Radio broadcast .. (1922-30)

Solving a Sometimes Baffling Problem — WHY A.F. TRANSFORMERS BURN OUT By HERBERT M. ISAACSON PERHAPS THERE are many who have wondered with the writer why trans- formers " burn out." Why the trans- formers in one set will stand up indefi- nitely, and yet transformers of the same make will break down frequently in a different set. That a transformer of "X" make should last longer than one manu- factured by "Y" might be accounted for in a number of ways. Fiber is almost al- ways used as the insulating material be- tween the core and the primary, and be- tween the primary and the secondary. In addition, the terminal mounting strip of many transformers is made of fiber; and fiber almost always contains traces of acid. "Y" transformers might invariably break down more frequently than "X ' transformers, because the fiber used in them contains more acid, which, of course, eats the fine copper wire away. But, that transformers of "X" make should on the average last five times longer when used in sets made by "A" than they do in sets made by "B", as has been found by the writer, indicates the presence of factors, outside of the transformers, that affect their life. Some time ago, the writer returned to the manufacturer for replacement, a trans- former with an open primary, mentioning that it had been in use in a conven- tional two-stage amplifier, using double- circuit jacks in both stages. The manu- facturer wrote back saying the open was undoubtedly caused by "surges' set up when the plug was removed from one jack and inserted in another; that if automatic filament jacks were not used, then the set should be shut off before plugging from one stage into another, to prevent these surges. Now, the fusing point of No. 40 wire, the size ordinarily used in a.f. transformers, is 1.85 amperes. The d.c. resistance of the average primary is 2000 ohms. To cause a current of 1.85 amperes to flow through this winding, a terminal voltage of 3700 would be necessary—6845 watts! [This is an interesting calculation but it neglects the possibility that comparatively high voltages may be developed across an inductance, such as an audio transformer, if the circuit is suddenly opened. These voltages do not depend upon the im- pressed voltage but are directly a function of the inductance of the circuit and the rate of change of current. If the circuit is opened quickly the rate of change in cur- rent will be high and comparatively large voltages will be produced across the transformer. The current would not be greater than the normal plate current of Mr. Isaccson, who is a member of the QRV Radio Service in New York City, has discovered a very interesting fact regarding audio transformers: namely, the way in which the unit is connected in the circuit has an important effect on Us life. Therefore, the next time one of your transformers "goes west," examine the diagram of the receiver before blaming the manufacturer. — THE EDITOR. Fig. 1—The point marked "X" in this drawing shows where elec- trolysis takes place in an a.f. transformer. the tube but the voltages will be greater and the transformer might arc over at some point. If this occurred frequently enough the conductor would finally break at the point where the arc takes place. This is undoubtedly the effect which the manufacturer had in mind when he spoke of " surges."— Editor.] Recently, the writer took apart a large number of defective transformers of a certain make. About half of them came from sets of one make and the other half from sets of a different make. In "A" sets the cores were grounded. In "B" sets they were not. The transformers in "B" sets had lasted about five times longer than those in "A" sets. In unwinding the primaries, it invariably was found that in those transformers that had their cores grounded, the first few layers of wire near- est the core were eaten away; in those transformers with ungrounded cores, there was no regularity in the position of the lesion. Apparently the grounding of the cores was an important factor in deter- mining the life of the transformer. But why? If we have two conductors at a potential difference, immersed in a conducting solu- tion known as an electrolyte, electrolysis will take place. The action is, briefly, as follows: The electrolyte is made up of positive and negative ions. Ions are atoms holding charges of electricity. The conductor which is maintained at a posi- tive potential with respect to the other is the anode. The negative terminal is the cathode. Due to the law of attraction of unlike charges, the positive ions are at- tracted to the cathode. When a positive ion reaches the cathode, its charge is neu- tralized by it, and it becomes an atom, which in the case of copper is deposited on the cathode. Under certain conditions the positive copper ions of the anode material go into the electrolyte and are carried to the cathode. The effect of all this then is a tearing down, a disintegrating of the anode, the positive conductor, and the building up of the cathode, the negative conductor. The process is electroplating and is also what takes place in our trans- former. The primary of our transformer is the anode. It is connected to the positive post of the B battery. Anything connected to the minus post, very often a metal chassis or metal panel, is the cathode. Moisture constitutes the conducting path, the electrolyte. As has just been shown, under the stimulus of the B voltage, the copper wire of the primary will be disintegrated and deposited on the metal chassis and panel. The rate of electrolysis is propor- tional to the current flow through the con- ducting moisture path. The current is in- versely proportional to the resistance of the path, which in turn is proportional to its length, assuming that the path is of uniform resistance. And here is where the grounded core enters. The grounded core is a cathode, separ- ated from the first few layers of the primary, an anode, by a very short distance (see Figs. 1 and 2). Assuming that it is twenty times nearer than any other cathode, electrolysis will then take place twenty times faster, which means that, other factors being the same, the trans- former with a grounded core will have a life only ^o as '° n g as °ne with an un- grounded core. Incidentally this electro- lysis goes on all the time the primary is maintained at a positive potential, which, in the case of a battery-operated set, is all the time, whether the set is turned on or not. In the case of sets securing B voltage from the house current, it takes place only while the set is turned on. Now that the cause of transformers becoming defective is known it is impor- tant to discover a satisfactory method of overcoming the trouble. It is well known that in many a.f. amplifier circuits it is essential to ground the cores of the trans- formers in order to obtain satisfactory operation. However, it will be found that in most cases the same stabilizing effect may be obtained by connecting the core to the positive B wire, and this would place the core at the same potential as the primary (inside) winding, thus effectively preventing electrolysis. The next time an a.f. transformer in your set "goes west," don't be too quick to blame it on a voltage surge; the trouble probably has been caused by the proces- sions of millions of atoms of copper, each bearing its charge of electricity, hurrying to a "Happy Hunting Ground" called the cathode—minus B in the vernacular— there to give up its charge, its mission having been accomplished. Electrolysis takes plan at this point Fig. 2 — The arrow in this circuit diagram indicates where trans- formers are apt to burn out. 164 JULY • 1929