An important characteristic to identify for transformers in power systems – both power transformers and instrument transformers – is polarity. At first it may seem inappropriate to speak of “polarity” when we know we are dealing with alternating voltages and currents, but what is really meant by this word is phasing. When multiple power transformers are interconnected in order to share load, or to form a three-phase transformer array from three single-phase transformer units, it is critical that the phase relationships between the transformer windings be known and clearly marked. Also, we need to know the phase relationship between the primary and secondary windings (coils) of an instrument transformer in order to properly connect it to a receiving instrument such as a protective relay. For some instruments such as simple indicating meters, polarity (phasing) is unimportant. For other instruments comparing the phase relationships of two or more received signals from instrument transformers, proper polarity (phasing) is critical.
Polarity markings for any transformer may be symbolized several different ways:
The marks should be interpreted in terms of voltage polarity, not current. To illustrate using a “test circuit” feeding a momentary pulse of DC to a transformer from a small battery:
Note how the secondary winding of the transformer develops the same polarity of voltage drop as is impressed across the primary winding by the DC pulse: for both the primary and secondary windings, the sides with the dots share the same positive potential.
If the battery were reversed and the test performed again, the side of each transformer winding with the dot would be negative:
If we reverse the secondary winding’s connection to the resistor and re-draw all voltages and currents, we see that the polarity dot always represents common voltage potential, regardless of source polarity:
It should be noted that this battery-and-switch method of testing should employ a fairly low- voltage battery in order to avoid leaving residual magnetism in the transformer’s core. A single 9-volt dry-cell battery works well given a sensitive meter.
Transformers with multiple secondary windings act the same, with each secondary winding’s polarity mark having the same polarity as every other winding:
To emphasize this important point again: transformer polarity dots always refer to voltage, never current. The polarity of voltage across a transformer winding will always match the polarity of every other winding on that same transformer in relation to the dots. The direction of current through a transformer winding, however, depends on whether the winding in question is functioning as a source or a load. This is why currents are seen to be in opposite directions (into the dot, out of the dot) from primary to secondary in all the previous examples shown while the voltage polarities all match the dots. A transformer’s primary winding functions as a load (conventional-flow current drawn flowing into the positive terminal) while its secondary winding functions as a source (conventional-flow current flowing out of the positive terminal).
Transformer polarity is very important in the electric power industry, and so terms have been coined for different polarity orientations of transformer windings. If polarity dots for primary and secondary windings lie on the same physical side of the transformer it means the primary and secondary windings are wrapped the same direction around the core, and this is called a subtractive transformer. If polarity dots lie on opposite sides of the transformer it means the primary and secondary windings are wrapped in opposite directions, and this is called an additive transformer. The terms “additive” and “subtractive” have more meaning when we view the effects of each configuration in a grounded AC power system. The following examples show how voltages may either add or subtract depending on the phase relationships of primary and secondary transformer windings:
Transformers operating at high voltages are typically designed with subtractive winding orientations, simply to minimize the dielectric stress placing on winding insulation from inter-winding voltages. Instrument transformers (PTs and CTs) by convention are always subtractive.
When three single-phase transformers are interconnected to form a three-phase transformer bank, the winding polarities must be properly oriented. Windings in a delta network must be connected such that the polarity marks of no two windings are common to each other. Curved arrows are drawn next to each winding to emphasize the phase relationships:
Windings in a wye network must be connected such that the polarity marks all face the same direction with respect to the center of the wye (typically, the polarity marks are all facing away from the center):
Failure to heed these phase relationships in a power transformer bank may result in catastrophic failure as soon as the transformers are energized!
The following photograph shows the diagram for a large utility power transformer equipped with a number of current transformers permanently installed in the bushings (the points at which power conductors penetrate the steel casing of the power transformer unit). Note the solid black squares marking one side of each CT secondary winding as well as one side of each primary and secondary winding in this three-phase power transformer. Comparing placement of these black squares we can tell all CTs as well as the power transformer itself are wound as subtractive devices:
An example of the importance of polarity marks to the connection of instrument transformers may be seen here, where a pair of current transformers with equal turns ratios are connected in parallel to drive a common instrument which is supposed to measure the difference in current entering and exiting a load:
Properly connected as shown above, the meter in the center of the circuit registers only the difference in current output by the two current transformers. If current into the load is precisely equal to current out of the load (which it should be), and the two CTs are precisely matched in their turns ratio, the meter will receive zero net current. If, however, a ground fault develops within the load causing more current to enter than to exit it, the imbalance in CT currents will be registered by the meter and thus indicate a fault condition in the load.
Let us suppose, though, that a technician mistakenly connected one of these CT units backwards. If we examine the resulting circuit, we see that the meter now senses the sum of the line currents rather than the difference as it should:
This will cause the meter to falsely indicate a current imbalance in the load when none exists.