Spaceplane Help

[Red Shirt]

I haven't messed with planes since 0.25. I did have a vertical launch SSTO in 0.90 that used ramjet and rapier engines for a completely reusable craft. The aero changes made this one obsolete. Change happens.

With that as background. I have built my first new plane using two ramjets and an aerospike. The COL is behind COM. Mechjeb says it has an SLT of over 2 in the SPH. First issue is kind of minor - on takeoff the craft, at first, flies like a dolphin swims by bobbing up and down. I checked the tail and it seems level. The tail is attached to the rudder - higher than COT, so I am guessing that is causing the behavior. Once I begin to climb it quickly smooths out. It is so smooth that climbing at around 45 degrees is hands off. It slowly levels at around 20k. Still I would like to understand what is happening at takeoff.

Next issue, at 20k I cannot build speed. About 800m/s is pretty much it. Using a slight dive I can get over 1000m/s but then I don't seem to have enough lift to start climbing again. I have to fire the aerospike to climb, but that means spending precious oxidizer too soon.

I have made an 80k Ap but ran out of O2 before being able to sufficiently raise Pe. This brings up my last issue - why are there no oxidizer only tanks? I have plenty of liquid fuel left. I guess I can try adding a short rocket fuel tank and removing the fuel. Seems like a poor fix.

Sorry I don't have a pic to post. I'm not on my KSP computer at the moment. Hopefully these are common issues that don't require a visual. Edited November 18, 2015 by Red Shirt

[Brainlord Mesomorph]

[quote name='Red Shirt']
I have made an 80k Ap but ran out of O2 before being able to sufficiently raise Pe. This brings up my last issue - why are there no oxidizer only tanks? I have plenty of liquid fuel left. I guess I can try adding a short rocket fuel tank and removing the fuel. Seems like a poor fix.[/QUOTE]

if you have leftover LF but not enough O, then don't ADD a tank (and remove LF) REPLACE one LF tank with an LFO tank.

[capi3101]

Regarding your fuel - these days, you want to try and design your spaceplanes to have about 1,800 m/s of delta-V solely for the rockets. Since you're using an aerospike, the vacuum Isp is 340 and Tsiolkovsky says:

dV = ln(M/Mo)*Isp*g = 1800
ln(M/Mo)*340*9.8 = 1800
ln(M/Mo) = 1800 / (340*9.8) = 0.540216
1-(M/Mo) = 1 - e^-0.54216 = 0.41738 = 41.7% <-- your propellant mass fraction (i.e. how much of your plane's mass at takeoff that has to be devoted solely to LF/LOX for the aerospike)

So assuming you're going with the old recommended value of 15 tonnes per turbojet, your plane has a maximum mass at takeoff of 30 tonnes - 12.51 tonnes of the plane has to be rocket fuel. Note that that's the mass of the FUEL, not the mass of the FUEL TANKS (subtle but important difference there).

Your turbojets need no more fuel than they used to - 150 units of LF per turbojet is plenty for the run-up, and since the Mk1 Fuselage holds 400 units these days, a single tank is good for two engines (and you have a little surplus). You don't need to airhog anymore though - one intake per engine is plenty.

These days you also need to begin flattening out and starting your run-up to speed a lot sooner than 20k. Most spaceplane jockety these days begin their runs in the 10-12k regime; flatten out and accelerate, keeping yourself 5-10 degrees above the horizon (generally rate of climb will be less than 100 m/s if you watch that gauge). If you start seeing heating bars, steepen the climb. You probably will see flames; that's natural these days. Make sure you're in a save you can revert the first few times...

Now, as far as the bobbing behavior is going on - what you've got there is a little bit of [url=http://code7700.com/stability.html]phugoid oscillation[/url]. I see it a fair amount my own self; the important thing is that it damps out with time. Edited November 17, 2015 by capi3101

[Red Shirt]

Please accept this in the humor it is intended.... In RL I am always considered the geek in the group. That said I have almost no clue what you just wrote! I did some tweeks - moved the tail wings down to the center and moved the turbojets outward to clear the tail. This stopped the oscillation. I am still having a big issue where between 15 and 20k I lose upward pitch. I use imbedded Mechjeb and it kind of acted like it was trying to takeover. So I disabled it. The same issue continues to occur. Maybe as it uses fuel it is getting nose heavy? Anyway, pic:

[img]http://i.imgur.com/5vY8qKe.jpg?1[/img]

[capi3101]

Which aero model are you using, stock or FAR? If FAR, I might think you were experiencing [url=https://en.wikipedia.org/wiki/Mach_tuck]Mach Tuck[/url], and the solution would be a bigger tailplane. I don't know if the same solution would apply in stock or not. One thing I do know is that you lose pitch authority if your tailplane is in line with the main wing (you don't get as much torque from the pitching surface); moving them downward may have killed the oscillation, but may have contributed to your current issue.

You shouldn't be burning enough fuel in atmo to cause that big of a shift in your CoM, just looking at the design of the plane.

Do you use RCS Build Aid? It would show you which way your CoM shifts as you burn fuel. Real handy tool for any serious spaceplane designer.

[Red Shirt]

Thanks for the reply. I am using stock 1.04. I'll try making the tail bigger and maybe raise it back up some. We'll see what happens. Oh, and it does appear I am a little shy of the 1800 dv in the aerospike stage. Looks like I may need that extra tank after all.

[capi3101]

I might also suggest you take a quick look at your engine setup - I fly FAR myself, but I keep track of the stock jockeys and their guidelines. Last I heard the guideline was one Turbojet and one rocket (in the neighborhood of 200 kN) per fifteen tonnes of mass at takeoff, or one RAPIER engine per thirteen tonnes at takeoff. Again, you'll want a little over two-fifths the total takeoff mass of your plane as rocket fuel if you stick with the Aerospikes (which are, in my experience anyway, good engines to couple with Turbojets).

[A_name]

Turbojets are not that good at providing thrust at the highest speeds and altitudes, but the RAPIERs excel at it. Before trying anything else, I would simply switch out the jets for RAPIERs, drop the aerospike and see how that goes. On days when my piloting skills are in the zone, I can make it to somewhere between 1400 and 1500 m/s on the air breathers (this is really cutting it thin in terms of heating, though), only needing between 200 and 250 units of Ox to complete the ascent. I use a Nerv to circularize, however, so you might need a bit more than that. Nevertheless, bottom line is that if you're breaking the sound barrier without issue you have enough engines, it's just you might be using the wrong ones. Edited November 17, 2015 by A_name

[A_name]

Additionally, I also second the comment that you're doing your speed run too high up. I'm not sure what your flight profile looks like as a whole, but you want to start pitching down gradually to gain speed as soon as you hit 8km, i.e. where the lightest blue bar on the atmosphere gauge ends. By 10km you should be well over supersonic, maybe around 400 - 500 m/s. I don't use Turbojets much for SSTO's due to the aforementioned poor performance at high speeds and altitudes, but for RAPIERs the sweet spot for max thrust seems to be between 12 and 15 km. You should be hitting 1000 m/s at around 15k. Also, you probably have noticed before that due to the curvature of Kerbin, when holding a certain heading at very high speeds the plane will "pitch up" relative to Kerbin’s horizon as the planet pulls away from under you (although the craft itself doesn’t really rotate at all). Use this effect to your advantage to get some vertical speed out of your plane before the air breathers flame out. How I usually do it is I stop giving the plane control inputs at around 15km altitude or so. So for example, if I was pointing 5° up at that point, by the time the air breathers flame out I'll be pointing roughly at 15°. This gain in vertical speed is at the expense of a bit of horizontal speed, but I prefer this way than having to actively pitch up, which causes drag and stress on my plane. This approach also allows you to rely more on wings than rocket thrust to counteract gravity, which might just be more efficient on the long run.

Hope this helps. Edited November 17, 2015 by A_name

[Red Shirt]

Well, giving up for the day. I put on a bigger tail and made Ap. Ran out of fuel. Added another tank of fuel/o2. Same ending. I'll remove the tank and try changing my ascent by leveling off at 10k. If that doesn't work - I'll loose the aerospike and switch the ramjets to rapiers. I will get to orbit and circularize. Thanks!

[eddiew]

20km is where you finish building speed, not where you hang around and gain it. Break mach 1 by 10km, and try to have maxed out your air breathing speed by 17-18km. After that, the thrust drops off so fast that you'll struggle just to keep the speed you already have :)

My rule of thumb for 1.0+ is go as shallow as you can without exploding. 100-200m/s vertical ascent is absolutely fine.

Check the spacecraft exchange forum and specifically the ssto threads for general design guidelines. Build things that look like other things that are known to work, and you won't go far wrong ^^

[capi3101]

Okay...just checking one last thing here real quick. Your plane - what's it's mass? I'm guesstimating a little over 30 tonnes; is that close? I'm also only seeing three rocket fuel parts. Assuming they're all full, your plane is carrying 8 tonnes of fuel specifically for the Aerospike.

8 / 30 = 26.7%. That ain't enough gas - not by a long shot.

I might also second the notion of using RAPIERs, assuming this is a sandbox plane and/or you've got them unlocked. Only catch there is that their vacuum Isp is only 305 seconds, so you'd need a greater propellant fraction:
1 - e ^ (-dV / Ve) = 1 - e ^ (-1800 / (9.8*305)) = 1 - e^ (-0.60221) = 45.24%

[Red Shirt]

plane is about 22.7 ton

[A_name]

[quote name='eddiew']20km is where you finish building speed, not where you hang around and gain it. Break mach 1 by 10km, and try to have maxed out your air breathing speed by 17-18km. After that, the thrust drops off so fast that you'll struggle just to keep the speed you already have :)

My rule of thumb for 1.0+ is go as shallow as you can without exploding. 100-200m/s vertical ascent is absolutely fine.

Check the spacecraft exchange forum and specifically the ssto threads for general design guidelines. Build things that look like other things that are known to work, and you won't go far wrong ^^[/QUOTE]

Do you usually keep going on the air breathers even when you're no longer gaining speed?

[Vegetal]

I didn't see anyone here stating the obvious from your craft pic:

1 - Circular intakes are bad for spaceplanes
2 - You have 2 big mono tanks causing lots of drag....whyyyy?
3 - Your tailplane seems to be so close to the CoM that your pitch authority is probably cr4p.

Just taking care of these issues should see you getting to orbit. And....is this still 1.0.4?

[eddiew]

[quote name='A_name']Do you usually keep going on the air breathers even when you're no longer gaining speed?[/QUOTE]

I keep going on air breathers while the AP is climbing ahead of me. Not accelerating - or even losing a percent or two of your speed - is fine if it means gaining several kilometres of altitude. Jet engines remain the most efficient form of thrust, and I'll stretch them to their limit :)

Works for me, anyway, mileages may vary.

[Val]

On 17/11/2015 at 7:24 PM, Red Shirt said:

Thanks for the reply. I am using stock 1.04. I'll try making the tail bigger and maybe raise it back up some. We'll see what happens. Oh, and it does appear I am a little shy of the 1800 dv in the aerospike stage. Looks like I may need that extra tank after all.

Since you're still on KSP 1.0.4. You can take a look at my D-2 Scout SSTOs.
Their size and propulsion is similar to yours, so you can compare stats with them and look at the Ascent Profile instructions.

It should be applicable to your craft.

Edited March 2, 2016 by Val
Forum update formatting fubar

[Red Shirt]

R.A.P.I.E.R. = E.Z. Orbit! Changed the intakes to shock cones. Raised the tail and used bigger sections. Lost the aerospike. Replaced the ramjets with rapiers. Straight to orbit with plenty of fuel left over. I did change my ascent to build speed between 8k and 15k. Pointed the nose at prograde around supersonic. Then began to climb slowly. As rapiers lost thrust switched to rocket mode. The most difficult part of the flight was reentry heat. Lost the RCS off the nose. Remembered Slashy saying he used swaying turns and I kept the nose about 20 degree above prograde. Overshot the runway by 150m.

Thanks all for the excellent advice.

[IMG]http://i.imgur.com/aFUos3R.jpg?1[/IMG] Edited November 18, 2015 by Red Shirt

[capi3101]

Glad to hear you made space. A couple of last things then - if you've got access to AIRBRAKEs, consider using them in your design; they'll help with the re-entry heat tremendously, as will keeping your nose pitched up during re-entry (i.e. high drag profile) for as long as your craft will tolerate it (if you've got leftover RCS set up for attitude control, you might consider using it for this purpose, but don't sweat it if you don't). If you use AIRBRAKEs, make sure you set them specifically for braking purposes only - i.e. don't use them as pitching or rolling surfaces - they'll just slow you down if you forget to set them up properly (particularly annoying during the speed-up part of the flight). Set them as soon as you're below 70k.

[Red Shirt]

I had airbrakes on an early attempt but didn't know I could set them for brake only. Shouldn't that be the default? (Just saying) I will give them another try now that I have a working craft. I have RCS tanks and did use them on reentry. When the nose RCS thruster exploded, the ship was surrounded in flame. I quickly looked to see if Jeb was still there in the lower right. So yeah, airbrakes.

Are my round RCS tanks really worse (drag vs weight) than adding another fuselage section for mono?

[capi3101]

They do add some unnecessary drag by the tail. If you want to keep them, I might suggest stowing them inside a cargo bay - you'll save some mass going that route, rather than including a dedicated monoprop fuselage section. Incidentally, the cargo bay would also be a good place to put a probe core for the craft if you ever want it to do unmanned operations (like rescues) if for some reason you didn't have access to Drone Cores (say in a mid-career save), and the walls of the cargo bay make excellent mounting points for prograde- and retrograde-firing RCS linear ports; you no doubt know already that places to put those can be tricky to find when it comes to spaceplanes. The cargo bay is just a suggestion in any event.

I myself haven't found much utility in AIRBRAKES other than as, well, brakes...and IMHO agree that the "just brake" setting should be the default. That is what it is, though.