I've been considering increasing efficiency of my early launch stages by adding jet engines. I found an old thread discussing the idea, and one post struck me as "make-or-break" of my idea:

IMO, using jets as a first stage just leads to forming bad habits. It only works because the current drag model doesn't work. As soon as there is a realistic representation of drag in the game, jet engine first stages won't work.

There's another, that explained why it worked pretty well:

Also, try not to use any intakes except the ram intakes, but especially not the radials or double-especially the engine nacelles. They all have intakeair quantities and drag determined by the size of the intake, except that there's a cap on how high the drag will rise...and the cap's the same on all the different types of intakes. They all max out at '2'. Not sure if that's the drag coefficient or what(If it is, the Nacelle and the Radial are both far heavier as well, so that'd be EVEN WORSE!) Anyway, what happens is once you get some speed up, all the intakes end up with the drag capped, and yet they all produce less intakeair than the ram intakes. The radials don't produce much, and the engine nacelles barely produce any at all, and yet end up with massive drag anyway.

My idea is very simple, increasing the complexity only marginally: I have my Asparagus first stage with "thrust plate". I can simply strap a couple jet engines with their air intakes right to the outer side of a tank powering given stage rocket, no extra decouplers, fuel lines, tanks.

They'd feed right off the same tank as the rocket, they'd switch off when "their" stage runs out of fuel, and they'd be discarded along with "their" stage. And for the sputter-out problem with lack of intake air, I'd simply not attach any to the stages that reach atmosphere too thin for jets. The only actual complexity-inducing change would be starting the jet engines and letting them spin up first before releasing the clamps and starting the rocket engines on the launchpad.

But the make-or-break thing is air drag. If the jet engines (and their intakes) introduce so much drag they would hobble the rockets instead of aiding them, this won't work. And since the entire air drag model changed with 1.0, I don't know if the "2" cap is still valid. The Air intake page of KSP wiki is scarce in details.

Since I'm still quite a bit away from the ram air intake in the science tree, I'd be most interested in drag of the Circular Intake, but if you can provide a more complex analysis (other intakes, jet engines themselves etc) that would be most welcome.

  • Life verified some of my ideas. I can't attach the engine directly, but I can attach a tiny tank (FL-T100 with 0 oxidizer, some 20% fuel); the idea of it switching off by itself goes out the window but I can still jettison the running engines (and the effects are quite spectacular as 6 angry little tanks start buzzing around). The thrust gain in the early flight (below 10000m) is immense; until 20000 still non-zero.
    – SF.
    May 29, 2015 at 9:47

2 Answers 2


Giving a blanket answer to "what's the drag value" for any part in KSP is difficult past 1.0 because drag does (and should) vary based on angle of attack (AoA). It used to be that drag was static based on AoA, so that "caps at 2" answer was valid. It also got changed several times after release, so that's probably why there wasn't reliable data out when you posted (late may).

According to this thread (which should be in 1.0.4), the axial intakes (one that sit at the 'end' of a stack, like the circular intake) seem to universally reduce drag, even when open, and even when compared to some nose cones. So definitely use the best intakes you have on the ends of added nacelles to reduce drag and suck in air.

The drag is going to increase as you add more radial intakes, though. I don't have a good, numerical answer for how much drag they're going to create. Even if there was, though, it would probably still be more effective to just titrate the number of radial intakes to effect, meaning keep adding them until you don't think the increase in intakes is effectively increasing the lifter's performance based on whatever parameters you want (power, duration of jet power, cost, etc).

Hope that helps.


As of 1.12.5:

Look up the actual drag in the debug menu

Hit ALT+F12 to open debug menu. Go to Physics > Aero.

Check 'Display Aero Data GUI' - this will show total actual drag, lift, thrust and other useful data like angle of attack in flight.

Check 'Display Aero Data in Action Menus' - this will show theoretical drag of each part in SPH and VAB and actual drag in flight. Right-click the part of interest.

Drag Cubes

tldr; check the excellent video of Lt_duckweed Kerbal University: The Drag Cube

Kerbal uses a simplified drag model where each part is represented by a cube. The faces of the cube have different area and drag coefficients. These are shown in the UI as the combined value A.Cd. Drag is calculated as

D = A.Cd * (p * v²)/2

D ... Drag

A.Cd ... Combination of Drag Coefficient and Area

p ... air density

v ... velocity

Additional transformations are applied based on which side faces the air stream, angle of attack, Mach number etc.

Now we can interpret the values shown in the action menu.

Drag: actual drag of the part in flight

XP,XN,YP,YN,ZP,ZN: this shows the area and the drag coefficient for each face of the cube.

Of most interest is YP (facing forward), followed by YN (facing backwards), assuming you don't fly with any sideslip angle.

OccA is the area of the part that is not occluded by another part attached to it axially by a node. E.g. a tank attached to the back of an intake of the same radial size will have YP OccA of 0 because it is fully covered by the intake. Try to keep this as small as possible.

WDrg is the weighted drag of this face. Depends on the part. Values below 0.4 - 0.3 are considered slippery and will lead to rapidly decreasing drag once supersonic.

Theoretical values

YP OccA WDrg YN OccA WDrg Aerodynamic Nose Cone 1.21 0.32 0.00 1.21 Shock Cone Intake 1.21 0.30 0.00 1.00 Adjustable Ramp Intake 1.22 0.38 0.01 0.00 Circular Intake 1.21 0.45 0.00 0.00

As you can see the higher tech intakes have better drag coefficients and therefore theoretically less drag. This is not the full truth.

Experimental values

Mach 0.5 0.9 1.0 1.2 1.4 2.0 3.0 5.0 Altitude (m) 200 600 750 1000 1300 2500 5600 18k AoA (deg) -2.1 -0.5 -0.8 -1.1 -1.3 -1.8 -1.9 -1.7

Drag Aerodynamic Nose Cone 0.70 1.76 1.05 1.32 1.43 1.25 1.13 0.32 Shock Cone Intake 0.82 1.96 1.37 1.53 1.65 1.43 1.28 0.74 Adjustable Ramp Intake 2.03 5.37 8.19 11.38 13.34 14.44 15.98 11.00 Circular Intake 1.24 3.45 3.38 3.93 4.32 3.98 3.80 1.31

Flow (U) Shock Cone Intake 112 202 230 259 283 341 316 46 Adjustable Ramp Intake 104 201 230 258 280 321 254 9 Circular Intake 95 166 189 208 216 188 57 4


Only use as many intakes as you need to feed your engines.

Use Shock Cones if available. Adjustable Ramp has good flow at high altitude but very high drag.

Use Aerodynamic Nose Cones to cap off open nodes.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .