IEEE C57.123-2002 pdf download IEEE Guide for Transformer Loss Measurement
3.2.6 Frequency Losses in the core have two main components: the hysteresis component and the eddy current component. The hysteresis loss component varies linearly with frequency. The eddy current component (containing both classical eddy losses and anomalous eddy losses) varies proportional to approximately the square of the frequency. The relative magnitudes of these two components are a function of the grade and thickness of steel used as well as the magnitude of the core flux density. Hence, these two parameters determine the magnitude of the effect of frequency on core losses. For examples, the 60 Hz to 50 Hz ratio of iron loss density at 1.5 T induction is typically 1.32 for 0.27 mm highly grain-oriented steel. This ratio is correspondingly equal to 1.26 for 0.23 mm regular-oriented steel at 1.75 T. Also, a frequency deviation of 0.5% corresponds to about 1% to 2% deviation in losses.
3.2.7 Workmanship The quality of workmanship in slitting, cutting, annealing, and handling of the individual core laminations and the quality of the assembly of the core have a direct effect on the magnitude of core losses. Quality of joints in the core also affects the value of core loss to a certain extent but usually has a greater effect on the magnitude of the exciting current. These factors can partially explain why loss measurements on essentially duplicate units can differ by a few percent.
3.2.8 Core temperature Core losses are affected to some degree by the temperature of the core at the time that losses are measured. Generally, core losses decrease with an increase in core temperature. This is due to a reduction of the eddy loss component of the core material iron loss caused by the higher resistivity of the material at higher temperatures. The calculation method to correct the measured values of core losses of distribution transformers to the reference temperature is given in 8.4 of IEEE Std C57.12.90-1999 and IEEE Std C57.12.91-2001. The magnitude of this effect is in reality a function of core design and core material. However, the effect is sufficiently small (about 1% for every 15 °C). In this case, using an average value of the correction factor would be satisfactory. The factor was chosen to be 0.065% per °C (0.00065 p.u. per °C). Its value was arrived at through consensus of the transformer industry and is based on typical values. Due to uncertainty in the actual value of the core temperature during operation, the reference temperature was chosen to be 20 °C for liquid-immersed transformers (per IEEE Std C57.12.00-2000). According to these standards, since core loss measurements on power transformers are typically made at, or near, room temperature, there is little need for applying temperature correction in this case.
3.2.9 Impulse tests No-load loss measurements taken directly after impulse tests are usually slightly higher (typically 1% to 3%) than those taken beforehand. Higher magnitudes of increased no-load loss have been experienced in some cases. This phenomenon is not fully understood at the present time. Existing data, however, shows that this increase is seldom permanent and usually diminishes with time (several hours).
3.2.10 Core stabilization When a transformer is energized for the purpose of no-load loss measurement, it may exhibit an initially high excitation current, a slightly higher core loss, and highly distorted voltage waveform. As the voltage is held constant, the current, loss, and distortion gradually decrease to the expected levels. The time period for this change to stabilize is typically a few seconds and may be longer for some transformer designs. The cause of this phenomenon is believed to be mainly due to the core residual magnetization phenomenon. To reduce the time to reach core stabilization, it is recommended that the core be excited first with higher flux density levels.
IEEE C57.123-2002 pdf download
PS:Thank you for your support!
