FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1948
1948 - 2198.PDF
760 FLIGHT DECEMBER 23RD, 1948 O-4 Powered Controls trophic Instances of accidents or near-accidents because ofthis are already on record. In consideration of the effects of manual reversion on thestability of the aircraft, the lecturer stated that, with pro- portional feed-back of "feel," the stabilities of the aircraftabout ali axes are unaffected by the powered controls; how- ever, with artificial feel, it is extremely unlikely that anypractical form will contain any element which gives a change of force on the controls with change in incidence. tf an automatic trimmer is not used, plausible stabilityconditions will have to be guaranteed in the emergency con- ditions with powered controls failed. It should be possible toallow certain relaxations on the degree of stability required in the emergencyconditions. For instance, a reduction in the directional stability pedal-free mightbe allowed when the powered control units failed as long as the stability wasstill positive. There should be no diffi- culty in preventing a reduction in thestick force / g; this can be done by set- ting a limit on the permissible forwardmovement of the manoeuvre point stick- free, that limit being adjusted by thedegree of increase in the stick forces due to the powered control unit failure. For some applications, notably onultra-high speed aircraft, duplicated powered controls must be regarded asessential. In other applications, especi- ally on very large aircraft, reversion tomanual control in the event of failure in the powered control is impracticableand thus duplication becomes necessary. In many systems, failure of the completeduplicated powered control system on any one surface still causes irreversi-bility to be lost on that surface and, in general, flutter will ensue and wreck theaircraft in a very short time. The degree of reliability that it isnecessary to obtain on military aircraft is difficult to ascertain, but with civilaircraft the factors are less variable. In fact, taking reasonable assumptions, itcan readily be shown that any one of the three main controls should not fail more than once in 4^ million hours' flying and,therefore, one individual control must not fail more than once in 14 million flying hours. In practice, it is likely, however,that cross-couplings between the halves of a duplicated power unit will exist and the reliability will thereby be decreased,necessitating a corresponding increase in the basic reliability of the half unit . .. — ^ :"• :•,..'•: : Ingenuity and Care On civil aircraft especially, therefore, it must not be assumedthat, if the powered control is duplicated, the possibility of control failure can automatically be forgotten. The ingenuityof the mechanical, hydraulic and electrical engineers will be taxed severely to provide the necessary reliability, and excep-tional care will have to be taken in the manner in which the half units are connected together, and in the measures usedto diagnose failure and switch the offending units out of action. Unfortunately, the failure rates required for civil air-craft are so small that, even in the most exacting and lengthy- laboratory tests there would be no chance of proving a com-plete system before it was required for installation. 1OO 2OO WING SPAN (FT.) 3OO Fig. 2. Typical variation with size of required control hinge moment coefficient of a powered flying control with emer- gency reversion to manual. EARTHED TO MAIN -" STRUCTURE -4- ARTIFICIAL-FEEL SPRINGS LOAD DETECTING DEVICE OPERATING SWITCHES SELECTING DIRECTION OF MOTION OF TAB TRIMMER' WHEEL Fig. 3. Schematic diagram illustrating the principle of an automatic trim tab on a powered flying system with artificial feel. The method of duplication which has attracted most atten-tion in this country is that of providing on each control surface two power systems which normally work together, but in theevent of failure of one, the other takes over complete charge. It is obvious that the ability of one unit to carry on satis-factorily after failure of the twin unit may be dependent on the correct operation of certain cut-out or locking devices.Some of the snags may be avoided by the use of mechanical irreversible units, but the one great drawback to the use ofunits of this type is that the efficiency of the powered system as a whole will fall and a greatly increased power input willbe necessary. There is a fundamentally different approach to this problem of duplication to be obtained by using whatmight be called a "multi-surface powered control system." This entails the sub-division of the total control surface areaacting about one axis of the aircraft into a number of completely separate sec-tions, all of roughly equal control power, and each driven by its ownpower unit through two duplicating irreversible units. The number of sections into which ixwould be necessary to divide the con- trol surface would be dependent on theamount of control that would be needed with one section failed in the mostadverse condition. Four would usually be sufficient, as a minimum of half theoriginal control power would then be available after failure of a single section.By keeping separate signal transmission systems to each section, and by usinga separate power source for each sec- tion, a system is given in which nocross-couplings exist between any of the sections and in which no reliance isplaced on subsidiary mechanisms. The sub-division of the control surfacewill provide an appreciable weight- saving, since it would be easier on thesmaller span sections to meet the tor- sional stiffness criteria necessary toprevent either flutter or any undue re- duction of control power. It is stronglyfelt that, even if there appeared to be a slight weight penalty to pay for usinga multi-surface system instead of the normal duplicated unit system, itwould be worth paying, especially on civil aircraft; because of the greater reliability and simplicity of layout. Referring to structural considerations, Mr. Lyons statedthat the required torsional stiffness of the main control surfaces themselves, when powered controls are used, depends on twomajor factors, either of which may be the critical one, viz.: (i) the tendency of the control surface to flutter, and (ity thetolerable loss in control power at the top limiting speed. If the powered control system is duplicated and all mass balanceweights are deleted, the torsional stiffness criteria will prob- ably be decided entirely from flutter considerations. The lecturer further suggested that, where powered controls,are used with an autornatic pilot, in general the automatic pilot will demand a better standard of response from thepowered control surface than that demanded by the human pilot. It is quite admissible to use the auto-pilot as part ofthe main powered control system, i.e., the pilot can select movement of the control surfaces using part of the auto-pilotto transmit his signals and the servo of the auto-pilot as the first amplifying stage of the powered control unit. Butit would probably be essential to motor the cockpit controls in phase with the control surfaces by an extra servo, or by a mechanical run from themain servo to the cockpit control in parallel with the electricaltransmission to the auto-pilot, in order to prevent difficulties inswitching the auto-pilot in and out. One of the advantages in con-necting the auto-pilot directly with the pilot's control run in thenormal way is that, if it is thought necessary to prevent the auto-pilotapplying dangerous amounts oi control, the existing safety devicesfitted to the cockpit controls to prevent the pilot causing struc-tural damage, can also be used to B 24 CABLES CARRYING CURRENT FOR UPWARD & DOWNWARD MOVEMENT OF TAB REVERSING MOTOR a JACK
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
