Control Arrius
Control Arrius
Control Arrius
I · 02
PAPER N" 87
SEPTEMBER 15-18,1992
AVIGNON, FRANCE
Abstract
Power shaft
throughshatt . - -
Fig. 2
single stage power turbine. ARRIUS control system
The gearbox macule is adapted to the specific The whole ARRIUS family is controlled by the
need of the different aircraft powered by the same fuel control system, which consists in a
ARRIUS: the ARRIUS 1 B has a power shalt with single channel Digital Electronic Control Unit
a 28° bevel gear. (DECU) associated with a fuel metering device
and a manual backup.
ARRIUS performance The DECU (see figure (3), fitted in the heli-
Based on the high efficiency of new components, copter, receives pilot commands and information
the operating cycle was optimized according to from engine sensors, then sends commands to the
the size of engines in order to have on one engine fuel system. Electrical power supply is
hand a high specific power and a low specific ensured by an alternator fitted in the engine.
weight, on the other hand, a low specific fuel
consumption at partial power.
Table 2 shows performance data of the ARRIUS
family for twin engine helicopters.
The overall performance of ARRIUS is enhanced
by its digital engine control unit (DECU)
Table 2
ARRIUS 1 I ARRIUS 2 o: tLEClRICAl
Pi)\( R SIJN'l¥
Maximum
continuous 472 567
Ratings Maximum
take oil 499 634
(SHP) Intermediate
contigency 499 634
!SA I SL ,__Maximum
I contigency 531 680
SFC I I
Deceleration control
The fuel flow is limited to prevent engine
K1: proportional gain of the idle flame-out.
droop law,
A : engine 2 stopped and engine 1 on "IDLE".
It is a theoretical point, which corresponds to
the max N1 on the integral part of the IDLE law.
Nl---- Stan- uo
PI) f1 acessor Ies
j j Stop/Idle/ - -
Fl\Qt\t selector L
control
_ _ _ __ J
"1:,-f _________'--------------
~ --{__ Start- uo control r
Engine Control
OEI selector
Fig. 4 87-3
Software description ARRIUS 1B
Control law Idle and Flight
F
(E)
60 70 80 90 100
N'2 ~~~
Fig. 5
To start the engine "IDLE" or "FLIGHT" have to For instance, on the 80108, starting the two
be selected and the DECU will start the engine. engines up to 100% NR (rotor speed), takes
This start is very consistent, temperature T4 exactly one minute.
is controlled and the risk of overheating is
nearly nil. Furthermore engine !imitations are presented on
one instrument since the TURBOMECA engines have
If "!OLEN is selected, the rotor will a single limiting parameter, which is Nl (gas
accelerate to (C) or (B). generator speed).
87-4
It is possible by trimming the engines Maximum contingency ls set by the DECU and the
differentially to match N1 or torque. pilot switches to intermediate contingency in
These settings remain constant and are inde- due time (time limit is signalled by the DECU) or
pendent from each other. when maximum contingency is no longer necessary.
Failure case operation This requires only a simple pilot action: switch
the flamed out engine's selector from "FLIGHT"
DECU failure to stop and then to" FLIGHT" again.
There are three types of DECU failure: The automatic sequence will restart the engine
much better than a pilot under stress.
• Redundancy: If one of the redundant trans-
ducers or circuits fails, the DECU switches Pilot training
automatically to the alternate. The pilot does
not know it and has nothing special to do. The Training for OEI operation
failure will be signalled at the end of the
flight for the maintenance crew. life of the engines and main gear box limits the
use of OEI ratings to real cases of OEI flight.
• Minor failure: Such a failure has no elfect
on the performance level of the engine, but may As OEI training is essential for pilot profi-
have an effect on handling possibilities. The ciency, the OEI ratings can be lowered. Associa-
pilot has to use the engine with care. The ted with lower grossweight, it is a represen-
signalled code numbers are listed in the Flight tative training: same rotor speed (NR} piloting
Manual with corresponding eventual procedure technique, same instrument indication.
• Major failure: In this case the fuel metering In case of "trained pilot"' error, if NR drops
system is frozen in the position it was in just too low, the idled engine will automatically
before the failure. Immediate pilot action is restore its power up to maximum OEI rating, if
nil or minimal. This failure is signalled and necessary.
the pilot can go on flying, while controlling
the engine manually: Trainino for simulated DECU failure
' the failed engine's manual lever is lit, The DECU can be lrozen by selecting "MANUAL"
to simulate its failure at any time. The
• the manual control has lui! authority to give engine can then be controlled manually to simu-
maximum power, to idle or to shut off the en- late for instance a landing with one engine in
gine, regardless of the fuel flow before the manual mode.
failure. It is also possible to relight the
engine on manual control. At any time, by switching the engine back to
NORMAL position, the pilot can restore
• some care is necessary when controlling normal operation.
manually. But in case of twin-engine heli-
copter, the other engine is still controlled Flioht safety
automatically.
Decreasing the pilot work load, having a sound
Enoine failure reaction to possible failure, with simple and
fool-proof action, making possible a realistic
All Engines Operative (AEO) ratings (lake-oil, pilot training as frequent as necessary with-
maximum continuous) are pilot controlled. In the out consuming high power life, all of those
case of One Engine Inoperative (OEI) operation concur to improve flight safety.
the DECU controls the engine, which delivers
its maximum contingency power and no more.
Thus the pilot focuses on piloting the helicopter,
controlling rotor speed sl"lgl1tly below normal
fiight value.
87-5
Advantages for the operator • no maintenance flight time is necessary for
check or adjustment of engine topping
Some of them have been already stated:
• no maintenance flight time is necessary for
• the automatic starting, with no overheating check or adjustment of a bleed valve or a flow
improves the real life of the engine, fence
• the training mode, while allowing very effi- • no maintenance flight time is necessary for
cient pilot training, is very thrifty on high adjustment of a mechanical pitch compensation
power hours. It is a real engine life time because the respective potentiometer which is
saver. used by the DECU can be precisely adjusted
on ground
• OEI ratings, being DECU limited, are never
exceeded. • the training mode can save main gear box
life,
Moreover the DECU being a powerful computer
can give additional precious help: • rotor efficiency can be improved by trimming
NR, manually for aerodynamics of the main
• Engine power check: The DECU can calculate rotor, or automatically, for instance by foot
the torque and turbine temperature that the pedal action, to improve lateral wind capability
minimum guaranteed engine should deliver in
those flight conditions, and compare them with • new functions can be introduced: even the
the actual values. control mode can be changed, for instance,
the control loop can change from proportional
• Health monitoring system: the DECU can log to integral when necessary,
hours, starts, cycles, to calculate crack and
creek elapsed life. • in case of a generalized management system,
all the engine parameters can be forwarded
• Help to maintenance: at the end of each by the DECU through a data link,
the DECU displays any control system defect,
so the maintenance crew can fix it without • the improvement of precision and versatility
losing a precious lime trouble shooting. of the control systems is such that new OEl
very high power ratings (30 s OEI rating) have
Increasing the life of the engine and saving been made possible. Such a rating has been
time for maintenance have direct financial already certified by TURBOMECA for the
advantages for the operator. MAKILA 1A2.
87-6
Ground and flight test program Test aircraft instrumentation
gine related tests are shown)
A total of 93 parameters have been measured
The ground/f!lght test program for a new engine during the first ground and flight tests of
with a digital engine control system has to check the 80 108 V2. However only 58 parameters were
some points which are related to the use of related to the testing of the new engines, the
of electronic equipment for engine control. other being necessary for general purpose
Safety aspects will define the sequence of tests or surveillance of the aircraft which was new
to be performed. The test steps listed here- and not identical to the first prototype V1.
after are shown in the sequence of priorities. More details about the instrumentation are given
hereafter
Configuration: Aircraft tied to ground,
engine cowlings removed, EPU connected Type of sensor Quantity
~
70
~
Torsional Stability
:f 50 ' toll. oil<
Torsional stability was an important objective 1i
I
of the first ground tests. Torsional stability
was tested with collective inputs from the pilot,
"'u "'
N
f.jrNif!\1
and with a sine of 3 to 7 Hz (collective axis)
+
,;: "'
20
I r'l' I
injected using a stimuli system and the AFCS J1 I
input of the hydraulic system. At the
beginning, torsional stability was found to be
10
u
marginal in OEI and poor in AEO conditions.
0 '
28 32
Optimisation of the respective low pass filter
8 12 16
"
in the DECU software was performed in two steps. Fig. 9
Fig. (7) shows the eng. torque response to a sinus Enoine installation:
sweep (collective) where the resonance phenomena A good efficiency of the ejector was found
with the eng. cowling installed and from there
Tor1lonal t!ab!llly 1.. 1 (So!!...-. 1.7) no problems whith the eng. compartment temp.
Slnu1 &xtllo!lon .._ 5 to 6 Hz col!edlv&
were found. Engine lubrication and oil
cooling worked well. Engine vibrations were
found to be well below the limits.
: +t--1-+-._;
VHF2 was performed prior to the first flight in
+
s I : :~ order to check the correct wiring and shielding
of the DECU system. For safety reasons the
20 ·,--I
i I I
!
I i I
output power of the transmitters was increased
to about 40 Watt (instead of 20) by means of an
10 ~---"!"~---1---l---'--~---.
amplifier for these tests.
" Tlm& (uc)
38
Fig. 7
67-8
Flight tests BO 108 ARRIUS 18 !rlm corK!.: ZP=£000 OAT=.3
The first ground test of the BO 108 with ARRIUS 105
- -
flight corK! ZP-18000 OAT--20 6
100
~~ • I
i
I I I 1
I
under the most severe conditions, like maximum
altitude combined with 200 A load on the
~ I
95
l\ I I
!
I
i
28 VDC generator and fast collective inputs !rom
['\; I i I i
I autorotation to AEO take off power or OEI max.
g 9o contingency power. Of a!! these tests none
~\
i
75
~ I 11 software. Fig. (12) shows a fast collective input
AEO !rom about 12% total torque to MCP with
~ I
70 I I I torque transients up to AEO take-off power. The
collective input was activated within 0.8 sec
92 9~ 96 98 100 102 10.(
and the rotor speed dropped down 10 96%. The
- 1:IAS=O OE! - ,,~,Jt~H'lro <•l - .3:1AS=60 0[!
- ~;IAS=60 A!:O - 5:1AS= 120 OEI - 6:1AS= 120 A!:O significant improvement between the pneumatic
engine control system and a sophisticated engine
Fig. 10 control syslem (like ARRIUS) can be shown by
The optimisation of the rotorspeed range was the comparison ol Fig. (12) and (13), showing
performed with the software versions 1.8 and 2.0. a measurement which was made 2 years earlier on
The first version had a steeper static droop !he lirst 80108 prototype (V1) which used a
line combined with lower gains in the N1-loop. conventional pneumatic control system. The pilot
The second version had a slightly lower static intended to perform a fast collective input from
droop line combined with higher gains in the low power to MCP (92% MT1+2). However due to the
N 1-loop. The second version was finaly chosen delayed acceleration of the engine and the
due to the improved acceleration characteristics absence of a torque limiting system he decided to
which could be demonstrated. The requirement to lower the collective pitch after some time even,
maintain the rotorspeed within ± 2% could be for a much slower and smaller input.
demonstrated with the exception of low power
settings in high altitude (N1 below 78%), which
are quite close to AR. 87-9
Engine occel•rolion ARRIUS 1 B (s ... 2.0) Eng, occ•l. 108 Vt (pn•um, •nQ. contr,)
AEO 5000ft/-2"C 60 Kt AEO 5000f110"C 70 Kl
110 110
100
l&i: I T T I.. 100
I 1 I T
g 90
' J.J-t+21 g I -t
• I 7 I I Cf11. P~'h I •
90
I I ,, I
I
80 80
';.
70
I I
'
7 T i i I " 70
I I ,+A I
I
0
u ! i ! I 0
u I I I/ I '
-
w '"'~
T ( ". "'
""
0
0
80
I I I
! !
I
I
!
i'
l
~
0
5()
1/
y /i
\i
I \
I
/
I '
MT\~2
"z' <O "7'
N 30
i I I I
I '
I ' I I N 30
'" i I ill I. . . . . V
+ I J I I I
i +
I I VI I I I I I
20 ' ~
" 10
T
I I
I'
i
! I'
I
I I i
"
20
10
I
' i
I
I
0 ' 0
0 2 10
'
Tim .. (,.,,,)
8 0 2 6
Tim<'l ("..,..._)
8
Fig. 12 Fig. 13
Fig (14) shows a fast collective input AEO from AR Fig. (15) shows approach and landing with eng. 1
lo MCP. This is a more severe test because the in manual mode (MT1 manually adjusted to 20%
compressor must accelerate from a very low N1. prior to the approach) and without further
The main part of the collective input was made corrections. The rotorspeed which is controlled
in about 1. 7 sec. The rotor speed dropped down to by engine No 2 only remains between 96 and 102%
94.7% which triggering the audio "low rotor-
speed" warning which starts at 95%.
Engln• occ•l•rotlon ARRIUS 1 S {Sw, 2.0) Approoch <:md londlnQ •nQ. 1 "MAN"
110
f rom AR AEO 5000f!l Z'C 60 Kl - 110
BO 108 ARRIUS 1 B
1 i I I I
I
I
I
100 100
!"- N-R TOR
E 90 ~·r .J .!.N,/, I ~ 90
I I I I
~ ! :eou.j pl!c~ I I
BO "00 80
" 70
7 r;;)li I I
'
i I "
"z' 70
! I !
0
u I I I I I I T I _f'-;t
60
7 { 60
rv
"s
0
"z'
N
50
"'
30
H 7
1/
I I i
I I' !
I i .1
!
I
i 0. 50
0
u
N•
"'
i ,.-,_ ~
I I
r-J"'.
/
~
\,../V
I/"' .\
I
I\\
\\
30
v
I I
"" \\
+
~ 20 I +
,
~ 20
z
10
I I I' I
10
I '\\ ~·
17 i \,\1/MTI 2
0 I I 0
I CaJLplkj>
0 2 60
' ' 8 10
" 0 20
Fig. 14 Fig. 15
;:' w
\'f\\/11\'JH ' I Ull.. 98
VI / /I
I \ I' ' I I, I i
~
0
97
I
" 50
I w I I i\ i .
0
~ i v VI 1/ I
<i +0
0
I
' I .I \ I. i
z'
"
95
/I v 1/1 I
"" I' I I
I
" I I I : \I '' I I
"+ 20
I I ~ "" I i 'I VTJ+2 I " I I I
~
10
0 I\ I I I '
I "
92
I I I I
0 18
' (Taus h~j
2 6 B 10 12 -2 2 6
Time (se<:) Pressure o!,ilu o (fl)
Fig. 16 Fig. 17
Variable rotorspeed Conclusion
The definition and test of a variable rotorspeed Testing of the ARRIUS 1B on BO 108 was very
is the main objective of the actual engine- successful. The cooperation between ECD and
related flight tests of the BO 108 V2. Future TURBOMECA was excellent and the result is
helicopters like the BO 108 must be remarkable under many aspects.
designed for the noise considerations of the • maximum comfort for the pilot with automatic
next 10 years and therefore should be developed starting, precise topping including lull OE!
today to meet future requirements. engine and gearbox protection, and sophisticated
The rotorspeed is (for a given main and tail OEI training features.
rotor) the most important parameter which • good engine installation with large margin
influences the noise emission of a helicopter. to the engine limitations (temperature and
1t is therefore important to use a reasonable vibrations) and excellent access to the engine
but low rotorspeed for the part of the flight lor maintenance
envelope where noise emission is a concern. • Modern futuristic variable rotorspeed control
On the other side it will be of interest to reducing noise and increasing high altitude
use the maximum allowable rotorspeed performance
for example lor hover at high altitude, in • safe and reliable concept with full engine
in order to reduce the main rotor torque and separation and no crosstalk between the two DECU
to get maximum thrust from the tail rotor.
A variable rotorspeed which varies the rotorspeed In conclusion, an engine with engine installation
by about 6% between low altitude/high density and optimisation all very promising for the
and high altitude/low density has been defined !uture.
by ECD and the respective software is just in
preparation at TURBOMECA. Flight tests of this
software will start in september 1992 .
Fig. (17) shows the N2/sigma function which will
be used for the first tests of a variable
rotorspeed.
87-11
Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.
Alternative Proxies: