FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1958
1958 - 0582.PDF
598 FLIGHT Same alternators controlled from a common constant-frequency reference. (". . . What a clearance of gad get ry that will be!") ELECTRICAL SYSTEMS IN THE AIR . . . Supposing I am right. What can they,do other than strip thewhole system out and replace it? Well, let us be constructive for a change. If we scrap the individual frequency references, thesynchronizing circuits and the load-sharing devices (phew! what a clearance of gadgetry that will be) and we reference all foursensitive governor controls to a single constant-frequency refer- ence, such as a tuning-fork generator, then, since all must thenkeep in step with the reference they must all stay in synchronism and at the same frequency regardless of whether they are paralleledor not; and, since they are at the same speed, they will equally share the load within a kilowatt or so according to the degreeof mismatching. But we are still not out of the wood. Reactive load-sharing is afunction of voltage regulation via equalizing lines on a similar plan to D.C. load equalization. Unfortunately, while on a D.C. systemwe have only one "phase" to monitor, on a three-phase system there are three. Which shall we monitor? Shall we use one phaseas we do on inverter circuits or shall we average the volts from all three phases? Unless we have balanced three-phase loads—andwho has on an aircraft?—we shall not keep more than one phase within the required voltage limits. One American proposal includesan overriding "highest voltage" reference with a mag-amplifier control circuit, but I still fail to see how it is possible even withthis to share reactive loads equally on all three phases from a single alternator field control. On conventional power-supply systemsthis problem does not arise, because distribution engineers are constantly monitoring the loads and, where great out-of-balanceexists, the single-phase loads are transferred to other phases on a broad pattern—with the overall result that, at the alternator, thedisparity of current on different phases is very small. On aircraft systems there may, on occasions, be as much as 50 per centdifference between the reactive currents in different phases under which condition reactive load-sharing becomes a farce. Let us assume that the impossible has been achieved. Are weout of the wood? Not on your life. Take protection. Merz Price systems have been lifted willy-nilly from power distribution andapplied to aircraft systems. But these depend for their operation on a return fault-path outside the normal distribution feeders. Ina three-phase delta E.H.T. system it is the earth via a dummy earth point in the transformer. In a domestic distribution systemit is the earth, since the out-of-balance current from mixed three- phase and single-phase systems is returned via the neutral. Thismust be earthed only at the sub-station. In an aircraft the neutral and the earth are one and the same, the aircraft structure. Howthen can a relay discriminate between a real fault and an out-of- balance current? Merz Price and similar balance protectivesystems were designed for use on long transmission lines. They have no real justification in an aircraft system, where overload isthe only fault condition to guard against. Take another aspect. A three-phase motor may find itself "two-phasing" through a feeder fault some distance back. It will run on two phases with some overheating. The danger here is that itwill impress, through transformer action, a voltage at some odd phase displacement, on to the third phase and, via the feeders,probably back to some other equipment such as an electronic navigational device, autopilot, etc., which is acutely phase-con-scious and so give misleading information until the fault is dis- covered. This is no theoretical supposition. It really can happenand results could be disastrous. I could go on indefinitely, but space forbids. What, then, of rectified A.C. systems? One has only to lookat the diagrams presented with the papers at the symposium to find the answer. Mag-amplifiers galore, compounding, cross-compounding, stabilizing and a host of other circuits all shrieking that the systems are full of palliatives for bad fundamental design.If we compare the systems as they were installed in the aircraft with those originally projected as "paper designs," the differenceis amazing. The realities are far more complex since many of these "palliatives" have been introduced in the laboratory and develop-ment stages to cover just those shortcomings and, all too often, fundamental inadequacies in the original design. What are the basic problems? First, to achieve within theimposed weight and space limitations an output from an alternator over a speed range which governs both alternator and its exciter.That is to say, the exciter must be able to generate full alternator load field current at the lowest speed range and yet be able tomaintain stability at no load on the alternator at maximum speed. On any alternator this is a tall bill to fill. On the limited parametersof an aircraft installation it is practically impossible of attainment. So we introduce compounding and stabilizing; we try out differenttypes of regulator and all too often we are forced to cut our losses by deleting the exciter and using a regulated voltage D.C. refer-ence supply for excitation. Since this is, on an all A.C. system, usually from the rectified output of the alternator, we have to adda permanent-magnet generator for starting up. As for the rectifiers, first we use selenium, which usually failson open circuit, then we transfer to germanium and silicon diodes, which fail on short circuit. Though the first is safest it carries apenalty of vast forced-cooling fans and ducts. Germanium and silicon diodes, for safety, are being run at less than 50 per centof their rated capacity and, even then, arc having to impose on all electronic components in the system a completely unacceptablelimitation on voltage transients which a D.C. generator would have taken in its stride. To say that the system, in the environmentalconditions of an aircraft, is vulnerable is the very least of the criticisms. The contributor of the original article mentions the startergenerator. Maybe he doesn't know that the project to which he refers has proved a failure. Starter generators have always beenrecognized as a poor compromise between the optimum perform- ances of separate generators and starters within the same framesize. In aircraft compromises cannot be allowed. Weights, space availabilities and torque and load requirement are all against it.Is it not significant that in the automobile industry, where such limitations did not exist, the starter generator went out soon afterthe bull-nosed Morris? And have you seen a generator starter circuit? ". . . and have you seen a generator starter circuit?" (There are four of these on a four-engined aircraft.)* *Mr. Hilt's intimidating example is reproduced at this size on the assumption that nobody could possibly have either the time or the inclination to study it closely—16.
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
