Magnetic properties deterioration of non-oriented steel due to laser cutting

Facing the global energy crisis, the energy efficiency of electrical equipment has received extensive attention. Statistics indicate that electric machines are generally responsible for about 2/3 of industrial power consumption. The stator and rotor cores are generally made of non-oriented electrical steels due to their approximate magnetic isotropy.^1,2 Laser cutting is widely used in the processing of electrical steels with the advantages of non-contact and high cutting speed. However, a large number of studies have shown that the stresses induced by laser cutting on the electrical steel sheets significantly affect the magnetic properties at the edge, resulting in a non-negligible increase in core losses.^3,4 Therefore, it is crucial to analyze the deterioration degree and mechanism of magnetic properties after laser cutting to improve the efficiency of electrical equipment.

The laser cutting process degrades the magnetic properties of electrical steels, which will change the microstructure and magnetic domain pattern near the edge of the electrical steels.⁵ The magnetic property deterioration due to laser cutting has been investigated previously. Researchers mostly use the single sheet tester to analyze the effect of cutting on the magnetic properties of materials.^6–8 The cutting quality of the samples is affected by cutting speed, laser frequency, pulse width, auxiliary gas and other factors, and the magnetic performance deteriorates more obviously as the number of cuts increases.^9–11 In addition, the laser cutting process involves melting and solidification, which will generate thermal stresses. The thermal stresses will produce domain pinning, leading to variations in the microstructure and magnetic domain patterns near the edge of the electrical steel sheets.^12–14 Therefore, the accurate measurement of the deterioration range of the magnetic properties caused by laser cutting and the analysis of the deterioration mechanism of non-oriented steels are also subjects that need to be studied.

In this research, the deterioration zone of non-oriented electrical steel sheets caused by laser cutting is calculated and compared by the local magnetic characteristic test system. Then, based on the magnetic domain observation technology, the magnetic domain patterns at the edge and those evolutions under the excitation of the magnetic field are investigated.

The samples are non-oriented (NO) 35WW250 steel sheets with 49.5 mm × 49.5 mm. The raw materials are annealed at 700 °C for 2 hours to relieve stresses induced due to prior processing and then cut by a high-power density laser beam along the rolling direction from the middle of the raw materials. The focused laser beam was directed at the magnetic materials resulting in the local melting of the steel. The physical parameters of the samples are listed in Table I.

Based on the local magnetic property testing system, the local magnetic characteristic of the non-oriented steel sample near the edge is tested under sinusoidal excitation and shown in Fig. 1(a). The local magnetic characteristic test system is mainly composed of the signal generation system, data acquisition system, B-H signal acquisition system, high-performance power amplifier, multi-channel differential amplifier circuits and local magnetic characteristic tester.

The mobile B-H sensor with B probes and H coils is shown in Fig. 1(b). Considering the signal instability and other factors, the H coils are made 6 mm × 6 mm. The front and rear double H coils are used for H measurement, and the average value of the measurement results of the two coils is taken as the measurement result. The distance between the two probes is 7 mm.

Under the excitation of 50 Hz frequency and magnetic flux density (B) of 1.6 T, the hysteresis loops at different positions along the rolling direction are shown in Fig. 2 (a). The laser cutting process deteriorates the magnetic properties, leading to the flattening of the hysteresis loops and the increase of the areas of the hysteresis loops. The required magnetic field strength H at the edge of the sample is larger than in the center of the sample when both are magnetized to the same magnetic flux density B. When the local B reaches up to 1.6 T, the applied H is 531 A/m in the center and 706 A/m in the zone 4 mm from the edge.

From the above analysis, it can be seen that the magnetic properties of the sample edge have been degraded by laser cutting. In order to quantitatively evaluate the range of the degradation of magnetic properties affected by the laser cutting, adjust the position knob of the three-dimensional mobile device to move the composite B-H sensor, so that the sensor is moved from the edge to the sample center gradually with a step of 2 mm until the measured local magnetic properties at different positions are basically consistent. Figure 2(b) compares the hysteresis loops at different distances from the edge. The closer to the edge, the worse the magnetic properties of the sample. The magnetic property in the zone within 18 mm from the edge is degraded significantly. Among them, the deterioration of magnetic properties is most obvious in the zone within 4 mm from the edge.

The core losses of non-oriented steel sheets at different frequencies and locations are illustrated in Fig. 3. With the excitation frequency of 50 Hz and B of 1.4 T, the loss at 4 mm from the edge of the sample is 2.19 W/kg and the loss at the center of the sample is 1.84 W/kg, with a difference of 13.5%. With the excitation frequency of 200 Hz and B of 1.4 T, the core losses at the two locations are 15.91 W/kg and 11.01 W/kg, respectively, with a difference of 44.5%. As a result, from the macroscopic perspective laser cutting leads to a significant increase in the losses at the edge, and the core losses are positively correlated to the working frequency. As the working frequency increases, the losses caused by laser cutting change more drastically.

The core losses analysis was performed for the 35WW250 steel according to the conventional loss separation theory, assuming that the total core losses are a summation of hysteresis loss and dynamic loss.^15,16

where W_hyst is hysteresis loss, W_dyn is the dynamic loss. k_h, k_dyn and α are hysteresis, dynamic and exponential of fitting coefficients, respectively.

Figure 4 illustrates the influence of the edge hysteresis and dynamic loss of the sample when the sample is magnetized to be 0.8 T at 50 Hz. The hysteresis loss and dynamic loss both increase significantly. Compared with the central zone, the dynamic loss increases by 31.19% while the hysteresis loss increases by 23.88% at 4mm from the edge. After reviewing the pieces of literature about magnetic domain patterns of silicon steels, we find that the hysteresis loss of ferromagnetic materials is originated from the movement of magnetic domains, and the laser cutting process increases the resistance of magnetic domain movement and thereby increases the hysteresis loss. The residual stress component perpendicular to the edge also increases hysteresis loss.⁸ The dynamic loss consists of classical eddy current loss and the excess loss that is caused by the uneven distribution of magnetic domain patterns. The increase in domain width may also lead to an increase in dynamic loss.^17,18

The laser cutting process causes an impulse heat source near the edge, resulting in thermal stress that finally changes the microstructure of the sample.⁸ In this paper, the domain patterns of non-oriented steels at different magnetization states are captured and shown in Fig. 5. The cord-shaped magnetic domains are visible away from the edge in the unaffected zone whereas the domain patterns near the edge are evolved due to laser cutting.

According to the spatial distribution of the magnetic domain pattern, it can be roughly divided into three zones as listed in Table II. The first zone consists of obvious cord-shaped magnetic domain patterns, the second zone consists of large and wide domain patterns following irregular distribution of magnetic domain patterns, and in the third zone, the magnetic domain patterns are ambiguous to observe.

Figures 5(d)–(i) show the variations of the magnetic domain patterns at different locations with the applied magnetic field. As shown in the circular mark, the cord-shaped magnetic domain pattern in the central zone has a threshold value of 111 Oe, while the magnetic domain pattern at zone 2 has a threshold value of 311 Oe. The sample at the edge is more difficult to be magnetized compared with the central position. It can be explained that the laser cutting process increases the resistance of magnetic domain movement and thereby increases the hysteresis loss.⁸ 

In addition, the speed of domain wall movement indicates the speed of magnetization of the sample.¹⁸ Assuming that there is N number of domain walls in the sample, the average core loss per unit volume is expressed as follows:

where w is the sample width, t_a and t_b are the corresponding time when the magnetic domain wall locates at positions a and b, respectively. v_i and α_i are the moving speed and damping coefficient of the magnetic domain wall, respectively.

It can be seen in Fig. 5 that the domain width increases from 0.8 ∼ 1 μm in the central zone of the rope-like domain to 2 ∼ 4 μm at the edge, and the domain width is the distance of the domain wall movement. Based on the distance of the magnetic domain wall movement and the time of the applied magnetic field, it can be calculated that the average movement rate of the magnetic domain wall increases from 0.11·∼ 0.13 μm/s to 0.55 ∼ 1.1 μm/s. A significant increase in the rate of motion of the magnetic domain walls at the edges can be observed, resulting in a significant increase in the core loss of the sample.

The effect of laser cutting on non-oriented electrical steel 35WW250 was investigated based on the microstructure and magnetic properties measurement and analysis. The laser cutting process leads to the degradation of the magnetic properties of the non-oriented steels, in which the transverse depth of deterioration reaches up to 18 mm from the cut edge. Compared with the central zone, the loss at the edge of the sample increases by 13.5%. The increase in frequency intensifies the edge effect, leading to an increase in the total losses. The internal stress introduced by laser cutting leads to a significant variation of the magnetic domain pattern at the edge, which increases the resistance to domain movement and leads to increasing losses in the non-oriented steel. This research can provide a guide for modeling and calculation of magnetic properties of non-oriented electrical steels used in electrical machines under working conditions.

This work was supported in part by the National Natural Science Foundation of China, (No. 52277008, 92066206, 52130710), and the Funds for Creative Research Groups of Hebei Province, (No. E2020202142).

The authors have no conflicts to disclose.

Changgeng Zhang: Conceptualization (lead); Methodology (equal); Writing – review & editing (equal). Yan Zhang: Investigation (equal); Writing – original draft (equal). Yongjian Li: Methodology (supporting). Qingxin Yang: Methodology (supporting).

The data that support the findings of this study are available from the corresponding author upon reasonable request.