Why is the tail-rotor of some helicopters (e.g., Boeing AH-64 Apache) made of non-orthogonal (scissor) blades?

What are the aerodynamic benefits (or any other benefits) compared to the orthogonal cruciform option?

enter image description here

  • 4
    $\begingroup$ The four-bladed fans once used in cars' radiators were often built so that the blades weren't orthogonal. The reason was (if my memory doesn't betray me...) that they were less noisy than orthogonal ones... $\endgroup$
    – xxavier
    Commented Aug 28, 2017 at 21:14
  • 3
    $\begingroup$ In the car and motorcycle community there is a thing about the sound of uneven firing angles, ranging from 45/60/95/90 degree V engines to 270/315 degree parallel twin to Yamaha's "none of the firing intervals between any of the 4 cylinders are the same" configuration. I believe same theory applies. $\endgroup$ Commented Aug 28, 2017 at 23:12
  • 1
    $\begingroup$ @xxavier: Apparently, the same is true for the cooling fans in some high-end laptops, although these have a lot more than four blades. $\endgroup$
    – Vikki
    Commented Jun 11, 2018 at 15:33

3 Answers 3


enter image description here

The comments and answers already say it's for noise reduction. What's more intriguing is the how:

The blades are not coplanar (image).

The tail-rotor of the Apache rotates clockwise (video) when viewed from the port side. The nearest blades (nearest plane) to the viewer are the leading blades of each group. A group is two close blades regardless of plane, e.g., 1 and 2 as labelled above are a group, 1 is leading.

The thrust of the tail-rotor is being directed away from the viewer (port to starboard), which means the viewer is looking at the pressure side of the rotor.

Below on the left is what it looks like from behind the Apache. This arrangement allows the trailing blades (2 and 4) to miss the wake from the ones ahead of- (and closest to-) them.

enter image description here
(Own work) Left: Apache; Right: coplanar normal spacing. View is from behind and thrust is left to right.

This L scissor configuration, as it is known, where the lower blade is leading yields more thrust due to the better interaction.

Based on extrapolation of the test results and calculations, the increase in thrust over a conventional four-bladed rotor of the same dimensions is about 10%. Rozhdestvensky attributes Bell's failure to come to similar conclusions to their use of a very small model.

enter image description here

And quieter, too

Noise measurements during flyovers were made with a Havoc having the four-bladed scissors tail rotor and the three-bladed tail rotor from the Hind. The two rotors made about the same noise at low speed but at high speed the scissors rotor was quieter. The primary reduction was in the "broad-band" frequency range from 600 Hz to 2000 Hz. Much of the noise at these frequencies is due to the effects of the tail rotor working in the confused wake of the main rotor. The scissors rotor apparently acts more like a two-bladed rotor in this situation.

(Helicopter Aerodynamics Volume II, p. 72.)

  • $\begingroup$ Nice input, looking at the picture found by Koyovis (and rudder trailing edge bending as a clue), i wonder if AoA (let's say in stationary flight) hasn't an opposite value (as the one in your bottom sketch), which means the blade ahead deflects air away from the following blade, and not towards it. $\endgroup$
    – user21228
    Commented Aug 29, 2017 at 15:10
  • $\begingroup$ (I assume vertical stabiliser trailing edge deflection allows zero tail rotor AoA in forward cruise flight, which reduces noise emissions even more) $\endgroup$
    – user21228
    Commented Aug 29, 2017 at 15:12
  • $\begingroup$ @qqjkztd an opposite value (as the one in your bottom sketch), which means the blade ahead deflects air away from the following blade, and not towards it. I'm not following. $\endgroup$
    – TomMcW
    Commented Aug 29, 2017 at 23:30
  • $\begingroup$ @TomMcW This illustration explains my observation. $\endgroup$
    – user21228
    Commented Aug 30, 2017 at 9:02
  • $\begingroup$ @ymb1 nice reference thank you, focusing more on the 10% thrust increase (rather than on noise reduction) I wonder if such tests have also been made on airplane fixed pitch propellers. (non coplanarity + scissor config) $\endgroup$
    – user21228
    Commented Aug 31, 2017 at 9:31

enter image description hereImage source

The tail rotor of the Apache consists of two 2-bladed teetering rotors, referred to as scissor rotors. This unusual configuration was first implemented by Hughes Helicopters in the late 1960s for reducing the noise of the OH-6 helicopter. The OH-6 had a single twin blade teetering tail rotor, which Hughes wanted to rotate slower for reducing noise levels.

In order to maintain the same thrust a second identical rotor was mounted, which could not be placed at 90 degrees due to interference. The resulting twin teetering scissors rotor had a good noise profile - our ears are less sensitive to lower frequencies, and the scissor rotor dominant frequency is two per revolution instead of 4 per revolution for the 90°.

The Hughes team minimised new inventions, and implemented this solution for the Advanced Attack Helicopter competition of 1972, which they won. The blades in the Apache tail rotor are mounted at 55 degrees, reducing the harmonics and the pressure levels in the sound profile.

Source (page 69)

From this reference:

The four-bladed tail rotor is unusual in that, rather than the blades being evenly spaced at 90° intervals, the blades are spaced at 55° and 125° angles. This allows for significant reductions in noise.

This Mindef reference describes experiences with the original configuration of the AH-64, which had a T-tail.


I believe its to reduce the noise, here is a partial info

The complex flowfield which is associated with a rotor wake gives rise to the multitude of aerodynamic interactions that may occur during rotorcraft operation. These interactions may give rise to undesirable noise and lead to an unacceptable performance degradation,


You must log in to answer this question.