Document
#-
RESEARCH MEMORANDUM
PRELIMINARY INVESTIGATION OF THE CONTROL OF A GAS-TURBINE ENGINE FOR A HELiCOPTER B y Richard P. Krebs Lewis Flight Propulsion Laboratory Cleveland, Ohio """""""" """"
UNCLASSIFIED
”” _” . . .
NATIONAL ADTTSORY C O W T E E X FOR AwONAU!PICS cu By Richard P. Rhbs (u c u (u , A preliminary investigation of the power plant control problem for the helicopter was made. The results obtained f m an analog indicated that current turbine-propeller. engine c m t r o l s are suitable for the helicopter. Rotor thrust or propeller torque could be increased frm one-ha= to rated mlue in less than 4 seconds. Control operation was satisfactory up to an altitude of l5,OOO feet.
IC7TRODlETION Success in recent military operations has demonstrated the utility of the helicopter and stimulated interest in new designa for the heli-
copter. A m o w these desigm m e several for a helicopter with higher
gross weight than those helicopters now flying and powered by a gas- turbine engine. The gas turbine drives either & lifting rotor or a
pair af forward-&iving propellers. men t h e propellers are connected
t o the englne, the rotor is In autorotation.
Use of a single gas-turbine engins to power a helicopter introduces certain ccmtrol problems.became the rotor and propellers have greatly different characteristics. The question arises as to whether the B&me type of control is suitable f o r a power plant, the dynamics of which change wlth a change in parer absorber as we11 as with a change Fn altitude. . .
Accordingly, the- NACA Lewis Laboratory has made an introductory study of the power plant control problem f o r the helicopter. The dynamics 0 f . a controlled gas-turbine engine with appropriate rotor and propellers were studied with an electronic analog. The d p m i c response of the engine and control, which maintained e r g h e speed by regulatbg the f u e l flaw, was. detemined for four dlff erent mBneuvers. Each of them maneuvers involved power increase fram one-half to full-rated power. In one af these maneuvers, t h e j m p take-off, the rotor was
UNCLASSIFIECI
2 - NACA RM E5lF19
connected .to the engine and the flight a l t i t u d e was. sea level. Ln the.
other three maneuvers, power was absorbed by the propellers at sea level, by the r o k a t an a l t i t u d e of 15,000 feet, and by the a t 15,000 feet. A.secandtype a € ' control in which'thespeed propellers was regulated by the variation of the blade angle was investigated for I . - the jump take-off maneuver.
r u N R) N The analog investigation waa based on engine and propeller
characteristicsobtained f r o m mnufactuTBrB' d a t a . The engine m a
assumed t o be e h i l a r t o a T40. Because the manufacturer recammends that t h i s engine be run at rated speed whenever possible, only constant speedcontrols were considered. -The rotor characteristics were scaled fram data available on a ty-pical helicopter rotor. Perf'omnasce data
on the engine, rotor, and propellers in the f o m of prtial derivati-rrea
among the p e r f o m n e variables evaluated at the rated. p o w e r operating point- were b s e r t e d in the electronic analog ccaqputer. The dynamic performance o f ' the helicopter power plant was expresrJed as the response of tseveral d the performasce v a r i a b l e s t o a step change in the poxer- eetting lever.
."
In the system for which the aynamic respome was first determined, the angine speed wa8 kept catllsta.nt by oontrolling the fuel flow and the power was determined by the blade angle actuated throug3.1 the collective pitchcontrol. This system which I s representative a t currentturblne- propeller. engFne control is s h m echematielly -figure 1. The engine . . . " speed N is compared t ot h e desired s e t speed Ns, and the differenbe, or speed error, serves as theinput for thecontrol. The nature of the control is such that, except f o r an assumed lag in the f u e l system
-
which has beenincluded as part of the control, the change in f u e l flaw . ..
I s pro2ortional to the sum Or the speed error and its time intern-l.
The inclusion of the time integral feature in &e control.aesures the absence a P any steady-state error In engine speed. A Lag was ale0 included between the collective pitch cantrol and the rotor blades.
The response of r o t o r thrust F, exhaust.gas temperature T, - engine speed- N, rotor blade -angle p, and engine f u e l flow Wf a r e " shown f o r a jump take-off in flgu& 2. The time base f o r a l l t r a c e s ie 20 seconds. For this sea-level mEbneuvB'r the collective pitch control . .
was rairjedInstantaneously f m m a positioncorrespondingto 50 percent ' power t o a position comeepondlng - t o full parer. Pertinent data f o r t h e i n t e r p r e t a t i m of figure 2 are as follme: " NACA RM E51F19 - .... -
Altitude. . . . . -. . . . . . . . . . . . . . . . .- . . . . . ... s e a level
Power absorber . . . . . . . . . . . . . . . . . . . . . . . . . r o t o r
Engine-rotor time constant. . . . . . . . . . . . . . . . . 1.5 seconds
Fuel system lag . . . . . . -. . . . . . . . . . . . . . . . 0.2 second
Blade system lag. . . . . . . . . . . . . . . . . . . . . . 1.0 second
Controlintegraltimeconstant. . . . . . . . . . . . . . . 2.0 seconds
Loop gain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .lo. 2
The loop gain is defined a s t h e change i n speed f o r a unit c h a w e i n cu N speed error when the control loop is open.
N co follows t h e change i n blade angle almost The r o t o r ' t h r u s t perfectly. Because of the lag i h t h e blade actuating m e c h a n i s m , the.
63 percent of their excursion in 1 secondand rotor blades execute complete 98 percent of. theirexcursion in 4 seconds. Likewise, t h e rotor thrust makes 98 percent of its change i n 4 seconds.
N e i t h e r the exhaustgas temperature nor t h e fuel flow overshoot t h e i r f i n a l v a l u e . The success of t h e systemand t h e a b i l i t y of t h e ' r o t o r t h r u s t t o follow the change in blade angle arises from the very .
nearu-constantrotationalspeed of t h e power plant. The dip i n speed amounts t o less than 1 percent of its rated value.
The response of the helicopter power plant at sea level when drivlng the propellers is shown i n figure 3. Again t h e power change i s from one-half t o f u l l rated power. The time abscissa for a l l traces is 20 seconds.Thrust data f o r thepr.opellers w e r e not available, and so propellertorque &p was substituted for the top trace.Pertinent data f o r figure 3 are:
Altitude. . . . . . . . . . . . . . . . . . . . . . . . . . . s e al e v e l
Power absorber. . . . . . . . . . . . . . . . . . . . . . . propellers
Engine-propeller time constant. . . . . . . . . . . . . . . 0.67 second
Fuel system l a g . . . . . . . . . . . . . . . . . . . . . . 0.2 second
B l a d e system lag. . . . . . . . . . . . . . . . . . . . . . 1.0 second
Controlintegraltimeconstant. . . . . . . . . . . . . . . 2.0 seconds
Loop g a i n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4
The responses shown i n figure 3 are almost i d e n t i c a l t o those shown i n figure 2. The propeller torque follows the change i n blade angle and very nearly attains its final v a l u e i n 4 seconds. Because t h e loop gain remained practically unchanged, it is possible t0 conrpare the results of figures 2 and 3 and observe that a two-to-one change i n power plant time constant has l i t t l e e f f e c t on theresponses. The engine speed deviated about 0.5 percent from i t s rated value.
The performance f o r t h e r o t o r drive and propeller drive at an &ti-
tude of l5,OOO f e e t is sham i n figures 4 and 5, respectively. A t alti-
tude- the pow- plant time constant i s increased by a factor of 1.68 and NACA R M E513719 the loop gain is increased-by a factor af.1.78 over the respective values at sea level. Thesechanges in.syst& paxamete2s have little apparent effect on .the dynamic performance of the system.
The results illustrated i n figures 2 t o 5 in which the control constants were not chaqged indicate that a control system s i m i l ~ t o the one illustrated i n figure 1 . i s . s a t i s f a c t o r y f o r controlling a gas- turbine engine when driving either a lifting rotor o r a p a i r of forward- driving propellers at a l t i t u d e s . u g t o 15 ,000 f e e t . N t u N c u Should it be possible to increase the speed. of r e q o n s e o F t h e bladeactuating mechanism, faster thrust responsecan be obtained. The r e s u l t s i l l u s t r a t e d i n figure 6 were obtained f o r a system simi-lar t o the one f o r which r e s u l t s are shown i n figure 2 except that the blade system l a g has beenreduced t o 0.5 second..The rotor thrust reaches 98 percent of its final value in 2 seconds, but the speed change has increased to-almost 1.9 percent o f i t s rated value. Any further attempts to increase the speed o f . response .oP the thrust would probably i n i t i a t e temperature overshoots.
.
Ap investigation was also made of the performance of the heUcopter power p l a n t i n which constant speed was maintained by regulating t h e . .
blade angle and power W&S set by the fuel f l o w t h r o t t l e . such a.con€-rol h is shilar t o the one recommended by the enginemanufacturer for turbine- propeller.service. The control system i s -shown schematically i n f i g u r e 7.
A speed error is determined a s ' i n the'first control by taking the differ- ence between theenginespeed and a set speed. The nature of the control is such that, except f o r the assumed lag i n the bladeactuating mechanism I which has been included as p a r t of the control, the change i n blade angle . .
i s proportional to the sum o f the speed error asd i t s time integral.. The f u e l system l a g has beenincl'udedbetween the t h r o t t l e and the e&ne.
Dynamic response of the gas-turbine engine driving 8 l i f t i n g r o t o r
at sea l e v e l and being regulated by a blade-angle -- epeed control i e shown
i n f i g u r e 8. Pertinent data for the power plant and control arei s e a l e v e l A l t i t u d e . . . . . . . . . . . . . . . . . . . . . . . . . .
rotor Power absorber . . . . . . . . . . . . . . . . . . . . . . . . . .
Engine-rotortimeconstant . . . . . . . . . . . . . . . . 1.5 seconds
Bladesystem lag . . . . . . . . . . . . . . .-. . . . . . . 1.0 secbad
Fuel system lag . . . . . . . . . . . . . . . . . . . . . . . 0.2 second
Control integral time constant . . . . . . . . . . . . . . 3.0 eeconds
. . . . . . . . . . . . . . . . . . . . . .5.2
Loop gain. . . . . . . ..
For an increase .in.gower from 50 percent t o r a t e d , the thrust reaches its r a t e d v a l u e i n 1 second, overshoots by 21 percent of its rated Value, returns and remains -canstant after about 8 seconds.' The speed also exceeds i t s rated value by 5 percent. Although the thrust
NACA RM E5lF19 ” . . “ e - 5
overshoot may not be objectionable, the speed overshoot w3.J-l endanger t h e engine. It is considered that a speedovershoot accampanying a Lower increase would be more detrimental to the engine than a speed overshoot accmanying a power decrease. The loopgainandcontrol i n t e g r a l time constant were chosen t o glve w h a t appeared t o be the most satisfactory speedand thrustresponse.Bny’furtherincrease i n the gain of the control aimed at reducingthe speedovershoot would render the thrust even more oscillatory and would introduce overshoots KJ in the temperature response.
N - N N
Ih the bIade-angle - speed control eyetEtm t h e r e a r e two lags of
aboutequal magnitude i n the control loop: a lag of 1.5 seconds f o r t h e power plant itself, and a lag of 1.0 second for the blade actuating mechanism. The presence of t h e t w o neaxlyequal lags i n t h e system i l l u s t r a t e d by figures 7 and 8 makes t h i s system more oscillatory with a loop gain of 5.2 than the system i l l u s t r a t e d i n figures 1 and 2 with a loopgain of 10.2. The decreasedloop gain,results i n an increased speed e r r o r .
A f’urther disadvantage o f - t h e blade-angle - speed control system f o r
helicopterservice is a mechnical one. When theengine is switched from rotor drive t o propeller drive, it would be necessary t o open the control loop. Special provision would have to be made during the change- over period to prevent the engine from making any radical change i n speed.
SUMMARY OF RESULTS A preliminary investigation o f the control problem of a gas-turbine powered helicopter indicated that ourrently proposed turbine- propeller engine controls are applicable f o r either lifting rotor drive or propeller drive. Power could be increased from one-kralf rated t o full power i n 1 e s s . t h a n 4 seconds. Satisfactory operation was indicated f o r a l t i t u d e s up t o 15,000 feet. The results w e r e obtained from an electronic analog computer.
Lewis Flight Propulsion Laboratory, National Advisory Connnittee.for Aeronautics, Cleveland, Ohio.
. ’
-
Wf
"
N
Ns x - CONTROL
.
T
ENGINE
P
F
F i g u r e 1 . - Schematic disgram of fuel-flow - s y e d control system.
ZZZZ . . .. . . " .
NACA RM E52319
-I
T
N
P
0 5 10 15 20 I I I 1 Time, sec Figure 2. - Controlled engine response for a step increase in .
m e r setting. F u e l - f l o w - speed control. Power absorber, rotors
altitude, sea level; blade-system lag, 1.0 second. .
T
0 5 IO 15
Time, sec Tgz&.7-
Figure 3. - Controlled englne reepqnse f o r a step increase In- p o ~
setting. F u e l - f l o w - sped control. B w e r absorber, -pmpellers; altitude, sea.leve1; blsik system lag, 1 . 0 second, .~ . . . .
. . .
- ..
-
.
: 0 5 10 15 20
.
Time, BBC v
Figure 4. - Controlled engine respame for a step increase i n p o w e r
setting. Fuel-flow - speed control.Parer absorber, rotor; altitude, W,ooO feet; blade system lag, 1.0 second.
t\3 N h)
f " Q p
1u
dN
NACA RM E5W19
t
Figure 6 . - Conkrolled engine respnse f o r a step Increase in
p o w e r setting. Fuel-flow - speed Cpntr~l. Foweer absorber, rotor; a l t l t u d e , sea level; blade system lag, 0.5 second.
. . . . . . .. .. . . . .. . . . . . . . .. . . . . . . .. . . . . . ..
I -
P
-CONTROL
N
T
ENGINE
F
Wf
>
I
I
I . . . . . .
NACA R4 i35LF19
n
F
N N cu c\l Y " 0 5 10 L5 20 I I I I I Time, aec
F i g w e 8. - Controlled engine r e v n s e for 8 step increase i n power
setting. Blade-angle - speed control. B w e r absorber, rotor; altitude, sea level; blade system lag, 1.0 second.
t .