Low exciting current is obtained by using high permeability material for the magnetic circuit, low flux density, short length of magnetic circuit, and greater number of turns per volt in the wind ings. Various alloys of iron (usually silicon-steel) are used on account of their high permeability.
(2) No-Load Loss or Core Loss.—The alternations of the mag netic flux in the transformer core waste energy (in the form of heat) due to (a) hysteresis and (b) eddy currents. Both the hys teresis loss and the eddy current loss are found to be very much less in silicon-steel than in many other kinds of iron alloys, and therefore this material is used exclusively for all power and dis tribution transformers. It also shows superior permeability (lower exciting current) compared with common forms of iron at mod erate flux densities. A further advantage of silicon steel is found in its non-aging quality, while common forms of iron increase in core loss with use. The eddy current loss is reduced by lami nating the core and insulating the laminations from each other either by oxide scale or preferably by thin enamel. The thicknesses of laminations used commercially are: for 5o to 6o cycle fre quencies about 0.014 in., for lower frequencies as much as 0.025 in., and for audio and radio frequencies 0.007 in., or smaller. Sili con steel being very hard and somewhat brittle, its rolling into thinner sheets than the above is difficult and expensive. In com mercial power transformers, the core loss varies from a quarter of 1% to several per cent.
(3) Load Loss or Copper Loss.—The load currents in the wind ings, flowing against the resistances of the conductors, produce an energy loss in them equal to PR. The resistance R is to be under stood as the alternating-current resistance, which usually is higher than the direct-current resistance of the winding, due to skin effect or non-uniform distribution of current among the various filaments of the conductor. The copper loss is tested or measured best by short-circuiting one winding of the transformer and putting excitation on the other winding sufficient to circulate the rated current in the windings. This test is sometimes spoken of as the impedance test, as it makes possible the measurement of the "im pedance volts" of the transformer, as well as its "impedance watts," which is another term used for the alternating-current cop per loss of the transformer.
(4) Leakage Reactance or Impedance.—Although the magneto motive forces in the primary and the secondary windings due to the load currents are equal and opposite to each other, and there fore neutralize each other in any magnetic circuit common to both —for example the steel core—yet they produce leakage fluxes in the spaces between the two coils, giving rise to a leakage reactance. This reactance, combined vectorially with the effective resistance of the windings, constitutes the leakage impedance of the trans former. Ordinarily, the reactance is many times the resistance of the windings, and hence substantially equal to the impedance. The leakage impedance of a constant potential transformer serves a protective purpose in limiting the short-circuit current of the transformer. However, it produces an undesirable effect in re ducing the secondary voltage. (See Voltage Regulations.) The reactance voltage, expressed as a percentage of the rated circuit voltage, varies from 3 to 15%, the lower values occurring in distribution transformers, the higher in the larger power trans formers. In order to hold the reactance of a transformer within the range indicated above, the primary and secondary windings have to be closely interlaced. If the windings were separated from each other, the reactance would be prohibitively high and the voltage regulation correspondingly poor.
(5) Voltage Regulation.—Reference to this has already been made. With a constant voltage maintained on the primary of the transformer, the no-load voltage of the secondary is higher than its full-load voltage. If the excitation is so adjusted as to make the secondary terminal voltage the full rated value of full-load, and then, without changing the primary impressed voltage, the load is removed, the secondary voltage will rise above the full-load value. This rise is called the regulation drop and is usually ex pressed as a percentage of the rated full-load voltage.