A detailed look at Kerbal's jet engines

[Iffn]

The in-game information as well as the current KSP wiki article on jet engines give some indication on the performance of jet engines in KSP. However they are usually not enough when building optimized SSTO's.

This is why I started analysing the vanilla engine components. So far, I've made the following discoveries: [Graphs in Imgur album]

Findings on thrust and speed:

• The maximum thrust of an engine is always relative to the current air speed.
  
• The Basic Jet works well at low speeds but is unsuitable for fast aircrafts, even if they fly at high altitude with minimum drag.
  
• Due to it's high thrust at low speeds, the basic jet can be used as a first stage engine for a ballistic rocket.
  
• Both the Turbo Jet as well as the RAPIER engine in jet mode peak at surface speeds of 1000m/s and work well beyond that. (Note: I've cut off the graph at 1500m/s, however, the curve would continue beyond that. The turbo jet will reach 0kN 2400m/s, while the rapier reaches 0kN at 2200m/s.
  

Findings on thrust and time:

• All jet engines have long start up times, which makes them difficult to use in VTOLs, since they need a short response time.
  
• According to my analysis, the Turbo Jet is able to increase it's thrust slightly faster then the Basic Jet despite a lower peak.
  (Measurements taken at fixed ground location)
  

Findings on fuel flow and altitude:

• Jet engines are very efficient compared to rocket engines.
  
• The fuel flow is only relative to altitude. That means that a fully spooled up Turbo Jet at maximum throttle at a certain altitude always uses the same amout of fuel per second, no matter how much thrust it produces.
  
• For those flying on Laythe: The graph below was made using data from Kerbin. I assume the ISP is actually calculated relative to air pressure instead of altitude above sea level. However, I haven't tested this so far. [For comparison]
  
• Compared to the other jet engines, the basic jet is very efficient when flying below 5000m. Above that, Turbo Jet and RAPIER engines are usually much more efficient since they're able to fly much faster due to the reduced drag at higher altitudes.
  
• The indicated ISP is also relative to altitude as seen here. However, there seems to be no direct connection between fuel flow, actual ISP and indicated ISP.
  
  
  Findings on intake air flow, speed and air pressure:
  • The intake air flow is relative to air pressure an speed.
    
  • The graphs above show the intake air flow relative to speed at Kerbin sea level. (= 1 atm)
    
  • The approximate formula of these is are as follow. They shouln't deviate more then 0.5% from the actual intake air flow : (Q = intake air flow, v = speed)
    - Ram Air Intake: Q = 0.096 * v + 58.7
    - Circular Intake: Q = 0.078 * v + 47.0
    - Engine Nacelle: Q = 0.049 * v + 29.3
    - XM-G50 Radial Air Intake: 0.039 * v +23.5
    
  • Due to the angle of attack, the movement speed will usually differ from the speed at the air intakes.
    The true intake speed can be viewed by right clicking the air intake and subtracting 100 from the indicated Air Speed.
    
  • These values can be multiplied by the current air pressure measured by the PresMat Barometer to get the intake air flow at higher altitudes. (Stock part, [wiki page])
    

  I hope this information will help some of you to better understand the different jet engines in KSP and to fly SSTOs at even higher speeds before switching to rocket power.

Edited July 13, 2014 by Iffn

[Justin Kerbice]

Good work , thanks.

Findings on ISP and altitude:

• The indicated Specific Impulse (=ISP) is only relative to altitude
  
• According to the KSP wiki, the ISP is calculated by dividing the engine thrust by the fuel consumption. however, there seem to be some problems with the current version as this picture shows.
  Fuel flow: 0.01348U, Thrust: 0.0kN, Specific Impulse: 1133.0s
  

Have you look deeper into how KSP really work vs real life ? ISP on wiki seems to be generic formula, not that's used in game as engines are defined by one or both of these two tables (TurboJet Engine example):

atmosphereCurve

{

key = 0 1200

key = 0.3 2500

key = 1 800

}

velocityCurve

{

key = 0 0.5 0 0

key = 1000 1 0 0

key = 2000 0.5 0 0

key = 2400 0 0 0

}

1st is ISP at a given altitude and 2nd is thrust at a given speed.

velocityCurve is used for jet/air breathing engines only.

Some explanations are here

[Iffn]

I've replaced the "Indicated ISP relative to altitude" part with a new "Fuel flow relative to altitude" graph, since there seems to be no direct connection between the indicated ISP and fuel flow.

@Justin Kerbice,

Thank you for the information. I assume those are the values KSP uses to create the shape of those graphs above. It would be interesting to know what the exact formulas are though. This would enable players to calculate, at what speed an altitude they need to fly their SSTO in jet mode without the engine stalling.

[eempc]

Very useful information, thanks for sharing this!

[Pondafarr]

Given 2 ram intakes (oriented towards airflow) and two circular intakes (at 90deg), how would I determine optimal altitude to switch from basic to turbo engines?

From one of these graphs, it appears that the turbo gets better ISP than the basic right about 250m/s, but I don't see anything about altitude?

[Pecan]

Fuel flow/altitude chart: "Compared to the other jet engines, the basic jet is very efficient when flying below 5000m."

So, presumably, the only time to bother with basic jets is flying <250m/s and <5km alt. The 'trainer' plane in my tutorial uses one, but mostly just to show how limited it is, otherwise I'd always get height above 5km in the first half-minute or so of a flight. The only other time I'll be anywhere in those constraints is when landing, in which case the engine won't be doing much more than idling anyway.

[Iffn]

Pecan is correct, a plane using Basic Jets is pretty limited above 250m/s and 10km altitude.

However, they are very useful when used as first stage engines for a ballistic rockets.

A single Basic Jet can accelerate a 13 tonne rocket to around 250m/s vertical speed and, when using 2 radial air intakes, an altitude of almost 30km before stalling. In my test, it only used around 25 units of fuel to do so, which makes it extreme efficient compared to normal rockets. A second stage using conventional rockets should then be able to bring the payload efficiently into orbit.

Compared to that, a Turbo Jet can only lift around 11 tonnes of rocket and uses a lot more fuel at low altitudes. While it would be theoretically able to accelerate to a higher speed due to the increasing thrust at higher speeds, vertically ascending rockets usually reach the stalling altitude before this comes into effect.

Finding added to the original post

Edited July 13, 2014 by Iffn

[Dispatcher]

Thanks for these graphs! Also for the posts & replies, everyone. I've experimented with mixed jet boosting systems, which I've found interesting. Using KER, the turbojet appears to show a higher dV than the basic jet or the RAPIER in air mode. Same with the TWR.

[einsteiner]

Would you mind posting or messaging me the raw data your excellent plots are made from? I would like to calculate the altitude/speed flameout limit for various engine configurations.

[Iffn]

Sure einsteiner, I put it on my Dropbox to download, created in Microsoft Excel 2013. [Link]

[O-Doc]

I use basics for most of my heavy lifter SSTOs. They have another property you have not discovered. They are faster below 10km than turbos. Take a plane w/ TWR > 1 and put one basic on it and observe your top speed at ground level. Now take those same parts and put two turbos on it. Now try three and four turbos. You'll notice a single basic will achieve a higher top speed. This an exceptional property for building up vertical speed at the start of your ascent profile. One basic with one turbo will always carry a 25T craft with a nuke into space. That's around 5km's dV@LKO, or more if you are spamming intakes. I scale that principle up for heavy lifters and use that engine combination as my design baseline.

[Iffn]

Simplified, an aircraft flying horizontally in KSP can accelerate until the engine thrust equals the drag. The Basic Jet has a higher thrust than the Turbo Jet below ~230m/s as seen on my 'Engine thrust relative to speed' diagram. An aircraft is usually slower at lower altitudes due to the higher air density resulting in a higher drag.

So I assume your aircraft is faster with Basic Jets because it is slower than 230m/s below 10km.

Edited July 25, 2014 by Iffn

[O-Doc]

Correct. However, my craft are flying vertically. Of course, everything is horizontal in KSP physics

I have done alot of experiments with engine setups and thrust does not stack which means that in KSP's simplistic drag model, thrust does not equal drag. But somehow, SQUAD has manged to reproduce pretty well effect of nozzle velocity in relation to static thrust in their engine speed curves.

[Wanderfound]

Anyone got an updated version of these graphs for the nerfed engines in the new FAR?

[RedJimi]

Any plans to renew this to accommodate for the current version, 1.0.2?