Document
N94-13297
A DESIGN CRITERIA FOR INTEGRATED FLIGHT/PROPULSION CONTROL SYSTEMS FOR STOVL FIGHTER AIRCRAFT James A. Franklin Group Leader V/STOL and High Speed Flight Dynamics NASA Ames Research Center Moffett Field, California ABSTRACT HGI hot gas ingestion HUD As part of NASA's program to develop technology head-up display for short takeoff and vertical landing (STOVL) fighter IGE aircraft, control system designs have been developed in-ground effect for a conceptual STOVL aircraft. This aircraft is IMC representative of the class of mixed-flow remote-lift instrument meteorological conditions concepts that was identified as the preferred design LIDS approach by the US/UK STOVL Joint Assessment and lift improvement devices Ranking Team. The control system designs have been OGE evaluated throughout the powered-lift flight envelope on out-of-ground effect Ames Research Center's Vertical Motion Simulator.
PIO Items assessed in the control system evaluation were: pilot-induced oscillation maximum control power used in transition and vertical SCAS flight, control system dynamic response associated with stabilization and command augmentation thrust transfer for attitude control, thrust margin in the system presence of ground effect and hot gas ingestion, and T dynamic thrust response for the engine core. Effects of propulsion system vertical thrust, lb wind, turbulence, and ship airwake disturbances are VC incorporated in the evaluation. Results provide the basis velocity command for a reassessment of existing flying qualities design W criteria applied to STOVL aircraft. gross weight, lb WOD wind over deck NOMENCLATURE AL lift increment referenced to out-of-ground AC attitude command effect conditions, lb AL/T FG gross thrust, lb normalized jet-induced aerodynamic ground effect acceleration due to gravity, ft/sec 2 g (_LjT)" normalized lift increment due to ground effect h landing gear wheel height above ground, ft and hot gas ingestion temperature ratio as a function of wheel height standard deviation Presented at Piloting Vertical Flight Aircraft: A Conference on Flying Qualities and Human Factors, San Francisco, California, 20-23 January, 1993.
INTRODUCTION This paper includes a description of the aircraft, the simulation facility and the experiments which were NASA has been involved in a collaborative program conducted. A summary of the results of these experi- with other government agencies in the United States and ments follows, including suggestions for revision or with the Ministry of Defence of the United Kingdom to modification of existing criteria.
develop technology for supersonic short takeoff and vertical landing (STOVL) aircraft. As a result of this effort, a wide variety of airframe and propulsion system MIXED-FLOW REMOTE-LIZI' AIRCRAFT concepts have been assessed through analytical studies, and critical technical issues have been identified for The design criteria presented in this paper are based investigation (Ref. 1). The preferred design approach on simulation experiments involving a mixed-flow identified by the US/UK STOVL Joint Assessment and remote-lift STOVL aircraft concept (Fig. 1). This concept Ranking Team for the airframe and propulsion system is is specifically referred to as mixed flow vectored thrust known as mixed-flow remote-lift, an example of which is (MFVT) and is described in further detail in Reference 2.
illustrated in Figure 1. This configuration features mixed The aircraft is a single-place, single-engine fighter/attack fan and core flows that can be directed forward or aft to aircraft with supersonic dash capability. It features a generate the lift and thrust forces and to provide (partially blended wing-body configuration with a canted or exclusively) control moments. The propulsion system empennage that provides longitudinal and directional will have forward thrust-producing device(s) that may control. The wing is characterized by a leading edge deflect as well as modulate that thrust component, a sweep of 50* and aspect ratio of 2.12. The propulsion variable area cruise nozzle that may provide thrust system concept uses a turbofan engine where the mixed deflection for pitch and yaw control, and real: lift fan and core streams are either ducted forward to the lift nozzle(s) that provide a thrust component for pitch nozzles or aft to a thrust deflecting cruise nozzle. A control and which may also deflect about the vertical.
ventral nozzle diverts some of the mixed flow to provide Combined with these propulsion components are the pitching moment to counter that of the lift nozzles. Lift aerodynamic surfaces that function during both wing- nozzle thrust can be deflected up to --20 ° about a nominal borne and jet-borne flight. These may include leading rearward cant angle of 8*. The cruise nozzle can be and trailing edge flaps on the wings, canards, ailerons, deflected laterally or vertically -*-20*. In conventional stabilators and rudders for lift and moment control.
flight, the mixed flow is directed aft through the cruise nozzle, whereas in hover it is diverted from the cruise Integration of these flight and propulsion controls nozzle to the forward lift nozzles, with a small portion has been identified as one of the critical technologies to reserved for the ventral nozzle. During transition from be developed for these aircraft. A program has been con- hover to conventional flight, the flow is smoothly ducted to define control concepts that combine the transferred from the lift to the cruise nozzle to provide acceleration.
various aerodynamic and propulsion control effectors with control laws designed to achieve fully satisfactory (Level 1) flying qualities throughout the powered-lift The basic flight control system uses a variety of flight envelope. Furthermore, criteria for the control control effectors: ailerons, a fully deflecting empennage, authority and dynamic response of the individual reaction control system nozzles located in the tail, differential thrust transfer between the lift nozzles and effectors have been explored. The control system designs have been evaluated throughout the powered-lift flight ventral nozzle, longitudinal deflection of lift nozzle envelope on Ames Research Center's Vertical Motion thrust, and vertical and lateral deflection of cruise nozzle Simulator. Included in the control system evaluation thrust. Pitch control is achieved by a combination of were assessments of maximum control power used in symmetric empennage deflection, reaction controls, transition and vertical flight, control system dynamic thrust transfer between the lift and ventral nozzles, and response associated with thrust transfer rates for attitude vertical deflection of the cruise nozzle. Roll control is control, thrust margin in the presence of ground effect produced by the ailerons and by lateral thrust transfer and hot gas ingestion, and dynamic thrust response for (differential lift nozzle thrust). Yaw control is derived the engine core. Effects of wind and turbulence and from the combination of differential empennage airwake disturbances from a ship are incorporated in the deflection, reaction control, and lateral cruise nozzle assessment. The purpose of this paper is to review these deflection. Longitudinal acceleration is achieved through assessments as a basis for possible revisions or exten- thrust transfer between the lift and cruise nozzles and by sions of flying qualities design criteria for this class of deflection of lift nozzle thrust.
aircraft.
Evaluation Tasks and Procedure To achieve the desired level of flying qualities during low-speed flight, stabilization and command The pilot's tasks for evaluation during the simulation augmentation modes were provided in the flight control were those considered the most demanding for precision system as noted in Table 1. During transition, either control of the aircraft---curved decelerating approaches attitude or flightpath SCAS mode was available. Both to hover followed by a vertical landing. For evaluation modes offer rate-command/attitude hold for pitch and roll control and dutch roll damping and turn coordination for purposes, the decelerating approach was initiated under instrument meteorological conditions (IMC) in level the yaw axis. When only the attitude SCAS is selected, the pilot must control thrust magnitude and deflection. flight at 1100 ft and 200 knots in the landing configura- When flightpath SCAS is engaged, the pilot commands tion. Capture of a 3* glide slope ensued, followed by flightpath angle and flightpath acceleration directly; the initiation of a 0.1 g deceleration, a turn to align with the final approach course, and acquisition of a stable hover control system coordinates thrust magnitude and deflection to achieve the desired response. Either the over the hover point. Vertical landings were accom- attitude or velocity SCAS may be selected in hover. Both plished either on a 100 by 200 ft landing zone marked on the airfield's main runway or on a 40 by 70 ft pad on the modes provide pitch and roll attitude command/attitude hold and yaw rate command. With attitude SCAS, the ship's aft deck. Six pilots with V/STOL and powered-lift pilot controls longitudinal and lateral translation through aircraft experience participated in the program.
changes in pitch attitude and bank angle. Thrust is used for height control. For the velocity SCAS, longitudinal, lateral, and vertical velocities are commanded directly. A Experiment Configurations thorough description of the control system is included in Reference 2.
Experiment variables for the decelerating approach and vertical landing included the control system config- A head-up display presented the primary flight uration, control system dynamics, thrust/weight ratio, information for these experiments. The display format jet-induced ground effect and hot-gas ingestion, and was a flightpath centered, pursuit presentation in tran- environmental conditions (wind, turbulence, and sea sition. In hover, the display switched to a format that condition). Both the attitude SCAS and attitude-plus- superimposed vertical and horizontal command and flightpath SCAS were investigated for the decelerating situation information in a pursuit tracking presentation. approach; attitude SCAS and attitude-plus-velocity A complete description of the display is included in SCAS were evaluated for the vertical landing. System Reference 3.
dynamics variations included control system authority, thrust transfer rates, engine core thrust response band- width and acceleration rate. Nine ground effect and SIMULATION EXPERIMENT ingestion profiles representative of a broad range of STOVL aircraft characteristics of lift and temperature Simulation Facility profile as a function of height (four of which were representative of the YAV-8B Harrier with LIDS on The experiments on which these criteria are based and off) were included for both airfield and shipboard landings. Wind conditions for the approach and airfield were conducted on the Vertical Motion Simulator (Fig. 2) at NASA Ames Research Center. This simulator provides landing were calm, 15 knots, and 34 knots, with six degree-of-freedom motion, with large excursions in crosswind components of 30* and 20", respectively, for the latter two wind conditions. Turbulence of 0, 3, and the vertical and longitudinal axes, and acceleration bandwidths in all axes that encompass the bandwidths of 6 ft/sec rms accompanied the respective wind cases.
motion that are expected to be of primary importance to Conditions for shipboard recovery included sea states of 0, 3, and 4 with wind over deck of 15, 27, and 46 knots the pilot in vertical flight tasks. A three-window, com- puter generated image system presented the external view from 30* to port.
to the pilot, which consisted of either an airfield scene or a shipboard scene consisting of a Spruance-class CONTROL POWER destroyer. An overhead optical combining glass projected the HUD for the pilot. Control inceptors consisted of a center stick, rudder pedals, and a left-hand quadrant that Existing design specifications and guidance for contained throttle and thrust vector deflection handles.
pitch, roll, and yaw control power for fixed-wing V/STOL aircraft are contained in References 4 and 5.
Additional information from STOL aircraft experience wind and turbulence for the tasks of transition, airfield that would apply to the V/STOL transition is provided in vertical landing, and shipboard landing is presented in Reference 6. Flight and simulation data on which these Figure 3. For the transition (Fig. 3a), results in calm air, publications are based date back to the late 1960s. Given which are indicative of maneuvering demands, show the present capability for achieving highly augmented that, for attitude command SCAS, pitch control power stability and control characteristics and the necessity for maximums fall within the range considered to be operating in IMC, it is worthwhile to reassess the validity satisfactory in Reference 5 for STOL operations (which of the control power requirements derived from the can be related to the transition phase of this simulation).
earlier data. The results which follow relate to control Two standard deviation levels are well below the power for maneuvering and for suppressing disturbances Reference 5 maximum. Peak values generally equate and have control required for trim removed. These results to 3---40 levels. The influence of turbulence on the are presented to reflect the influence of flight phase, additional control required for disturbance suppression is including effects of control augmentation and magnitude apparent. For rms turbulence of 6 ft/sec (Turb6), a few of atmospheric disturbance. The breakdown related to instances of control usage exceed the maximum flight phase is important not only because of the differ- recommended level of Reference 5. Thus, to cater for ence in the pilot's tasks, but because of the demands maneuvering and the effects of turbulence, a control placed on different control effectors (aerodynamic power of 0.2-0.25 rad/sec 2 would provide for at least surfaces and propulsion system components) that, in turn, 99% of all demands encountered.
place different demands on the aircraft's design. Control power usage is presented in terms of individual maxi- Results for the attitude-plus-flightpath SCAS are mum values (plus or minus about the mean value) for comparable to those for the attitude SCAS, reflecting the each run and an aggregate value of two standard fact that the pilot's pitch control task is similar for the deviations for the ensemble at that condition. For a two systems during transition. The pilot uses pitch Gaussian distribution of frequency of occurrence of attitude changes for flightpath control during the early control use, expected maximum values would be three to stages of the approach, where :a frontside control four times the standard deviation. Two standard deviat- technique is appropriate, as well as to regulate against ions represents a level of control use that is exceeded disturbances arising from wind and turbulence.
4.6% of the time over the ensemble of data runs. Aircraft response specifications of References 4 and 6 were Pitch control during the vertical landing with the translated to measures of control power for direct attitude SCAS (Fig. 3b) shows levels of peak control comparison with the current resulis. 'r'hese Criteria were usage that are less than the requirements of References 4 converted from attitude change in 1 sec using an attitude and 5. The maximum control required was 0.27 rad/sec 2 control bandwidth of 2 rad/sec for an attitude command (3-4o values of 0.14-0.18 rad/sec2). Turbulence response that is critically damped, or using a first-order disturbances did not impose additional demands on response with a time constant appropriate to the axis control authority. Consequently, control authority of being controlled. 0.14-0.27 tad/see 2 would accommodate most of the demands for the attitude SCAS. By comparison, the 3* Maximum demands for pitch control during hover attitude change in 1 sec required by Reference 4 converts and vertical landing are pertinent to sizing requirements to a peak pitch control power of 0.29 rad/sec 2 for a for the aircraft's reaction control system or for thrust 2 rad/sec attitude command bandwidth.
transfer between components of the propulsion system.
Demands for roll control generally size the amount of With the velocity command SCAS, even less pitch thrust transfer required between the lift nozzles. Yaw control is required, reflecting the difference in the pitch demands contribute to sizing of the reaction control control task between the two SCAS configurations. With system. During transhiofi, the requirements on control attitude S(_As alone, control of longitudinal position and sizing would incorporate both the propulsion system and velocity in hover is accomplished through modulation of the aerodynamic effectors.
pitch attitude. When the velocity command system is engaged, control of the longitudinal axis is achieved through deflection of the thrust vector with attitude fixed.
Pitch Control In this case, the vertical landing can require a control authority of 0.17 rad/sec 2, independent of winds and turbulence.
Effect of Flight Phase. A collection of results of pitch control usage for both attitude command and attitude-plus-flightpath command SCAS over a range of the attitude SCAS and considerably more than adequate
Results for hover andverticallanding aboard ship
withattitude command alone (Fig.3c)arecomparable to for control with the velocity command SCAS.
thecriteria of Reference 5 andLevel1 handling values
in Reference 4 (although neither criterion applies to
Roll Control
shipboard operation, butrather to hover out-of-ground
effect). Peak control usage is 0.38tad/see 2 or less, with
3--40levels being0.12-0.16 rad/sec 2.Fortheattitude- Effect of Flight Phase. Roll control use for the different flight phases, SCAS modes, and turbulence is
plus-velocity command system, peakcontroluseis
shown in Figure 4. Maximum roll control use for
approximately two-thirds of thatfor attitude command
maneuvering in calm air during transition (Fig. 4a)
alone, reflecting, asin theairfieldvertical landing, the
substantially exceeds that called for in Reference 5, with
different taskrequired for the pitchaxis.Forneither
peaks of 0.4-0.9 rad/sec 2. However, the 3--40 levels of
system doeswind overdeckseemto influence the
0.3-0.4 rad/sec 2 are more in line with the criteria. For
amount of controlrequired for thelanding. Thus,for
control in the heaviest turbulence, demands for as much
shipboard operations, thecontrol power requirement of
as 1.2 rad/sec 2 occur, although the range is more
References 4 ar/d5 appear appropriate with attitude
SCAS alone, anda requirement for0.2rad/sec 2 should typically 0.6-0.9 rad/sec 2, which is consistent with suffice fortheattitude-plus-velocity command SCAS. 3-40 values. As a further comparison, the Level 1 requirement of Reference 6 for maneuver control during Summary of Pitch Control Requirements. A STOL operations provides for 30* of bank angle change in 2.4 sec, which is satisfied by a control authority of summary of the required pitch control authority deter- 0.55 rad/sec 2 for a roll damping time constant of 0.5 sec.
mined from these STOVL aircraft simulation results, compared to (1) the Level 1 criteria of References 4, 5, The latter requirement represents a more specific criterion for operation during transition, particularly and 6, (2) available control power for some relevant where that phase consists of precision path tracking in V/STOL fighter aircraft designs (Refs. 7-9), and forward flight during instrument flight conditions in (3) earlier fixed-base simulation results for the E-7A adverse weather. Based on the results of this STOVL STOVL concept (Ref. 10), is presented in Table 2. For aircraft simulation, a roll control authority of the transition phase, the pertinent criteria are those of References 5 and 6; no control power data are available 0.9-1.2 rad/sec 2 would be necessary to satisfy demands for maneuvering and control in turbulence.
for the individual aircraft. For the vertical landing, References 4 and 5 apply; the total available control power has been tabulated for the Harrier and VAK-191. Control use for the vertical landing, shown in Figure 4b, is consistently less than the Reference 4 In the transition phase, the highest value of the requirement, and falls within the range suggested in criteria of Reference 5 does not quite accommodate the Reference 5. Peak maneuvering demands for attitude command SCAS range from 0.1 to 0.3 rad/sec 2, and are peak control use in turbulence noted for this experiment (MFVT STOVL). Maximum control experienced during comparable to 3---40 values. The heaviest turbulence increases these levels modestly to 0.2-0.4 rad/sec 2. For the E-7A STOVL simulation was considerably greater, both for maneuvering and control in turbulence, and is the attitude-plus-velocity SCAS, which provides lateral more in line with the requirement of Reference 6. For the velocity command through bank angle control, calm air vertical landing, both References 4 and 5 appear to be too maneuvering control use is somewhat less than for attitude SCAS alone; however, in turbulence the demanding. The current results indicate that less control power is used, especially with a velocity command demands for the two systems are similar.
system that employs thrust deflection for longitudinal Results for shipboard recovery are generally in control. No criteria are available for shipboard opera- tions. Values shown for the Harrier and VAK-191 agreement with the criteria of References 4 and 5, except aircraft represent total control authority available for trim for high wind over deck conditions (Fig. 4c). In light winds, the peaks vary from 0.2 to 0.4 rad/sec 2. In the and maneuvering; actual control used by these aircraft is heaviest winds, maximum control of 0.9-1.1 rad/sec 2 not available. By comparison, the total control available for the MFVT STOVL aircraft is 0.42 rad/sec 2 in hover, was observed for the attitude command SCAS; for the with 0.08 rad/sec 2 of that being used on the average for lateral velocity command SCAS, maximums ranged from 1.3 up to 2.0 rad/sec 2. Based on pilot comments from the trim in winds up to 34 knots. Thus, the pitch control for this aircraft was adequate to handle the measured trim subject simulation experiments, operation aboard ship and maneuver demands in hover and vertical landing for would be precluded at higher sea states because of the limit on capability to recover to a more actively moving
deck. If shipboard operations atthese extreme conditions
controls were employed for roll control, this increment of
areanticipated, roll controlauthority in excess of that
control power would demand 0.7 lb/sec of bleed flow.
givenin References 4 and5 must beprovided. Further, The bleed flow values are based on an assumption of
lateral velocity command capability will demand more
90 ib of reaction control thrust per pounds per second of
controlauthority thanthatused for attitude command
bleed flow rate (Ref. 12), and on minimal nozzle flow alone. Thelattertwo conclusions arecontingent both losses or adverse jet interference. If the latter two
on thevalidityof theshipairwake model used in this
influences are not optimized, bleed flow requirements would increase.
experiment (Ref.11)andontheaircraft's sensitivity to
airwake disturbancesand should be qualified
accordingly.
Yaw Control
Summary of Roll Control Requirements. Table 3
presents a summary of the required roll control authority Effect of Flight Phase. Yaw control use shown in determined from these simulation results, compared to Figure 5 is considerably less than the criteria of Refer- the Level 1 criteria of References 4, 5, and 6, to available ences 4 and 5 for any flight phase. For the transition control power for the V/STOL fighters, and to the E-TA (Fig. 5a), peak demands in calm air range from 0.02 to STOVL concept. For the transition phase, the pertinent 0.04 rad/sec 2. In the heaviest turbulence, maximum criteria again are those of References 5 and 6. In the control usage of 0.04-0.14 rad/sec 2 was observed, with hover and vertical landing, References 4 and 5 are the most confined to the range of 0.05--0.07 rad/sec 2, within applicable documents.
the 3--40 band. In contrast, the recommended range is 0.15-0.25 tad/see 2 from Reference 5. As a further During transition, References 5 and 6 accommodate example, the requirement of Reference 6 for a 15" head- the level of roll control required for maneuvering in calm ing change in 2.2 see translates into a maximum yaw air, but call for an insufficient level of control to handle control power of 0.22 rad/sec 2 for a yaw damping time the current STOVL configuration in turbulence up to the constant of I sec. The disparity between these two level shown. Considering experience of the Harrier criteria for yaw control and the recent simulation design evolution, the dominant requirement for roll experience is likely attributable to good yaw stability control during transition may well be associated with augmentation employed and the lower sensitivity to countering sideslip excursions. The AV-8B has sufficient disturbances for the recent STOVL fighter concepts lateral control to trim with sideslip angles of 15" or more compared to the collection of aircraft on which the earlier criteria were based.
during transition. The current MFVT configuration can achieve lateral trim with sideslip of 10" or greater over the low speed flight envelope. Criteria of References 4 Maximum yaw control for the vertical landing and 5 are about right for the vertical landing. No criteria (Fig. 5b) is comparable to that for the transition.
are available for shipboard operations. Total control Maximum maneuvering control in calm air varies from authority available for trim and maneuvering is shown 0.015 to 0.065 rad/sec"2; control in turbulence increases for the Harrier and VAK-191. Total cOntrol available for somewhat with an occasional peak excursion as large as the current STOVL aircraft in its basic configuration in 0.1 rad/sec 2. The maximum range in turbulence corre- hover is 1.1 rad/sec 2, which was adequate for disturbance sponds to 3--4o values. The Reference 4 requirement for suppression and more than adequate for control of the a heading change of 6* in 1 see converts to a maximum vertical landing. However, it was necessary to augment control power of 0.28 rad/sec 2 for a yaw time constant of the baseline roll control system with reaction control to 1 see. For the shipboard landing (Fig. 5c), maximum provide Sufficient Cohtrb| power to handle thehighest control use is similar to that for the runway landing, with wind over deck for recovery to the ship. In the latter case, peaks to 0.1 tad/see 2 for the highest wind over deck.
the total control power was 2.15 tad/see 2. Control used = for maneuvering in calm air and control needed in Summary of Yaw Control Requirements. Yaw turbulence for the E-7A were less than those required for control summaries of authority determined from these the MFVT STOVL and more in line with the criteria of STOVL aircraft simulation results, compared to the References 5 and 6. It should be noted that for the MFVT Level 1 criteria of References 4, 5, and 6, to available STOVL design every 0.1 tad/see 2 of additional roll control power for other V/STOL fighter designs, and to control power would require an additional .,-170 Ib of the E-7A, are provided in Table 4. For the transition differential thrust at the lift nozzles in the hover phase, the pertinent criteria once more are those of Refer- condition, or 2.4 ib/sec of reaction control bleed at the ences 5 and 6. For the vertical landing, References 4 tail mounted reaction control nozzles. If wing tip reaction and 5 are the pertinent criteria.
this aircraft configuration, where 4 klb/sec is equivalent For the transition and vertical landing, the criteria of to 1 rad/sec 3. In turn, the maximum rate of change of References 4, 5, and 6 all exceed the current experience control power can be used to define the relationship for yaw control use to a significant degree. Based on the between peak control usage and the effective bandwidth current experience, yaw control power for maneuvering of control that can be achieved without encountering the and turbulence suppression could be considerably control rate limit. For example, a maximum thrust reduced. As before, shipboard operations are not covered transfer rate of 2 klb/sec (corresponding to a rate of by the existing criteria. Total control authority for the Harrier and VAK-191 are somewhat in excess of that for change of angular acceleration of 0.5 rad/sec 3) and a peak control usage of 0.05 rad/sec 2 (representative of lo the current STOVL design (0.28 rad/sec2). Control used level of control use for closed-loop regulation) would by the E-TA in the fixed-base simulation experiment is imply a rate limit free control bandwidth of 10 rad/sec.
comparable to that for the MFVT STOVL tested on the VMS. For this STOVL aircraft design, every 0.1 rad/sec 2 Conversely, for the same thrust transfer rate and a representative control bandwidth of 5 rad/sec, rate limit reduction in yaw control power would reduce the reaction control bleed at the tail mounted reaction control free operation could be sustained up to a control authority of 0.1 rad/sec 2.
nozzles by 4.8 Ib/sec.
Roll Control THRUST TRANSFER RATES Effect of Flight Phase. In Figure 7, the rates of Ability to achieve adequate rates of thrust transfer thrust transfer employed for roll control are indicated for between propulsion system components for pitch and roll the different flight phases. Throughout the transition control is an important aspect of control system dynamic response. Maximum thrust transfer rates observed for the (Fig. 7a), typical maximum rates for maneuver control different tasks in the simulation program are documented ranged from 1 to 2 klb/sec with the exception of two cases which demanded 4.5-6.5 klb/sec. In the heaviest in this section. Results are presented both as maximum turbulence, rates of 3-4 kib/see occur frequently, with rate of change of thrust and, more generally, as the rate of occasional peaks from 5 to 8 klb/sec. For roll control, change of pitch and roll angular acceleration. Implica- tions for thrust control bandwidth are also noted. a thrust transfer rate of 10 klb/sec is equivalent to 3 rad/sec 3.
Pitch Control Maneuver control rates for the runway vertical landing (Fig. 7b) generally ranged from 2 to 4 klb/sec.
Turbulence did not affect control rates up to the Effect of Flight Phase. Thrust transfer rates magnitude of disturbances evaluated. For shipboard for pitch control are documented in Figure 6. During landings (Fig. 7c), peak rates of 7--8 klb/sec are observed the transition (Fig. 6a), maneuvering control in for the attitude SCAS with significant wind over deck calm air produces peak rates ranging from 0.2 to 1.3 kilopounds (klb)/see for the attitude command and represent a substantial increase over other phases of SCAS. Maximum rates of 1.5-3.3 klb/sec are reached operation. With the attitude-plus-velocity SCAS, wind over deck has a strong influence on thrust transfer rates, under the highest wind and turbulence condition. This with peaks of 10 klb/sec (3 rad/sec 3) reached on occasion maximum range exceeds that for 3--4_ values. Results are independent of SCAS mode. Runway vertical for the highest wind over deck. In lighter winds, transfer landings appear to be more demanding on maneuver rates are comparable for the two SCAS modes.
control rates than the previous flight phase, but with no As an example for roll control, a maximum thrust influence of SCAS mode (Fig. 6b). Peak rates ranging from 1 to 2.6 klb/sec are observed in the data. Turbulence transfer rate of 5 klb/sec (corresponding to a rate of has no influence on the rate of control use. The most change of angular acceleration of 1.5 rad/sec 3) and a peak control usage of 0.2 rad/sec 2 would imply a rate significant control rates appear for the shipboard landings limit free control bandwidth of 7.5 rad/sec. For the same (Fig. 6c). Maximum rates of 3--.4 klb/see with attitude thrust transfer rate and a bandwidth of 5 tad/see, a peak command and 3--6 klb/sec with longitudinal velocity control authority of 0.3 rad/sec 2 could be achieved command SCAS occur at the highest wind over deck.
without reaching the control rate limit.
To generalize these results, thrust transfer rates can be expressed in time rate of change of control power for THRUST CONTROL the ratio 15:43 to bring it into conformity with the definition of mean ground effect used herein.
Influence of Ground Effect and Ingestion The shape of the boundaries is established by height Vertical axis control power in vertical flight is control out-of-ground effect for positive ground effect, associated with the margin of thrust in excess of that on abort capability at decision height for neutral to required to equilibrate the aircraft's weight. The require- moderately negative ground effect and ingestion, and on ments for thrust margin during vertical landing are control of sink rate and hover position to touchdown for influenced by the disturbances imposed by jet-induced larger negative ground effect. Results from simulation evaluation of the YAV-SB Harrier are somewhat less aerodynamic forces in proximity to the ground and degradation in engine thrust that results from temperature conservative than the boundary derived from the evalu- rise at the engine inlet due to the recirculation of hot gas ation of generalized ground effect and are consistent with exhaust from the propulsion system. Experiments have Harrier flight experience as described in the aircraft's been conducted on the VMS to evaluate in general the operations manuals (Refs. 13 and 14). The boundary influence of ground effect and hot gas ingestion on tkrust correlates over much of its range with an analytical margin necessary to control height and sink rate during prediction of the trend of thrust/weight with mean ground airfield vertical landings (Ref. 2). In turn, these results effect required to arrest a nominal sink rate of 4 ft/sec were validated with specific simulation assessments of prior to touchdown with an application of maximum vertical landings with the YAV-8B Harrier, an aircraft thrust at an altitude of 21 ft. This analytical relationship whose vertical landing characteristics are well known and is expressed as have been related to the simulation experience. Results from these simulations are presented in Figure 8. The 43 h i boundaries shown define acceptable and unacceptable (h2- 112)/2g=£ (AI_./T)' d h +£(AT/W )d h regions for combinations of meangroun d effect and ingestion and thrust/weight ratio. One boundary was and can be used in synthesis of new STOVL designs to extracted from the generalized evaluations reported in detcrmine the required thrust margin for anticipated Reference 2. Data from the YAV-8B ground effect levels of mean ground effect and ingestion. Finally, evaluation are also presented with an appropriate fairing bascd on the results of Reference 2, it was noted that the to illustrate the trend. The YAV-SB data correspond to employment of a vertical velocity command control did configurations with and without lift improvement devices not shift the boundary shown in Figure 8, which was (LIDS) and for two levels of hot gas ingestion, and span obtained for attitude SCAS alone. However, as noted in the range of mean ground effect covered in the previous Reference 2, vertical velocity command does reduce the generalized investigations. Thrust/weight ratio is chance for abuse of sink rate control during the descent determined out-of-ground effect. Mean ground effect to landing and, hence, improves the control margin for and ingestion are defined here by the relationship vertical landing' 4-_f043 (AI../T)'d h Influence of Engine Dynamics where (AL/T)" incorporates jet induced aerodynamic Effects of thrust response dynamics on the pilot's ground effect as well as thrust variations with inlet assessment of control of the vertical landing are shown in temperature and is defined as Figure 9. These data come from Reference 2 and apply to manual control of thrust With only attitude SCAS (ZERO" = {[1 + zt.rr][1 + (AFG/A0)(A0/W)] - 1} available. It is apparent that bandwidth of thrust response of the engine core of 4-5 rad/sec is sufficient to achieve The altitude range over which the mean ground effect satisfactory ratings for height and sink rate control. For and ingestion are based is 43 ft and represents the range bandwidths below 3 tad/see, the control task deteriorates over which ground effect exists for the Harrier. For the rapidly. Both the transition and hover point acquisition earlier generalized ground effect simulation, the integral tasks were less sensitive tO variations in thrust control defining mean ground effect was based on an altitude bandwidth than was tlie vertical landing (Ref. 2). Vertical range of 15 ft, where ground effect did not vary above velocity command in addition to attitude SCAS insulates that altitude. The mean ground effect that defined the the pilot from the dynamics of the propulsion system boundary for that experiment (Ref. 2) was adjusted by response and results in toleration of slower engine For the transition and vertical landing, the existing response (providing the overall airframe response is not criteria all exceed the current experience for yaw control altered) than for attitude SCAS alone.
use. As before, shipboard operations are not covered by the existing criteria.
To a point, the vertical landing is insensitive to maximum rate of change of core thrust, which is Thrust transfer rates for pitch and roll control were associated with engine acceleration limits imposed by maximum allowable temperatures in the core. Thrust observed to be greatest for shipboard operations, with the decelerating transition placing the next greatest demand.
rates varying from 25% of maximum thrust/sec down Control mode did not have a strong influence on these to nearly 10%/see were tolerable for height control.
results.
However, at about 10%/see, thrust rate limiting and loss of control were encountered on occasion for such slow acceleration characteristics. These acceleration rate limits Thrust margins for vertical landing in the presence of ground effect and hot gas ingestion were defined based can be related to surge margin in design of the propulsion on results from simulation of the YAV-8B Harrier. The system control. Deceleration rate limits are important to shape of the boundaries is established by height control the ability to rapidly reduce thrust at touchdown, as well out-of-ground effect for positive ground effect, on abort as to the dynamic control of vertical velocity in the capability at decision height for neutral to moderately hover. Vertical velocity command does not seem to alter these results. negative ground effect and ingestion, and on control of sink rate and hover position to touchdown for larger negative ground effect. The boundary correlates with an CONCLUSIONS analytical prediction of the trend of thrust/weight with mean ground effect required to arrest a nominal sink rate with an application of maximum thrust at decision height.
A program has been conducted to define and The employment of a vertical velocity command control experimentally evaluate control system concepts for does not alter the thrust margin requirement.
STOVL fighter aircraft in powered-lift flight. The control system designs have been evaluated in Ames Research Center's Vertical Motion Simulator. Items assessed in the Bandwidth of thrust response of the engine core of 4-5 rad/sec is sufficient to achieve satisfactory ratings program were maximum control power, control system for height and sink rate control. For bandwidths below dynamic response associated with thrust transfer for 3 tad/see, the control task deteriorates rapidly. Vertical attitude control, thrust margin in the presence of ground velocity command systems can tolerate somewhat slower effect and hot gas ingestion, and dynamic thrust response engine response (providing the overall airframe response for the engine core. Results provide the basis for a is not altered) than can be accepted by the pilot for reassessment of existing flying qualities design criteria for this class of aircraft.
manual control of thrust. To a point, the vertical landing is insensitive to maximum rate of change of core thrust; This experience shows that pitch control power used however, loss of control appears at the lowest thrust in transition is in general accord with existing criteria, transfer rates. Vertical velocity command does not seem to alter these results.
whereas that used for vertical landing is somewhat lower.
When a translational velocity command system using deflected thrust for longitudinal force control is REFERENCES employed, pitch control use is considerably less than the criteria suggest. No criteria, except that for hover, exist for shipboard recovery. 1. Levine, J. and Inglis, M., "US/UK Advanced Short Takeoff and Vertical Landing Program," AIAA Paper 89-2039, July 1989.
In the roll axis, control power recommended by current design criteria is insufficient to cover demands 2. Franklin, J. A., Stortz, M. W., Engelland, S. A., for transition. Agreement is good with criteria for vertical landing. Again, no criteria are available for shipboard Hardy, G. H., and Martin, J. L., "Moving Base operations. For these operations, lateral velocity Simulation Evaluation of Control System Concepts and command through bank angle control typically used Design Criteria for STOVL Aircraft," NASA TM- 103843, June 1991.
greater control power than did an attitude command system alone.
3. Merrick, V. K., Farris, G. G., and Vanags, A. A., 9. Fortenbaugh, R., Hutchins, D., Miller, D., "A Head-Up Display Format for Application to V/STOL Schweinfurth, R., Traskos, R., and Tumilowicz, Approach and Landing," NASA TM-102262, Sept. 1989.
R.,"USN/FMOD FRG VAK-191B Joint Flight Test Program - Stability and Control," Final Report, Vol. 4, NAVAIR-4R-76, Aug. 1976.
4. Chalk, C. R., Key, D. L., Kroll, J. Jr., Wasscrman, R., and Radford, R. C., "Background Information and 10. Franklin, J. A., Stortz, M. W., Gerdes, R. M., Hardy, User Guide for MIL-F-83300 - Military Specifications - G. H., Martin, J. L., and Engelland, S. A., "Simulation Flying Qualities for Piloted V/STOL Aircraft," Evaluation of Transition and Hover Flying Qualities AFFDL-TR-70-88, Nov. 1971.
of the E-7A STOVL Aircraft," NASA TM-IO1015, Aug. 1988.
5. "V/STOL Handling, I - Criteria and Discussion," AGARD R-577-70, Dec. 1970.
11. Fortenbaugh, R. L., "Mathematical Models for the Aircraft Operational Environment of DD-963 Class Ships," Vought Corp. Rep. 2-55800/8R-3500, Sept.
6. lnnis, R. C., Holzhauser, C. A., and Quiglcy, H. C., 1978.
"Airworthiness Considerations for STOL Aircraft," NASA TN D-5594, Jan. 1970.
12. Hange, C. and Gelhausen, P., "An Investigation Into Sizing the Reaction Control System in Conceptual 7 Anon., "AV-8B Aircraft Stability, Control and STOVL Aircraft," NASA TM-103990, 1993.
Flying Qualities," McDonnell Douglas Corporation Report MDC A5192, Aug. 1981.
13. Anon., "NATOPS Flight Manual Navy Model AV-8B," AI-AVgB-NFM-000, Dec. 1984.
8. Lacey, T. R., "MIL-F-83300, View from an Aircraft 14. Anon., "NATOPS Flight Manual Navy Designer," Proceedings of the Navy/NASA VSTOL Model AV-8A and TAV-8A," NAVAIR 01-AVgA-1, Flying Qualities Workshop, April 1977.
June 1978.
Table 1. Flight Control Modes Transition Hover Control axis Attitude SCAS FlightpathSCAS Attitude SCAS Velocity SCAS Pitch/roll Rate command-attitude Rate command-attitude Attitude command- Attitude command- hold hold attitude hold attitude hold Yaw Turn coordination Turn coordination Yaw rate command Yaw rate command Vertical Thrust magnitude Flightpath command Thrust magnitude Velocity command Longitudinal Thrust deflection Acceleration command- Thrust deflection Velocity command velocity hold Lateral Velocity command Nil nnmn I i Inlnl i inl n Table 2. Comparison of Pitch Control Power Criteria with STOVL Aircraft Designs i MIL-F AGARD NASA AV-8B AV-8A VAK-191 Recent STOVL Concepts Flight 83300 R-577 TN 5594 MFVT E-7A (Ref. 10) phase Ref. 4 Ref. 5 Ref. 6 Ref. 7 Ref. 8 Ref. 9 Maneuver Turb6 Maneuver Turb6 0.5 0.15-0.19 0.2-0.25 0.6 0.6 Transition 0.05- 0.2 0.53 0.8 1.0 0.16-0.27 0.16-0.27 Vertical 0.29 0.1-0.3 (AC) landing -0.83 -0.75 0.17 0.17 (VC) WOD 15 WOD 46 WOD 15 WOD 34 0.3 0.4 0.53 0.8 0.31 0.37 Shipboard -0.83 -0.75 0.22 0.22 (VC) landing All values expressed in terms of control power in rad/sec 2.
Notes: (1) Reference 7 and 9 requirements converted from attitude response based on a time constant of 0.5 sec for (2) rate command systems or a natural frequency of 2 rad/sec for a critically damped attitude command system.
(3) Control power for actual aircraft represent total available in hover; transition values not available.
(4) Control power for MFVT and E-7A represent maximum used.
Table 3. Comparison of Roll Control Power Criteria with STOVL Aircraft Designs MIL-F AGARD NASA AV-8B AV-8A VAK-191 Recent STOVLConcepts Flight 83300 R-577 TN 5594 MFVT E-7A (Ref. 10) phase Ref. 5 Ref. 6 Ref. 7 Ref. 8 Ref. 9 Maneuver Turb6 Maneuver Turb6 Ref. 4 0.25 0.6 Transition O.1-0.6 0.55 0.3-0.4 0.9-1.2 0.38 0.2-0.4 2.2 1.73 1.4 0.1-0.3 0.2-0.4 Vertical landing WOD 15 WOD 46 WOD 15 WOD 34 0.2-0.4 0.9-1.1 0,55 1.8 2.2 1.73 Shipboard (AC) landing ! .3-2.0 (VC) Notes: (1) All values expressed in terms of control power in tad/see 2.
(2) Reference 7 and 9 requirements converted from attitude response based on a time constant of 0.5 scc for rate command systems or a natural frequency of 2 rad/sec for a critically damped attitude command system., (3) Control power for actual aircraft represent total available in hover; transition values not available.
(4) Control power for MFVT and E-TA represent maximum used.
Table 4. Comparison of Yaw Control Power Criteria with STOVL Aircraft Designs ii iir'T Flight MIL-F AGARD NASA AV-8B AV-8A VAK-191 Recent STOVLConcepts phase 83300 R-577 TN 5594 MFVT E-7A (Ref. 10) Ref. 4 Ref. 5 Ref. 6 Ref. 7 Ref. 8 Ref. 9 Maneuver Turb6 Maneuver Turb6 Transition 0.15- 0.22 0.02-0.04 0.05-0.07 0.04 0.04 0.25 Vertical 0.28 0.1-0.5 0.43 0.46 0.4 0.15- 0.1 landing 0.065 WOD 15 WOD 46 WOD 15 WOD 34 Shipboard 0.43 0.46 0.065 0.1 0.05 0.12 landing Notes: (1) All values expressed in terms of control power in rad/sec 2.
(2) Reference 7 and 9 requirements converted from attitude response based on a time constant of 1 sec for rate command systems.
(3) Control power for actual aircraft represent total available in hover; transition values not available.
(4) Control power for MFVT and E-7A represent maximum used.
RCS nozzles RCS nozzles Cruise nozzle Trim lift nozzle Main lift nozzles Figure 1. Mixed-Flow Remote Lift STOVL Aircraft \ \\ Vertical Motion Simulator Figure 2.
.4 m - Attitude command 0 Maximum Attitude + flight path command • 20 Q ,3 ® (9 Pitch control ////// usage _ (rad/sec 2)
l °
(9 @ .1
-
_1 I I I _1 I I I 0 2.5 5.0 7.5 0 2.5 5.0 7.5 RMS turbulence (ft/sec) (a) Transition .4_ Attitude command 0 Maximum - Attitude + velocity command • 20 -- Ref. 4 .3 ,I///// Z.L/.ZJ_L ® Q ® Q Pitch Q control .2 B e usage IX (9 (rad/sec 2) • • B -(9 • _1 t 1 i _1 I I I 0 2.5 5.0 7.5 0 2.5 5.0 7,5 RMS turbulence (ft/sec) (b) Vertical Landing Figure 3. Influence of SCAS Configuration and Wind Environment on Pitch Control Use.
!
.4 - Attitude command 0 Maximum - Attitude + velocity command • 20 Q 0 ® ® 0 Ref. 4 ////// .3 ® J_zJ_L_ O ® Pitch ® in control .2 ,_ ® usage B: (rad/sec 2) Q Q
! _
,1 - I I I I _1 I I O-I 0 20 40 60 0 20 40 60 Wlnd over deck (knots) (c) Shipboard Landing Figure 3. Concluded.
0 Maximum 1.25 -- Attitude + fllghtpath command -- Attitude command • 20 Q ® Ref. 5 m m 1.00 O O O Q m .75 Roll
o ®
control usage ////// O (rad/sec 2) 0 Ref. 6 ® .50 -® ® Q .25 _1 I I I _1 l I I 0 2.5 5.0 7,5 0 2.5 5.0 7.5 RMS turbulence (it/sac) (a) Transition Figure 4. Influence of SCAS Configuration and Wind Environment on Roll Control Use.
1.0 Attitude command 0 Maximum -- Attitude + velocity command Q 20' .75 Roll control .5O usage (rad/sec 2) Ref. 4 ®
o
® o 8
,25
- o
®
o _i
1 I I I I o 2.5 5.0 7,5 0 2.5 5.0 7.5 RMS turbulence (ft/sec) (b) Vertical Landing 2.0 -- Attitude command 0 Maximum - Attitude + velocity command • 20 ,,,,, 1.5 ® Roll control 1.0 Q usage (rad/sec 2) .5 Ref. 5 Ref, 4 ////// ////// _1 I I _[ I I I 2O 4O 6O 0 2O 4O 6O Wind over deck (knots) (c) Shipboard Landing Figure 4. Concluded.
D ,2O Attitude command O Max .2O Attitude command O Max • 20 • 2(7 Ref. 4, 0.28 red/see2 Ref. 5, 0.15 - 0.25 red/nc2 Ref. 5, 0.1 - 0.5 rad/eec2 Ref. 6, 0.22 rad/sec2 B ,lS .15 ® Yaw Ysw control control .10 ® .10 usage usage (r=d_ (r_I.¢2) ® ® ® .0S .O5 ® • ® [ I I _l I I J 0 7.5 0 2.5 5.0 7.5 o 2.5 5.0 RMS turbulence (ft/sec) RMS turbulence (tt/sec) (a) Transition (b) Vertical Landing .2O - Attitude command O Max • 2a Raf. 4, 0.28 rad/sec2 Ref. 5, 0.1 - 0.5 rad/sec2 .15 Yaw control .10 ® usage (rad/sec2) ® .05 (9 0 _1 I I I 0 20 40 60 Wind over deck (knots) (c) Shipboard Landing Figure 5. Influence o[ Wind Environment on Yaw Control Use.
m Attitude command - Attitude + flightpath command ® Q Ventral nozzle thrust transfer 2
- 8
rate (klb/sec) Q Q n
_qP I
1 I I I 0 2.5 5.0 7.5 0 2.5 5.0 7.5 RMS turbulence (Wsec) (a) Transition - Attitude command O Max -- Attitude + velocity command • 2o" m ® ® Ventral G nozzle thrust Q
_¢
m transfer 2 ® ® rate (klb/sec) ® @
o
® Q Q ® ® _1 i I I _1 1 I I 0 2.5 5.0 7.5 0 2.5 5.0 7.5 RMS turbulence (ft/sec) (b) Vertical Landing Figure 6. Influence of SCAS Configuration and Wind Environment on Thrust Transfer Rate for Pitch Control.
10.0 Attitude command 0 Max - Attitude + velocity command • 2_ n 7.5 Ventral nozzle thrust transfer 5.0 rate (klb/sec)
o
o 6
2.5 ® _1 I I I _1 I 1 I 0 2O 40 60 0 20 40 60 Wind over deck (knots) (c) Shipboard Landing Figure 6. Concluded.
i0.0 -- Attitude command 0 Max ® 7.5 ® ® Lift nozzle ® thrust D 5.0 transfer ® rate (klb/sec) 2.5 ® • • • _l I I 0 2.5 5.0 7,5 RMS turbulence (ft/sec) (a) Transition Figure 7. Influence of SCAS Configuration and Wind Environment on Thrust Transfer Rates for Roll Control.
10.0 - Attitude command - Attitude + velocity command i 7.5 E) Lift ® nozzle thrust m transfer 5.0 rate ® ® (k_b/aec)
@
$
_® --E) 2.5 ® ® ® ® E) _1 I I _1 I I I 0 2.5 5,0 7.5 0 2.5 5.0 7.5 RMS turbulence (ft/sec) (b) Vertical Landing 10.0 - Attitude command 0 Max - Attitude + velocity commend 6) • 2o 6)
e
Q _ (9 B ® 7.5
®
® Lift nozzle ® thrust m transfer 5,0 ® rate Q (klb/sec) E) w 2.5 _1 ] l I _ I I I I 0 20 4o 6O 0 20 40 6O Wind over deck (knots) (c) Shipboard Landing Figure 7. Concluded.
YAV-8B .025 Ref. 20GE • On Nominal • Off Nominal LIDS HGI • On Increased Mean • Off Increased ground effect and Ingestion (g'e) 1 43 / ho=0 I I I I I 1.025 1.050 1.075 1.100 1.125 Thrust/weight (OGE) Figure 8. Influence of Ground Effect and Hot Gas Ingestion on Thrust Margin for Vertical Landing Pilot 10 - rating Inadequate 6- Adequate - 4 • Satisfactory 2 _1 I I I _l I I I 0 5 10 15 0 10 20 30 Thrust response Maximum thrust bandwidth responsible rate (red/sac) (%/sac) Figure 9.
Effect of Thrust Response Bandwidth and Response Rate on Control of Vertical Landing