**6.1 Samples, PWM voltage application and space charge measurement timing during PWM application**

All measurements were carried out at 80 °C using the high temperature PEA system which is already described in **Figure 3** at chapter 3. In this section, we used commercially available polyamide-imide (PAI) films as measurement samples, which are actually used as the covering insulating materials for the general motor windings. Since the PAI are usually distributed in varnish form, we made a film shape sample from the varnish. The thicknesses of the films were about 100 μm.

**Figure 9 (a), (b)** and **(c)** show a schematic model of the applied half-wave of AC, unipolar square rectified wave and unipolar square rectified wave with surge shape voltages to the samples, respectively. In this figure, the timings of the application of pulse voltage, which is applied to the sample to obtain the measurement signal, are also described using green lines. The peak voltage of it was corresponding to the average electric field of 50 kV/mm in each measurement. Also, the peak surge voltage of it was corresponding to the electric field of 60 kV/mm in each measurement. Frequency of the half-wave of AC, unipolar square rectified wave, unipolar surge square wave rectified wave voltage was 50 Hz. Duty ratio of the square wave voltage was basically

*Space Charge Accumulation Phenomena in PI under Various Practicable Environment DOI: http://dx.doi.org/10.5772/intechopen.96786*

**Figure 9.**

*Time sequence of various wave voltages and pulse voltage applications for PEA measurement. ((a) Half-wave of AC, (b) square wave and (c) square wave with surge shape).*

50%. To generate waveforms mentioned above, low voltage waveforms are creates digitally using a computer at first, then the waves are amplified using a high voltage amplifier. Therefore, the timing of the pulse voltage application is also controlled using the computer. The pulse voltage to observe the space charge distribution was applied to the sample under the voltage application ("volt-on" duration) and the short circuit ("volt-off" duration) conditions alternatively. To obtain a clear space charge distribution, it is usually used an averaging technique to reduce noise. In this measurement, we tried to create one space charge distribution under voltage application by averaging the sequential 200 signals obtained at "volt-on" durations. In the case to obtain one space charge distribution under short circuit condition, 200 sequential "volt-off" signals were also averaged. Since the frequency of the square wave voltage was 50 Hz, one averaged signal was composed of 200 signals during 4 s. It means that the observation interval of the space charge distribution was 4 s under "volt-on" and "volt-off" conditions, alternatively. To simulate a practical test condition, some measurements were also using carried out samples, which were kept at a high temperature (80 °C) with a high humidity (80%) condition for 60 min before the measurement.

### **6.2 Space charge accumulation results under various voltage wave form**

**Figure 10 (A), (B), (C)** and **(D)** show the measurement results of time dependent space charge distributions in PAI by applying DC, half-wave of AC, unipolar square rectified wave and square rectified wave with surge shape voltages, respectively. In this figure, (a), (b), (c), (d) and (e) show a time dependent charge density distribution using a color chart at "volt-on", a time dependent charge density distribution using a color chart at "volt-off", profiles of a charge density distribution at "volt-on", a charge density distribution at "volt-off", and electric field distribution at "volt-on", respectively. In these figures, red, black and blue lines ware obtained at just after (4 s), 2 min and 5 min later after the beginning of the voltage application.

### *Polyimide for Electronic and Electrical Engineering Applications*

**Figure 10.**

*Space charge distributions in PAI films under various voltage application (50 kV/mm).*

As shown in **Figure 10(A-d)**, **(C-d)** and **(D-d)**, in the cases by applying the DC, the square wave and surge voltages, it is found that homogeneous negative charge accumulations across the bulk of the sample were observed from 2 min later after the voltage applications, and it was obviously recognized from the result shown in **Figures 10(A-a)**, **(C-a)** and **(D-a)**. As shown in **Figure 10(A-a)**, the accumulation of the negative charge seemed to be stable after 5 min later. By the negative charge accumulation, the electric field was distorted near the anode as shown in **Figure 10(A-e)**. On the other hand, in the case of result obtained by applying the half-wave of AC voltage, while a negative charge accumulation near the cathode was observed as shown in **Figure 10(B-b, d)**, the amount of it was small and it was only located near the cathode. By the negative charge accumulation, the electric field was distorted near the anode as shown in **Figure 10 (B-e)**, but it was very tiny.

**Figure 11 (a)** and **(b)** show the maximum electric field and amount of charge in the sample calculated from the distributions obtained by applying DC, *Space Charge Accumulation Phenomena in PI under Various Practicable Environment DOI: http://dx.doi.org/10.5772/intechopen.96786*

### **Figure 11.**

*Maximum electric field and amount of accumulated charge.*

half-wave of AC, square wave, square wave with surge shape voltages. As shown in **Figure 11 (a)**, it was found that the maximum electric field value is larger in the order of DC, square wave with surge shape, square wave and half-wave of AC voltages. Furthermore, it was found that the charge accumulation amounts were large in the order of square wave with surge shape, DC, square wave and half-wave of AC. However, there was almost no difference among them expect for that obtained by applying half-wave of AC. From the above, it is clear that the space charge is not accumulated, nor consequently the electric field is not distorted by applying the half-wave of AC to the sample.
