Category Helicopter Performance, Stability, and Control

The tail rotor can be installed either as a pusher or as a tractor. Test results given in Chapter 4 show that the pusher is more effective since it is less interfered with by the vertical fin. Nevertheless, there may be reasons for using the tractor installation. For example, on the Sikorsky UH-60, the designers decided to tilt the tail rotor shaft to take advantage of a vertical component of tail rotor thrust. To obtain adequate clearances as a pusher, the tail rotor shaft would have had to be very long. As a tractor, the installation was much lighter and more compact.

The longitudinal and vertical location of the tail rotor with respect to the main rotor affects the mutual interference in hover, sideward flight, and forward flight. Reference.10.7 discusses test results concerning these interferences in hover. As might be expected, these tests show minimum interferences when the gap between the main rotor and tail rotor discs is large.

Reference 10.8 cites experience indicating that left sideward flight was smoother and the Dutch roll made more stable when the tail rotor was raised on the Hughes AH-64 during its flight test development phase.

Once a hub design with hinge offset has been selected, the question becomes, "How much?” In most cases, the decision is based on mechanical or structural considerations rather than on performance or handling qualities. There is, however, some advice that can be given to the designer in this matter. Large hinge offsets produce penalties in the form of large, draggy hubs and the increased ability to accidentally roll the helicopter over on the ground. Small offsets, on the other hand, may produce marginal control power especially when the rotor is unloaded

such as in pushovers to low load factors. A minimum offset can be defined to satisfy a flying quality requirement in the form of: "The control power during flight at zero load factor shall be no less than one half (or some other fraction) of the control power in level flight.” A requirement such as this will help avoid "mast bumping” or "droop stop pounding” when the pilot tries to control the helicopter under conditions of low’g’s”.

In Chapter 7, it was shown that the control power is:

where the first term is the hub moment due to hinge offset:

dMM _ 3 e AbpR(ClRya dah 4 R у

If the control power at zero load factor is to be a fraction, l/l + K, of what it is in level flight, then:

( dMC G

da l J о 1 1

dMCG 1 + К ^ + ЮЖ.

da s J icvci dMM

da і

LJ

With a little algebraic manipulation, the required hinge offset ratio is:

/ 2D CT/o lcvd

/ e _ 3 R a

{r/щ k

But both in hover and in forward flight, the coning is approximately:

2 у

“o = ~~ Ct/gicvei, radians

3 a

Thus the required hinge offset ratio can be related to the coning in level flight and to the rotor height above the center of gravity:

The main rotor, of course, is the dominant component of the helicopter and therefore receives most of the designer’s attention. As several projects have shown, however, unfortunate decisions about tail rotors can jeopardize the operational success of the aircraft. Tail rotor parameters of some current helicopters are listed in Appendix B.

For best performance in forward flight, the fuselage angle of attack should correspond to its minimum drag condition. Since the tip path plane angle of attack must be nose-down to produce the proper balance of forces, a fuselage whose longitudinal axis is parallel to the rotor may be generating more drag than it would if designed to fly more level. To minimize fuselage drag, many helicopters are built with nose-down shaft tilt. A typical value. is —5°. Higher values might be desirable for very high speed helicopters, but the designer must consider the effects on hover attitude where the rotor must be horizontal. A helicopter with high shaft tilt would hover with the fuselage tilted far nose up, with possible effects on field of view and seat comfort.

The aerodynamicist and the weights engineer could save themselves considerable later work if they could convince the designer to make his layout drawings with the waterlines perpendicular to the rotor shaft. Drag optimization would then be done with fuselage tilt with respect to the waterlines instead of shaft tilt. This would simplify analyses in which both vertical and longitudinal positions of the center of gravity are used.

Flare Angle

A parameter that is affected by the nose-down shaft (or nose-up fuselage) tilt is the maximum fuselage angle that can be achieved during a landing flare without doing structural damage by striking the ground with the fuselage aft end or the tail rotor. Unless special provisions are made, such as a sturdy tail boom and a shock­absorbing tail landing gear, the maximum shaft flare angle is that reached as the main landing gear and the aft end of the tailboom or tail wheel/skid touch simultaneously. The higher this angle, the better will be the final deceleration capability. A survey of many helicopter side views such as those in Appendix В indicates that designers believe that a flare angle of at least 8° is desirable. Note that some designers—such as those of the Hughes AH-64—have preferred simply to make their tail booms sturdy enough to withstand a significant landing impact.

Most modern American helicopters have rorors that turn so that the advancing blade is on the right, but many European helicopters use the other rotation. At one time, it was thought that a pilot trained in one type would have trouble making the transition to the other because the pedal motion required to compensate for changes in rotor torque would be different. Today, enough pilots can fly both types on the same day that the argument seems no longer valid. The explanation is that the pilot instinctively does the right thing with his feet to hold heading. This is certainly true near the ground with good visibility, where the directional cues are very strong. Only at high altitude or in degraded visibility conditions might he revert, when making power changes, to pedal actions more appropriate to the helicopter he learned to fly than to the one he is now flying. It is thought that this should not be a strong factor and that the designer should feel free to choose the direction of rotation that will result in the lightest transmission, especially if the design also includes mechanical coupling between main and tail rotor pitch to minimize the pilot workload, as is true of many of the recent Sikorsky designs.

Inertia in the main rotor is valuable for two purposes: to prevent the rotor from decelerating too quickly following an unexpected loss of power, and to provide a source of energy for making a landing flare at the end of an autorotational descent. In either case, the level of rotor inertia is more important to a single-engine helicopter than to a multiengine machine, since a sudden power failure of two or more engines is unlikely and there is not the requirement to practice auto­rotational landings continuously as there is with single-engine helicopters.

Flight test experience with a number of single-engine helicopters has led to the conclusion that the rotor kinetic energy stored in the rotor at normal rotor speed should be sufficient to provide the equivalent of at least one and a half seconds of hover time to insure a satisfactory flare capability. It is believed that an equivalent hover time of 0.75 seconds is sufficient for a twin-engine helicopter. The methods for calculating the equivalent hover time are developed in Chapter 5.

The main rotor pitch travels must be adequate to trim the helicopter throughout its flight envelope while leaving suitable margins for maneuvering. The methods of Chapters 3 and 8 can be used to calculate the trim values at the extreme flight conditions. Then these should be increased by 10 to 20 percent. The configuration lists of Appendix В show what designers of some existing helicopters have chosen.

The minimum collective setting must be that which will correspond to 100 percent rotor speed in autorotation at the most extreme combination of gross weight and altitude in the expected flight envelope. This is usually at the lowest gross weight and at sea level, since this is where the CT/a is the lowest.

As a general rule, the maximum collective pitch is required when making a flare from autorotation to allow all the available rotor energy to be used as the rotor slows down.

The cyclic pitch requirement for trim is unsymmetrical. For example, the forward longitudinal cyclic pitch required at the maximum forward speed is much higher than the aft cyclic pitch required at the maximum allowable rearward speed. By recognizing this fact, the designer can save control system weight and space compared to designing for as much aft travel as forward. Similarly, the total lateral cyclic travel an be minimized by biasing toward the high-speed trim value.

The choice of an airfoil—or a series of airfoils—for the main rotor is another exercise in compromise. The ideal airfoil should simultaneously have a high maximum lift coefficient and a high drag divergence Mach number. A study of airfoil characteristics show that these two characteristics do not go together in any one airfoil. If the information in Chapter 6 does not provide enough guidance for this difficult decision, the blade designer will have to rely on the results of later analytical and experimental studies.

Tip Shape

Tip shapes other than square may be selected for a variety of reasons. The most common is the reduction of compressibility effects on the advancing tip at high forward speeds through the use of leading edge sweep. This, of course, is the same reason that jet transports use swept-back wings. There are two compressibility effects that have proved to be significant: the generation of high blade torsion and control loads through Mach tuck (discussed in Chapter 6) and the generation of noise by propagated shock waves. By sweeping the leading edge of the tip, both of these disturbing phenomena can be delayed to forward speeds above the helicopter’s normal speed range.

Another motivation for non-square blade tips is to reduce the blade loads and noise generated when a blade passes through, or close to, the concentrated tip vortex left by a preceding blade. If the tip vortex could be spread out by using a special tip shape, the argument goes, the subsequent interaction should be less violent. At this writing, this remains a reasonable but as yet unverified hypothesis.

Swept-back tips have yet another potential advantage: dynamic twist. At high forward speeds, most blades with twist carry a nonproductive downward load on

the advancing tip. If the tip is swept back, this download acts behind the structural axis and tends to twist the blade nose up, thus reducing the download and its aerodynamic penalty. On the retreating side, the upload on the swept tip twists the blade nose down and alleviates retreating blade stall. The effect also works in hover, where the upload increases the twist, which is beneficial for hover performance.

As with most concepts in helicopter aerodynamics, the use of a swept blade tip is not as straightforward as it might seem. Tests reported in reference 10.6 show a lag in the compressibility effects such that they peak after the blade enters the front half of the disc. This is where a straight blade naturally takes on a swept characteristic from the combination of rotational and forward speed—whereas the swept blade is being aerodynamically unswept by the same effects. Thus it is ■ possible that the swept tip could suffer more from compressibility than the straight one in this region. This possibility has not prevented designers from using swept tips. Figure 10.6 shows several used on contemporary helicopters.

Unfortunately, nothing comes free. Swept tips complicate the structural design of the blade, doubly so if they must be replaceable in the Field when damaged. The cost of designing, testing, and building these blades is significantly higher than for straight blades.

From the earliest days of helicopter development, it was known that blade taper improved hovering performance by unloading the tips to achieve a more uniform

Source: Arcidiacono & Zincone, “Titanium UTTAS Main Rotor Blade,” JAHS 21-2, 1976.

induced velocity distribution across the disc Because of this, many early helicopters had tapered blades. Generally, these blades were built like airplane wings with a spar, many ribs, and a covering of plywood and fabric. When blades began to be made of sheet metal, it was easier to use a constant chord and to rely on twist to make the induced velocity distribution uniform. The next innovation in blade construction was the use of fiberglass or another composite material. For these blades the fabrication engineers had no difficulty with taper, so it о ice again became feasible and has been used on several programs.

Source: Fradenburgh, “Aerodynamic Design of the Sikorsky S-76 Helicopter,” JAHS 24-4, 1979.

Several factors, however, should be considered before deciding whether to use taper. For small helicopters, taper may drive the tip chord to such a small value that the tip airfoil will suffer penalties in drag and in maximum lift coefficient because of low Reynolds numbers. A second consideration has to do with tip weights. Most rotor blades have tip weights either to improve the autorotational capability or to control dynamic characteristics by placing the frequency of the second blade bending mode below three times per revolution (3/rev). Planform taper, especially when combined with thin tip airfoil sections, may not provide enough physical volume for the required weights. Finally, a tapered blade needs more area than a straight blade to produce the thrust required by a high load factor maneuver. Thus it may weigh more, which will subtract from whatever payload advantage resulted from its increased aerodynamic performance in hover and vertical climb. (Some preliminary studies indicate that perhaps inverse taper holds some promise in this regard.)

High blade twist produces good hovering performance and delays retreating blade stall at high forward speeds, but it also produces high vibration in forward flight as it causes the blades to bend as they go from the advancing to the retreating side. The effects on performance in hover and forward flight can be deduced from the discussions in Chapters 1 and 3, but the question of vibration is not easily quantified. At this writing, designers are generally selecting values of main rotor blade twist in the —8° to —14° range. If the dynamicists can find ways of reducing the vibration felt in the aircraft, the aerodynamicists will undoubtedly recommend higher values of twist.

Variations on the conventional twist distribution should be considered for special reasons. High twist that is good for hovering out of ground effect was shown to be too high for efficient hover in ground effect in reference 10.3. In addition, twist that is beneficial in powered flight is detrimental in autorotation. Thus the decision about what twist to use on a new design may depend on its projected use. In Chapter 1 there is a discussion of the tip vortex interference problem in hover. At least one helicopter design has attempted to deal with this by using a special nonlinear twist at the tip, as reported in reference 10.4. This twist distribution is shown in Figure 10.5. Yet another variation has been used to reduce the negative angle of attack in the reverse-flow region at high speed. This is shown in Figure 10.6 and is from reference 10.5.

Once the blade area has been selected, the rotor designer must chose the number of blades. Since, for this choice, the aerodynamic considerations are relatively secondary, he will usually make his choice based on the conflicting recommenda­tions of those specialists concerned with vibration, noise, weight, cost, and operational suitability. Some of their concerns can be organized in terms of the advantages shown below.

Advantages of Low Number of Blades Low rotor wreight Low rotor cost Ease of folding or storing Low vulnerability to combat damage High blade torsional stiffness

If these considerations do not lead to an immediate decision, the aerodynamicist might be asked for some recommendations. Unfortunately, these will also be conflicting; both as applied to hover and to forward flight.

As discussed in Chapter 1, a performance penalty in hover is caused by blade – vortex interaction. For a small number of blades, this should be less because the tip

vortex has a chance to get out of the way before the next blade passes by. This should provide a slight power advantage in hover. This advantage, however, might be lost if the small number of blades results in such a stubby shape that the tip loss region is a significant portion of the blade area. On the other hand, splitting the required blade area into many small chord blades might introduce Reynolds number penalties on maximum lift coefficient and skin friction if the chords are less than about 5 inches.

Some design teams limit the minimum aspect ratio of the blade outboard of the flapping hinge to about 12. This is done to ensure that the natural frequency of the second flapping mode will be below 3 cycles/revolution.

In forward flight, a potential advantage of a low number of blades—such as two—is in lower hub drag, as indicated by the hub drag survey presented in Chapter 4. Against this advantage is the disadvantage that the wake left by a rotor with a small number of blades is pulsating, which generates a higher induced power than the smoother wake left by a multibladed rotor. Some basis for this observation will be found in reference 10.2.