Jet engine thrust

[chrisl]

I'm trying to design some space planes that use turbojets to get the plane up to 20km-30km altitude, then switch to rockets to finish orbital insertion. I've got Mechjeb2 installed but I am only using it for the information that it provides. Anyway, I've got several designs that seem to work well but one thing I keep noticing is that the Surface TWR and Max thrust that Mechjeb2 lists while I'm on the runway or in flight are about half what is reported when I'm in the SPH. I know that jet thrust is variable based on speed (and probably other factors) but I never seem to be able to get more then about 1/2 the thrust rating for a specific jet engine. Is this a bug or am I doing something wrong?

[Hodo]

I'm trying to design some space planes that use turbojets to get the plane up to 20km-30km altitude, then switch to rockets to finish orbital insertion. I've got Mechjeb2 installed but I am only using it for the information that it provides. Anyway, I've got several designs that seem to work well but one thing I keep noticing is that the Surface TWR and Max thrust that Mechjeb2 lists while I'm on the runway or in flight are about half what is reported when I'm in the SPH. I know that jet thrust is variable based on speed (and probably other factors) but I never seem to be able to get more then about 1/2 the thrust rating for a specific jet engine. Is this a bug or am I doing something wrong?

Are you using FAR?

If so FAR nerfs air breathing engines to a little better than half of stock power output.

[mhoram]

I believe this works as intended in stock KSP - thrust also depends on altitude.

Have a look at this thread for graphs and descriptions of thrust curves:

Fuel consumption as a function of atmospheric pressure

[Laie]

Surface TWR and Max thrust that Mechjeb2 lists while I'm on the runway or in flight are about half what is reported when I'm in the SPH. I know that jet thrust is variable based on speed

That's it, exactly. Mechjeb assumes the nominal (=peak) thrust in the SPH, but shows the actual, current thrust while in flight. As it happens, Turbojets deliver exactly 50% of their nominal thrust while standing still. Peak value is at 1000m/s btw, then it drops back to 50% at 2000m/s, and 0% at 2400m/s. For rapiers, it's the same up to 2000m/s, but they drop to zero by 2200m/s.

This is stock behaviour. I don't know about FAR.

[chrisl]

No, I'm not using FAR. I so use Deady Reentry but I'm not sure if that effect jet characteristics.

I do use Mechjeb to monitor thrust and such, as I stated above, but I've also kept the stock information window open on one my turbojets from take off to flame out. Altitudes from ground level to just shy of 30km. And speeds of around 2000m/s and I've never see the jets come anywhere near max thrust according to their stock rating.

Also, that really seems weird. Why would jets have to reach such a high speed in order to get even close to their thrust rating (and, again, I've never gotten even 75% thrust regardless of altitude or speed)? You basically have to fly with twice as many jets as you really need in order to get up to speed, and then have twice as much thrust as you need once you're at speed. Itseems counter intuitive (not that I know much about real jet engines). Wouldn't you start off with high thrust but lose it as you got faster and higher?

Edited October 29, 2014 by chrisl

[NathanKell]

Jets are not rockets. Jets work by compressing and burning the incoming air. If you have more incoming air (i.e. you're going faster) then you will get more thrust out of the jet.

Note that what should also be true (but is not in KSP) is that thrust will decrease as the air gets thinner (what matters isn't the volume of air taken into the jet but the *mass*), so thrust (should be) a function of airspeed and altitude, but as it stands, KSP only simulates half of that.

[Laie]

Altitudes from ground level to just shy of 30km. And speeds of around 2000m/s and I've never see the jets come anywhere near max thrust according to their stock rating.

Yes you do -- but only briefly. As you approach 1000m/s, your thrust improves dramatically, which also means that you can quickly accelerate past that mark and proceed into regions where your thrust goes back down again. This also happens to be a phase of the flight where you have to pay attention to other matters. So the moment of peak thrust is easy to miss.

But yes, for all practical purposes it's better to assume that you only have half as much power as it says in the stats, because that's what you really have for most of the flight.

[Red Iron Crown]

Kerbal Engineer Redux gives tweakable figures for air-breathing engines in the VAB. You can adjust the atmospheric parameters to see what the TWR will be at a standing start on the runway or at higher speed and altitude.

[KerikBalm]

Check at around 1,000 m/s whe nyour turbojets arent starved for intake air, they should produce their rated thrust at that point.

Basic jets should produce their rated thrust from a standstill (but give them a little time to spool up)

[Wanderfound]

Also worth keeping in mind: spaceplanes don't need a whole lot of thrust (it's fun to have it, but you don't need it).

Basically, if you can get it off the runway, then it has enough thrust to go to orbit (so long as you're running Turbojets + rocket or RAPIERs). Getting off the runway typically only requires 70-120m/s (depending on design; streamlined things with minimalist wings require more speed than broadwinged craft).

Lift with your wings rather than your engines, and don't leave the breathable atmosphere until you've cranked up as much horizontal velocity as possible.

[chrisl]

Also worth keeping in mind: spaceplanes don't need a whole lot of thrust (it's fun to have it, but you don't need it).

Basically, if you can get it off the runway, then it has enough thrust to go to orbit (so long as you're running Turbojets + rocket or RAPIERs). Getting off the runway typically only requires 70-120m/s (depending on design; streamlined things with minimalist wings require more speed than broadwinged craft).

Lift with your wings rather than your engines, and don't leave the breathable atmosphere until you've cranked up as much horizontal velocity as possible.

I have a small spaceplane that runs on two RAPIER engines. It's based on the example design from Spaceplane Plus. It works great. I can get it into space and have it rendesvous with my orbiting space station/refueling station with delta-V to spare.

Now I'm working on a larger plane that I want to use to go capture an incoming asteroid. I know it's going to need more delta-V once it catches the asteroid so I'm building it with about 30-40% more rocket fuel. I also know it's going to need more thrust in space once it catches the asteroid so I've been designing it around a pair of rocket engines with turbojets to push it in atmosphere. Because of the extra weight from it's larger size and the extra rocket fuel (not to mention having seperate rockets) it's take off TWR is only around .9 with two turbojet engines. I can get the plane off the ground. But I'm running into an issue where pitch is pegged up but climb rate is 0 or dropping. If I put 4 turbojets on the plane, that issue goes away but then I run the risk of torquing if I don't shut down at the right time.

If my problem is really lack of lift instead of lack of thrust, then how much lift should I be aiming for per ton of plane? I'm using Procedural Wings (mostly so I could keep part count down) to create the wings and control surfaces. The wings I've currently got designed say they give a 15 lift each and if I add up the lift rating on all the attached parts, I end up with somewhere around a 40 lift rating. The plane itself has a takeoff weight of just under 29 tons. How much more lift do I need if I only want to use two turbojets? Also, these wings I've currently got have the plane's lift rating slightly in front of it's center of mass. I know high speed planes usually work better when lift is just behind CoM, but I can't seem to do that without making the wings long and narrow or sticking off the back of the plane.

Should also say that pretty much all of my planes, large or small, large wing area or small, have to use the entire runway to take off. Every so often I'll have a design that can lift off early, but most don't seem to get any effect from pitching up until the end of the runway no matter how fast they are going. Had one design that got up to close to 180m/s and still couldn't take off until the end of the runway but it flew perfectly and made orbit just fine. I figured this was a bug, though.

[Laie]

I have a small spaceplane [...] take off TWR is only around .9 with two turbojet engines. [...] If my problem is really lack of lift instead of lack of thrust,

Planes can take off with a very low TWR on the order of 0.2 or less (Ion gliders for example). That's only a question of having enough wings, and/or angling the wings. Climbing will take ages, though.

By an altitude of 10-12km, things will become a whole lot easier both because of the peculiarities of KSP's aerodynamic model, and because your Jets give more thrust the faster they go. This works until you run out of thrust because of either going very fast, or lacking intake air. By the time this happens, more wing will do you no good (or rather, the huge amount of wing you'd need would become a problem in it's own right). More thrust is the easier answer at that point, and more intakes is the most lightweight approach.

Not sure if I recommend it, though. Asteroid capture isn't exactly where spaceplanes shine; and getting an overweight, bogged down plane into orbit is no fun, even if you eventually succeed. At any rate, things will be a whole lot easier if you manage to have ay TWR >1 to begin with.

As to generic airplane woes (needing the entire runway &c), there's this nice primer.

Edited October 29, 2014 by Laie

[Mystique]

Should also say that pretty much all of my planes, large or small, large wing area or small, have to use the entire runway to take off. Every so often I'll have a design that can lift off early, but most don't seem to get any effect from pitching up until the end of the runway no matter how fast they are going. Had one design that got up to close to 180m/s and still couldn't take off until the end of the runway but it flew perfectly and made orbit just fine. I figured this was a bug, though.

Check out position of your rear landing gear, perhaps they're too far behind from center of mass and plane just can't tilt nose up, because rear gears just don't allow its stern to go down.

[Wanderfound]

I have a small spaceplane that runs on two RAPIER engines. It's based on the example design from Spaceplane Plus. It works great. I can get it into space and have it rendesvous with my orbiting space station/refueling station with delta-V to spare.

Now I'm working on a larger plane that I want to use to go capture an incoming asteroid. I know it's going to need more delta-V once it catches the asteroid so I'm building it with about 30-40% more rocket fuel. I also know it's going to need more thrust in space once it catches the asteroid so I've been designing it around a pair of rocket engines with turbojets to push it in atmosphere. Because of the extra weight from it's larger size and the extra rocket fuel (not to mention having seperate rockets) it's take off TWR is only around .9 with two turbojet engines. I can get the plane off the ground. But I'm running into an issue where pitch is pegged up but climb rate is 0 or dropping. If I put 4 turbojets on the plane, that issue goes away but then I run the risk of torquing if I don't shut down at the right time.

General rule: any time that you're asking for advice on an aircraft build, provide screenshots. We can't diagnose what we can't see, and you'll get better answers if you provide the needed information.

TWR of 0.9 is actually quite high in aircraft terms. Only a handful of fighter jets do better than that in the real world.

If my problem is really lack of lift instead of lack of thrust, then how much lift should I be aiming for per ton of plane? I'm using Procedural Wings (mostly so I could keep part count down) to create the wings and control surfaces. The wings I've currently got designed say they give a 15 lift each and if I add up the lift rating on all the attached parts, I end up with somewhere around a 40 lift rating. The plane itself has a takeoff weight of just under 29 tons. How much more lift do I need if I only want to use two turbojets? Also, these wings I've currently got have the plane's lift rating slightly in front of it's center of mass. I know high speed planes usually work better when lift is just behind CoM, but I can't seem to do that without making the wings long and narrow or sticking off the back of the plane.

How much wing is always a "how long is a piece of string" thing; it depends on too many other factors. In stock aero, generally speaking, the more wing the better. In more realistic aero (FAR/NEAR), you rapidly hit a point where it becomes counterproductive.

The Centre of Lift is also known as the Centre of Drag, and it needs to go behind Centre of Mass. An object moving through the air tends to rotate so that its centre of drag is behind its centre of mass; think of the feathers on the back of a heavy-tipped dart. In KSP, overpowered SAS units let you get away with a bit of CoL/CoM misalignment, but especially for a heavy lift vehicle you want CoL behind CoM.

Hit the Spacecraft Exchange (or the Akademy Awards: http://forum.kerbalspaceprogram.com/threads/93779-SSTO-Spaceplane-Airplane-Design-Contest-II-Akademy-Awards), grab some other people's craft files, take 'em to the SPH and rip 'em apart to see how they manage to put things together with weight and drag in the right places. Then use that knowledge when you're building your own.

Should also say that pretty much all of my planes, large or small, large wing area or small, have to use the entire runway to take off. Every so often I'll have a design that can lift off early, but most don't seem to get any effect from pitching up until the end of the runway no matter how fast they are going. Had one design that got up to close to 180m/s and still couldn't take off until the end of the runway but it flew perfectly and made orbit just fine. I figured this was a bug, though.

Have a look through the illustrated tutorial at http://forum.kerbalspaceprogram.com/threads/52080-Basic-Aircraft-Design-Explained-Simply-With-Pictures

We can't be certain without screenshots, but from the sound of it you're placing your landing gear too far towards the rear. Apart from special cases like high-set tailstrike guards or ships that are build with strong static pitch, your rearmost landing gear should be just behind your CoM.

[chrisl]

Thank you all for your advise. I think I finally managed a setup that works. It's not perfect yet and I'm not sure I'll actually be able to use it to capture asteroids, but it's tought me alot about building spaceplanes. I've included some screenshots below.

Figured out that the issue I was having with the jets is related to the Karbonite mod I'm using. The liquid fuel, inline turbojet in the Karbonite Mod uses a different thrust/velocity range then the standard turbojet. You don't reach peak thrust until 1350m/s and it drops to nothing at something like 1522m/s. I tweaked it's numbers to match the standard turbojets (mostly cause I prefer how they look) and my planes performance changed dramatically. The plane in the below screen shots managed to reach a stable 150km orbit using less then 130m/s deltaV from it's rockets.

I'm using a bunch of mods but the ones that play a role in this spaceplane would be Procedural Wings (to limit part count), Karbonite and Deadly Reentry. I've not yet tried using FAR, mostly because I figure if I couldn't get spaceplanes to work in the (mostly) default game, I probably wouldn't have any luck when using a mod that had more realistic flight characteristics.

[Srpadget]

Congrats on your successful design! Based on the pics posted above, though, I do have one piece of additional advice: lose the bi-adapter and the two rockets. Those big LVs are adding a bunch of mass to give you vacuum thrust that you don't need. All you need back there is a single LV-909 (really, all you truly NEED is a dinky little Rockomax 48-7S for circularization at apo and for your deorbit burn, but nobody ever believes that until they've done it themselves).

But that's fine-tuning. You designed a spaceplane capable of making orbit, and successfully piloted it--that's a major accomplishment! Well done!

[numerobis]

The tails mounted on the wing are a nice visual touch. Pity they're so heavy (400 kg each!)

[How2FoldSoup]

If you still want the oomph of bigger rockets, switch to aerospikes..they follow the rule of cool plus, if I remember correctly, they have better ISP than those engines you currently have. Additionally, you won't have to worry about tailstrikes on takeoff with the shorter engines.

[chrisl]

Congrats on your successful design! Based on the pics posted above, though, I do have one piece of additional advice: lose the bi-adapter and the two rockets. Those big LVs are adding a bunch of mass to give you vacuum thrust that you don't need. All you need back there is a single LV-909 (really, all you truly NEED is a dinky little Rockomax 48-7S for circularization at apo and for your deorbit burn, but nobody ever believes that until they've done it themselves).

But that's fine-tuning. You designed a spaceplane capable of making orbit, and successfully piloted it--that's a major accomplishment! Well done!

I need more then just the tiny Rockomax engine since this plane is going to try and capture and adjust the orbit of an asteroid. The plane I just put into orbit based on the above design replaces those LV-T45 engines with LV-N Atomic engines and replaces the LF+O Short section with a second Long section. I'm worried that even the combined 120 thrust isn't going to be enough once I've captured the asteroid but I'm hoping the greatly increased isp will at least let me get out there and back. We'll see. That said, your suggestion for the 48-7S engines is something I'll have to try on my other designs. Mostly I avoid that particular enging because the isp is so poor, though.

The tails mounted on the wing are a nice visual touch. Pity they're so heavy (400 kg each!)

Yeah. Can't figure out why that tail section weighs so much when all the other plane parts are so comparatively light. A pair of stacked Structural Fuselage units are nearly as long as the tail piece, but only weigh a combined 0.2 so I'm not really sure why the tail peice is so much heavier. But in the long run I figured 4 tons of weight was worth the aesthetics of the finished product. Plus, the next update to Kerbal may see the part weights change.

If you still want the oomph of bigger rockets, switch to aerospikes..they follow the rule of cool plus, if I remember correctly, they have better ISP than those engines you currently have. Additionally, you won't have to worry about tailstrikes on takeoff with the shorter engines.

I used the aerospikes on an earlier model but the reduced thrust I thought was being an issue. Turned out the problem was I simply didn't have enough lift and air intake to allow my jets to run long enough. With the heavier plane I just put into orbit (heavier because it has LV-N engines for the isp) I managed to keep the turbojets running until I was somewhere above 35km. All told I managed to get into an 87km stable polar orbit using 427units of Liquid Fuel and 190 units of Oxydizer. Mechjeb says I've still got about 2470 deltaV remaining which I'm hoping is enough to push me into an 8.7mil meter orbit, capture an asteroid and then adjust the roids orbit so it'll stay in Kerbin's SOI. Fingers crossed.

[numerobis]

Two notes:

1. A pair of Mk2 short tanks have less drag than a single Mk2 long tank, for some reason.

2. The rocket-fuel tanks have a higher fuel percentage: every tonne of tank can hold 8t of fuel, as opposed to just 6t for the Mk2 tanks.

[Red Iron Crown]

Two notes:

1. A pair of Mk2 short tanks have less drag than a single Mk2 long tank, for some reason.

2. The rocket-fuel tanks have a higher fuel percentage: every tonne of tank can hold 8t of fuel, as opposed to just 6t for the Mk2 tanks.

1. How did you measure that? Their cfgs indicate the same Cd and exactly half the mass, so one would think the drag would be equivalent.

2. That makes some sense, the Mk2 shape has more surface area per unit of volume than a cylinder shape.

[Redshift OTF]

The amount of thrust you have in space isn't very relevant, it only affects how long a burn takes or whether you can land or take off from a planet or moon. LV-N's are a better bet as you get more DeltaV out of them and you need a lot of DeltaV to re-direct asteroids.

[numerobis]

1. How did you measure that? Their cfgs indicate the same Cd and exactly half the mass, so one would think the drag would be equivalent.

The long fuselage has:

    dragAtMaxAoA = 0.3
    dragAtMinAoA = 0.1

The short fuselage has:

    dragAtMaxAoA = 0.15
    dragAtMinAoA = 0.1

They also have maximum_drag settings, but those are overridden by the ModuleLiftingSurface.

I posted more details in the general discussion.

[Red Iron Crown]

Twice the drag for twice the lift at max AoA, seems equivalent to me.

/me goes to look for your other thread.

Edit: Here it is. I did some tests and posted them there, there is definitely lower drag on the smaller tank.

Edited November 4, 2014 by Red Iron Crown

[chrisl]

Just an update for those who are interested. I managed to meet up with, capture and redirect the orbit of a 524.88t asteroid. Admittedly the plane I used was barely able to nudge the asteroid (think I went from 900 deltaV potential before contact to 22 deltaV after contact and had to use 15 of that for the redirect maneuver) but it was enough to put it into an eliptical orbit (8.7m - 82.3m) around Kerbin. Plus I was able to return to KSC. Of course, the return required something like six aerobrake maneurvers and it was sheer luck that the last maneuver happened to take me close enough to KSC to make the landing. But for this flight, I'll be happy with luck. I've got a screenshot of the finished plane here. Forgot to take a shot of it just after landing.

My next plane I'm going to try replacing the LV-N rockets with several PB-ION engines. I'll replace the long LF+O sections with the large cargo hold so I can put Xenon tanks inside. And I'll remove all the oxidizer from the nose and tail section. I'm not exactly sure if Ion engines will actually have the oomph to push the plane into a stable orbit after the turbojets give out around 40km but if they can I'm hoping the the extremely high isp will let me adjust the orbit of my captured asteroid to a more significant degree. Of course, it's going to be extremely long burn times but can't think of a way around that. I do have all the "Near Future" mods so I could try one of those "electrical" engines but most require so much Ec that I don't know if they'd be viable. Either way, I'll update this thread with my progress for anyone who is interested.