

Huh? How is it that your 1.25 m Passenger cabin has a door/hatch at the top? Last time I used one for (sub)orbital tourist contracts inline in a rocket, everybody was trapped until recovering the craft, because both the front and back hatch have been obstructed of course...

[sardia]

Its a mod, though it's aero performance is the same. For others using stock, they can transfer using the cockpit,iirc. 

Based on the drag profiles, weight, and heat tolerance, I think RAM intakes are the best nose cones.  They weigh less than shock cones,  are equally heat tolerant and just as not draggy. 

Unfortunately, I can't find a good way to terminate a Mk 2 body into a 1.25 m engine. The drag from not matching sizes is a concern. 

Jeb & Bob, I hope you're still with us, and that we aren't going too fast for you. 

[Tex_NL]

2 hours ago, Aerospike said:

Huh? How is it that your 1.25 m Passenger cabin has a door/hatch at the top? ...

 

29 minutes ago, sardia said:

Its a mod, though it's aero performance is the same. ...

He asks what it is. You know what it is. For gods sake, just provide the link.

 

[Kryten 2X4B 523P]

21 hours ago, Rune said:

That is very sciency of you. You'll get far in this game thinking that way!

Yeah., shocking I know :-)

[AeroGav]

On 21/12/2016 at 2:10 AM, Rune said:

 

-The flight path for the engine is crucial. With a RAPIER, you basically want to do a straight line to orbit, pitching up/down only slightly, or not at all (10º over the horizon is climbing rapidly if your speed is 1,000m/s, and kerbin curves under you), and getting to ~400m/s near sea level, in order to "wake up" the engines.

 

 

Agree with most of what Rune said,  low drag rules.   However, I'd like to expand on that "get to 400 m/s at sea level" point, because it doesn't apply to every aircraft, and can cause trouble.

The long version -

1.  Between 240 m/s and 440 m/s is the high drag transonic region.  After 440 m/s drag is worse than below 240, but not as bad as being transonic.   You don't want to spend any longer in the transonic region  than necessary.    Try to be faster or slower than transonic. When the time has  come to cross the transonic zone,  don't climb.    Fly level if you're down on the deck,  if at altitude a shallow dive may be a good idea.

2. As you get higher, drag falls quicker than thrust, so it is easier to achieve high speed in level flight.  Well , up to  a point that is - beyond 14km (Panther in afterburner)  17km (whiplash) or 22km (rapier) engines die off really quick.

3. Don't raise your nose more than 5 degrees above prograde or you'll have excessive drag.

So,  after takeoff you should climb until it is no longer possible to get higher without your speed either exceeding 240 m/s or the nose getting more than 5 degrees above prograde.  At this point,  lower the nose and accelerate through to 440 m/s in a shallow dive.

Your next goal is to reach your best speed altitude (see point 2) and fly level to get as much speed as possible on air breathing engines.  When you are no longer accelerating properly , pitch up so the nose is at 5 degrees above prograde (best lift:drag ratio here) and activate rockets.  Hold that angle of attack till you leave the atmosphere.

Case in point -   someone downloaded one of my career mode planes and complained they couldn't get over 330m/s.     They were trying to do it at sea level,  because everyone told him that's how you fly an SSTO.

https://kerbalx.com/AeroGav/Basic-Tourist-SSTO

I had to add some pictures to explain how to get supersonic on this aircraft -

[AeroGav]

On 21/12/2016 at 7:34 PM, sardia said:

 Where are these pics? And do you have any liquid fuel only space planes with full complement of all the usual goodies (RCS, docking port, chutes, cargo, and passengers?)

If you look through my KerbalX profile, you'll see that I've built many.     MK2 and mk3 parts are not available in LF only form so my larger craft tend to have small amounts of oxidizer which reduces the amount of nukes to be carried.  However, these are my recent LF-only craft :https://kerbalx.com/AeroGav/Firefox-II

This is a mk2 cargo mid-career ship h one Whiplash and 2 Nukes.   Can be used to deliver probes/satellites to orbit, lighter loads (science juniors and other instruments) could be carried all the way to Minmus.

In Sandbox mode I have this -

https://kerbalx.com/AeroGav/ASES-SP2-Xkos-Dual-Nukes

Thanks to the Rapier engine (higher airbreathing max speed) and the ability to store fuel in the wings and strakes, it is easily able to leave Kerbin SOI despite hardly any of the fuselage being used for fuel tankage.  6 seats, RCS, docking adapter, small cargo bay, generous provision for monoprop.

Finally there's this speedbuild of a LF-only mk1 SSTO 

 

[Aerospike]

@Tex_NL Thank you for the link!

Since the thread was about Stock SSTOs, I just wanted to make sure this is not some kind of stock option that I somehow missed.

Now I have to debate myself if this is an acceptable mod for a "stock parts only" career game, but this is becoming off-topic now.

[Leafbaron]

Honestly, When i was making taking my first baby steps with SSTO's I built rocket planes. only used the swivel or thud's as the engine, as its isp and thrust in atmosphere is "acceptable". They couldn't do much except get to orbit and shuttle maybe 2 kerbals, No cargo capabilities. But it was extremely easy, relatively speaking. Maybe try making a rocket but adding wings and launching it like a plane. you'll need about 3700-4000 m/s of dv to reach orbit with a rocket plane.