HOVER TESTING OF A 1/6^THMACH-SCALE CH-47D BLADE WITH INTEGRAL TWIST ACTUATION

John P. Rodgers, Senior Engineer, Midé Technology Corp., Cambridge, MA, 02142, (617) 252-0660, research performed while Research Assistant, Active Materials and Structures Laboratory, MIT

Nesbitt W. Hagood, Associate Professor, Active Materials and Structures Laboratory, MIT 37-315, 77 Massachusetts Avenue, Cambridge, MA 02139, (617) 253-2738

Presented at the 9th International Conference on Adaptive Structures and Technology, Cambridge, MA, 1998

Keywords: Helicopter, Rotor, Individual blade control, Active fiber composite, Higher harmonic control, Piezoelectric, Servo-flap, Vibrations

ABSTRACT

An integral twist-actuated rotor blade has been developed for helicopter Individual Blade Control (IBC) applications. A 1/6th Mach scale CH-47D blade was designed, fabricated, and tested in hover at the MIT Hover Test Stand Facility. The design incorporates Active Fiber Composite (AFC) actuators within the composite spar to induce a distributed twisting moment along the span. The anisotropic actuators are oriented at 45° to the blade axis to maximize the shear stresses generated. The twist actuation performance of the active blade was evaluated over a range of rotor speeds, actuation frequencies, and blade loading conditions in hover. Transfer function data were collected from input voltage to blade twist and induced vertical hub shear. Changing test conditions had little affect on the measured performance, though blade flapping mode dynamics had a significant effect. This project successfully demonstrated the effectiveness of integral twist actuation in Mach-scale hover testing, supporting the need for further investigation of the concept for IBC.

INTRODUCTION

Helicopter rotor blades experience significant vibration and noise levels as a result of variations in rotor blade aerodynamic loads with blade azimuth angle. Actively controlled rotor blades are currently being investigated as a means of reducing the detrimental vibrations and noise. Higher Harmonic Control (HHC) has been applied using a swashplate to control the pitch of the blades (Shaw, 1985). A more advantageous approach is Individual Blade Control (IBC), utilizing blade mounted actuators to control blade pitch. Providing the blade designer with active twist distribution control offers other possible advantages including increased induced power efficiency, payload, and cruise speed.

The majority of current research in the field of HHC has focused on trailing edge flap actuation as the means of pitch control (Prechtl and Hall, 1998; Bernhard and Chopra, 1997; Straub and Hassan, 1996). Several other studies have also investigated IBC concepts in hover testing (Millot and Friedmann, 1992; Jacklin et al, 1994; Fulton and Ormiston, 1998). Another concept which offers a simpler actuation scheme is integral twist actuation. While flap concepts rely on aerodynamics to induce a twisting moment, direct twist actuation of the rotor blade uses the actuators distributed along the span of the blade to deform the relatively stiff blade structure. The direct twist concept has been tested using monolithic piezoceramic actuators in a Froude-scale system (Chen and Chopra, 1997).

The integral twist concept described in this work offers an alternative approach for direct twist actuation. Active Fiber Composites (AFC) may be integrated within a composite rotor blade to induce a distributed twisting moment as depicted in Figure 1. The AFC actuators are implemented in the form of active plies within the composite spar of the rotor blade. These anisotropic active plies can be oriented at a 45° angle to the blade span in order to induce shear stresses and a distributed twisting moment along the blade. The result is a twisting of the blade and a control input for IBC.

The AFC offers a number advantages over trailing edge flap and other active twist rotor concepts. The AFC is an anisotropic, conformable actuator which can be integrated with a passive structure. The actuators are distributed throughout the structure providing redundancy in operation. The active blade requires no articulating components, thus eliminating the need for an efficient actuation amplification device and complex flap driving mechanisms. The integral concept does not increase the profile drag of the blade unlike servoflap concepts. In addition, the integral blade can be designed to allow for both bend and twist control as well as additional spanwise degrees of freedom. A major challenge with the integral blade is to develop a design with sufficient authority to twist the relatively stiff blade structure without sacrificing the structural integrity of the system.

The actuator consists of continuous, aligned, electroceramic fibers in an epoxy-based matrix which is sandwiched between two layers of polyimid film which have a conductive inner surface for applying the driving electric field. More recently, the performance of the AFC system was greatly improved with the change to an Interdigitated Electrode (IDE) pattern, which orients the applied electric field along the active fibers, enabling the use of the primary piezoelectric effect (Bent and Hagood, 1995). A diagram of the actuator is shown in Figure 2.

The primary objective of this research is to develop a twist-actuated rotor blade for helicopter vibration control. This project will demonstrate the integral twist actuation concept and its advantages over other actuation methods in Mach-scale hover testing. A secondary goal of the project is to demonstrate the effectiveness of active fiber composites in a large-scale application with a realistic loading environment.

Several previous studies have formed the background for the current research. In order to evaluate the feasibility of the integral twist actuation concept, a modified Rehfield-type single-cell composite beam model was developed (du Plessis, 1995). Interdigitated electrode piezoelectric fiber composite actuators were selected and used in a 1/16^th scale benchtop twist demonstration. A more advanced rotor dynamic analysis of the integral actuation scheme was later performed (Derham, 1996). This included a systems-level cost-benefit analysis and demonstrated the potential impact of the integral actuation concept. The design of the integral blade and the development of the actuators, including rigorous structural integrity testing, has been previously described (Rodgers et al, 1997). Results from the testing of half-span blade sections and preliminary hover data have also been presented (Rodgers and Hagood, 1998). A complete report of the integral blade project has also been completed (Rodgers and Hagood,1998b). This paper will highlight some of the key findings in the hover testing and the analysis of the results.

INTEGRAL BLADE DESIGN

The approach used to design the active rotor blade was to select an existing rotor blade design as a baseline configuration and then modify it to incorporate active plies. The baseline configuration selected was a 1/6th Mach-scale CH-47D blade developed for wind-tunnel testing at Boeing. This configuration was selected because it was an appropriate size for the anticipated testing and because of the significant experimental data and manufacturing experience available at Boeing for this model system. Additional details of the design and the models used can be found in previous work (Rodgers et al, 1997).

The model CH-47D blade, shown in Figure 3, has a span of 60.619 measured from the center of rotation) and a chord of 5.388 inches. It is designed to be used on a fully articulated hub with a single pin located at 0.15R (15% radius). The blade has built-in 12° linear twist and tapers from a VR7 airfoil at 0.85R to a VR8 airfoil at the tip. The primary structural member of the model blade is a co-cured "D" spar , while the aft fairing is added in a secondary cure. For the active blade design, several of the materials used were updated to reflect current best practices. E-glass fabric, S-glass unidirectional, and IM7 unidirectional tapes are used with a Rohacell foam core.

The baseline model blade is Mach scaled from the Boeing CH-47D with a geometric scaling of 1:5.939 (approximately 1/6th scale). Mach scaling was selected to provide actuator performance data which would be the most applicable to the development of actuators for the full scale blade. The model blade is aerodynamically similar to the full scale blade. The mass distribution and torsional stiffness properties were allowed to vary in order to achieve the design goals for twist, while the other stiffness values were maintained. The Lock number of the active model blade as tested was 9.32 or 99.8% of the full scale CH-47D blade.

A configuration with three active plies in the spar laminate was selected for the integral blade. The active plies are uniformly distributed between 0.27R and 0.95R and are divided into 7 spanwise AFC segments or packs. The composite lay-up was designed to meet the stiffness and inertial requirements as well as strength requirements for 3g maneuver loads. The basic spar laminate lay-up consists of an inner graphite unidirectional 0° ply and a total of 4 glass fiber plies sandwiching the three active plies. The second of the glass plies, located between the outer pair of active plies, is a unidirectional S-glass, while the other three are ±45° E-glass fabric plies. A flexible circuit located along the web of the spar is used to deliver power independently to each of the 42 packs. The AFC packs and the blade were fabricated in the Active Materials and Structures Lab at MIT.

HOVER TESTING

Spin testing of the integral blade was performed at the MIT Hover Test Stand Facility. A passive blade of similar geometry was used as a balance in a two-bladed configuration. The blade is connected to the outboard end of the pitch shaft through a vertical pin located at 0.15R. The inboard end of the pitch shaft connects to the hub through a horizontal pin at 0.028R. The pitch shaft assemblies allow for manual pitch adjustments between tests. Below the hub, a 6-axis JR3 load cell interfaces with the main shaft and a leads shell houses all sensor lead connections. The stand was designed for testing the 1/6 th

Mach-scale CH-47D blade system at 1336 RPM. A slipring with 138 sensor channels is used for transferring data from the rotating frame load cell and internal blade-mounted strain gages.

The primary sensors used in the analysis of the twist actuation performance are the vertical hub shear (lift) and torsional strain measurements. The vertical hub shear is presented in the form of the nondimensionalized coefficient of thrust, C_T,

where F_z is the thrust, r is the air density, A is the rotor disk area, W is the rotor speed, and R is the radius (Johnson, 1980). Figure 4 shows the entire blade in position on the hover test stand.

An asymmetry in the testing room was found to cause a nonuniform inflow into the rotor disk during testing. This manifested as 2/rev and 4/rev disturbances in the thrust data channel. In order to improve the efficiency and confidence in the data collection, a real-time Fourier coefficient analysis technique was used. Transfer function (frequency response) data were collected from the voltage input to the actuators to each of the various sensors over a range of actuation frequencies from 10 to 150 Hz. At a given single frequency in the range, data were continuously averaged until the desired accuracy of 5% was achieved with 90% confidence, or a maximum number of averages of 5000 was reached.

Several parameters were varied in order to evaluate the performance of the twist actuation. Data were collected over a range of rotor speeds from 200 to 1336 RPM and for collective pitch angles of 0°, 4°, and 8°. The variation in the rotor speeds allows for the evaluation of the effects of the centrifugal loading and airspeed on the twist performance. Varying the collective pitch shows the effect of varying the blade loading from 0 to 0.07.

Two manufacturing flaws limited the twist actuation capability of the integral blade during the testing. First, several electrical connections to the actuators failed during the secondary fairing cure of the blade resulting in the disconnecting of 11 of 42 packs. Secondly, a cyanoacrylate adhesive used in the internal strain gages outgassed during the spar cure, resulting in the formation of large core voids and localized delaminations in the spar laminate. This damage is believed to have caused the electrical failure of several additional packs at higher voltages. Therefore, the operating voltage of the blade was limited to 2000 Vpp or 50% of the design level. The loss of the actuators resulted in an overall reduction in authority from design targets. However, the authority was sufficient to obtain significant data on the performance of the system as well as for correlation with model predictions.

HOVER DATA

This section will present several data sets collected during the hover testing of the integral blade. The first data set collected on the integral blade included the measurement of the twist actuation from 0 RPM to 1336 RPM at 8° collective pitch. The induced vertical hub shear was also recorded. This data was presented in a preliminary report on this work (Rodgers and Hagood, 1998) and is not repeated here. However, the quasi-steady actuator performance measured during this test segment will be discussed in the analysis discussion of the next section. Since the maximum number of actuators were functional during this test, the measured performance was also a maximum. In testing at 1920 Vpp and 10 Hz actuation at 1336 RPM, a

torsional strain amplitude of approximately 40 microstrain and an induced thrust amplitude of 12 lbs. were measured. These data were collected with 30 of 42 packs operational. Attempts to increase the operating voltage resulted in significant electrical failures attributed to the manufacturing defects described in the previous section. As a result, the transfer function data presented were collected with approximately 18 packs operating.

Rpm Sweep

Transfer function data from voltage in to the actuators to the torsional strain measured at 0.9R and to the coefficient of thrust are presented in Figures 5 and 6, respectively.

The data were collected for a range of rotor speeds at 8° collective pitch. The torsional strain data provide a direct measure of the twist actuation capability during the tests. The transfer function data show the dynamic response to the first torsional mode near 110 Hz (5/rev). The quasi-steady torsional response shows no significant change during changes in rotor speed. However, the torsional resonance becomes critically damped at full speed.

The transfer function from voltage to the induced change in the coefficient of thrust, presented in Figure 6, also shows significant system dynamic effects. In this case, the first three elastic flapping mode resonances are clearly visible. With increasing rotor speed, the resonant frequencies increase as expected and aerodynamic damping effects increase as well. At 1336 RPM, the amplitude of the induced coefficient of thrust is fairly constant through 4/rev. The quasi-steady amplitude of the response remains fairly constant (approx. 1.3x10^-4) with changes in rotor speed.

Collective Sweep

Data for the collective sweep are presented in Figures 7 and 8 for 1336 RPM and 0°, 4°, and 8° collective pitch. Transfer function data from voltage to the torsional strain at 0.8R (Figure 7) show a uniform offset in the response at different collectives. However, the variation does not correlate with blade loading, but instead illustrates the effect of a two actuator failures. The data set collected at 8°, the maximum blade loading, was collected first and thus demonstrates the greatest response. The next data set was collected at 0° and thus shows a small reduction in performance resulting from a single failure. The final data set collected at 4° shows a similar decline resulting from a second failure. Blade loading was not found to decrease the effective twist actuation.

Similar trends are apparent to a lesser extent in the coefficient of thrust data presented in Figure 8. The greatest magnitude of response was measured at the greatest blade loading condition. The system dynamics remain fairly constant between the tests.

Confidence

As a result of the aforementioned nonuniform inflow effects from the testing room asymmetry, the confidence in the coefficient of thrust data was reduced during high blade loading tests. This is illustrated in Figures 9 and 10.

Figure 9 shows the transfer function data collected at 1336 RPM and 0° collective. In this case, the inflow is rather small and thus the 2/rev and 4/rev noise peaks are reduced. The 90% confidence intervals plotted around the data illustrate the high accuracy below 4/rev where the response is greater than 10^-5. In contrast, the 90% confidence bands around the data in Figure 10 show a significantly reduced accuracy. While the accuracy of these data are poor, the trends in the data collected during the rotor speed and collective sweeps clearly illustrate the system dynamics and provide a sufficient measure of the integral blade performance. Recent modifications to the MIT Hover Test Stand Facility have reduced the nonuniform inflow effects by an order of magnitude for future tests.

ANALYSIS

In this section, the measured twist actuation and induced vertical hub shear forces will be compared with model predictions. A more detailed analysis including the description of the active beam model can be found in the reference (Rodgers and Hagood, 1998b).

Twist Actuation

Twist actuation data for the full integral blade are presented in Figures 11 and 12. The data are peak-to-peak benchtop tip twist measurements collected at 10 Hz using Keyence LB12/72 laser displacement sensors.

The data are compared with the predicted twist from a modified Rehfield, single-cell beam model of the blade (du Plessis and Hagood, 1995). In this model, additional stiffness terms are lumped on the diagonal elements of the stiffness matrix to approximately account for the unmodeled core and fairing. The model accounts for the packs which were operational during the tests and also matches the experimentally determined torsional stiffness. Data from the proof testing of the AFC packs at 10 Hz were used to estimate the effective high-field, linear piezoelectric properties used in the model (1168 microstrain at 4 kV).

In Figure 11, the nonlinearity of the actuation is evident as function of the applied voltage. This nonlinear behavior is typical of piezoelectrics in which the effective coupling is increased at higher field levels. Since the estimated properties were calculated using high field properties, the linear estimate more accurately models the response at the higher voltages. The model predicts 0.74° at 1920 Vpp which is 5% below the measured 0.78°. Figure 12 illustrates the spanwise variation in the beam model, accounting for variations in lay-up and in the number of active packs. The data are plotted as a line since the twist was only measured at the tip. In the model, the blade was discretized according to spanwise pack groups and experimental torsional stiffness estimates, as summarized in Table I.

Spanwise Position (r/R)	Packs Operational	GJ (Nm^2)	Description
0.000-0.273	N/A	250	Root section
0.273-0.338	5	177	Pack group 1
0.338-0.442	4	160	Pack group 2, ply drop
0.442-0.546	3	160	Pack group 3
0.546-0.610	3	160	Pack group 4
0.610-0.714	4	148	Pack group 5, ply change
0.714-0.818	6	148	Pack group 6
0.818-0.961	5	116	Pack group 7, taper
0.961-1.000	N/A	100	taper

TABLE I. Model Input Parameters

In general, the predictive capability for the induced twist rate appears to be fairly accurate if the experimentally determined torsional stiffness and pack properties are considered. The current method of using an estimate of the average pack free strain properties has been successful for predicting the high field level performance of the blade.

Induced Vertical Hub Shear

A simple aerodynamic model was used to obtain a rough estimate of the induced thrust for a 0.4° tip twist at 1336 RPM and 8° collective pitch. The tip twist was first estimated from a correlation between the strain gage data collected in hover with identical benchtop data collected with tip displacement sensors. Using the steady aerodynamic calculation,

L = qcbC_Laa

where q is the dynamic pressure, b is the span, C_La is the lift curve slope, and a is the angle of attack, the change in lift or thrust was estimated. A linear twist was assumed between 0.27R and 0.95R. Four additional parameters were included in the calculation (Johnson, 1980). An estimate of 5.7 was used for the lift curve slope. Blade element theory was used to include a uniform inflow effect (l=0.045) for a rotor in hover which changed the effective radial velocity distribution slightly. A ground effect factor was also included using an estimate based on the method of images (T/T¥ =1.029). A blade tip loss factor was also incorporated (B=0.97). The resulting thrust prediction is 11.4 lbs. This is 10% less than the measured thrust of 12.6 lbs. amplitude.

CONCLUSIONS

This research demonstrated the effectiveness of the integral twist actuation in Mach-scaled hover tests. Although model blade manufacturing difficulties resulted in twist performance significantly below design levels, previous tests on half-span sections successfully demonstrated twist at full authority. Excellent correlation with model predictions for tip twist was also demonstrated. Correcting the manufacturing problems is expected to result in greater than 2° of peak-to-peak tip twist with the current design. A reduction of the torsional stiffness of the blade to target levels would increase the performance further. Even with the reduced twist, the demonstrated hub load generation at all blade loading conditions and rotor speeds was consistent and near predicted levels. The data suggest that the integral twist actuation concept is suitable for IBC and is worthy of further investigation.

ACKNOWLEDGMENTS

This work was supported by DARPA under the Smart Structures for Rotor Control contract with Dr. Spencer Wu of AFOSR and Dr. Robert Crowe of DARPA as the technical contract monitors. Additional support was received from the ARO with Gary Anderson as the technical contract monitor. The authors acknowledge Douglas B. Weems of Boeing Helicopter (Philadelphia) for contributions to the blade design and analysis. Robert Derham and Richard Bussom also supported this project at Boeing Helicopter (Philadelphia). Special thanks to Eric Prechtl, Paul Bauer, SangJoon Shin, Kamyar Ghandi, Alessandro Pizzochero, Seward Pulitzer, Jaymee Johnson, and Margee Best for their assistance with this project at MIT.

REFERENCES

Bent, A. A. and N. W. Hagood, “Improved Performance in Piezoelectric Fiber Composites using Interdigitated Electrodes,” SPIE Paper No. 2441-50, Proceedings of the 1995 North American Conference on Smart Structures and Materials, San Diego, CA, 1995.

Bernhard, A. P. F. and I. Chopra, "Development of a Smart Moving Blade Tip and an Active Twist Rotor Blade Driven by a Piezo-Induced Bending-Torsion Coupled Beam", SPIE’s 1997 Symposium on Smart Structures and Materials, San Diego, CA, 1997.

Chen, Peter C. and I. Chopra, "Hover Testing of Smart Rotor with Induced-Strain Actuation of Blade Twist", AIAA Journal 35(1), January, 1997.

Derham, Robert C. and Nesbitt W. Hagood, “Rotor Design Using Smart Materials to Actively Twist Blades”, Proceedings of the American Helicopter Society 52nd Annual Forum, Washington, DC, 1996.

du Plessis, A. J. and N. W. Hagood, “Performance Investigation of Twist Actuated Single Cell Composite Beams for Helicopter Blade Control,” 6th International Conference on Adaptive Structures Technology, Key West, FL, 1995.

Fulton, Mark V., and Ormiston, Robert A., "Small-Scale Rotor Experiments with On-Blade Elevons to Reduce Blade Vibratory Loads in Forward Flight," AHS 54th Annual Forum, Washington, DC, 1998.

Jacklin, S. A, Nguyen, K. Q., Blaas, A., and Richter, P., "Full-Scale Wind Tunnel Test of a Helicopter Individual Blade Control System," Proceedings of the 50th Annual Forum of the American Helicopter Society, Washington, DC, May 11-13, 1994.

Johnson, W., Helicopter Theory, Princeton U. Press, 1980.

Millott, T. and Friedmann, P., "Vibration Reduction in Helicopter Rotors Using an Active Control Surface Located on the Blade," AIAA Paper 92-2451-CP, Proceedings of the 33rd AIAA/ASME/AHS/ASC Structures, Structural Dynamics and Materials Conference, Dallas, TX, April 13-15, 1992.

Prechtl, E. F. and S. R. Hall, "An X-Frame Actuator Servo-Flap Actuation System for Rotor Control", Proceedings of SPIE’s 1998 Symposium on Smart Structures and Materials, San Diego, CA, 1998.

Rodgers, J. P., Hagood, N. W., and Douglas B. Weems, “Design and Manufacture of an Integral Twist-Actuated Rotor Blade”, AIAA Paper No. 97-1264, presented at the 38th AIAA/ASME/AHS Adaptive Structures Forum, Kissimmee, FL, 1997.

Rodgers, J. P., and N. W. Hagood, “Preliminary Mach-Scale Hover Testing of an Integral Twist-Actuated Rotor Blade”, SPIE Paper 3329-32, Proceedings of SPIE’s 1998 Symposium on Smart Materials and Structures, San Diego, CA, March 1998.

Rodgers, J. P., and N. W. Hagood, “Development of an Integral Twist-Actuated Rotor Blade for Individual Blade Control”, Active Materials and Structures Laboratory Report #98-6, MIT, 1998.

Shaw, J., Albion, N., Hanker, E. J., and Teal, R. S., "Higher Harmonic Control: Wind Tunnel Demonstration of Fully Effective Vibratory Hub Force Suppression," J. AHS 34(1), pp. 14-25, January 1989.

Straub, Friedrich K. and Ahmed A. Hassan, "Aeromechanic Considerations in the Design of a Rotor with Smart Material Actuated Trailing Edge Flaps", Proceedings of the AHS 52nd Annual Forum, June 1996.