Space plane piloting question: Kerbin Orbit

[sloth]

So I've been trying to create a plane that can take off from the runway, get into a stable orbit around kerbin, get out of it and land on kerbin. (I know, a whole universe of planets out there now and Im playing around with Kerbin)

Up to now the best I've managed is getting up there but I dont have (even close to) enough fuel to circularise my orbit. Now I'm sure the design can still use some work but thats not what I'm currently looking for advise on. I'm assuming my piloting can be made more efficient as well since ive been more or less having a gutt feeling go at it. So basically, whats the most efficient way of flying a space plane up?

My current plan is get into a 45 degree angle off the runway, hitting SAS and waiting for 13.5-14 km where I switch from the 2 jet engines to the aerospike and increase my climbing angle to about 70-80 and wait for the projected apo to hit 70k at which point I throttle down until Im near the apo (65km ish) and then start burning again level with the horizon to circularise. Then I run out of fuel shortly after.

Surely there are a myriad of improvements to be made.

[Fuzzy Dunlop]

Firstly the jet engines can still supply useful thrust up till about 20km (and even after then keeping them burning will at least shed weight) so don't turn them off too soon.

Secondly it sounds like your trajectory is too steep. I think you should try keeping the climbing angle at about 45 degrees after areospike ignition. Then gradually level off from about 20km onwards so by the time you pass 35-40km you should be more or less horizontal.

Thirdly it sound life you may be trying to build a single stage to orbit (SSTO) design. This isn't easy, but you can look on the K-prize thread in the challanges forum for inspiration. Or look for SSTO in the spacecraft exchange.

[PigletPG]

Moar fuel tanks! Moar rockets! Moar everything!

[EndlessWaves]

If you haven't seen it then the K-prize challenge is many other people's attempt at this:

http://kerbalspaceprogram.com/forum/showthread.php/11214-The-K-Prize-100-reusable-spaceplane-to-orbit-and-back

25-30° seems to be the optimal angle for wings but jet engines are so efficient it doesn't matter much - you often run out of altitude before emptying a single fuel tank - and I guess your problem is that rockets and their fuel don't make up a large enough proportion of the machine by weight.

[sloth]

Thanks for the advise.

I hadnt heard of the K-Prize yet but it looks like I wont qualify since I use non-stock parts (need landing lights!). Maybe I'll have a go at a stock plane after I get it done with my first design.

@EndlessWaves and Fuzzy Dunlop: I'm sure you're right and all, but can you explain why less angle is better? In my lamen opinion that just means itll take longer to get there and thus more fuel. Though I must admit I do not really understand how aerospikes work and what effect lift/glide has on fuel consumption when climbing.

Is the idea to always burn full throttle or can you dial it back in higher atmo?

(Also, more of a side question, but does half throttle mean half thrust and half consumption or is it not that simple?)

[EndlessWaves]

Wings angled at 25-30° to the horizontal seem to produce the most downward lift from my experiments (what speed does a plane need to take off with different wing angles). Additionally wings act as force multipliers, they don't just redirect your thrust downwards but seem to add to it themselves.

However, I don't know if it's actually possible to use wings to get to orbit using less fuel than going straight up, as far as I know none of the mods yet give thrust not in the direction of travel and without real time numbers it's a bigger project to find out.

But as I said, the jets should be a small fraction of the weight so they won't have too big an impact.

Aerospikes are just rockets that don't burn extra fuel in atmosphere like most do.

[vexx32]

Yeah, the lift model's pretty broken, considering anything over 16 degrees will cause a stall. Also, it's not the angle to the horizontal that matters, it's the angle to the relative airflow. In KSP's case, that means it's the angle relative to the motion of your craft, since there's no wind or any turbulence.

[sloth]

Aerospikes are just rockets that don't burn extra fuel in atmosphere like most do.

So aerospikes are just completely imbalanced rocket engines?

[vexx32]

Yeah, pretty much.

[Suzaku]

Yeah, pretty much.

Though Aerospikes are just that efficient in real life as they are in-game.

[bsalis]

I hadnt heard of the K-Prize yet but it looks like I wont qualify since I use non-stock parts (need landing lights!). Maybe I'll have a go at a stock plane after I get it done with my first design.

I recommend reading the K-Prize post. Landing light mod parts are fine to use. I also suggest you should be wary of 0.16 designs due to the fuel bug, since many SSTO designs used fuel crossflow which was impacted by the bug. Remember that there are only 3 ways to increase the delta-V of an SSTO; a) increase the full:dry mass ratio, use higher Isp engines - aerospikes are your friend, and c) avoid fuel waste through better design - ie. excessive gravity loss from poor trust-to-weight.

[sloth]

I recommend reading the K-Prize post. Landing light mod parts are fine to use. I also suggest you should be wary of 0.16 designs due to the fuel bug, since many SSTO designs used fuel crossflow which was impacted by the bug. Remember that there are only 3 ways to increase the delta-V of an SSTO; a) increase the full:dry mass ratio, use higher Isp engines - aerospikes are your friend, and c) avoid fuel waste through better design - ie. excessive gravity loss from poor trust-to-weight.

Thanks for the advise. I'll definitely check it out at some point to optimize my design and try a fully K-Prize compliant one (I use a few other parts as well). But for now I prefer to try muddling along on my own (design-wise that is, piloting wise I need all the help I can get. Jeb I ain't).

[EndlessWaves]

Yeah, the lift model's pretty broken, considering anything over 16 degrees will cause a stall.

I've not noticed any effects above 16°, certainly not loss of control as in an aircraft stall. What do you mean by a stall?

So aerospikes are just completely imbalanced rocket engines?

Well, they do have disadvantages like no thrust vectoring and placement restrictions (no decouplers/tricouplers) but given their high thrust to weight ratio they are the engine of choice in more circumstances than they probably should be.

The key thing is to optimise weight, adding a little extra to the top stage means a lot extra required further down the rocket. I find the kerbal engineer redux mod good for showing DeltaV.

[T-Pilot395]

I've not noticed any effects above 16°, certainly not loss of control as in an aircraft stall. What do you mean by a stall?

A stall in real life aircraft (applies to KSP), is a situation where the angle of attack (A.K.A how high is the wing tilted) is too high, to the point where the wings no longer are able to produce lift. This causes the plane to slow to a point where it will start falling out of the sky like a giant brick.

I typed the whole thing out from experience. I'm a flight simulator enthusiast you see.

To be honest stalls rarely happens to me in KSP.

[Fuzzy Dunlop]

Stall is where the angle of attack of a wing gets so large that air stops flowing smoothly around it. It becomes turbulent and the lift drops to near zero and the drag also increases. That doesn't happen in KSP as the air doesn't flow so much as bounce off the wing (I think thats how the lift model works).

I did some theory work on a very basic lift model. I reckon lift peaks at 45 degrees (makes sense since then the air gets bounced directly downwards). Drag increases exponentially from 0 at 0 degrees to a maximum at 90 degrees. So I guess you could say the wings "stall" at 45 degrees.

[Fuzzy Dunlop]

Thanks for the advise.

I hadnt heard of the K-Prize yet but it looks like I wont qualify since I use non-stock parts (need landing lights!). Maybe I'll have a go at a stock plane after I get it done with my first design.

@EndlessWaves and Fuzzy Dunlop: I'm sure you're right and all, but can you explain why less angle is better? In my lamen opinion that just means itll take longer to get there and thus more fuel. Though I must admit I do not really understand how aerospikes work and what effect lift/glide has on fuel consumption when climbing.

Is the idea to always burn full throttle or can you dial it back in higher atmo?

(Also, more of a side question, but does half throttle mean half thrust and half consumption or is it not that simple?)

The point about less angle wasn't about planes exclusivley. Remember getting to orbit is all about -horizontal- velocity. The reason you go vertical at first is to give yourself enough time to build up the required horizontal velocity before you hit the ground, also to get you above the drag of the atmosphere (on bodies with atmospheres).

Now on Kerbin there isn't much drag above 30km so once your that high you should just burn horizontal untill you get most of the required velocity. This will give you an apogee half-way around Kerbin, once you have that you can just coast up to it and circularize. What you don't want to do is just coast vertically upwards to 70 km, since gravity is leaching away your energy all the time your not in orbit.

While you can just turn 90 degrees from vertical to horizontal its more efficient to level out in a smooth curve. The optimal curve depends on your design. With a spaceplane the wings generate lift so you don't even have to start vetical.

And yes half throttle gives half thrust and half fuel consumption. It can be slightly more efficient to dial back at some points in the assent but it doesn't make a very big difference.

[sloth]

Success! At 260/70km it was far from circular but at least it was an orbit. Time to start practising getting it perfect, I need to start going vertical sooner I guess.

I had literally half a second of aerospike burn left to deorbit, had to do most of it with RCS. And my landing was nowhere near KSC (A continent away). But not a crash so good enough for now.

Thanks for all the advice. Now to reproduce it with stock parts...

(My kerbal for this mission was Jeb himself, I'm sure that helped)

[Yourself]

That would be a very basic lift model. In real life an airfoil will stall between about 10-20 degrees angle of attack. 45 degrees is almost certainly well past anything resembling the stall point of even a flat plate. I hope KSP's lift model doesn't follow this, because that would be quite weird.