s-,)u --9y· E-Pt:>~f"

NASA Contractor Report 19104 7

Army Research Laboratory Contractor Report ARL-CR-13

Hot Gas Ingestion Effects on Fuel Control Surge Recovery and AH-1 Rotor Drive Train Torque Spikes

Frank Tokarski, Mihir Desai, Martin Books, and Raymond Zagranski Coltec Industries West Hartford, Connecticut

April 1994

Prepared for Lewis Research Center Under Contract NAS3-26075

National Aeronautics and Space Administration RESEARCH LABORATORY

ABSTRACf

This report summarizes the work accomplished through computer simulation to understand the impact of the hydromechanir.al n.u:bine assembly (TA) fuel control on rocket gas ingestion induced engine surges on the AH-1 (Cobra) helicopter. These surges excite the lightly damped torsional modes of the Cobra rotor drive ti3in and can cause overtorqueing of the tail rotor shaft.

The simulation studies show that the hydromechanical TA control has a negligible effect on drive train resonances because its response is sufficiently attenuated at the resonant frequencies. However, a digital electronic control working through the TA control's sepuate, emergency fuel metering system has been identified as a solution to the overtorqueing problem. State-of-the-art software within the electronic control can provide active damping of the rotor drive train to eliminate excessive torque spikes due to any disturbances including engine surges and aggressive helicopter maneuvers.

Modifications to the existing TA hydromechanical control are relatively minor, and existing engine sensors can be utilized by the electronic control. Therefore, it is concluded that the combination of full authority digital electronic control (FADEC), with hydromechanical backup using the existing TA control enhances flight safety, improves helicopter perfonnance, reduces pilot workload. . .and provides a substantial payback for very little investment

PREFACE

This is the final report covering work completed under Contract NAS3-26075 during the period September, 1992 to October, 1992.

The program was conducted under the technical direction of Messrs. George BobuJa, U.S. Army Propulsion Directorate and Mike McCall, U.S. Army Aviation Systems Command, for the NASA Lewis Research Center, Qeveland, Ohio. The work was per:fonned by Coltec Industries, Chandler Evans Control Systems Division. Mr. Ray Zagranski was Program Manager and Frank Tokarski, Mihir Desai, and Martin Books were the principal engineers.

ii

TABLE OF CONTENTS

S~Y .................................................... 1

IN'IRODUCTION . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

SWULATION MODEl. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Fllel Coottol . . . . . . . . . . . . . . . . . • . • . . • . . . . • . . • • . . . • . . . . . . . . . . . . . 3 &lgirle • . . . . . • • . . • . . . . . . • . • . . . • • . • . • • • . . . . • . . . . . • • . . • . . . . . . . 7 Rotor Dri:ve Tmin. . . . . . . . . • . • • . . . . • • • . • . . • . . . . . • . . . . . . • • . • . . . • . . 10 Airfraine . . . . . . . . . . . . . . . . . . . . . • • • . . . . . • . . . . . . . . . . . . . • . . . • . . . 10

DISCUSSION OF RESUL'I'S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Simulation of Rocket Fire SlJlies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Potential Solution Using the Existing TA Hydromechanical Coottol . . . . . . . . . . . . . . . . 21 Active Damping via Electronic Coottol . . . . . • . . • . . . . • . • . . . • . . . . • • . . • . . . . 25 Perfonnance Benefits . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . • . . 31

PRELIMINARY DESIGN ... FADEC SYSTEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 E'l.ec:trorlic Coottol Unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 TA Coottol Modifications ....•....•.•..•...... ·• . . . . • . . • . . • • • . . . . . . 39 Fngine and Airvehicle Modifications . . . . • . • • . . . . . • . . . . . • . . . • . . . • . . . . . . . 41

CONCLUSIONS AND RECOMMENDATIONS . • . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

APPENDIX A Ust of Symbols and Acronyms • • • • • . . . . • • • • • • • • • . • • . . • • • • • • • • . . • . . . • 46 Ust of Figures, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

iii

SUMMARY

A non-linear computer simulation of the Chandler Evans TA hydromechanic:al fuel control Lyco~g !53~~ and Bell Helicop_ter AH-1 Cobra airvehicle WclS contigured on a 386 based PC. ~ sunulation IS ~le of modeling the engine and helicopter throughout its operating r.ange from engme startup to maxunum power and from sea level to 20,00> ft. altitude.

MK66 rocket firings were simulated by disturbing the engine inlet conditions with pressure (Pl) and temperature (Tl) pulses representative of hot gas ingestions. These pulses were applied as single shots and as multiple firings at the two pilot selectable frequencies (6.3 Hz and 8.3 Hz).

It WclS shown that the hydromechanic:al fuel control filtered out the high frequency Pl and Tl distuibances from its metered fuel fiow and compressor geometry control outputs. Therefore, the dominant effect to the helicopter WclS detennined to be the repeaterl engine surges that occur during multiple rocket firings. These result in torque disturbances to the rotor drive train at frequencies very near the tail rotor resc:>nant mode (7.3 Hz). Under these conditions, the lightly damped rotor drive train resonates beyond the nonnal operating torque limits of the tail rotor shaft.

Methods of reducing engine power via fuel fiow and engine geometry in an attempt to lower the nominal value of drive train torque and thus the peak torque oscillation were unsuccessful. The simulation revealed that these potentially corrective inputs needed to be applied very rapidly and that these inputs acted as additional disturbances to the rotor drive train. Therefore, larger resonant torque spikes were caused as compared to doing nothing.

A solution to this dilemma was found to be the incorporation of "active damping" of the rotor drive train resonances via the fi1st combustive torque response of the engine. Combustive torque is that thermodynamic component of engine output torque that is achieved via additional engine burn flow at a constant gas generator speed. Since the gas generator does not have to change speed, combustive torque response is fust. Active damping applies combustive torque in direct opposition to drive train oscillations in order to reduce their magnitude.

An electronic control with multivariable control algorithms specifically sized to damp the rotor drive train resonances via the engine's combustive torque WclS simulated and found to keep tail rotor drive shaft torque spikes within nonnal operating limits. This perfonnance was achieved during single as well as multiple rocket firings. This controller also provides an inherently fuster engine and rotor speed control loop for nonnal operation. Thus, tr.ansient rotor speed droop and engine torque overshoots during aggressive maneuvers are significantly reduced compared to the existing bill-of-material control. Thus, a side benefit is increased helicopter agility with less pilot workload.

-1-

A jxeliminary design of the electronic controller WclS made. An airframe mounted full authority digital electronic control (FADEC) commwrl.cating via a small electric motor interfilce to the existing TA hydromechanical control was the most straight forward and cost effective insta11ation. ·-The electric motor rotates the separate emetgency fuel valve in the TA control while the primary hydromechanical control remains operational with its fuel valve not in control of engine fuel flow. If milure of the electronic control should occur, fuelfiaw remains fixed and the pilot can transfer to the primary hydromechanical control. Thus, the state-of-the-art electronic control is backed up by the existing TA control. Flight safety is improved as well as helicopter perfonnance.

The estimated cost of adding the FADEC and making modifications to theTA hydromechanical assembly is equal to approximately the cost of overhauling the edsting TA control. Thus a substantial payback in helicopter safety and performance can be realized for very little investment

INTRODUCTION

The AH-1 Cobra helicopter has the capability to launch 2. 75 inch folding fin aerial rockets (FFARs) equipped with Mark 66 (MK66) rocket motors to engage battlefield taigets. Following a class A accident in which an AH-1 suffered. a tail rotor drive shaft milure while launching MK66 FFARs, a joint engineering investigation of the AH-1F helicopter was conducted by the Airworthiness Qualification Test Directorate (RQTD) of the U.S. Army Aviation Technical Test Center and the U.S. Army Aviation Applied Technology Directorate (AA1D). The results of this investigation which recommend an engine inlet shield to defiect rocket exhaust gases away from the engine are oontained in a U.S. Army Test and Evaluation Command (TECOM) Fmal Report, TECOM Project No. 4-C0-2~16, dated June 1991.

Subsequent to this work, the U.S. Army Avi3tion Systems Command (AVSCOM) in conjunction with the NASA Lewis Research Center directed Chandler Evans, the engine fuel control supplier for the AH-1 helicopter, to investigate potential modifications to the fuel control to alleviate the rocket fire surge problem. The work at Chandler Evans WclS based on a computer simulation of the engine, fuel control, and AH-1 helicopter.

The objective of the investigation was not to preclude engine smge following hot gas ingestion but to recover smoothly from surge and avoid overtorqueing the rotor drive system. This objective, if achieved, would have application not only to rocket firings, but to helicopter operation in general where inlet distortion effects and engine deterioration can also cause the engine to surge and potentially cause damage to the rotor drive train.

-2-

\

SIMULATION MOPEL

The simulation model is a non-linear representation of the complete AH-lff53 engine control system as illustrated in Figure 1. The 1500 SHP, 1'53-L-703 tuiboshaft engine is coupled to the Bell Helicopter AH-lF Cobra Gunship's rotor drive train. The Chandler Evans TA-7 hydromechanical control is intermced to the Cobra·cockpit and controls engine power in response to pilot commands. Rotor speed governing as well as engine overfueling, surge, and blowout protection are provided by the TA control. Details of the complete engine, control and airframe models are included below.

Fuel Control

The Chandler Evans TA control is a full authority hydromechanical assembly that schedules engine fuel fiow, compressor inlet guide vane position, and interstage bleed band position for safe engine opemtion throughout its operating envelope. ·

A simplified functional block diagram of the TA hydromechanical computer is illustrated in Figure 2. The control law is based on Wflo, where o = Pl/14.7. Therefore, the proportional power turbine and gas generator speed governors both moduJate Wf/o. The governor requesting the lowest Wf/o is in control of the engine.

Typically, the pilot sets the gas generator governor's PIA to maximum. This provides a maximum power available setting to protect the engine against

~

INLET AIR

FLOW

p, SENSE

L_

PAM&

TA-7 HYDROMECHANICAL

CONTROL

T53 ENGINE

WF FUEL FLOW

OUTPUT SHAFT TORQUE 1 '• , o---[roiiolor _____ .,...

Drive Train)

I I I I I I I I I

POWER LEVER I ~!:E_I~I,!

I VI

I

NF* (BEEPER)

POWER TURBINE GOVERNOR

W~NCPTG l;~~R•fS ~YEA NF~ ~

GAS GENERATOR GOVERNOR I L

ACCEL SCHEDULE

T1SENSE

DECEL SCHEDULE

0 w E s T

w I N s

WF/8

T1SENSE

H I G H

WF E s 8 T

6 w I N s

SERVO

BLEED BAND CONTROL

MAX STOP

m 1:1

HYSTERESIS

L 0 w E s T

IZ s

BLEED BAND POSITION

IGV SCHEDULE

N1

T•SENSE

WFMV (METERING

VALVE)

ll . 1 N1 IGV

ACTUATOR POSITION

WF (FUELF\.0\'If

'tPTG • .25 -.. 1.5 SEC. (low-High Power)

'tN1 • .020 SEC.

'tooo = .045 SEC. 'tn • 10 SEC.

'tP1 • .010 SEC.

'tWF • .015SEC

'taa • .08-.. .25 SEC (Open • Close)

TA CONTROL SYSTEM ... SIMPLIFIED FUNCTIONAL BLOCK DIAGRAM Figure 2

TM53 A006811 R

The engine is further protected from blowout and overfueling by minimum and maXimum metering valve stops. Fuel shutoff is accomplished via a separate PLA actuated valve not shown in the simplified ftmctional block diagram of Figure 2.

The compressor inlet geometty actuator position is scheduled as a function of gas generator speed and Tl. At low speed, the IGVs are closed to provide smge margin. At high speed, the IGVs are open to provide maximum engine efficiency. Actuator dynamics are filst and have been neglected for this study.

The nw position (open or closed) compressor bleed band is controlled as a ftmction ofWf/o and gas generator speed. Inlet temperature biases the Nl trip point to effectively result in a corrected NI/Ve schedule. For low Wf/8 and low speeds, the bleed band remains open to preclude steady state engine swge. For high Wf/8 and high speed, the bleed band is closed to provide maximum engine efficiency. During acceleration transients (i.e., high Wf/8 and low speed) the bleed band is open to preclude transient swge.

The separate, emetgency backup fuel metering system has not been modeled. This system was not in effect during the rocket fire problems encountered in the field, and theiefore including it in this study was deemed to be out of scope. The eme~gency backup fuel metering system consists of a separate metering valve, separate metering head pressure regu1ator and a solenoid actuated transfer valve that allows the pilot to switch between the primary and backup systems. The backup fuel metering valve is mechanically linked directly to the PI.A lever. A more detailed description of this system is contained in the Preliminary Design Section as it provides a convenient int.er&ce to a FADEC system that can potentially solve the rocket fire, transient torque spike problem.

In summary, all of the features of the primary engine control have been modeled. In addition, the essential dynamics of the primary hydromechanical computer and fuel handling section have also been modeled. These are incorporated as non-linear functions in the actual simulation code but are depicted as linearized time constants in Figure 2 to give the reader a quick reference as to the response of the various control loops.

-6-

Filgine

The Textron Lycoming T53-L-703 engine is a 1500 SHP tUiboshaft engine consisting of a gas generator section and a power (or free) tUibine that is coupled to the helicopter rotor drive train. Figure 1 shows a schematic of the system and defines speed, pressure, and temperature stations throughout the engine.

A functional block diagram of the TS3 core (or gas generator) section of the engine is illustrated in Figure 3. This engine model is commonly tefened by the industry as an excess torque model. Metered fuel tlow is burned through relatively fast combustor dynamics and is compuecl to the engine steady state required to run fueltlow. Any difference generates excess torque and thus NDOT that accelerates or decelerates the gas turbine to a new operating condition.

The thermodynamic torque that is available to be applied to the free tutbine, QF, is a sum of the core engine's steady state output torque and the fast transient torque that is available in the burned excess fueltlow that is not used by the gas turbine.

The engine model, as shown in Figure 3, is fully non-linear with engine performance a function of NG, BT EED, and IGV position.

The model is corrected for inlet air pressure, P1, and temperature, T1. Therefore, the effects of roc1cet fire ingestion can be readily simulated by disturbing the model with Pl and Tl profiles as a function of time.

Swge is also modeled on an open loop basis by matching observed engine performance during an actual surge. The reduction in output torque, QF is simulated as shown in Figure 4. Incidence of hot gases at the inlet triggers the surge which then affects engine output for approximately 0.25 seconds.

-7-

I 00 I

COMBUSTOR

WF .I 1 I WFE-t 'tcS + 1

GAS GENERATOR TURBINE

f(NG)/.[9

.Jif SURGE

MULTIPLIER

STEADY RUNNING LINE

~

'-----8

NG

STEADY STATE TORQUE

a 6

I • •IFT-LB I~

JGV BLEED

'----.[ff ~---6

+ +

SURGE MULTIPLIER

6 • Pl/14.7

QF

9 •(T1+460)/519

T53 CORE ENGINE ... FUNCTIONAL BLOCK DIAGRAM Figure 3

u. d w a

$ en 0..

0 .2

17

16

15

14

13

12 0 .2

150

100 0 .2

.4

.4

.6 .8 1.0

TIME ·SEC.

.6 .8 1.0 TIME·SEC.

.6 .8 1.0 TIME·SEC.

1.2 1.4 1.6

1.2 u 1.6

1.2

1.0 irr!4W :(JtH~B!-3 ·-··-~··:···-:--·:-~···-+·-,.··-:---:V-;·-~--~·-:·-:--:··· ... ---"!-·-:-···:-;•••!

0 0 - .. L .... l .. -~--.;. .... ; .... L .. l .. ..J ..... l.._L .... .:.... .. .L_l _ _L __ L_J. __ i __ i ...... L .. l ........ ~---; ..... .;_ ..... ~ • : : : : : SURGE MULTIPLIER ... NOOT : : : : : : :

~Efm±~Sffi~!~f!I~J 0 .2 .4 .6 .8 1.0

TIME:SEC. 1.2 1.4 1.6

ROCKET FIRE ENGINE SURGE SIMULATION Figure 4

-9-

:Rotor Drive Train

The rotor diM train consisting of the T53 power turbine, Cobra main rotor and tail rotor, and associated driveshaft and gearing is illustrated schematically in Figure 5. This diagram shows a 4 inertia/3 spring system that accurately models the main and tail rotor torsional modes. Since these modes are the prevalent frequencies observed on tlight trace recordings of rocket fire surges, the simplified drive train model was deemed sufficient for the purposes of investigating fuel control effects.

A functional block diagram of the rotor drive train model is given in Figure 6. The model is non-linear in that the main rotor, tail rotor and free turbine damping terms are a function of tlight condition. However, the block diagram is illustratfd in a linearized 13shion, to give the reader a quick reference as to the relative magnitude of damping terms present in the drive train.

For the sea level hover, rocket fire condition, a linearized bOde plot of the rotor drive train is given in Figure 7. This shows that the rotor system is very underdamped with a.main rotor torsional mode of 2.9 Hz and 5.8% critical damping. The tail rotor torsional mode is 7.3 Hz and 2.5% of critical damping. Therefore, it should be no surprise that rapid distUibances to this system can excite the torsional modes and result in transient torque spikes.

Airframe

A three degree of. freedom airframe model including longitudinal, vertical and yaw degrees of freedom was utilized in the study. The airftame model also includes an automatic heading control that simulates tJ:le reaction of a human pilot. This model is proprietary to Chandler Evans, therefore, a more detailed description is not provided herein.

The airftarne model was configured to the AH-1 helicopter and a few simulated tlight maneuvers are enclosed to demonstrate that the model approximates AH-1 climb rate, descent rate, max airspeed, and heading control performance. These traces are enclosed in Figures 8 and 9. Based on these traces, the airframe model was deemed sufficient to evaluate the effect of rocket firings on the body states of the airvehicle.

-10-

I ..... -I

p,

IGV BLEED

COP

ENGINE

ADOT

XDOT 1

_,.. _,..

vLUivn _,.. _,.. _,.. _,.. _,.. _,.. _,.. OF

INERTIAS JF•1.010 SLUG. fT2 JGO• .074 JR• 5.900 JT• .180

SHAFTS Ks• 10,000 FT-LBI RAO KR·355.6 KT. 316.8

DAMPING BF • .493 FT-LB I RAO I SEC BR-2.26 BT •• 709

Ors

MAIN ROTOR

JR BR

CIP

GEAR I I KT BOX

Jao Nao

lOOo/o SPEEDS NF • 6568 RPM (SHAFT)

21,018 RPM (TURBINE) NR • 320RPM Nr • 1651RPM

AIRSPEEDS AOOT • CLIMB RATE XOOT • FORWARD SPEED

NOTE: ALL INERTIAS, SHAFTS, DAMPING REFERRED TO POWER TURBINE OUTPUT SHAFT SPEED. DAMPING CALCULA TEO AT 98% No

AH-1/ T53 ROTOR DRIVE TRAIN· SCHEMATIC Figure 5

TNI'53 A0381113A

+

o----- as

CLUTCH

+

+

+

~----~-J~--·~

+ Ors

+ Nr

~-------+---Nr

o--Ors

AH·1/T53 ROTOR DRIVE TRAIN • FUNCTIONAL BLOCK DIAGRAM Figure 6

-12-

TA A0177112C

0 I ! I ! I I ! ! ! I I I I I I I I I I I I I I I I I I j240

MAIN

-20 I GAIN -,__ .. , ... , R~~OR 180

-4o I I =>--"" 1;1 ""'-\ I -"~ .. !12o

~ . . . . . .. . . . . . \ . . . . .. ~ ~-we I 11 ' r ~ ~ 8 -so ~- .. -~ . . . . . ~ . 1 ~ o & I £ . • •.. :: : : :: : . . . :::: ~ -100 -60

-120 -120

-140 I I I I I I I I I I I I I I I I I I I I I I I I I I I I- .. 30

1 o-• 1 ()0 1 01 1 02 Frequency (rod/sec)

AH-1/T53 ROTOR DRIVE TRAIN FREQUENCY RESPONSE • NF/QF

Figure 7

0 w w Q.

~ ;c· X

~

Ot-wz a:w wo ~(I) ow a.O

-14-

z ::::) a: z ::::)

0

a: w > ::::) w Zco

~ 3: c: s m c "'11 r-C5 ::1:

-n-t .!... (Q' 3: v. c: 'h.

I CD z com

c: < m :a

C) c: z :a c: z

PEDAL POSITION (INCHES)

HEADING (DEGREES)

ALTITUDE (FEET)

AIRSPEED (MPH)

o8311So § loasii§ I I : !' I i I I l . . . : ., I i : ' : ' ... ···t·••t•··r- ···r,·· c·r· ... ....... ... .. ·+·+· ... ·+ ...... , ... ; 'i"·t·+·j""i"·

._._. ::: ::: :::rr:i:::,:::rr: ._._._. :::j:::f' .. ::: -.-.-. ~ (3 :: ::: ·:t:t.tt t:rt:1c:L

......... i ........ , ... i .. +++· "'J"' '""ll" ... ,I~~..... . "'l"'l"+"i· l ... j ... , .. , ... J .. ... ... ... ,.+ ..... : ... t ... l ... f... .. ............... ~ :o .. ·· .. 'i' .. t .. Y·l· '!"·j' .. ·t--·· .. + ........... T ..... 1 .. +-y+.. ... .......... · ···•-t g:.. ... ...i ... ; ... i ... ; ·1··· .... , ... l.. .. i-

1 I i I ' - : . I l ' I • ' • _._._. _._._.::f. _._._. _._._.!_._._.,_._._. -.-.-:~::.- _. i ~ ::. ::_ .... 1

i ... 1i ... j::. _._._. :::rr:j::::r: t:rrtr:

........................... :... sc ... ... .. ... ... ... ... ·+"i"·t·++·' .; ... J.-1. .. _._._. _._._. -.-.-1'.-.-.- _._._._. _._._.l_._._.t.-.-r:rr. i i _._._. _._._. _ _. _. _._._.j_._._.l_._._.l_._._._ :::rrt::r.-.-,r.-::1:.-r.::.-+

. ::: ::. :::::: :::::: ·::.1::: ::::,::: .. ~ ...... :::1::::: ·::. ::: ·::. ::: l" "fl" ... "l"t-"1 .. ......... ,' .......... 1 .. ·1·-·l·-·· .. - ... J .. -1 ....... , ... r.-r-+ .. t··· ... :::.,r:::j:::,:::.,·:::r:J.:x::t:::r . ........................ r ... ,l... . .............. 1·· r·J... ... ... .. ....... t ...... ~ ... · I I ' . I

..... ....

.,... ...'

s: .. . i· ... ,.

..... -~--- ......... , .... 1... .. ... --- --- .. T ..... 1 .. -t ... l ... -- ...... 1 .... 1 ... 5 .. .... ~ ~ ~.. .. ........... , ... t---LJ .... ,.++-_· ... ... ... ... .. ... .... ... ... .. ......... ! ... j.... ... . .. < .... 3 '"I ~. ""t"' ......... , ....... ""1"'1"' - ,... ... ... ... ... ... .. ... ... ... .. ................... 1..:~;;. . ... .. ... ~.. ... !IE. ..., ... "f''"' ... "'!"' ...... .. : !: ••• ·••· ••• ··•· ••• -·••. ••• ·•·. ·••• ••• .•.. •·• •••· ••· ~; • . ·••· ••• ·••· X ·••· ••• ··•. :1·•• . · rtF Ill aL................... • _ ...... LL ............ L .. 11... LL.J .. L

THRUST VECTOR (DEG. FROU VERTICAL)

o 8 3 I I 8 I ! I ! ! . ! I I

~ COLLECTIVE PITCH

(INCHES)

O..a.Nc.t .. UI

I I I I I I I I

DISCUSSION OF RESULJS

The simuJation model described in the previous section was used to evaluate the effects of rocket fire induced engine smges on the Cobra rotor drive train. The results are discussed in the following sections.

Simulation of Rocket F'll'e Surges

Baseline flight test data was extracted from a U.S. Anny Test and Ewluation Command (TECOM) final report TECOM Prqect No. 4-CQ-23().(XX).()16, "AH-1F Helioopler/Marlc 66 2.75 im1 Folding Fin Aerial Rocket Engineering Investigation", June 1991. FJ.gUre E-1 from this report is enclosed in Figure 10. This data shows a single engine smge as the result of a simultaneous firing of a pair of rockets (one from each pod). Engine inlet pressure and temperature, engine fuel flow and compressor discharge pressure, and rotor drive train torque spikes are all shown. Thus, a significant amount of data is present to correlate the simulation with test.

Figure 11 shows the simulation of the same event, and correlation with critical parameters is summarized in the table below.

Parameter

Main Rotor Torque

Tail Rotor Torque

Engine Shaft Torque

Fuel Flow Oscillation

Compressor Discharge Pressure Drop

Test Data

Damping Hertz Peak Ratio Hertz

2.7 1400 ft-lb 3.2% 2.9

7.0 340 ft-lb 3.4% 7.3

2.7 72psi 3.5% 2.9

2.7 ±35pph 3.5% 2.9

N/A 90% N/A N/A

Note: { = 100 In (Xol 21CN XN

wlwre: { = dllmping ratio, ~ of criticDl dllmping XN • peak-peak amplitude of cycle N X0 = peak-peak amplitutk of first cyck N = number of cycks ( 10 typicDl)

Table I

Simulation

Damping Peak Ratio

1700 ft-lb 4.7%

350 ft-lb 4.5%

90psi 4.4%

± 40pph 4.4%

90% N/A

CoiTelation of Flight Test and Simulation of Rocket Fire Induced Fngine Surge

-16-

AVG CG • TRIM AVG GROSS LOCATION DENSITY OAT ROCKET INLET FLIGHT WEIGHT LONG LAT ALTITUDE FIRED SHIELD CONDITION

(LS) (FS) (BL) (FEEl) (DEG.C) 9670 197.6(MIO) 0.1 RT 1300 26.0 2 NONE 5 KT L T SIDEWARD

SO'F

NOTES: 1. MARK 66 ROCKET MOTORS USED. 2. M261 PODS INBOARD AND M65 LAUNCHERS OUTBOARD. 3. ENVIRONMENTAL CONTROL UNIT· OFF. 4. SKID HEIGHT ·10FT. 5. AH·1 F USA SIN 76·22600

SOLID LINE SHOAT DASH MAIN ROTOR TORQUE

(TOROUES AT 6568 RPM)

40.000

c: ~":" 30.000

g~~ :l g)(

20.000 ~t;§" ~~~ 10.000

I I ! ' I

I ,.1

" !,. I , i I :I " \ ,., " • I '· . II II

II \_, ... v .. , , '1 -I I I 'I II .; '.! I

0

30.000 TAIL ROTOR TORQUE

w ~8- 20.000 ba:CII a:g~ 10.000 ..JI-:. -~n

~~ 0 - .~ tf I • lilt! jill ..... !--'"' .. f-v II u.:~ ~· . 'J ,- - ;-.

·10,000

ENGINE OUTPUT SHAFT TORQUE eo w

~ eo a: 0:-1-0

40 w!!:. z

~ .! I.

h. I I ~~' II \. ,, ,., r''..l' ,. ... ........ 1-- -·- -·-f-- 1.1 I I v ....,, \ '• , " . '

\t • 0 20 z ..,

0

a:W oa: eo ~i ~w eo Q.f§' ~Win 40 ~!!:. we

20 ~%

~~ 0 w~-

COMPRESSOR DISCHARGE PRESSURE I .1

if I"" ~NQI~E r SURGE

I I I I leG 1400

cnw

w~~ 140 !~- 1200

1- (.) Z.Jj 120 ! ffi s 1000 0~ Zw ~~e. w;:)_ 100 1&. ~I!! 100

eo 100

~ELIFLJ /~ V\.. ~ .._ I-"

GAS TEMP - - -- -- -- ---+---+---+---- -- --

I I l INLET CONDmONS

f\ I ~ +,

;,,

12 J 0 3 4

TIME • SECOND6

MK66 ROCKET FIRE FLIGHT TEST DATA (1 PAIR) Figure 10

-17-

1140

! 1230 ;;;

CD .... 120 ,.:.

\!:. 410

0

1142

421 'Iii' Ill ....

2" ,.:. \!:..

0

-2!4

!511

1137 ;;;

751 Ill .... ,.:. \!:.

310

0

15

75

~ !55 :n !!:. 35

15

1012 ~

us § 0

1St ~ ;

132 % Q. 505

!!:.

171

140 ~

122 cj w e. ..

32

(/) -3: c:

E 6 z 0 "'11 3: ~ en ::0

:II 00 I CQ

- c " qo ~ m ......... ..... !!

::0 m "'11 r-s :!i -1

m -...L "tJ ~ ::0 -

ENGINE INLET TEMPERATURE

("F)

,0 t; s i i § o I ! ! !

... , ...... , ... , .... , l··t·······t··· w-t··· ··· ··· ·t·! ·+·l·+·,···

~~:J-::. :::~:::r~ :~LI:::t::r: f ····:··· ..f.+· ···!···!··+·+·· ; · :::'::: -~··:'~::~ -~··:tr:t ~ ... ··~ ............ :; ... ! .. .!.. .. , .. . til"' ............... ··1··+-·t' ·+·· ·········· ...... ······r···l··· . -~:::,::: ·::: ::: ·::: :::;::r::j::: :::_

·················+··1 ..... +. ::::::::: t:l:

~:;- :;:;; j;l;f; .. ..,_._-1:-'--...,._......-+--'--+--w • ~ •

ENGINE INLET PRESSURE

(PSIA)

ENGINE COMPRESSOR ENGINE FUEL FLOW OtSCHARGE PRESSURE

(PPH X 10'1) (PSIA) ENGINE TORQUE

(PSI)

ol!lli!SSol!lliss gol!lli!s8

: l::Lt:J:::!: :t:-·1 tt:: • l I I o • • • •

lltfJJlll ···j··++·· ... ·t·+·-1 "''··+·· . I I •• I • ······t···:·+··l"'"l···t· ··+·· ·······1··-~·-+····r··-~-· ·+·· ···!··+····+·,·····+ ···-i··· ··f···,···t··J ........ "!" ... , .. .

•···4···1··-r-+·-r··,· ... :. t··· .. . :::ltj·:::lt: :::.jt:::cl :i:: ·::_

\i:OLJ.(:••

·:::tFJ::::I:::I:::. ·::: :::~ :lt:::::: ............ ········· ....... . ..............................

···•·t .. l·-··1··· ... ...; ··t···· .. . '-···l···l···l··· ........ ·i ....... . • I I I I ···1··+-·r· .. , ·~···!··! ··~- .. , ...

.. 1 .. ·i···l···f···j···~ ... ·r-j-·· ·· r·t···l··t-·l··t·. ··j···t··· ... : ....•... : ....... .!. .. : ....•.......•...

i ~ i i i : : :

:;;_L[L-LLLLL · : t:.tt.Lttt:t: ···f··i·+·i···i i···i···f

~.... i ~ i i + ..... ·+·-~···j· ··+·t i···t··· ... 1··•

. I ... I . ···l···r····-.. , .... r. I···· ···I··· . ··+-·· ··-j--·t······ ... , ... tl···l .. ··r·l··-r-·1~ m .. , ... ···r . ···t···l···t··· ;a ~ ·· ··· ···I · ·+··i···l·-~~ ~ ·t-·r·l · ···t···t····l···l ..•... , .... .

··+··t··+··l-···1'··· ··+·· ..... . ··yl·-~··r ··· ····r1··· ··· :::::::::·::r:·:::::,:::·::::: ·:!:::1:: :j.:j.: ::t·::: :: ~::: "l··f··· ···•····•··· ....... !... ..

··-r··t··•t••·t···j··-r-··1···1···1··· ···t···, ....... 'l ... , .. t .......... . ·· ··· ··· ······ ···r··•···t·· ··· ··l··i·· .. t···l···l---~··+··t··· ... :::j:::f:::j:::t::J:: :::t:j::: ::., .. .I ....•.... J . .! ...... .I.. ...... .

I I . ~ I ! . ! . t

::t:::l::·l:::i:: ::.I:::L:L:!:::

iliff Ol;ll!li ··t·t···r··-~····:·· • ..

..l ... . _... ...... ~,jo::':'l.

IJI::iill ... !...., ... , ... ;- ··r·rl- ·•··· ···t···!··-j--·-' . ···t .. Y·t·+·· ···· ··Yr-1' , ··-t··+-·•··-j--······ .. j···t···i···. ·-t--·j···l··t ..

:::!l:::r:,t:j:: :::l:::rrr ::: .. ::r:;::r: .rr:1::+:: ·····~···j .... '··-j--· ···l··t· ···I··· :_,_::r:j:::r. :::r:r:I::Y: ::t:t::~:::t:: :::t:::/::t::L . .... ,l . ..il ... t ... j ... +·+·t···i··· t . 0 • • : ··::r::l::rr :::r::rr:r:: ···1··l+··i·· ··+·r··t·+·· . .. .... t .• .! ... .L .•.••• l ... t .. J ... .t .. . 0 lllal~~~

(FT-LBS x 1 o·?

TAIL ROTOR.TORQUE (FT·LBSx 10·?

~ool!l~t;

::11~1; llll~lil~l :::: .. ~ .... ::~:::~·-~::: ::rt:. ::r:t·t··· ...... : .. :,:::j:::.t:::

. I . . I . I ::~rT-·r··-: ·,···r:~::r:~::: ··r·r··r· .... , ... ~"' .. !···1··· ::i:rt :-::l::rr ::r:· ·lJ··•fll

::: :::{:::1::: .:: :::t:::. :::J::t:.

::;:l1_·;l' .. ·ji·:. 'it.l:l:;·t,::;l(if ............... , ......... ···t·· ....... , ...... ··················· i··t··l+·· ........ , ... t-···1···

'-J:::ror.:: ::r:rr:r:r: I • • • I I ••

~~--r~:.t~~~r~~- ~~J~~~l-~.- .t:~J:~.I~_- __ ,

MAIN ROTOR TORQUE (FT·LBS X 10'1)

0 t; s i i § I i I ! ' i ::rrt:; ::c.l::rrr: • I I : I . I • I ··r·j· r···~ ·r·····t··t··r

liftll!l .1\' : ;i-,]J I. I .. I I . ..... ,... . ... , ... !" , .. , ...

...... ... ; .... ·'····t--r+ ....... t... ... i···t···t···i·· .. :::!::::::. ·::1:::.:::,:::1":::::: ............. , ....... ···j··· .. . ............. ·t··· ........... .

... II··· ... ···, .. ·t··· ··t·:·· .. . :::·::.::::::.·:.::::::,::r_::: .:::.1::::::.::: +::::rr::l:::

itf!jt[, ··-r·l··+· ·r· ··+-·r·-J--· ::r:r:::j:::; ::r :::t·::.r:

1'::.

.. l .. t .. ·r·:···•···f··· ··Y ....

...•... .!.. .•.... .!.. .... : ...•......

Peak tail rotor drive shaft torque, ~on frequency, and damping ratio match almost exactly with the test data. These are the critical parameters, since a tail rotor drive shaft fililure has oc:curred in the field as a result of rocket firings.

'fhe main rotor torsional mode frequency and damping ratio match well considering the margin of enor in reading the compressed time scale of the flight t:races. However, the peak main rotor torque oscillation is 20% higher in the simulation. This also affects the simulated shaft ton}ue which is the sum of main and tail rotor torques. Since the damping ratios of the actual vs. simulated drive train are close, the di1ferences in the initial peak torque can be attributed to the simulated effect of engine surge on output torque.

The surge model illustrated in Figure 4 was left unchanged because the compressor discharge pressure drop as a result of surge and the resulting tail rotor spike due to the sudden disturbance to the rotor drive train were matched quite accurately. These are the most critical parameters for the study.

The single surge t:races show that the rotor drive train is very lightly damped. The engine disturbance causes the drive train to ring with significant ton}ue oscillations that require approximately 7 seconds to die out This makes the rotor susceptible to a second or thiid surge disturbance that can occur due to multiple rocket firings. Furthermore, multiple rocket firings are perfonned at two pilot selectable frequencies that bracket the tail rotor resonant mode as shown in the table below.

Parameter Frequency (Hz)

Main Rotor Torsional Mode 2.9

Slow Rocket Fire 6.3

Tail Rotor Anti-Resonance 6.7

Tail Rotor Torsiooal Mode 7.3

Fast Rocket Fire 8.3

TableD . Rocket F1re Frequency l'So Rotor Drive Train Torsional Modes

Therefore, multiple rocket firings have a very good chance of amplifying torque spikes. Figurel2 illustrates a simulated, multiple rocket firing. Three rocket fires at the slow frequency result in two engine surges. As expected, the torsional oscillations are amplified beyond the nonnal tail rotor operating shaft torque limit

The effect of the fuel control and modifications to the control system to attenuate these ton}ue spikes . are discussed in the following section.

-19-

en -3: c r-~ 0 z 0 ""1'1

3: c :n~ co-

I C , N -, r-9 ~m

I\) :a ·o

0

" ~ :!! :::D m en -w , ~ -:a -

ENGINE INLET TEMPERATURE

('F)

0 8 8 i g §

-::f::~:::~::~:::·· .. ::::::. ::: 0

t .. '·JttttttJ·I -····•···•l·l·Jj ............ .

........ . ......... , ............ ..

...................... , ........ .. ................................

i 1·:: :::::: ·::::::· .::::::: :::,::: ........... ! ............ "'1"'1 .. . HI ............................. . I~ ::: :::. ::: ·::. ::: ·:: ·::: ::: :::F

...........................................

...............................................

..............................................

• !.'!.'!'!'J'j w • .,. ......

ENGINE INLET PRESSURE

(PSIA)

ENGINE COMPRESSOR ENGINE FUEL FLOW DISCHARGE PRESSURE ENGINE TORQUE

(PPH X 10'1) (PSIA) (PSI)

o 1!J 8 8! go I!J 8 8! So I!J 8 8 8 S r-.--. . . I ' I . I • I I I

::t:l+t·: :::j::-1: :(-::j ··l···•····t···t·· ··l···t····l···t···

:::j:::f:::j:::f'. ::j'::f: j:::f: .

·::!:+:r::!:: .·:::~:::r·t·:l:: .. -:-·-r··-:··· ··· ~ ... J. ... L.r:.

: : • i -··11··--..-·-.. : JJ::::·::~::·.::.; ... : ....... ~:::

I I ' 1 l ::r: .. ·::: ::::·:--: .. -.. ·::.;:::i:::r::: ... f ... 1...r ......... J ... j .... f .. . ... , ... ; ... fl .. .' ... ~ .......... f' .. .

t : • ....... r ... , ..... , .. '~" ........ :::l::rrl .. :j:::~l::: :::f::. ... , .... , ... l ... t ... ·j ............. . "'t"' ....... , ... , ..... , ....... . .... , ... , ... ·+" .. ! .. +f ... l"' .... , ... , ...... y ... !...}·+ ... .. ... l ... j .... , ... i ..... + .. ·+ .. j"' ...... t ... : .. l .... ·r· .... , .. T ... :·+·++· ' .. t+"!' ... ! ... -+ .. i ... I·..... . .. r+ .... +. .... , ... , ...... t ..... , .. + .... , ... ·+·J'·· ... , ... ·+·++ +· ... , ... , ... ,, ... , ... "'l"'t"'l'"+ ... I .. T·:·'J" ... l .. l'·· .. ·t· ••• J •••• : ••• : ••• .!. •••••• J. •••••• J .... t. •••

o~GI~iji

(FT·LBS X 10'~

TAIL ROTOR TORQUE (FT·LBS X 10'1)

i>ooi!Jl!l8 ! : i i i ! ! i t···s-······s···t····•···t : : : : I I '"l' .. , ... t· ... ·t

I"T" "i"'l"+·l

[f~'];JJ] : I I I I '

rT -:::r::r:rJ 1 -·•-------·---~---

............. .:-:r;---····t··· ! • 0

i i :

I · ................ : .... · .. t:,:::i::: ;or+-i...i-1 I •

:::,r:::r ............. ::r:r ... ... . " ... , ... 1' "f " ....................... , ...

IJ·~ .:_ til! I . :::1:::1:::

1::: ·:. ::::Fl. ~ ~

"'!"""""' '""'!"l"l"l' ... , ... t ......... "'l"'f'" "'l"' ... \ ...... , ........ , .... , ... j ...... . ... i.+ .. t' ...... , ... t' .. j ... t' ...... .

I • • I .... , ... , ... , ..... l""!"'i'":'"!"'

:;rrr .. ~:.Lr.r:r.L

~- ' ~- ;-

-~ ~·

MAIN R(TOR TORQUE (FT·'!..BS X 10'') .,

0 t 8 i i § : , •• I ; • I · ·j .. ·y·: ·r .. r ... , .. '1" .. 1 ...

:::r1':::rl ::i:::i::t:rr:

..., ....... , .......... , ... j. ... , ... f .. . • i I : I r I ..+ .. 1

... ~ ...... l ... i .. f"·! ..... ... , ....... , ..... , ... , .... ,., .. +

.+l;l+ill I o : ~

~-1-:':":'f'··r-··

-·-·· •. ..: ..•. 1 ..• ; .••. ---···

··::: ::: ':::': .... · ":t·::.j:::j::t:: ........ ........ t"'"'i"""' ·::. ::: ::: ....... ·:::1-:::I:::II::: ::: ······ ····· -····· ... ···-·-······ ............... ·-······· . ............................ .

. ........ + .... ·++·+"''' ...... l"l" ll"l'"l"" ... ::r::::r: ... :FI:::·::::::::: ::: :::r:ll::::.::r:::,::::.:::l::: ::: . ...... , .......... l ... l ...... 1 ..... .. "' "0 "'tU• """ ,,_ "' "" '"

:::LLL· :: ·:::~::: ·::: :::j::: ... , ... t . .' ... j ..... "" "' ....... . ........ L.J...l JJ ..... J ... .

IWential Solution Using the Existing TA Hydromechanical Control

The contribution of the bill-of-material TA conqol on rotor dri~ train torque spikes is demonstrated by comparing the simulated flight test data of Figure 11 and the results of Figure 13 where fuel flow is held consaant throughout the rocket fire transient. The drive train responses are very similar with peak torque oscillations nearly identical.

The TA control attenuates torsional frequencies which are present in the power twbine speed signal by virtue of its 1 second hydraulic Jag in the power tuibine governor. The control also does not respond to the short duration P1 and T1 inlet disturbances and does not directly sense engine compressor discharge pressure and shaft torque which drop off suddenly during an engine surge. Therefore, it can be concluded that the TA control has a negligible effect on rotor drive train torque spikes caused by rocket fire engine surges.

The fuel control system must be modified to actively prevent engine surge or to ~ to engine surge in some fashion to attenuate rotor drive train torque spikes.

In an attempt to prevent surge, it was assumed that closure of the compressor inlet guide vanes (IGVs), when synchronized with rocket firings, could go a long way to block the amount of hot gases· entering the engine and thereby preclude surge. The IGV s must be modulated between the open and closed position for each rocket fire because closing down the IGVs for a prolonged time period would degrade engine power and cause loss of rotor speed and helicopter lift.

Figure 14 illustrates the effect of modulating IGVs for a 19 shot, multiple rocket fire salvo. Engine surge is not triggered. However, the repeated IGV torque distuibances near the tail rotor frequency excite the lightly damped Cobra rotor drive train. Tail rotor torque spikes are amplified to an unacceptable level. Therefore, the rapid modulation of IGVs to preclude surge is not recommended as a viable solution.

Another possible alternative to alleviating the rocket fire torque spikes via the TA fuel control is to recover from engine surge in a smooth filshion. Engine output torque could be held dO\W upon detection of surge, thereby reducing the tail rotor drive shaft torque spike.

Figure 15 illustrates the simulation of a single rocket fire with IGV closed immediately upon detection of surge. The resulting tail rotor torque spike is similar (within 15%) to the simulated

. flight test data of Figure 11.

The IGV input is essentially too late, the rotor drive ~ resonates at too high a frequency (7 .3 Hz). Shaft torque rebounds from the initial drop in torque due to engine surge before the IGV s can do much about it

-21-

en -i: c: s 0 z 0 _.,.

~:D :;:o "110

"TIC" 1 ca·!. m

N C "'I-f N .., -I CD 0 "'II .... ~(..) :D &>m

::I "'II UJI ;r-::::~Cl .e~

-t

m -...r. -o ,.. -:D -

ENGINE INLET TEMPERATURE

rF) ~ ~

•I I I I I I I I I I I - - - - - -w .. Ul • .... • ENGINE INLET

PRESSURE (PSIA)

ENGINE COMPRESSOR ENGINE FUEL FLOW DISCHARGE PRESSURE

(PPH X 10'1) (PSIA)

o 8 a s 8 § o 8 a s 8 § I ! ! . ! . ! .• ! . I I . ! ' I I I t

ll-Jltlll•.

.:tf.j'f):F: li ···f··+·· ··+··· ·y·j ·+··L ··r··r····~··t·C! r-·1 .. r···l··· ::j:::t :J: I!! :::.i J::t +l· ··-l;!! ···l·.L+. ·+··l······l" i ···i +··t··· ::r:I:::.~:J: ~ :::1 :rr ::r:r: ::: ·: ~ t::.~ ::j·::tl::: ··· ... ! ........ ~r-··t········· :::l::ll:::. ::: ::·t···l':::j :: :::,::: ··l"········j····1······j····r :::l:::.f::::::t,:::r:: ::1:::.,···· ···t······t···f··t+··i········· ·++··t···:··J-··Y·i ··r·+-· ··y+-····+·+··l···l···l··l· :::L.Lc.Ltt:.! .. L.F

• I I : I I : :

llllllTli ::l::f:::l:::l:::t:.. :rr. !:::. """l""""'""ll' ;n·,···j·· ... ··T+·TT~ z: , .. ··· ···t···j···t- "'""~ e~···t··· ··· ·::.t:::r::t:::j'::~ ~ :::·::: :::

I I . I I ... , ... ·····t-+·1··· ······ ... ... , .. .f..·J···r-·Yl·· ··l·· · · ···t···l···!·······t···t··y+· ...

::11! ;,~t~lFJ••• •••·

.... ::: :::rt'···l·-· ........ . Y·l··+t-··+··1 .... 1

... l:::tt::jt:l:::: ::.l:::.f:::l··· .... ::::::,::::::::::::::x::r::: .. , .. t .... l-f-···+··1······ ... 1 ... li .. +.T ......... fl··+·- .. . ·········j'·--r+··j'·······l····· ································ ...

ENGINE TORQUE (PSI)

0 8 & 8 8 g

Hlllllf ··l···l····t···t··· ···1···1····1···1···

1111.\UJ :::cr············.· ......•... t .. .f:.

I I : : ' : '

Tfli:iJJl ···1-·t··j···r- ··J···Y·!··f .. :::.tl:::i:Jl .. ··ttt:,::: I . . . . :::l::t:l:::j::· :::j:::f::j:::i:::

I • • • I .

::: ::T: :::r: ::rri:F ····. ···!··-l--r· ····i···•···f··+-· :::j:::j::::il:::r: ::lt::':::t'::: ··t-l·,··T· ···i··+··l···~··· :;_fj,· IL J,--JLI

I : : ! I ;

:::L.LLL J::r.F:.F o~s:c;~~~~

(FT·LBS X 10'?

TAIL ROTOR TORQUE (FT·LBS X 10'?

0 0 0 8 1!1 $ I ! . • ! I I

;:;···, ; ·-~ ·rr:·;:r;r:· I I , f I

···~·-·•···' ;···t···f··· . •··· x ~ ....... ··r·+ ;n Q > m ··I t··+··j· -8 cn::u I . . e-~ ~ :: : .. ··:r:r:J: ~ f ~ ::u .. j... l+· . ii1 :::1 i5 !!: + :·T·1· .j. ~ 2 iDe t· ·· ····· ~en fnl!! --r·· .......... ~~ m n ··1· ··· ··· ···\

1

·•· c I! =112····· ·········· m!; ~~·+······· ... :::~ :-i p :r: ::cr::: .. :..~. ··y-j .... 1······-I·········!···L ···~···L ...... , ... f···~····+·· ·-·i·-·i···l ......... j .... ! .. .l.. .... .

ilffllffJ ... i. .. .l ... i:. .... J. .. i. ... i .. J ... L.

MAIN ROTOR TORQUE (FT·LBS X 1CJ1)

o a s i i § ' I I • • • I :

.. j···1' ··+·; ·+-· 1 ••• ,.l .. i··· ··t··· ··-j-·; tr-··:·t+·· ·:::j:::f::r: ::rr:rrl:: . +++··i .l...j ... t···l···j'···

: : I : t I • :

JlHT!TFl . : i

- ...... ···t···l···l····

::'j::::j ... : .... : .. : .. ··::tLt:: • • • • • I f • .. t··y· .. . ·t· "I""T t··

··t·j··· ·:·+··j···j··t··j ···1··-r-····· ··+··r1······· ... ···t··.f. .. ' .. ···jl···r··· ... j .. • l I l i :::r: :::,:: .. :rrr:l':::. :::

::::·r::: ::]::: .t:::l:::r::: ::: ::: ::r:.t:::j::: ·::1::::::::::1::: :: ·· r-·-1-·t· ·. ··1···t··· ···r·· ···

::lll::; :r:l:: ·::r: ::· ·--r+··j···, ·t···j····,·l·l··· ···t·+·l··· .. , ........... , ..... .

. . I I l ···j··r· ···· .................. . ··t·rt·,·'·········.······ ····!···J··t··f ··f····!···j····l··· ... ·+··t·T··j·· ······~·-·j···j··· :::r .. ::t:, :::ct:l:·:j:,:.:::

N ~ I

en -s: c E

~g :!:o Cl"'Q

Ul

B,..... 1%. oo ....... «"' g~ ~It _,_ ~

1%UI o« (1);:) (I)~ ~UI ~I%-~s (,)~~

1%-

~~ 0~ z_ UIQ

~ u.r -10 Ulo-;:)IJC

:!if ~!!:. z Ul

200~~~~~~~~~~~~~~~~~~~~~~~~-----

:: ~rr+r-F±ffit ... i11" .r!-,,-,A,,-+::""~ .·i-. +++-+-+-+-!-H~-+_..~....;..-+....;... ..... ..;_.;_;....: ~ ! lHV: \ill ..i. ;_

~ f ... W-;...+-L-l...;...j-+-i.J..j. PEAK OSCILLATION SIMILAR LJ......;._;_. 30+---+ : : ~ : : : '~: TOFLIGHTTEST (Figure 11): ' : :

i __ •· : : : ' : : : : ~: : 0 • ; : •• 0 •• : • ' • : : : :

20 ~:+tt+-t=rt· : ~ :~"l~~ .~ )1\; ~ i : : : : .:: ~· .·II, HJ.+li-~L.+~;;...;~..::r-"" "':·-'L-'!-~ ... _..iV-+:-+ ": ...!..-i--+-..;....._+-+-'-;...;.....;-'-~t-~-~

100

eo ~-~r--:~-~,_~_l ___ ~, _____ ~,~, -,-l.=,~jr~-~~=t;~tt=·~~~-~~~~~~t!:t~::~Et~ eo

~

20

0

; : .'SURQ!~~·~:~~~~~~~~-'-~~~4-~~~~

~-~~'-+-~-+-+-+-+~-+-+-+~-+-+~-+-+-+-+-+~-+-+-~ ~ i ~ 100

80

eo

~

. . . . . . . ~ ........ ~~~~ .... ~~~---+--+-+~---+--+-..;.....;...~. : : : :

20

0

·2.0 -1-'-' -i-: _ . .._: ....._.;....' ~; _._~:_..: ....;

In the simulation results of Figure 15, zero time delay was assumed for surge detection and the IGV actuator was assumed to be instantaneous. In actual practice, a minimum of 0.1 second of reaction time must be added which corresponds to 3/4 of a tail rotor resonant cycle. This would degrade the results of Figure 15 even more. Therefore, it can be concluded that. rear,tino: to surge is too slow to preclude tail rotor drive shaft torque spikes.

In summary, the attempts to alleviate the drive train resonances in an "open loop" fashion by triggering off the rocket fire signal or by detecting surge are not viable solutions. The corrective action taken by the control must be very fast and even more importantly, it must be in direct opposition to the drive train resonances. If the corrective actions are not phased properly, they could do more harm than good by exciting the lightly damped torsional modes of the rotor drive train.

A "closed loop" method of attenuating drive train resonances is preferred. This requires a high bandwidth electronic control that eliminates the phase shift due to hydraulic servo lags that are present in the bill-of-material TA control.

Active Damping via Flectronic Control

The architecture for a state feedback controller is illustrated in Figure 16. Additional states from the rotor drive train including main and tail rotor speeds and shaft torques are combined with typical engine control signals to create a fuel fiow input that is in direct opposition to drive train resonances. The fuel fiow input works through the fast combustive torque path of the engine as described in Figure 3 to modulate gas torque, QF, to provide damping at the 2.9 and 7.3 Hz torsional modes of the rotor drive train.

A frequency response plot of the AH-1/T53 rotor drive train, as defined in Figures 5 and 6, is included in Figure 17. As shown by the frequency response, the drive train is very lightly damped at the main and tail rotor torsional modes. With the electronic, combustive damping loop in effect however, the torsional modes are significantly attenuated. The performance of this controller during rocket fires is illustrated in Figures 18, 19, and 20.

For the single rocket fire in Figure 18 that emulates the flight test data, the peak tail rotor torque spike is significantly reduced. But more importantly, the drive train oscillations are damped out in one cycle. The.refort; multiple rockets can be fired off without fear of amplifying torsional oscillations.

Figure 19 shows the perfonnance of the electronic controller during multiple rocket firings. The two rotor torque spikes caused by the two engine surges are contained within normal operating limits and the drive train oscillations are damped out in time for the next rocket fire. A bit more rotor speed droop (2%) than the baseline TA control is realized because the electronic control transiently lowers fuel fiow to fight the drive train overtorque condition. The additional transient rotor speed droop should be of little consequence to the pilot since rotor speed recovers smoothly and quickly to the reference speed •.. as compared to theTA hydromechanical control which tends to overshoot and cause additional pilot workload.

-25-

LOAD ANTICIPATORS

C/PRATE NR DECAY RATE

+

WIDE BAND GOVERNOR

100% + ( ~ NFEM .I INTEGRAL • PTG

N t:{'

NF

+ ... FUEL CONTROL

ENGINE

, r GAIN] I LMATRIX IC=::X FEEDBACK COMBUSTIVE DAMPING

GAIN MATRIX~:): l

WFMv No NF as NR OR Nr Or

NFEM

1lNID PPHIPPH PPHio/o PPHio/o PPHIFT-LB PPHio/o PPHIFT-LB PPHI'Yo PPHIFT-LB PPHISEC/o/o

GEAR BOX

COMBUSTIVE DAMPING BLOCK DIAGRAM Figure 16

NR OR

Or Nr

MAIN ROTOR

TNT53 A03WI7 0

I

~

-10 r---~---r~~-r~~-----r--~~,-rT~T-----~~~~~~~ MAIN

ROTOR

-20 ~----------------~-----------------+--~ffi-----------~

TAIL -30 I I / \ .... ________ .. , I \ ROTOR .

7 ' 1 r, 1 I

m-4o I '-'-50 I \ II \

. , I "0 '-" c 0 . • • .. ~,c

• . . . c~(/.0 i\~e -so 1 • #t-V::~0~f

1

~ ~~~~" . "~~ .. cv .

-70 ~~---------------+------------------r-----------~-----;

-80 L---~~~~~~~~----~--._~~~~~----~~~~~~~

10-1 100 101 Frequency (rod/sec)

EFFECT OF ACTIVE DAMPING ON AH-1ff53 ROTOR TORSIONALS TORQUE DISTURBANCE REJECTION - NF/QD

Figure17

102

w

~:' oo .... -a:• g~ ~It z-j

w :::l o_ a:-:-oo ... -a:• g! 0~ a:t _,-~

w a a: gii w!:. z ~ w

a:W oa: en:::l en en wen a:W Q.a:c-~Q.-0 wen uee:. wa: ~~ e~ ffia

~ U.:' -lb w-:::l>C u.:r WQ. ZQ. a-z w

200 •••~•••••••o••••••-••••-••••u••-•:u•~•••••••-•••;••••••••o••••.•••••••••:••••~••••••••:••ooh•':"'""""""'""""-:-'-•'''""'""""'""-"'"'''"'''"•••-•••-•••• ... ~--- ......... - ....•... ~. ··:· .. :--· ~- .. -~ ... ; ... ·.· ... : ... ~ ....... ~- .. -~ ... ;· .. -. --~. --~- ···• ... ~- .. .;., . -~- .. -... ; ... ~- .. -~ ....... ~-. ·;··. , .... .

1110

120

• • • ; ...................... - ................. ; •.• ;,.. •• -~ .............. ; ••• .:.. ••• ~ •• -~--- -~ •. l• •• : .. ••• : •• ,.;., •••••• : • ••• : •••••••• - ••••••• ;, . --~ •••••••• :. ••••••. ,;_

... ~-- --~ ... ~ ... ~- ... ' ........ ; ... : ... ~ ... ; ....... -~ ....... -~ ... ; ... ~- .. i ... ;,, . -~ ....... -~ ... ; ... ~-- --~ ... ~- .. ~ ............ -~- .. 1. .. _ ....... ·- ........... .

... ~ ... .;. ....... ~-. -~- .. •· .. -~ ... ; ... .;.... .. : ... · . -~ ... ; .... ; ... ; ... ~ ... : ... l .... ~ ... : ... ~ .. ,; ... l .. ..i ..• ; ............. .:. ... L .. i .... : •... ,; ... ~ ... ~- ........ : ... ;, . -~·· ; .... : .... ~ .. ·-.. -~ ... : ... ~ .... ; .. +· .. ~. ~ .. ··~·· .; .... : ... -~--. ~ .... ~···: .... ~ .. + .. + .. + .. L..-... : .... : ... ·! ... ~-- -~ ... ~ .... ~ .. -~ ... ~- .. -~

10

40

0

:::;::::c::::=::::::::::::.:::::::~:::' ::r:::t:::::::.t=.~.::;:::;::::::::l:::::::;:::r::r:;:::::::::::=:::c;::::~::·.::::r::::::::.::: :::t::t::::::=:::::::::::·::·:::::: : 1 :::r::r:c:::::::::r:::::::c:t:::::::L:;:::~::::::::::::::::::::::::::::~:::::::::::::::::::::::.•

v

100

10

eo

40

20

0

100

80 ···~···:·······"·"-"''*"''·''" ......... _ ... ·-··:-·--~ .. -~ ... - ....... "!"'"''" .... ..

80

40

20

0

:1t:tHili1diL~.~S~~;~..,:::E·ti : . .

2 3 4 5 7 8 I TIME ·SECONDS

SIMULATION OF ROCKET FIRE FLIGHT TEST (1 PAIR) (With Electronic Combustive Damping)

Figure 18

-28-

20 ... i ... ..:. ... ~ ......... ; ... L ... ~ •.. : ... · ··-~···-~·-····-~···: ... ~.---~---~- .. :

103

~~~~~~~~~~~~~~--~~~~~~~~~~~~~~-0 2 3 4 5 II 7 • • TIME ·SECONDS

::"" 0 -.. (/J

~ ..,:. 1:!:.

SIMULATION OF MULTIPLE ROCKET FIRES (2 PAIR) (With Electronic Combustive Damping)

Figure 19

-29-

"' a .... g! a::• e= 0~ a::t: ~-

~

SINGLE ROCKET FIRE (1 PAIR)

~~~~~~~~--~~~~~~~~~~--~~--~~~~~

» ++·::~.:·+ .i.::: "t=h -~.· :-__ ~.: ·:__,:··.~ .• _,_: '.: . : ... :, ;-t-+-+T-~-t···~··•-·i __ ;. .. -. .............. _ .. .;. .. ~ .. -~........ . ...... ------~--+.....;.....;.,; -~.~ ! .: ;..+..:t::r~:~==.~ =~

20 •uL, ••~--~•••.;_••~-~~. _l.u -:-·-···;·~-.. -;-~--;--;- .;._.;.. .. ;. .. --~

10

f ... _ .. ___ .... _ ....... -......... .

0 ........ ..:. ... _ ..• ~-........ _..:.. ........ ..

·10~·_ .. _ ... _.-~ .. _·-_ ... -_ .... _._ .... _._.~_ .. ~.; .. _._-·_ .. ~~--~--~·--·-+--~--..... --~~--~--~--~~·~:~:

4 5 II 7 8 9 TIME • SECONDS

MULTIPLE ROCKET FIRES (2 PAIR)

(

L_~

A summary of electronic combustive damping performance is given in Figure 20. Peak tail rotor torque spikes are attenuated by 25-40% for single and multiple rocket firings as compared to the baseline system. And, most importantly, tail rotor resonances are damped out quickly such that repeated rocket firings do not have a chance to amplify the drive shaft natural resonances and potentially damage the drive system.

Perl'onnance Benefits

In additioo to damping out the rotor drive train resonances, the electronic control provides a fast rotor speed control for normal operation. Figure 21 compares the frequency responses of the electronic Wideband Power Thrbine Governor (PTG) with the bill-of-material TA hydromechanical control.

The NF/QD frequency response illustrates the improved disturbance rejection capability of the Wideband PIG. The effect of a ramp disturbance to the rotor drive train (i.e., collective pull/drop) causes rotor speed under Wideband PTG control to change by nearly 1/10 the amount of theTA control. The NF/NF frequency response shows the improved response of the speed control loop. The Wideband governor corrects speed ermrs 50% faster than theTA control.

The effect of improved disturbance rejection and faster speed control results in a more agile and easier to fly helicopter as demonstrated by Figures 22 and 23.

Figure 22 illustrates a 2 second collective pull from a low power descent into a high power climb with theTA control on board. A 7.5% transient rotor speed droop is caused by a slow transition from Power Thrbine Governor to the engine's accel fuel flow limit. This is primarily due to the 0.2 to 1.0 second hydraulic lag in the PIG servo that is designed to filter out torsional resonances and thereby maintain closed loop stability. This lag also causes the speed control loop to hunt a bit as it settles into a new operating condition. This is evident by the 20% torque overshoot and subsequent settling oscillations at high power.

The net result of this speed and torque control performance is degraded helicopter maneuverability and increased pilot workload. Six (6) seconds are required to arrest the descent. And, the nose of the helicopter swings back and forth as the pilot is working the pedals to maintain heading. With this type of performance, it should be no surprise that the Cobra gets out-maneuvered in air-air combat against FADEC equipped airvehicles.

Figure 23 illustrates the same maneuver with the electronic Wideband PIG on-board. A fast transition from power turbine governing to the engine's accel limit is achieved which results in minimal rotor speed droop (1 %). Torque overshoot is also minimal (4%) with no settling oscillation. Therefore, the descent is arrested in 3.5 seconds, and heading is maintained with little pilot workload.

These traces demonstrate that a significant improvement in aircraft maneuverability and handling qualities can be achieved by adding a FADEC system to the Cobra helicopter.

-31-

I

L .... -~

10 r-------------~~------------~--------------~

JS - 20 ~------------~----------~-r-ft-7~~----~--~ u. z ~ -30 ~------------~r------------r~-------+~~~~ z

-~ L---J-~~-L~~~--J-~~-L~LU----~~~J-~LW

10-1

- 10

-20

ro -3o ~ z

.....

X

~ a:-WU) >w _x 1-(.J uz w:::::. ..... 5 (.J

w :J 0 c: 0::=-1-U)

w~ z a z w

0 w w a.-en~ c:a. o!;. 1-0 c:

~--0 !::w cnw oc: a.Cl 3:W

I·

...J

:r u 1-a:-WUJ >w _:z: 1-U uz w:. ...J

5 u

w a cc 0= 1-UJ w~ z a z w

3: o _ ...J:r U..a. ...J Q. LU -:::> lJ..

< Ul CD ...J ,..:. !:..

---~ -~ ---·-~ ~ --------·--------·-·~·--· --~-

PRELIMINARY DESIGN ... FADEC SYSTEM

The results of the previous section clearly point to an electronic solution of the rocket fire surge problem. Engine fuel flow must be modulated very rapidly using sophisticated control laws in order to kill ctnve train oscillations before they have a chance to build up.

Therefore, a microprocessor based controller is required in order to incorporate the sophisticated logic in software.

TheTA hydromechanical control contains a completely separate fuel metering system for emergency operation. This metering system can be readily int.erfuced to a FADEC system thereby taking full advantage of the electronic control. A single channel FADEC can be used, thereby limiting the modifications to the aircraft. Electronic sensors can be shared with existing cockpit instruments. And, flight safety is enhanced because the primary hydromechanical TA control remains on board to serve as a backup in case of failure of the electronic control.

Electronic Control Unit

Hardware

Chandler Evans is at the present time completing a full-scale development and certification program of a "generic" FADEC which is shown in Figure 24. The generic FADEC was designed from the onset to be adaptable to many applications with minimal design changes. It has considerable computing power, extensive input/output signal processing ~ility, is modular in design, and has provision for an application boord to accommodate the unique requirements of each particular installation.

Existing sensors are utilized where possible and shared with cockpit gages as necessary. Therefore, airframe modifications are minimized. The FADEC is powered from the aircraft's 28V OC bus and requires approximately 100 watts under peak load conditions.

Separate overspeed protection is provided within the FADEC enclosure. An analog overspeed system with an independent power supply operates through the existing engine mounted overspeed solenoid to chop fuel flow to minimum flow on detection of an overspeed condition.

The FADEC 1/0 lines are extensively shielded and filtered to provide lightning protection and EMI insensitivity to DO 1 &X:, 200 V 1M fields. Therefore, the primary control as well as the overspeed system are insensitive to external upsets.

-35-

'

L ---

>-

Sofuwre

The software architecture is illustrated in Figure 25. Range, rate, and difference tests are applied as well as reasonability tests based on Kalman Filter errors. These errors are the difference between computed states and actual sensor measurements. Only validated signals are passed on to the engine control laws.

The engine and airvehicle control functions provided by the FADEC system are illustrated in Figure 25 and are described below.

Wukband Power TUrbine Governor A high response governor that minimizes transient rotor speed droop and torque overshoots during maneuvering flight and damps out rotor drive train oscillalions during rocket fire surge. Isochronous (constant speed) governing is provided in steady state.

Gas Generator Governor Provides isochronous NG governing as a function of PLA position between ground idle and maximum power.

T4.5limiter Provides gas temperature limiting during starting to ~lude hot starts and for engine protection at high power settings.

Torque limiter Provides output shaft torque limiting to preclude overstressing the rotor drive train during steady state as well as transient conditions. Allows the pilot to tly the helicopter at or near the torque limit for improved maneuvenlbility.

NDOT Schedule The start engine control Jaw is based on NDOT, i.e., gas generator acceleration rate. This allows closed loop control of engine starts which is more accurate than the open loop Wf/8 control law employed in the hydromechanical TA control. Therefore, rep-amble engine starts can be achieved without overtemperaturing the engine. During starting, the engine will achieve idle under varying ambient conditions wi~t hanging because of the closed loop control of acceleration rate, NDOT.

Accel 'Kfla Schedule The existing Wf/8 open loop acceleration schedules that are implemented in theTA hydromechanical control are Ietained in the FADEC implementation. The engine has been qualified with these limits, therefore, there is no need to change these schedules. Furthermore, the open loop schedules will be more accurate with the digital electronic implementation, thereby giving more repeatable transient performance.

-37-

~ qo

GAS GENERATOR GOVERNOR

T45 LIMITER

TORQUE LIMITER

START NDOT

SCHEDULE

-

-L 0 - w E s T NOOT

w f- I

N s

--

COMPUTED STATES

tt s. B Ora ~

Ft-i

ROTOR STATE

ESTIMATOR AND

KALMAN FILTER

II r--

ACCR~ WF/6 SCHEDULE L

& 0 w E

I

NDOT WF s WF GOVERNOR T

w I

N WIDEBAND s ~OWER

TURBINE GOVERNOR

....__

DECEL ~ WF/6 SCHEDULE

&

p, Tt

CIP PLA Tu Os NG NF :::=~

-1-

H I G H E s T

w I N s

......_

INPUT VALIDATION

~

FUEL LIMITS RANGE

& RATE

BLEED BAND

CONTROL

1st PASS FAULT DETECT

FAULT DETECTION

AND I 2nd PASS J REDUNDANCY FAULT DETECT MANAGEMENT

INPUT PROCE~ING

I

WF

~,

Enable/ Oiaable WFuv and Bleed Valve Control

BRUSHLESS D.C.

MOTOR CONTROL r

WFuv

J

METERING VALVE

POSITION

BLEED SOLENOID BLEED VALVE

Open/Closed) (On/Off) DRIVER

~

~

T53 FADEC ••• SOFTWARE ARCHITECTURE Figure25

'Fuv Internal Diagnostics

TNT53

--"

Decel WJ'~ SchRdule The ecisting Wf/fJ open loop deceleration schedules that are implemented in theTA hydromechanical control are also retained in the FADEC for the same reasons as above.

Fuel Limits Range and Rate Absolute fuel flow limits and rate limits are imposed on the final Wf demand. These preclude overfueling and flameout and provide further checking of processor functionality before actually positioning the fuel metering valve.

Bleed Band Control F ADEC control of the pneumatic bleed band actuator is provided via an on/off solenoid valve. The existing logic that is implemented· in the TA hydromechanical control will be utilized for the same reasons as above. However, this interfilce is available to improve surge ret:.CJVerY with additional logic if engine/flight test shows that it is of benefit.

In summary, sophisticated engine and rotor speed control is provided to enhance flight safety (rocket fire toique spikes are eliminated); improve engine performance (repeatable starts, temperature and torque limiting); and enhance helicopter maneuverability and handling qualities (reduced rotor speed droop and torque overshoot with less pilot workload). Furthermore,. extensive fiwlt c:overage is provided such that if an electronic control or sensor fil.ilure should occur, the fuel metering value fiWs fixed at its current position. An indication is given to the pilot at which time he or she can transfer to the bill-of-material TA hydromechanical control for a safe continuation of the flight.

The fuel metering section of theTA hydromechanical control is illustrated in Figure 26. Two separate metering systems consisting of a metering valve and cb:licat.ed head regulator are provided. The backup metering valve is positioned by the hydromechanical TA oomputer consisting of speed, temperature, and pressure servos and 3-D cams (not shown). The primary metering valve is a rotary valve, positioned by an electric motor. A transfer valve and solenoid which is actuated by a cockpit switch detennines which metering system is in control of engine fuel flow. The motor and controller are designed to mil-fixed, thereby giving the pilot time to manually transfer to the TA hydromechanical control for safe continuation of flight

-39-

$

CASE PRESSURE

REGULATOR

IIACKUI' IEI'EIINQ

HEAD REGULATOR

FAOMTA HYDAOMECHANICAL

COMPUTER

1 {A)PLA

l94 t-TONG GOVERNOR

'v--::- ,Pf I p >A~ PF

FUEL PRESSURES:

P8·800ST PC-CASE PF ·PUMP DISCHARGE

Uz5?Lz z., PM PIIMARY

IEI'EIINQ HEAD REGULATOR AND

PRUIURI REUEP VALVE

PM· METERING VALVE DISCHARGE PN • ENGWE NOZZLE P8 ·SERVO 8UPPl Y

PC

DROP _ ~ TIGHT

~ PM ~FOOT > > I 1 7 h VALVE

PRIMARY METE liNG

VALVE

MODEL TA-7 "MODIFIED" FUEL METERIN(J SECTION

Figure 26

MOTOR AND

GEARHEAD

PN

TAIT53 A0318113A

A 1ayout dmwing of the DC motor and gearhead interfclce to theTA control is shown in Figure 27.

The new hardware fits in a module placed between the power turbine governor and the cover for the main hydromechanical computer. The change involves one new casting for the motor module. In addition, the PTG housing cover must be redesigned for IIlOWlting purposes. Aside fror:n a slight 0.5 inch movement of the fuel dischalge port, the only change affecting instalJation is the addition of an electrical connector. All other interfilces remain unchanged. Inspection of the actual AH-1 aircraft instalJation indicates that these hardware changes are feasible within the current space ~~~. '

Fngine and Ain'ehicle Modifications

The modificati~ to the AH-1tr53 airvehicle for the single channel FADEC system with the existing TA hydromechanical control as backup are summarized below.

Cockpit 1) No additional inputs are required. The existing emetgency reversion switch will be used to

tiansfer between FADEC and theTA control. And, the rotor operates at constant, 100% rpm when on the power turbine governor, therefore, a speed set adjustment is not requin:d for electronic control.

2) Install a FADEC tault lamp and an oveispeed test tault Jamp in the cockpit. These give the pilot an indication of FADEC status and the results of the automatic overspeed test which is performed during an engine start. Based on this information, the pilot can abort a flight or tiansfer to theTA hydromechanical control while in the air to continue the mission.

Engine 1) Install a T1 RTD at the engine inlet. 2) Install a gear-up mechanism and a magnetic speed pickup, one for Ng and one for Np between

the existing tach generator used for cockpit disp1ay and the engine drive p!d. 3) Install bleed band control and transfer solenoids in the pneumatic line feeding theTA control.

The transfer solenoid is activated by the existing emetgency reversion switch that is used to transfer between FADEC and theTA control. The bleed band control solenoid is modulated by the FADEC. Both solenoids are wired such that loss of electrical power causes reversion to theTA hydromechanical control.

4) Provide an electrical harness for the above engine sensors, the bleed solenoids, the shared QS and T4.5 sensors, and theTA control to the FADEC.

-41-

EXISTING HARDWARE

MMVLINK

SHAFT SPLIT HERE

I I

POWER TURBINE GOVERNOR

REDESIGNED HOUSING COVER

FACE ATIACHMENT

FADEC FUEL METERING INTERFACE

HOUSING (NEW)

RELOCATED FUEL DISCHARGE PORT

(WITHIN .s· OF CURRENT LOCATION)

CONNECTOR

SEGMENT GEAR

MMV FEEDBACK

NEW HARDWARE

MODEL TA-7 CONTROL MODIFIED FOR FADEC INTERFACE

Figure 27

-42-

MOTOR

Aiiframe 1) lns1al1 collecti.ve pitch and PLA RVDTs on the linkages between the pilot's stick and theTA

control. -2) Provide an electrical harness from the above sensors, the shared rotor speed sensor, and 28V

DC power to 'the airframe mounted FADEC.

As shown above, cockpit and airframe modifications are minor and engine modifications are straightforward. The major cost of the FADEC system is the FADEC unit itself with its electric motor interfuce to theTA control. It is estimated that the cost of the FADEC hardware (electronic control and modifications to the hydromechanical control) is approximately equal to the cost of a TA hydromechanical control overhaul.

CONCLUSIONS AND RECOMMENDATIONS

1) The AH-1tr53 Cobra rotor drive train is very lightly damped. Thelefore, any significant disturbances such as rocket fire swges or potential fuel control or IGV corrections that are not phased properly tend to excite the torsional modes of the drive train and cause transient overtorques.

2) An electronic control, however, modulating engine fuel flow and torque in direct opposition to drive train resonances has been shown by simulation to damp Out torque spikes and hold them within nonnal operating limits. The electronic control can be readily inter&ced to the existing TA hydromechanical unit via its separate emergency fuel valve. This controller also provides an inherently faster engine and rotor speed control loop for nonnal operation. Thus, transient rotor speed droop and engine torque overshoots during aggressive maneuvers are significantly reduced as compared to the existing bill-of-material control. Thus, a side benefit of increased helicopter agility with less pilot workload is realized.

3) Because the above work is based on a simulation of the engine, it is recommended that an engine test be perfonned whereby high frequency fuel flow inputs are applied to the engine. The objectives of this test are to:

a) Verify that rapid fuel flow modulation will not adversely affect the TS3 engine.

Previous work pet formed under U.S. Army contract showed that this concept is feasible using the 250 engine (Reference AVSCOM Technical Reports: USAAVRADCOM-1R-83-D-l, Adaptive Fuel Control Feasibility Investigation, dated 1983; USAAVSCOM-1R-86-D-14, Adaptive Flectronic Fuel Control for Helicopters. .. 250 Engine Testing and 206L Helicopter Flight Test, dated 1986). Therefore, a high confidence level exists that the rocket fire problem can be solved by modulating fuel ftow as described herein.

-43-

b) Verify the engine's tast path fuel fiow to torque relationship. Combustive damping of the drive train resonances is highly dependent on this gain. The final -control design requires an accurate definition of this path.

c) Verify the engine's combustor and gas generator small signal response characteristics beyond the tail rotor resonant frequency of approximately 10 Hz. Again, the final control design requires a good definition of the core engine response.

The engine test can be performed with the existing TA control and an off-the-shelf PC based controller operating a high response fuel valve that is switched in at various operating conditions of the engine to modulate Wf about the steady running condition.

4) Depending oo the Slxx:ess of the above engine test, an AH-l!I'53 Cobra flight test demoostratioo is recommended. A single channel FADEC intermced to the modified TA control is recommended to insure flight worthy hard\WI'e. The objectives of this test are:

a) Demonstrate combustive damping of the rotor drive train. Torsionals can be easily induced on the Cobra drive train by modulating PLA.

b) Demonstrate enhanced maneuvering C3JlWility and handling qualities.

· c) Demonstrate insensitivity of the system to rocket fires.

5) Depending on the success of the flight test demonstration and a life cycle cost study, it is recommended that the FADEC system with TA hydromechanical backup be considered for incorporation into a U.S. Army or National Guard fieet of Cobra helicopters. Or, the advanced control concepts can be incorporated into modem FADEC equipped helicopters such as Comanche.

-44-

··~

APPENDIX A

-45-

List of Symbok and Acronyms

AA'ID

AOOT

AVSCOM

BEEP

BF

BLEED

BR

Br

CIP

CDP

FADEC

IGV

JF

JGB

JR

IT

KR

KS

KT

Nl

NASA

NDOT

NF

NG

NGB

NR

Nf

Pl

Aviation Applied Technology Diiectmate (U.S. Anny)

Helicopter Vertical Velocity

Aviation Systems Command (U.S. Anny)

Pilot's Rotor Speed Trim to TA Control

Power (Free) Tutbine Damping

Engine Bleed Band Position

Main Rotor Damping

Tail Rotor Damping

Pilot's Collective Pitch Input

Compressor Discharge Pressure

Full Authority Digital Flectronic Control

Engine Inlet Guide Vane Position

Power (Free) 1\nbine Inertia

Gear Box Inertia

Main Rotor Inertia

Tail Rotor Inertia

Main Rotor Mast Spring Rate

Engine Output Shaft Spring Rate

Tail Rotor Shaft Spring Rate

Gas Genemtor Speed Computer Output (TA Control)

National Aeronautics and Space Administration

Gas Generator Acceleration

Power (Free) Turbine Speed

Gas Generator Speed

Gear Box (Fngine to Rotor Drive Train) Speed

Main Rotor Speed

Tail Rotor Speed

Engine Inlet Pressure

-46-

PAMB Ambient Pressure

PB Boost Pump Pressure (fA Control)

PC Case Pressure (fA Control)

PEDAlS Pilot's Tail Rotor Pitch Input

PLA Power Lever Angle (fA Control)

PM Metering Valve Discharge Pressure (TA Control)

PN Engine Nozzle Fuel Pressure

PS Servo Pressure (fA Control)

PIG Power Turbine Governor

QD External Torque Disturl>ance to Rotor Drive Train

QF Engine Gas Torque

QRS Main Rotor Mast Torque

QS Engine Output Shaft Torque

QTS Tail Rotor Shaft Torque

SHP Engine Shaft Horsepower

T1 Engine Inlet Temperature

T1SENSE Lagged Tl Sensor Output (fA Control)

T4.5 Interstage Turbine Temperature

TA Turbine Assem~ly

TECOM Test and Evaluation Command (U.S. Army)

WF Engine Fuel Flow

WFSS Engine Steady State Fuel Flow

WFMV Metering Valve Position (TA Control)

xoar Helicopter Forward Velocity YAW Helicopter Heading Change

0 P1/14.7

e (f1 + 460)/519 T Linearized First Order Tm1e Constant

-47-

Figure 1

Figure 2

Figure 3

Figure4

Figure 5

Figure 6

Figure7

FigureS

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure17

Figure 18

Figure 19

Figure20

AH-1tr53 Engine Control System

TA Control System ..• FWlCtional Block Diagram

T53 Core Engine. •• Fwtetional Block Diagram

Rocket Fire Engine Swge Simulation

AH-1tr53 Rotor Drive Train- Schematic

AH-1fl'53 Rotor Drive Train - Ftmetional Block Diagram

AH-lfl'53 Rotor Drive Train Frequency Response - NF/QF

Definition of Flight Maneuver .•• GlUl Run

Simulated Flight Maneuver ... GWl Run

MK66 Rocket Fue Flight Test Data (1 Pdir)

Simulation of MK66 Rocket Fue Flight Test (1 PcUr)

Simula!ion of Multiple Rocket Fll'eS (3 Pair)

Simulation of Rocket Fue Flight Test (1 PcUr) ••• With Fuel Flow Constant

. Simulation of Multiple (19) Rocket Fues. •. With IGV Synched to Rocket Fire

Simulation of Rocket Fue Flight Test (1 Pair) ••• With IGV Synched to Sw:ge

Combustive Damping Block Diagram

Effect of Active Damping on AH-1fl'53 Rotor 1'orsionals Torque DistUibance Rejection NF/QD

Simulation of Rocket Fue Flight Test (1 PcUr) ••• With Bectronic Combustive Damping

Simulation of Multiple Rocket Fll'eS (3 Pdir) ••• With Bectronic Combustive Damping

Effect of Combustive Damping on AH-1 Rotor Drive Train Torsionals

-48-

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure26

Figure 27

Frequency Response Comparison of Power Thrbine Speed Control- NFINF Wldeband Governor vs. TA Control . Torque Distw:bance Rejection Comparison - NF/QD Wideband Governor vs. TA Control

TA Hydromechanical PI'G Perfonnance. .. 2 Seoot1U LOht...."tive Pull from Low to High Power

Wideband Electronic PTG Perfonnance •.. 2 Second Collective Pull from Low to High Power

Chandler Evans "Generic" FADEC

1'53 FADEC ... Software Architecture

Model TA-7 "Modified" Fuel Metering Secti.on .•. Schematic

Model TA-7 Control Modified for FADEC Interface •.. J..ayout Drawing

-49-

REPORT DOCUMENTATION PAGE Form Approved OMB No. 0704-0188

Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503.

1. AGENCY USE ONLY (Leave blank) 12. REPORT DATE 13. REPORT TYPE AND DATES COVERED

April1994 Final Contractor Report 4. TITLE AND SUBTITLE 5. FUNDING NUMBERS

Hot Gas Ingestion Effects On Fuel Control Surge Recovery and AH-1 Rotor Drive Train Torque Spikes WU-None

C-NAS3-26075 6. AUTHOR($) 1Ll62211A47

Frank Tokarski, Mihir Desai, Martin Books, and Raymond Zagranski

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION REPORT NUMBER

Coltec Industries Chandler Evans Control Systems Division West Hartford, Connecticut E-8638

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORING/MONITORING AGENCY REPORT NUMBER

National Aeronautics and Space Administration NASA CR-191047 Lewis Research Center

Cleveland, Ohio 44135-3191 ARL-CR-13

11. SUPPLEMENTARY NOTES

Project Manager, George A. Bobula, Vehicle Propulsion Directorate, organization code 0300, NASA Lewis Research Center, (216) 433-3698.

12a. DISTRIBUTION/AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE

Unclassified-Unlimited Subject Category 07

13. ABSTRACT (Maximum 200 words)

This report summarizes the work accomplished through computer simulation to understand the impact of the hydro-mechanical turbine assembly (TA) fuel control on rocket gas ingestion induced engine surges on the AH-1 (Cobra) helicopter. These surges excite the lightly damped torsional modes of the Cobra rotor drive train and can cause overtorqueing of the tail rotor shaft. The simulation studies show that the hydromechanical TA control has a negligible effect on drive train resonances because its response is sufficiently attenuated at the resonant frequencies. However, a digital electronic control working through theTA control's separate, emergency fuel metering system has been identi-fied as a solution to the overtorqueing problem. State-of-the-art software within the electronic control can provide active damping of the rotor drive train to eliminate excessive torque spikes due to any disturbances including engine surges and aggressive helicopter maneuvers. Modifications to the existing TA hydromechanical control are relatively minor, and existing engine sensors can be utilized by the electronic control. Therefore, it is concluded that the combination of full authority digital electronic control (FADEC), with hydromechanical backup using the existing TA control enhances flight safety, improves helicopter performance, reduces pilot workload ... and provides a substantial payback for very little investment.

14. SUBJECT TERMS

Engine surge; Torque spike; Turbeshaft; Hot gas; Drivetrain oscillation; Active damping; Fuel control

17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIRCATION OF REPORT OFTHISPAGE OF ABSTRACT

Unclassified Unclassified Unclassified

NSN 7540-01-280-5500

15. NUMBER OF PAGES 54

16. PRICE CODE

A04 20. LIMITATION OF ABSTRACT

Standard Form 298 (Rev. 2-89) Prescribed by ANSI Std. Z39-18 298-102