Publisher’s version / Version de l'éditeur:
National Research Council of Canada Aeronautical Report, 1968-02
READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE.
https://nrc-publications.canada.ca/eng/copyright
Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca.
Questions? Contact the NRC Publications Archive team at
PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. If you wish to email the authors directly, please see the first page of the publication for their contact information.
NRC Publications Archive
Archives des publications du CNRC
For the publisher’s version, please access the DOI link below./ Pour consulter la version de l’éditeur, utilisez le lien DOI ci-dessous.
https://doi.org/10.4224/21272068
Access and use of this website and the material on it are subject to the Terms and Conditions set forth at
Description of a four degrees of freedom, V/STOL aircraft, airborne
simulator
Daw, D. F.; Lum, K.; McGregor, D. M.
https://publications-cnrc.canada.ca/fra/droits
L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.
NRC Publications Record / Notice d'Archives des publications de CNRC:
https://nrc-publications.canada.ca/eng/view/object/?id=5989cb18-a63d-44b4-b479-765790d80bcc https://publications-cnrc.canada.ca/fra/voir/objet/?id=5989cb18-a63d-44b4-b479-765790d80bcc
' I ! / '• ; r ·< .. · ' • ·
..
,, , (..
ｾＮ＠ '-. I· ' ｾ＠...
..
Ｎ ｾ＠.... · ... · f ' I .:.-"' ｾ＠,·
I tl r ..·
'" ,. I / ;I·:I
NATIOriAL .
ｾ ｅｓｅａｾ
ｃ ｈ＠
CQ.UNCIL
OF
CANADA
':,,! · '. , .r· • . ! ' ' ,j . ' ., Ｍｾ Ｎ＠
·.r .
·:AERONAUTI.CAL REPORT LR-499 ; ',• ' ' I ｾ＠ ,. _ r . • ( I /:. I I I ' I / f , / J ｾ＠ I f • I I ; ' I.. I
' I •I I' I : I ·, / ｾ＠ Ｌｾ＠ ... . _., It' ' /,.
IｏＧｅｓｃｒｉｐＮｔｩｴｈｾ
Ｈ ｏＧｆ＠
A .F.OUR ·DEG·REES
OF.
FREEDOM,
I ' ' '. ' , I I '
V/STOL f'IRCRAFT,
ａｉｒｽＳＰｒｎｾ＠
SIMULATOR
. . . ., / : . . • r ,• •I I, ·, /
.
'.·
' • I I I r ( I 'r I • .: ID. F. ｾａｗＬ＠ !(: LUM AND D. M. McGREGOR . NATIONAL AERONAUTICAL ESTABLISHMENT
. .
,,
' ' lilTTAW!' I FEBRUARY 1968. ·, '.: ·'' {.
I.·
/.· ' I I ,,.J.: · · · I I.
/ : ./ NRC NO.- lp182 ;, ; f '! / I ' 'TABLE OF CONTENTS Page SUMMARY
... ... ...
(iii) TABLE • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 0 • • • • • • • (iv) ILLUSTRATIONS . . . . (iv) SYMBOLS . . . . (v) APPENDICES...
(vi) 1. 0 INTRODUCTION 1 2.0 GENERAL DESCRIPTION OF THE SIMULATOR 1 3. 0 DESCRIPTION OF EQUIPMENT 3 3 . 1 Analogue Computer . . . . 33. 2 Equipment Associated with the Computer ... .. ... .. . . . 4
3. 3 Autopilot . . . · . . . . 5
3. 3. 1 Lead or Pre-emphasis ... . . . 5
3. 3. 2 Control Loop Compensation . . . . 5
3.3.3 Positional Servo Actuators . . . .- . . . . 6
3. 3. 4 Feedback Sensors . . . . 7
v . v . · J.. ＮｊＮｾ ＮＮ｟ＮＮＬ＠ ... .._..._.._ ... ｾｾ＠ ._.,._. __ ....,....,_ 3. 3 . 5 Filtering . . . : : : : : : : :
1
3 . 4 Tape Recorder . . . . 7
3 . 5 Synthetic Instrument Landing System (I. L. S.) . . . . 7
4 . 0 CONCLUDING REMARKS . . . . 7
5 . 0 REFERENCES . . . . 8
TABLE Table Page 1 Undamped Natural Frequency for Four Control Loops 13 ILLUSTRATIONS Figure Page 1 Four Degrees of Freedom V/STOL Airborne Simulator . . . . . . . . . 15
2 ''Model-Controlled 11 Simulation Method. . . . . . . . . . . . . . . . 16
(iv) Figure 3 4 5 6 7 8 9a 9b 9c 9d 10 11 12 13 Symbol
c
E F.M. Hz ips K ILLUSTRATIONS (Cont'd) Page View of the Simulator Cockpit from the Right Side . . . . . . . . . . . 17The Analogue Computer Used for the Model . . . . . . . . . . . . . . . 18
View of the Simula tor Cockpit from the Left Side . . . . . . . . . . 19
Orange Sub-Cockpit Installation for Simulated Instrument 20 Flying . . . . Power Spectrum of Simulated Atmospheric Turbulence . . . . . . . . . . . . 21
Autopilot Schematic • • • • • • • • • • • • • • • • • • 0 • • 0 • • • • • • • • • • • 0 • • • • • • • • 22 Frequency Response for Yaw Control Loop . . .
0..
23Frequency Response for Pitch Control Loop . . . . . . . . . . . . . . . . 24
Frequency Response for Roll Control Loop . . . . . . . . . . . . . . 25
Frequency Response for Heave Control Loop . . . . . . . . . . . . . . . . . 26
Analogue Computer Study on the Effect of Lead Network on Helicopter Transient Response . . . . . . . . . . . . 27
Autopilot Actuator System . . . . . . . . . . . . . . . . . . 28
Synthetic Instrument Landing System . . . . . . . . . . . . . . . . . 29
Illustrative Computer Wiring Diagram for the Directional Equation of Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
SYMBOLS Definition
Yawing moment of inertia, slug rt2
Product of inertia, slug ft2
Frequency modulated Cycles per sec
Inches per sec
Gain in autopilot compensation and Gain in turbulence expression
Symbol L N p r
s
t Tv
(3 0 a 0 r cp ((.) SYMBOLS (Cont'd) Definition Scale length of turbulence, ftYawing angula r acceleration per unit subscript, rad/ sec2 /unit subscript
Angular rate of roll, rad/ sec Angul ar rate of y aw , racl/ sec Laplace opera tor
Time , sec
Time constru1t in autopilot compensation Flight velocity, ft/ sec
Angl e of attack, rad Angle of side slip, racl
Pilot ' s roll control deflection, in Pil ot' s yaw control deflection, in
2
Angul ar acceleration in yaw , rad/ sec
2
Angul ar accel eration in roll, rad/ sec Frequency, rad/ sec
Unda mp ed na t ur a l frequency, rad/ sec
Subscripts
H Helicopter
Other sub scripts a re defined above.
APPENDICES
App . Page
A App e ndix A 0 • • • • • • • • • • • • • • • • • •• • • • • 0 • • • • • • • • • • • • • • • • • • • • • • 31
(vi)
1
-DESCRIPTION OF A FOUR DEGREES OF FRE E DOM , V/ STOL AIRCR AFT, AIHB ORNE SIMULAT OH
1. 0 INTHODUCTION
It is wid ely r e cogni zed t ha t th e dev elopm.e nt of ve rtical and short take -off
and la nding a ircr a ft is still in its r e l a tiv e infancy. Ther e r e m a in m a ny unknowns in establishing optimum or, in fa ct , ac ceptable handling qu a litie s , the most suitable modes of flight ope r a tions, pa rticul a rly unde r instrume nt conditi ons, s a tisfactory flight instruments to allow th e pilot to utili ze th e a ircr a ft's pote nti als fully, m ea ningful desig11 criteria, and other fa c e ts of their d e sig11 and us e. Current he licopter a nd
V/STOL aircraft sp ecifications (He f. 1 ru1d 2) wer e f ormul a ted l a r ge ly from exp eri enc e
with he licopters tha t, in ge ne r al , wer e v e ry differ e nt from the pre s e nt a nd th e proposed g ene r a tion of V/STOL aircraft.
Cl e a rly, a ne ed exists for a ve rsatile airborne simula tor c apabl e of reproclueing and simulating, with high fidelity, a wid e r a nge of flying char a cteristic s app li -cable to t llis new bre ed of a ircr·:tft. Such a res earch tool m.al< e s i t pos s ible to inves-tiga te the above problems in a systema tic and controlled m a nner in the proper a ir-borne envirom11 ent .
The first a ttempt , by the Nationa l Aeronautical Establishm e nt, to fi ll tllis requirement r esulted in the de velopm e nt of the vari able stability helicopte r described in Heference 3 . The flight chara cteristics of this simulator coul d be a lte r ed s i gnifi-cantly about its three rota tional d e gree s of freedom by use of the model-controlled method of simulation, and e nabled a va riety of handling qualities investiga tions to be
successfully compl eted (e. g . Hef. 4 and 5).
With the experience of building and operating the three degrees of fr eedom simulator, the developm e nt of a v a stly improved v a riable stability helicopter wa s undertaken incorporati ng the c a pability to alter its chara cteristics in the v ertic a l ,
or hea ve chmmel, as well as the three rotational modes of motion . An ele ctro-hydraulic autopilot and an increased computing capacit-y ha ve gT eatly enha nced its us e fulness
and ha ve yie lded a reliable a nd yet flexible simula tor. Tllis is d e monstr a ted in the
three progTams reported in Heferences 6 , 7, and 8 - one of a general res earch nahtre
a nd the other two simulations of particula r aircraft .
This Heport describes the equipment installed in the helicopter to a c hieve its variable stability capability and indicates how it is used to simul a te various flight chara cteristics .
The Appendix contains an illustration of how the equations of motion are programmed for use with the simulator.
2. 0 GENEHAL DESCRIPTION OF THE SIMULATOR
The basic vehicle used for this airborne simul ator is a light, single rotor
helicopter, a Bell 47G-3Bl (Fig. 1) with a maximum gross weight of 2950 pounds and
2
-the full engine power in summer months during -the prolonged hovering and low speed flight necessary to evaluate certain V/STOL aircraft characteristics.
It will be noted from Figure 1 that the stabilizer bar, one fuel tank, and the
b a ttery have been removed to lighten the aircraft, allowing the installation of more computing components and project equipment. Even with the fuel capacity reduced in this way, the available testing time is still in excess of 1! hours.
Since the simulator system ultimately relies on the basic helicopter controls to produce the desired motions, the maximum accelerations (control powers) employed in simulations are limited to those attainable by the helicopter. These are quite high, how eve r, when compared with existing V/STOL aircraft in the low speed flig·ht regime, and have been measured to be 2. 4 rad/ sec2 in yaw, 1. 2 rad/ sec2 in pitch, 2. 2 rad/ sec2 in roll, and 8 ft/ s e c2 in he ave .
Two pilots occupy the simulator during a ll flight tests, with the one on the l eft side a cting as safety pilot a s well as program rnanager and the one on the right flying the aircraft tlrrough the 11 fly-by-wire 11 system while assessing the handling qualities being simulated. A screen is installed between the two cockpits to prevent the evaluation pilot from seeing the s afety pilot's control mov e m e nts, which in some cases are entir ely different from his own and, hence, are quite disconcerting if in his field of view.
The simula tion method used is the "model-controlled" method (Ref. 9), which was demonstrated by the original simulator to achieve accurate and predictable flight characteristics without a detailed knowledge of the stability and control charac-teristics of the basic vehicle. With this approach to simulation (Fig . 2) , movements of the evalua tion pilot's el ectric flying controls (Fig. 3) supply sig·nals proportional
' ＮＬ Ｌ ＬＮｾ Ｌ＠ ｾＱ ｾ Ｎｮ Ｂ Ｂ Ｇ Ｍ Ｚ Ｂ ＢＧ ＢＧ＠ J-,., i·h'"' ...,,..., ln o1 1P roll1nut.er . :I:he computer (Fio·. 4) is "Ratched"
to control deflections to the ana ogue compu ·er . Tlie computer (Ylg . '±) 1s " paLc;neu ..
for the characteristics of the e qu atio ns of motion to be simulated and yields signals tha t are properly scaled to r ep res ent the yaw, pitch, and roll a ngular rate s, and the normal accel era tion of the simulated vehicle. These calculated motions a re compared with thos e of the helicopter, a s sensed by angular rate gyros and normal accelero-meters, a nd the autopilot, operating in a closed-loop fashion, forces the helicopter to follow the motions prescribed by the computer, Other inputs to the equ a tions of motion such as synthetic ilubulence, simulated steady winds, and angles of attack and sideslip as measured by vanes, can be utili zed to produce realistic simulations of very hi gh fide lity. Tins s i mul ation method has the adva ntage s of pr eventing the trim changes occurring on the test air craft from being f elt by the evaluation pilot and re-sisting· disturbances in the controlled de gree s of fr eedom tha t a r e due to external tur-bulence.
All four of the autopilot acLctators are connected in parallel with the safety pilot's controls . In the main rotor collective, a nd both the longitudinal and l ateral cyclic systems they are loc ated befor e the irr eversible hydr aulic boost cylinders normally part of the basic helicop ter. The inherent dir ectional control forces are very light; conse-quently, the yaw a utopilot actuator was connected directly to the tail rotor collective pitch sys tem in parallel with the safety pilot's pedals . Since the autopilot ha s 100 percent authority, this method of attac hm ent was nece ss a r y to keep the s afety pilot informed of the true control positions and en abl e him to a ssume positive control at any time.
The a nalogue computer a nd autopilot a r e controlled, trimmed, programmed
3
-a nd monitored by the s-vvitc he s, g:-ain-setting potentiometers, function switches, -a nd
ｮｾ･ｴ･ｲＬ＠ mounted on the console SItuated b etween the pilots in the cockpit and shown in Figure 5.
. . The ･ｶ｡ｬｵ｡ｾｩｯｮ＠ pilot's :ontrol l eveFs can assume any form providing el ec
-tncal signals proportiOnal to then deflections can be obtained. 11Pow er l ever' ' and
11 collective 11 height controls have be en utili zed as well as a conventional control column
ｾ､＠ ｾ＠ wheel for pitch and _roll control. . Spring feel with viscous da mping and a djustable
fncti_on ｨ ｡ｶｾ＠ been used ｗｉｾｨ＠ the roll, pitch, and yaw controls, while adjustab le friction
only IS provided for the height control. Both "beep " and "push-to-r e lease " trimmers
ｾｲ･＠ ｡Ｚｶ｡ｩｬ｡｢ｬｾ＠ for the ｾｯｮｴｲｯｬ＠ c_olm:rm. A more fl exible el ectro-hydr aulic feel system IS bemg designed for msta lla t10n m the near future.
Two pods are mounted ex t ernally, a s shown in Figure 1, with th e one on the port side housing the m ain portion of the ana lo gue computer (Fig . 4) while th at on the sta rboar? side accommodates various power supplies, a 14channel tape r e -corder, plus miscellane ous el ec tronic equipment including the circuitry as socia t ed with the engine speed governor.
The e_ngine speed, and consequently the rotor speed, are governed by an electro-hydr aulic control system a llowing duplication of conditions existing in modern V/STOL aircraft equipped with free turb ine engines. The evalua tion pilot has been provided with a "b eep" trimmer for this system, which is normally insta lled on Ius power lever or collective control.
An i nstrument flight panel containing the six b asic Hight instruments,
ｴｯｾ･ｴｨ･ｲ＠ with two engine instruments and the instrument landing system cross-pointer driV en by the ground installation described in Sec tion 3. 5 , is situa ted in fron t of the
ｾｶ｡ ｬｵ ｡ｴ ｩｯｮ＠ pilot (Fig. 3). The se instruments can be r earra nged , replaced, or other mstruments added, a s was done for the r e s earch of Reference 8. During instrument
flight ･ｶｾｬｵ｡ｴｩｯｮ＠ t a sks, an orange sub-cockpit is insta lled on the right side (Fi g . 6)
a nd the pilot dons blue goggles, tins combination pr eventing lum from viewing th e exte rnal world wlule he r efers only to his instruments for flight information.
3. 0 DESCRIPTION OF EQUIPMENT 3. 1 Analogue Computer
The heart of this simulation method is, of course, the computer used to build up the model of the aircraft to b e simulated . Standard electrical a na logue plug-in
components, which ar e direc tly interchangeable with l aboratory units, hav e b een mounted in the fibreglass pods for this installation (Fig. 4). The b a sic computer consists of:
(i) 76 operational amplifiers (16 of which ar e used in the autopilot) ·
(ii) 30 integTating n etworks or non-lin ear elements such as diode function
generators, multipliers, etc:
(iii) 50 attenua tors or gain-setting potentiometers (20 of winch a r e sihtated on
the centre console in the cockpit to enab le the safety pilot to change charac-teri stics without unstrapping);
(iv) (v) (vi) (vii) (vi.ii) (ix) (x) (xi) (xii) 4
-12 function switches, also available to the s afety pilot to preset changes of
conditions such as stability augmentation failur e s (Ref. 7 and 8);
2 r elays with 4 double throw positions, which can be actuated by a switch on the
s afety pilot1 s collective handle to effect various changes in a stepwise fashion;
a transport time delay unit with a capability of inserting delays from 0. 1 sec
to 160 sec ;
2 ntrack-and-hold11 circuits that a r e used to eliminate transients as the vanes
are switched in a nd out of the model when the aircraft departs and approaches
hovering flight (see Para 3. 2 (i)) ;
60 balancing controls located in the cockpit, one for each amplifier used in
the model;
GO ov erload lights situated in front of the safety pilot that indicate when the
amplifiers exceed their limits of linear operation;
an ov er load horn that sounds in the safety pilot's left ear when any amplifier re aches its limit;
pa tching points to conne ct the computing components together and to introduce the sig11als from the various transducers, described in the next Se ction, into the electrical analogue;
pow er supplies that have been e specially designed and packaged to minimize both weight and space.
All these components can b e 11patched11 together to simulate quite complex
nxcd-point equa tions of motion in th e four controlled degrees of freedom. 3. 2 Equipment Associated with the Computer
A va r iety of tr ansducers a nd simulating elements have been installed in the airc raft for u sc in the equa tions of motion. These include:
(i ) Angle of a tta ck and angle of sideslip v anes that are used in forward flight above 25 knots to drive potentiom et ers yi e lding signals proportional to these para-m eters . Th ese vanes arc norpara-ma lly switched into the sipara-mulation during acceleration from th e hover (around 30 knots indicat e d airspeed) and switched out as the helicopter
decele r ates tlu·ough 20 - 25 knots on the a pproach to landing. Appropriate 11
track-and-hold 11 circ uit s preve nt tr ansi ents a s the changes are made.
(ii) A directional gyro that yields sig11als proportional to (a) the .heading of the
air cra ft , (b) th e change of an gl e from the heading at which the system lS .engaged, and
(c) the sine and consine of this change of heading, which can be used to s1mulate the c iTccts of a wind on the lon gitudinal and lateral velocities as the helicopter is turned in the hov er .
(iii) A ver tic al gyro that yi elds sig11als proportional to the roll and ｰｾｴ｣ｨ＠ angles
and has proven us eful in simulating various attihlde hold features of stab1hty augmen-ta tion systems.
5
-(iv) T\VO l a teral accelerometers situat ed with resp e ct to the centr e of gravity to
minimize the e ffe cts of yawing a nd rolling angular accel erations. The s e a ccelerometers
were used in the r e sea rch of Referenc e 8 as one sourc e of directional stabili zation.
(v) Three statistically independent synthetic turbulence generators tha t may b e
prog-ra1ru11ed a t any g·ust intensity l evel to give r ealistic flight disturbance s. These devices conta in Geiger tub e s a nd r adioa ctive ele m ents to provide sourc e s of random
pulses (Ref. 10), and their outputs can b e shape d by activ e filters to yi eld power
spec-tra approximating thos e of a tmospheric turbulence . A typic al shape us ed is shown in
Figure 7.
3 . 3 Autopilot
The desire d helicopter r espons e s a r e calculated by the ana log·ue computer
and, as indic a ted by the diagT a m of Fig·ur e 2, ar e compared with the actua l h e licopter
motions measured by angular rate gyros and normal acce lerom e te rs mounted on th e aircraft. To m ak e the autopilot control loops operate in the optimum. fashion, each
consists of a 11lead 11 term (or pre -empha sis), a compensation network, a hydr aulic
positional servo actuator, the basic aircraft , a f eedb ack sensor, and a filter n e twork
(Fig. 8) .
Va rious safety devices are built into the system to allow th e safety pilot to assume positive control of the aircraft should a dangerous situ a tion a ris e . One of these is a circuit that automatically disengage s the autopilot following a failure of a ny one of six critical voltages in the system, and simultaneously sounds an audio a l a rm
in the safety pilot1 s earphone.
The frequency response for each autopilot control loop, shown in Figur e 9,
was obt:'lined by determining the tr ansfer function for each system component. It was
not practical to determine the actual frequency response of each control loop experi-menta lly since this would h ave required forcing the aircraft at fr equencies tha t would excite struchtral modes, c ausing damage to the helicopter. Hence, only the undamped nat1.1ral frequencies were obta ined by increasing the loop gains until sustained oscil-l ations occurred. In most ca ses the experimentaoscil-l r e suoscil-lts shown in Taboscil-le I a r e in fair agreeme nt with those determined from transfer function a nalysis.
3. 3. 1 Le ad or Pr e -emphasis
In order to improve the response of the simula tor, a 11l ead 11 comp ensa tion
n etwork has b een pla ced between the mod el and the hydraulic servo in each chmmel (Fig. 8) . The s e circuits a ttempt to eliminate the l ag s due to th e basic helicopter by pre-e mpha siz ing the mod el command sig11al, and hav e succ eeded in enhancing the system performance as indicated in the frequency and lime domains in Figures 9 and
10. The insertion of these le ad terms does not affect the closed loop system stability,
sensitivity to external disturb anc es , etc., but only pre-empha size s the model outvuts to the autopilot.
3. 3. 2 Control Loop Compensation
Good model-following chara cteristics of the airborne simulator r equire the control system to hav e high closed loop gain a nd wide b a ndwidth relative to the band-width of the signa l from the model. High gain is essential for fast transient response,
6
-but the usable gain is limited by system overshoot and stability .
. . K (1 + TS) . . .
Compensation circmts of the form S have been apphed to the
angula r rate control loops ｾｯ＠ make_ ｴｨｾｬｾ＠ Type _1 ｳｾｳｴ･ｾｮｳＮ＠ _ This ensures ｩｮｦｩＮＱｾｴ･＠ _open
loop gains for de signals w1thout s1gmflcantly mer easmg t:1e ｰｨ｡ｳｾ＠ lags at h1gher
fre-quencies that would t end to reduce the undamp ed natural fr equenc1es.
For the heave control loop the compensation is of the lag-lead-lag form,
to cancel out the zero a t
s
= 0 term in the transfer function de scribing the he licoptervertical acceleration to collective control inputs.
H was found necessary to incorpor ate a gain change circuit in the pitch control loop because of the increasing control se.nsi1.ivity of the basic helicopter with
increasing speed resulting from the aerodynan:-1c effec t of the ｣ｯｮｴｲｯｬｬｾ｢ｬ･＠ ･ ｬ Ｎ･ ｶｾｴｯｲＮ＠
In order to maintain an opti-mum closed-loop gam_ throu?hout, the autop1lot gam ｾｳ＠
high at low speed and vice versa, with the alter a bon bemg made smoothly by a Ｎｬｾｧｨｴ ﾭ
sensitive resistance illuminated by a modulated la mp actuated by a speed-sens1tlve,
spring-loaded vane at the front of the helicopter. Thi_s ｾ｡ｩｮ＠ ｣ Ｎｨ ｡ ｮｧ ｾ Ｎ＠occu.rs a t
app.r?xi-mately 45 knots. The other thr ee control loops r e mam mv a n ant w1th fl1ght conchbon. 3. 3 . 3 . Positional Servo Actuators
ｾ｡｣ｨ＠ autopilot channel is connected to the appropriate helicopter control
through a hydraulic actuator (Fi g . 11) . To keep the frequency ｲ･ ｳｾｯＺＱｳ ･＠ of ｴｨｾ＠ .overall
system high these actuators hav e been m ade positional controls, w1th the pos1t1.on feedback being obtained from film potentiometers mounted on the achtator body.
Sin ce it i s necessary to en gage the autopilot in a variety of flight conditions
from still air hover to high speed flight, tr ack- and-hold circuits a r e installed in each
feedback path, which keeps the position sig11als zero until the autopilot is ｾｮｧ｡＿･､ Ｎ＠
Actuator changes from their initial positions ar e then me a sured and used ll1 ｴｾｮｳ＠
con-trol loop. These circ uits are dis engaged when the safety pilot actu.ates a sprmg-loadecl button on Lhe top of the control column, altering the actuator to an mteg-ratmg type,
thus enabling him to accur ately trim the positional loop before leaving ｴｾ･＠ ground.
While this button is engaged the integTating capacitors in the comp ens a twn networks o.re short-circuited, pr eventing any sig11als from the feedback transducers from aff ecti ng the rrctuator trim.
Hioh response electro-hydraulic valves control the flow to the actuators, vvhilc both an"' engin e -driv e n pump and a pump dr iven by an el ectTic motor during ground Lesting arc avai l ab l e to supply the 250 psi hydraulic pressure for system op-eration.
The safety pilot can override th e syste m by overpowering the autopilot acluators with Iorces of 17 lb in the longill.tdinal cyclic, 14 lb in the lateral cyclic, 25 lb on the trril rotor pedals, and 13 lb on the m ain rotor collective. Che ck valves
(Fi g. 11) have bee n incorporated around each achtator to m ake it possible for the pilot to mov e the controls by exerting forces just slightly higher tha n those above,
e\ e n if the el ec tro-hydr aulic valve fails in the fully closed position. The entire
autopilot (and e ngine speed governor) can be dis engaged by either pilot through buttons loc ated on their control columns, which a llows hydraulic fluid to be pumped freely from
7
-one side of the actuator to the other through the "by-pass" valve shovvn in Figure 11. 3. 3. 4 Feedback Sensors
The angular velocity of the simulator in eac h rotation al degT ee of freedom
is sensed by a high natur a l fr equ e ncy (30 Hz) r ate g·yro with a lin ear range of ± 60
deg/ sec. The three gyros that provide the e l ectric al signals for compa rison with the angular rates calculated in the model arc mounted on a m ac hin ed block that ha s been carefully aligned with the aircraft axes.
Two linear accelerometers, mounted with r esp ec t to the centre of gravity to eliminate signa ls du e to pitching and rollin g acce l erations, provide the feedback signal to be compared with the calculated norma l acce lerations from the mod e l. These
d evice s are subminiatl.tre 11force-balance" servo systems that have natural frequenci e s
of 100 Hz. 3. 3. 5 Filtering
Since the tr a nsduc ers in the autopilot feedback loops respond to motions of the simulator, they also a r e affec t ed by the r el a tiv e ly high vibration level inherent in the helicopter caused by rotating components such as the main ::U1d tail rotors and the engine . To prevent this nois e from limiting the us eabl e gain in eac h a utopilot closed loop, the signal from eac h tr a nsduc er is filtered by a network designed to e lim-inate the bothersome signals and yet minimiz e the phase lag a t low frequencies. The fr equency response information contained in Figure 9 was obta ined including the filters.
3. 4 Tape Recorder
A 14-channel magn etic ta pe r ecord er is installed in the starboard pod and is currently s e t up to r ecord 13 chmmels of F. M. data a nd one channel of audio at 1-7/8 ips . Up to 13 par a meters in the equa tions of motion or the r es ponses of the simulator may b e r ecord ed while the audio channel is used to record the pilot's comments. Three calibration pulses on each F. M. channel are automatically r ecord ed after the pilot s e -lects the r ec order on .
3. 5 Synthe tic Instrument Landing System (I . L. S.)
Using the orange-bllte system de scribed previously, this vehicle can be us e d to simulate instrument flying conditions. To provide a flexible, meaningful, instrument task, a synthetic instrument landing system has been developed tha t allows s el e ction of a ny loca lizer angle a nd any glide path angle up to approximately 30 deg . In a ddition, the localize r and glide path beam sensitivities can be adjusted over a wide r a nge.
The system, shown pictorially in Figure 12, involves an operator on the ground optically tracking the helicopter in both azimuth and elevation, while the errors from the desired flight path are transmitted continuously to the helicopter and displ aye d to the evalua tion pilot on a cross-pointer type of instrument.
4. 0 CONCLUDING REMARKS
8
-proved to be a valuable airborne research tool. The results from the several proo:rams completed to date, utilizing the greatly improved capabilities of the four degrees ;f freedom simulator described in this Report, show tha t it provides an even more versa-tile means of conducting experiments relating to the stability and control of VTOL and STOL aircraft. 5. 0 REFERENCES 1. 2. 3 . Da w, D. McGr e gor, D.M . 4 . McGregor, D.M . 5 . Smith, R . E . 6. McGr cg·or, D. M. 7 . McGr e gor, D. M . 8 . Mc Gr e gor, D. M.
U.S . Department of Defense Military
Specifica-tion, Helicopter Flying and Ground Handlino· b
Qualities: General Requirements. MIL-H-8501A, 7 Sept . 1961.
Recommendations for V/STOL Handling Qualities. AGARD Report 408, Advisory Group for Aero-nautical Research and Development (NATO), October 1962 .
Development of a Model-Controlled V/STOL Airborne Simulator.
NRC, NAE A ero. Report LR-352, National Research Council of Canada, Aug·ust 1962 . An Investigation of the Effects of Lateral-Direc-tional Control Cross-Coupling on F l ying Qualities using a V/STOL Airborne Simulator.
NRC, NAE Aero. Report LR-390, National
,.... Ｇ ｾ＠ ｾｾ Ｂ＠ ｾ＠ l , ,.-. ,.., , " ,., ; l n f
E
'111 ') r.h D P r P nlhP1' 1 9G ')R e search Council of anada, De c e ml)f:l' l91:]:J:
A Comparison of V/STOL Aircraft Directional Ha ndling Qualities Crite ria for Visual and Instru-m e nt Flight using an Airborne Sirnulator.
NH.C, NAE Ae ro . R e port LR- 465, National Research Council of Ca na da. Sept . 1966.
The Influe nce of Aircr a ft Size on Control Power and Control Se nsitivity Requir e m ents.
A Comparison of R e sults from Tvvo Variable Stability Helicopters .
NH.C, NAE Aero . R eport LR- 459, National R e s earch Council of Canada, July 1966 . Simula tion of the Ca naclair CL - 84 Tilt-vVing Aircraft using an Airborne V/STOL Simulator. NRC, NAE Ae ro. Report LR-t135, National R e sea rch Council of Canada, July 1965 . A Flight Inv e stigation of Various StabilHy Augme nta tion Sy stems for a J e t-Lift V/STOL Aircr a ft (Hawker Siddeley Pll27) using an Air-borne Simula tor.
(NH.C, NAE Ae ro. R e port LR-500, National R e s earch Council of Ca nada, Feb. 1968 .
9 . Gould, D . G.
10 .
9
-The Mod el Controlled Me thod for th e Dev elopment of Va riable StnbilHy \ircr a ft.
NRC, NAE Ae ro. Re port LR34 5 , Nation a l R e -s earch Council of Ca nacb, Jun 1962.
Random Nois e Gene rator for Simula tion St1.1dies. Goody ear Aircr aft Corp., Akron, Ohio, Report GER-6436, De cen1b e r 1954.
-
11-\PP EN DDC A
As a n e x a n1pl c of how t he m od l i s pr ogr a mm ed a nd scale d f or us e in this
m odel-contr olled m e thod, c on sider how ｾｾ＠ ty pical dir ec t io na l e qu[ltion of motion is
s i mub te d. Let th e y:L \\ in g acc cl r:1 ti on b e
ti - N-0 r · 6 + N r 6 a 6 + N a r r + N p · p +N (3
wh e r e each of the te rms is d efine d in the list of sy mbols.
(3 + E (j)
c
?
The sta bility d eriv a tiv e s ha ve dime ns i ons of r a d/ ｳ ･ ｣ ｾ＠ per unit varia ble
(e . g. N 6 , r
I
2 r ad se c ) l t · 1 b 1· ·d· tl · t · t · 11 .1 f del o J ·a u1ec " c 1v1 me: ·1e " awmg 1non1en · [Jer un1 · v arw J e
1nc 1 o· · ru e r J " J ｾ＠
by the yaw in g mom e nt of inertia of the simula t ed aircr aft.
Ea ch v a ri a ble is sca l ed in the norma l a nalogu e computer fa shion to utili ze a s much of the linear r a nge of the amplifiers as possible . a llowing for the anticip a t ed maximum vari a tions . The only volta ge scaling tha t ca nnot be alte r ed is at the fin a l ste p, where the mod el is compa r ed vvith the helicopter a ngul ar r a te sinc e the helicop-ter sca ling from the rate gyro to the a mplifier us ed a s the compa r a tor i n the autopilot is det e rmined by the autopilot closed loop gain.
Using the optim.iz ed scaling terms as defined in the list of sym.bols , and c alculating the yaw r a te instead of the acceler a tion, give s the following sca l ed equ a tion
· ( K ) t K r
ｾ＠
= I.r r ,_ _ N -oj
IC o · o r o r r 6 clt + r ) t. No
f
Ko . o
:l a o a t dt + (N )j '
K r r 0(K )
t (K
)
t(K E)
+ K r N Jj
K J • pelt + K r · N(3j
l\.(3 • (3 dt +I/
i
K J · p p l o l (3 o p I. · r cl twhe r e the te rms in the brack e ts become the settings on the potentiome ters scaling the va rious t e rms in the equa tion, as indica ted in F i gure 13. The c a lcula ted rate of yaw is compared with the actua l helicopter angula r r a te by introducing it into the comparator with a resista nce given by
R IN MODEL where TLN a nd K , - -l H E LICOPTER rH K = r X R K rH INHELICOPTER . 1
1
radthe y a w r a te gyro scahng (vo ts - - ) ,
12
-invariant. Simila r ly, as the model output scaling is alte r ed , the l ead or pre- empha sis scaling must be changed inversely to maintain proper overall operation .
Figure 13 indicates how the equation is "patche d" on the computer . In any
given program the parameters that a r e to be vari ed are wired into the g·ain setting potentiometers mounted in the cockpit , so that the safety pilot can a lte r them without l eaving his seat .
.
'
\
1
j
13
-UNDAMPED NATURAL FREQUENCY,wN IN RAD/SEC
CHANNEL
DETERM INED FROM DETERMINED FROM
GRAPHICAL ANALYSIS ACTUAL FLIGHT TEST
HOVER 40.5 31.8 YAW 50 KNOTS 30 HOVER 9 .2 9.5 PITCH 50 10 KNOTS 12 HOVER 10 .5 13 ROL L 50 KNO TS II. I 11 .2 HOVER 46 21 HEAVE 50 45 28 KNOTS
TABLE I
FIG . I ｾ＠
i
!' \
f
ｾﾷＢＧ＠
ｾ ｩ ｘｦＮｎ Ｍ ＨＮＺ ｉ Ｉｬ＠ ｾ ｬｌ｀ Ｚ ｉｩＺＬ ｔ＠ ｴ ｾ ｅ ｟ ｓ ｦ ﾷａＡｾｾ ﾷ ｡＠ ｓ Ｑ ｅ］ｴＢｬｫｬｾｍ＠ tFOUR DEGREES OF FREEDOM V/STOL AIRBORNE SIMULATOR
;_...
f-'
!TASK
I
I FIG . 2 FIG. 3 SIMULATED ATMOSPHER IC TURBULENCE, a AND {3 VANE,SIMULATOR PITCH ATTITUDE, ETC.
'LEAD
I
I
CIRCUITI
•
SAFETY PILOT ---r-1 It
tl..l
NORMAL !ACTUATOR HELICOPTER ｲＭＫｩｈｅｌｉｃｏｐｔｅｒｬｾ＠ CONTROLS ELECTRIC ANALOGUE + ELECTRIC OF EQUATIONS OFFLYING MOTION OF THE
J \
-CONTROLS SIMULATED AIRCRAFj
C
ERROR
ｾ＠
PILOTDESIRED SIMULATOR
ｌ｟ＺｾｾｾｾｾｾｾｾｾｾＺＡＮＧＺ｟＠
__________
l
MOTION
ACTUAL SIMULATOR MOTION
11
MODEL-CONTROLLED11 SIMULATION METHOD
EVALUATION PILOT'S
ELECTRIC FL Yl NG CONTROLS
ｉ ｾ＠
VIEW OF THE SIMULATOR COCKPIT FROM THE RIGHT SIDE
,...
>-' -J >--' 0 )-FIG . 4 THE ANALOGUE COMPUTER USED FOR THE MODEL
FIG. 5 VIEW OF THE SIMULATOR COCKPIT FROM THE LEFT SIDE
ｾ＠
I-' CfJ
I-' C!)
FIG . 6 IIl 0
-ｾ＠ ::J c:r: 1-u w Q.. (f) w > - 1-<! ....J w c:r: FIG. 7ORANGE SUB-COCKPIT INSTALLATION FOR SIMULATED INSTRUMENT FLYING
,,
ＲＰＮＭＭＭＭＭＮＭＭＭＭＭＭＬＮＭＭＭＭＭｾＭＭＭＭＮＭＭＭＭＭＭＭＮＭＭＭ ＭＭ ｾＭＭＭＭｾＭＭＭＭＭＭｾＭＭＭＭｾＭＭＭＭｾＭＭＭＭＭＭｾＭＭＭＭＬ＠
10
I I I I I I I If
ATMOSPHERIC :URBUL:NC: SPECTRUM
I
I +3 (
1.339v
w)ol//J_----10
1
I, ｾ＠ + ( 1.339 ｾ＠ w)2]'';:
6- v- - -...;:
FOR L = 200 FEETSIMULATED TURBULENCE
_/l
V = 30 KNOTSKw2
ｾ＠ ＫＨＰ ＮｾＱＸ＠
)2
]
ｾ＠
+(ＰｾＵ＠
)2]
ｾ＠
+ (ｾ＠
t]
'\;: -201 I I I I I I I I I I
ｔｾ＠
'\ ＭＳＰｾＭＭＭＭｾＭＭＭＭＭＭｾＭＭＭＭｾＭＭＭＭｾＭＭＭＭＭＭｾＭＭＭＭｾＭＭＭＭｾＭＭＭＭＭＭｾＭＭＭＭｾＭＭＭＭｾＭＭＭＭＭＭｾＭＭｾ＠ 0 .001 0.01 0 .1 1.0 10FREQUENCY, (RAD /SEC)
POWER SPECTRUM OF SIMULATED ATMOSPHERIC TURBULENCE
l "
0
l'V
CAL CUL ATED FIG. 8 ｾ＠ ro 0 w 0 ::> t-..J a.. ｾ＠ <{ ｾ＠ (/) w w cr: (!) w 0 ｾ＠ w ..J (!)
z
<{ w (/) <{ 20 0 -20 0 -100 I -200 a.. -0 .1 FIG . 9a--
-+
L E AD OR PRE-EMPHASI S RESPONSE sI .,.., COMP ENSATION A CT UATOR HE LICOP TER I • ' ...
-
---
-
--
-AUTOPILOT SC HEMATIC--
1- .-
.---
-...--
'-
--""'
'""'
F EEDBAC K TRA NSDUCER I - - - OPEN LOOP - - - CLOSED LOOPI
CLOS ED LO OP WIT H L E AD COMPENS ATION-
...ｾ＠
t--- ----
r---_
...---
- - - -
-r-- .--...
...-
r----
...-
..._----
---':'---... ｾ＠ ... ·-..::::::, - - - -1.0 10 .0FREQ UE NCY (RAD / SEC)
FREQUENCY RESPONSE FOR YAW CONTROL LOOP
l'v l'V
I
tv
"'Tl G) c.D ( ) "'Tl :::0 rn 0
c
rnz
('")-<
::0 m (J) ""0 0z
(J) rn "'Tl 0 :::0 ::0 0 r r ('") 0z
-l :::0 0 r r 0 0 ""0 AMPLITUDE (DB) PHASE ANGLE (DEGREES)1 I (.)J 1\) ..!.. 0 0 0 90 0 0 1\) +:> 0 0 0 0
I
'I
I
I
' I II
I
I II
I
I
I;
I
II
II
I II
II
I
II
IｾＰＱ＠
I
I
0I
II
cI
I
fTlI
z 0 -<I
I
II
I I II
:::0 I I )> 0ｾ＠
I
I
I
0 ｾ＠I
I
I
I
I
I
I / 0 01 1 ,/tI
'I
I
II
'I
I
I .I
II
v
I ,/ I
I I II
I
.:;/
//»
() (") (") 0 o r r -o :s: o 0 fTl -o (}) (}) z/
fTl z O o r JTl JTl (}) 0I
J> r r o -iO 0 -o - o o I ｾ＠ -o -oI /
ｾ＠ -i1,/
I II
_;>;
v
:r: r fTl )> 0/ I
'/
(JJ ｾ＠ Ｐｾ ｾＭＭＭＭＭＭ ＭＭ ｾ ｾ ｾ ＭＭＭＭＭＭＭＭ ＫＭＭI 6 0 0 0PHASE ANGLE (D EGREE S)
I;
I
I
ｾ＠
gz -0 1\) 0 AMPLITUDE (DB) +:> 0 I..
"'Tl G) <.0 0"".,
:::0 rn 0 c rnz
('")-<
:::0 fTl (J)-u
0z
(J) rn..,.,
0 :::0-u
----j 0 I 0 0z
-l :::0 0 r5
0-u
PHASE ANGL E (DEGREES) I 1\) ...!... 0 0 9 0 0 0
I
'I
I
I
I
I
""TlOI
:::0I
fTlI
0 c fTlI
I
z
IJ
0 -< ｾ＠ :::0 'I
)> 0 ... I (f)I
fTl 0 ｾ＠I
I
I
II
I
/ I/
I
j / I I ' AMP LI TUDE (DB) 0 1I
I
II
II
\
\I
I1\
I II
I Ill
I
II //
I
I.
1\) 0 1/ I ' 'I
'I
II
II
I
I
I ,: I
: i
(") () 0 r r -o 0 0 fTl (}) (f) z f1l f1l o o r 0 r r o 0 0 -o 0 0 -o -oI
6 01I
I I II
I
/(
::! -i :r: rI
fTl )> 'I 0ｾ＠
(") 0 :s: (JJ ｾ ｌ｟ＭＭｾ ｾ ＭＭＭＭＭＭＭＭＭＭ ＫｾＭＭＭＭＭ 0 1\) -0 0 0 0PHASE ANG LE (DEGREES)
ｾ＠
r;7
I ｾ＠vz
-0 1\) 0 AMP LIT UDE (DB)-o fTl z (}) )> -i 0 z - · · I +:> 0 I
I
+:> 0/
I
OJ 0 20 w 0 0 => !::: _j Q_ ｾ＠ <( -20 0 - 100 I -(/) w w 0::: (.9 w
9
w _j <..!)z
<( ,_ w -200 (f) <( I Q_ 0 .\ F IG. 9d- - -
--
-
-
----
--
--
----
---
---- ----
---
1---I---
ｾＭ- - -
-
=-==-=---t\
---
-- ---- ---- --
- - - -
- --
----
...___
Ｍｾ＠ - - - OPE N LOOP - - - CL OS ED L OOPCLOSE D LO OP WITH LEA D COMPE NS AT ION
\.0 10.0
FR EQUENCY, (RAD I SEC )
FREQUENCY RESPONSE FOR HEAVE CO NTROL LOOP
STE P RESPONSE OF Is! ORDER MOD EL, 0 . 1 Se c T IM E CON STANT
/"""'-HELICOPTE R RESPONS E TO MODEL IN PUT YAW ＯＯｾ＠
...
--:;.
"""'
_ ...,--
----\. o
I
.Y-
--:
>-..? ·==---
..
-PITCH 1.01 7':&o --... ... ' ... :>:C ;-:...-._===
t - - - -- \ SecI
RO LL{ WIT H LE A D NET WORK WITHOU T LEAD NET WORK
TIM E
-",
""
""
FIG . 10
ANALOGUE COMPUTER STUDY ON THE EFFECT OF LEAD NETWORK
ON HELICOPTER TRANSIENT RESPONSE
' '
""'
l" m I l " ....:)l
TR I MI
AUTOP ILOT COMPENSATION LEAD OR r--PRE - EMPHA SI S . _ FIG . II FIG. 12 SUMMI NG AMPLIFIERｾ＠
v
FEEDBA CK POTENTI OM ETER TRACK AND HOLD -CIRCUIT SYSTEM RETURN PRESSURE EL ECTRO- HYDRAU LI C VALVE ACTUATOR--..
ｾ＠
/
ｾ＠ DISENGAGE ｾ＠ ｾ＠ BY -PA SS VALVE..
TO HELI COPTER CONTRO L ECK VALVES AT 250 ps iAUTOPILOT ACTUATOR SYSTEM
GLIDE PATH ERROR
'
"""'-LINE OFｾｾｉｇｈｔ＠
..
TRANSMITTED TO HELICOPTER LOCALIZER ERROR GLIDE PATH ERROR OPERATOR TRACKS HELICOPTER THROUGH THEODOLITESYNTHETIC INSTRUMENT LANDING SYSTEM
l"
CJJ
I
!:-.:>
EVALUATION PILO T 'S
RUDDER CONTROL ,
Or
K8r ·
Or
EVALUATION PILOT'S
A ILERON CON T ROL,
8
0K8
8
0 0 HOVER RELAY 0 FORWARD FLIGHTI . . TO THE ROLL EQUAT ION
Kr. r CALCULATED YAW RATE Kp.P FROM THE CALCULATION OF THE ROLLING VELOCITY . LEAD OR PRE-EMPHASIS TO AUTOPILOT ACTUATOR COMPENSATION
