JPH0197173A - High frequency pwm full-bridge power converter - Google Patents

Abstract

PURPOSE:To improve power conversion efficiency, by combining an element having low switching loss and an element having low conduction loss. CONSTITUTION:In a high frequency PWM inverter, BPT(bipolar transistors) are employed as switching elements constituting switching sections 2-1, 2-2 for carrying out low speed switching with a signal wave, while SIT(static induction transistors) are employed as switching elements constituting switching sections 3-1, 3-2 for carrying out high speed switching with a modulated wave. Although loss determined from switching loss and conduction loss is a function of operating frequency, total power conversion efficiency of a PWM full-bridge circuit is maximized when a low speed element having a low ON-voltage is combined with a high speed element having a relatively high ON-voltage.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention provides a high frequency PWM full bridge power conversion device that obtains high conversion efficiency by configuring a bridge by combining switching elements with different switching losses and conduction losses. It is about the forward transformation,
The present invention relates to power conversion devices such as inverse conversion, forward/reverse conversion, reactive power compensation, and active filters.

[Background of the invention]

Conventionally, relatively low-speed power self-extinguishing elements such as bipolar transistors (hereinafter referred to as rBPTJ) and GTO have been used as switching elements in PWM full-bridge power conversion devices such as PWM inverters.
The modulation frequency was 500 Hz to 2 kHz, and the number of PWM pulses was a combination of several pulses, ten or more pulses, and relatively long pulses.

In the PWM waveform consisting of a pulse train of these long pulses, in order to keep the inverter output waveform as a sine wave, it is necessary to
A CC common circuit filter was required.

By the way, with the advent of high-speed switching elements for power such as static induction transistors (SIT) and static induction thyristors (Sr thyristors), the modulation frequency of PWM has increased to several tens of kHz.
z It has become possible to combine short pulses of several hundred pulses per PWM pulse, and by increasing the PWM modulation frequency, it has become possible to obtain an undistorted sine wave as the output waveform of the inverter.

However, the conversion efficiency of a PWM full-bridge power converter such as a high-frequency PWM inverter inevitably decreases as the modulation frequency becomes higher due to switching loss of switching elements. Self-extinguishing switching elements used in high-frequency PWM full-bridge power converters are:
There is a trade-off relationship between switching time and on-voltage, and conventional high-frequency PWM full-bridge power converters have been designed to achieve high speed at the expense of the on-resistance of the switching elements used to increase the carrier frequency. The device has been used in all full-bridge devices.

As a result, attempts have been made to increase the frequency of the PWM system, but due to the high on-resistance of the elements used, the loss of the device is large, and the path to higher efficiency has been blocked.

Therefore, by using a device with a high switching speed, the switching loss can be reduced, but on the other hand, the conversion loss will increase, and the overall conversion efficiency will be limited by t. Specifically, semiconductor switching elements include bipolar transistors, MOS transistors, SrT, Sl thyristors, IGB71GTO, and the like.

To explain the characteristics of the element based on specific numerical values, in the case of an electrostatic induction thyristor (Sl thyristor), 20k 1-1
In high-frequency devices that operate at 100kHz to 100kHz, the on-voltage is as high as 3 to 5V, but due to the reduction of the Tilmi current and tail time during switching, the turn-off time is 1200μs or less.
Obtained with a 300A class element. This is the case when lifetime control such as pt%Au, electron beam irradiation, proton irradiation, etc. is performed, but on the other hand, when such lifetime control is not performed, the on-voltage is 1.5V or less at 300A, and the turn-off is The time is 15μs measured at 10% to 90%
~20 μs. That is, in the example of a 1200V-300A class SI thyristor, lifetime control is performed to increase the speed and increase the on-voltage, but without lifetime control, the device becomes an element with an extremely low on-voltage.

Figure 1 shows the on-voltage Vox and turn-off time (10
%~90%) trade-off relationship of 1200Vtf&S
These are the results of research on elements with different anode side element structures of the r thalistor. No lifetime control is performed on the Sl thyristor. In FIG. 1, A-F indicates the difference in element structure, and FIG. 2 shows the structure. This shows that the trade-off relationship between voh and j ofF changes greatly depending on the difference in the anode side structure. Figure 1 shows 12
IGB as 00v system, 1700V system, 180OV system
The trade-off relationship between Vl1% and j oH at T is also shown.

Roughly comparing MOS transistors and bipolar transistors, MOS transistors are suitable for high speed applications, while bipolar transistors are suitable for low speed and low on-voltage applications. The Sl thyristor can be changed depending on the element design, whether it is for low speed and low on-voltage or for high speed. Comparing normally-on type SITs and bipolar transistors, SITs are suitable for high speed applications, while bipolar transistors are suitable for low speeds and low on-voltage applications.

The normally-off type SIT is for low on-voltage, and has a lower on-voltage than a bipolar transistor, and is faster. Comparing Sr thyristors and IGBTs, as shown in Figure 1, in the same 1200mm series, Sl thyristors have better performance without lifetime control.
The trade-off for vohto pups is excellent. Also, if we look at the field of high power and compare Sl thyristors and GTOs, we can say that Sl thyristors are for high speeds and GTOs are for low speeds.

Therefore, by using semiconductor switching elements, which have trade-offs in terms of element characteristics, by using the right materials in the right places, the overall P.
It is possible to increase the efficiency of the WM full-bridge power converter.

[Purpose of the invention]

An object of the present invention is to reduce the loss of a high-frequency PWM full-bridge power converter and to improve efficiency, and to provide a low-loss, high-efficiency, high-frequency PWM full-bridge power converter. be. More specifically, one of the objects of the present invention is to provide a high-frequency PWM inverter that achieves high conversion efficiency, which has been made in consideration of minimizing the total loss generated in switching elements. It is.

[Summary of the invention]

This invention provides a bridge circuit for a PWM full-bridge power converter such as a high-frequency PWM inverter, in which a switching element with small conduction loss is installed in an arm that performs low-speed switching using a signal wave, and a switching element with low switching loss is installed in an arm that performs high-speed switching using a modulated wave. High conversion efficiency is achieved by using switching elements.

[Embodiments of the invention]

Prior to detailed description of the present invention, a high frequency PWM inverter used in a demonstration test of the present invention will be described as an example.

Figure 3 shows the main circuit configuration of a commercial wave PWM inverter.
The M drive control circuit is shown in FIG. 4, and the waveforms of each part are shown in FIG.

Switch section 2-1 that performs low-speed switching using signal waves.
The switching element constituting 2-2 was a BPT. The BPT used in the demonstration test has an on-voltage of 0.28V/20
A, and the conduction loss is very small.

On the other hand, switch section 3- which performs high-speed switching using modulated waves
The switching element constituting 1.3-2 was an SIT. The SIT used in the demonstration test had no storage time due to its operating principle, and was a high-speed one with typical switching times of 250 ns for turn-on and 300 ns for turn-off.

In FIG. 3, 1 indicates a voltage source E. In FIG. That is, a voltage type inverter will be explained as an example. ■ and ■ indicate driver signals for switching element 2-1.2-2, respectively, and ■ and ■ indicate switching element 3-1.3-2, respectively.
shows the driver signal for 4 is a low-pass filter for the PWM output waveform (7) (see FIG. 5), which outputs a sine wave waveform shown in FIG. 5 (8) to the output terminal 5. In FIG. 4, 6 is a sine wave oscillator, and the frequency is variable in the range of 30H2 to 400H2, for example. 7 is a triangular wave pulse generator, for example, the frequency is 10
It is selected to be variable in the range of kH2 to 300k+2. 8 is a square wave generator. 9 is a full-wave rectifier circuit, and 11 is a comparator. As shown in Figure 5 (1), the full-wave rectified waveform and the triangular pulse are compared, and the fifth
Create the PWM signal shown in Figure (2). 12 is a switching circuit section, and 13 is a waveform shaping section, which outputs the rectangular waves shown in FIGS. 5 (3) and (4) and the PWM signal waves shown in FIGS. 5 (5) and (6), respectively. 14 is a put-time setting circuit for preventing short circuit between the upper and lower switching elements in FIG. With the above configuration, the driver signal waveforms for ■, ■, ■, ■, (3), (4), (5) in FIG.
), (6) is (q).

As mentioned above, the SIT used in the demonstration test had a turn-on time of 2.
50 ns and turn-off time of 300 ns, but the forward voltage drop is 2°3V (V,, -+0
．． 7V), which is relatively high.

In contrast, the bipolar transistor used in the demonstration test had a forward voltage drop of only 0.28V at 20A.
Although it is extremely small, the turn-off time is extremely long at 12 to 14 μs. A case where switching elements having these two characteristics are configured as a PWM inverter as shown in Fig. 3 using the same element, and 3-1, 3 which operate with a high-speed PWM signal.
- Using SIT as the switching element of 2, and BPT as the switching elements of 2-1 and 2-2 that operate with low-speed signals.
FIG. 6 shows the experimental results of the power conversion efficiency η from DC to AC in each case. In FIG. 6, the power conversion efficiency 7 is plotted as a function of the output AC power Po.

What is SIT type? 2-1.2-2.3-1.3 in Figure 3
-2 is a PWM inverter made entirely of the above-mentioned SIT, and the BPT type is a PWM inverter made entirely of the above-mentioned BPT. On the other hand, SIT
, BPT combination type consists of the low-speed switching element 2-1.2-2 in Fig. 3 with BPT, and 3-1.3-
This is an example in which the second high-speed switching element is configured with SIT. In both cases, the carrier frequency is f0 = 25kHz, the output limiting wave is fO = 50H2, and the modulation degree m-
It is 0.80.

As is clear from FIG. 6, the combination type of SIT and BPT is superior in terms of conversion efficiency. That is, it can be seen that the combination of BPT, which is a low-speed switching element with a low on-voltage, and SIT, which is a high-speed switching element with a high relative Beon voltage, is the most excellent. This indicates that it is preferable to use low on-voltage low-speed switching and high-speed switching elements as the low-speed arm and high-speed arm of the PWM inverter, respectively. In both cases, on-voltage vo7 and turn-off time 1. This shows that in any given trade-off relationship, there is a combination that minimizes the overall power conversion loss by selecting elements. This will be concretely shown using calculated values.

Based on the switching time and on-resistance of the elements used in high-speed switching, the switching time of the elements used in low-speed switching is 5 times that of the elements used in high-speed switching,
Combinations of 10 times, 50 times, 100 times, and 200 times, on-resistance of 1/2 times, 115 times, 1/10 times, 1/20 times, and 1/100 times, output same wave number 50H2, modulation frequency @ When power conversion loss is calculated using a 25kHz high-frequency PWM method, the number of low-speed side switching is 1 for every 250 high-speed side switchings, but the switching loss due to the increase in the switching time of the low-speed side switching element is The variation exhibits an increasing trend, shown in FIG. 7a, proportional to the switching time relative to the reference value 1.

In addition, the on-resistance of the bridge circuit is determined from the series resistance between the low-speed switching element and the high-speed switching element, so the change in on-loss due to the reduction in the on-resistance of the low-speed switching element is as shown in Figure 7 with respect to the reference value 1. It shows a gradual decreasing trend.

Now, when the switching time is 5 times, the on resistance is 1/2 times, when the switching time is 10 times, the on resistance is 115 times, when the switching time is 50 times, the on resistance is 1/10 times, and the switching time is 1
00 times, on-resistance 1/20 times, switching time 200 times
The change in loss when the on-resistance is 1/100 times the standard value is as shown in Figure 7C, compared to the standard value 1. From the combination of 10 times the switching time and 115 times the on-resistance, the on-resistance is 1/10 times the switching time 50 times as shown in Figure 7C. There is a region where the loss is the lowest by about 23% with respect to the reference value for combinations of the following.

In the case of a PWM inverter, the carrier frequency f.

and the switching speed to2, t of the switching element
+c = 22 squares... (1) 2TP TP-t oH + t d +j on + j o + td
Ten tons 1. The following relationship exists, and the maximum operating frequency f determined by the equation (1) is determined by the following relationship. Here, m is the modulation degree, tQ is the turn-on time, 0sF is the turn-off time, and td is the get time. In the case of PWM inverter, Ki1? An element that operates with PWM high frequency pulses at rear frequency CH and low frequency frequency CH. The elements that operate in this way will be combined. Therefore, as an element that operates at high speed, it is desirable to use a high-speed switching element because it is necessary to increase the maximum operating frequency determined by 0=I-27P, and it is also desirable that the on-state voltage is low.

On the other hand, for an element that operates at low speed, it is most desirable to have a low on-state voltage. FIG. 8 shows the on-voltage and on-current characteristics of switching devices for SIT and BPT used in PWM inverters. S.I.T.
has a switching performance that is one order of magnitude faster than that of BPT, but the on-voltage at 11 A is 2.61 V, which is higher than that of BPT. On the other hand, BPT is 0°192V at 11A and S
It can be seen that the Veon voltage relative to IT is low.

Figure 6 shows that by combining SIT and BPT like this, P
This shows that the overall efficiency as a WM inverter is improved.

Note that the present invention is not limited to the high-frequency PWM inverter that combines the SIT and BPT of the above embodiment, but may be applied to other power semiconductor devices as described above, such as a combination of a high-speed SI thyristor' and a low-speed Sl thyristor. It's okay,
In some cases, Sl thyristors are used as the main element with low on-voltage characteristics, and STT or MOSFET
It may be a combination of using as a high-speed element, or it may be a combination of GTO and Sr thyristor,
Furthermore, IGBT and MOSFET or IGBT and SI
It is clear that a combination of T may also be used. Furthermore, even if the switching elements are of the same type, the switching time
It is clear that the overall power conversion efficiency can be increased by combining different on-resistances.

The present invention relates to a PWM full-bridge power conversion device, and is applicable not only to forward conversion but also to power conversion devices such as inverse conversion, forward and reverse side conversion, reactive power compensation, and active filters.

The loss determined by switching loss and conduction loss is a function of the operating frequency, but as in the above example, by combining a slow element with a low on-voltage and a high-speed element with a relatively high on-voltage, it is possible to create a PWM full-bridge circuit. The present invention has shown that the overall power conversion efficiency is maximized.

〔Effect of the invention〕

According to the present invention, by combining an element with low switching loss and an element with low conduction loss, the decrease in power conversion efficiency accompanying the increase in frequency of PWM full-bridge power conversion devices such as PWM inverters is suppressed, and power with high power conversion efficiency is achieved. A conversion device can be obtained.

[Brief explanation of the drawing]

Figure 1 is a diagram plotting the trade relationship between turn-off time and on-voltage VO for 1200V system SI risk, and at the same time 1200V and 1700V1
A diagram showing a 1800V system IGBT, Figures 2A to F are diagrams showing Sl thyristors with different anode side element structures, Figure 3 is a diagram showing the main circuit block of a PWM inverter, and Figure 4 is a diagram showing control of the PWM inverter. Figure 5 is a diagram showing the circuit block, Figure 5 is a diagram showing operating waveforms as a sine wave PWM inverter, Figure 6 is a diagram showing experimental results of conversion efficiency η and output power P6, Figure 7 is an example of calculation of loss increase/decrease. FIG. 8 is a diagram showing the relationship between on-voltage and current of a switching device. 2nd figure 3 Figure 4 On-current IoJcr^) Procedures Amendment Writing method) Director General of the Japan Patent Office Kunio Ogawa 1, Indication of the case Patent Application No. 252180 of 1988 2, Name of the invention High frequency PWM Full-bridge power converter 3, relationship with the case of the person making the amendment Patent applicant address: 4 Kawauchi, Sendai City, Miyagi region (no address), shipping mill with the date of the amendment order) December 22, 1988 "Drawing (No. Figure 2 (corrected)゛・'4+,,,"l'6
, Contents of the amendment as shown in Attachment 7, List of attached documents (1) Drawings 1 copy 1, "In Figure 2" stated on page 5, line 10 of the specification of the present application [
2, Ibid., No. 1
“Figure 2 A to F” written on page 7, line 10 has been changed to “Figure 2 (a
') to (f)'' 3. Figure 2 of the drawing is amended as shown in the attached drawing. Il! 2 figure

Claims (1)

[Claims] In the switching elements that make up the bridge of a PWM full-bridge power converter, an element with low conduction loss is used in the arm that switches with a signal wave, and an element with low switching loss is used in the arm that switches with a modulated wave. A high frequency PWM full bridge power conversion device characterized in that it is used.