WO2013091315A1 - 一种变压器用磁心 - Google Patents

一种变压器用磁心 Download PDF

Info

Publication number
WO2013091315A1
WO2013091315A1 PCT/CN2012/072656 CN2012072656W WO2013091315A1 WO 2013091315 A1 WO2013091315 A1 WO 2013091315A1 CN 2012072656 W CN2012072656 W CN 2012072656W WO 2013091315 A1 WO2013091315 A1 WO 2013091315A1
Authority
WO
WIPO (PCT)
Prior art keywords
magnetic
core
detail
present
sectional area
Prior art date
Application number
PCT/CN2012/072656
Other languages
English (en)
French (fr)
Inventor
王保均
Original Assignee
广州金升阳科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Family has litigation
First worldwide family litigation filed litigation Critical https://patents.darts-ip.com/?family=45984915&utm_source=google_patent&utm_medium=platform_link&utm_campaign=public_patent_search&patent=WO2013091315(A1) "Global patent litigation dataset” by Darts-ip is licensed under a Creative Commons Attribution 4.0 International License.
Application filed by 广州金升阳科技有限公司 filed Critical 广州金升阳科技有限公司
Priority to US14/356,886 priority Critical patent/US10008312B2/en
Publication of WO2013091315A1 publication Critical patent/WO2013091315A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01F—MAGNETS; INDUCTANCES; TRANSFORMERS; SELECTION OF MATERIALS FOR THEIR MAGNETIC PROPERTIES
    • H01F3/00—Cores, Yokes, or armatures
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01F—MAGNETS; INDUCTANCES; TRANSFORMERS; SELECTION OF MATERIALS FOR THEIR MAGNETIC PROPERTIES
    • H01F3/00—Cores, Yokes, or armatures
    • H01F3/10—Composite arrangements of magnetic circuits
    • H01F3/14—Constrictions; Gaps, e.g. air-gaps
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02M—APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF
    • H02M3/00—Conversion of dc power input into dc power output
    • H02M3/22—Conversion of dc power input into dc power output with intermediate conversion into ac
    • H02M3/24—Conversion of dc power input into dc power output with intermediate conversion into ac by static converters
    • H02M3/28—Conversion of dc power input into dc power output with intermediate conversion into ac by static converters using discharge tubes with control electrode or semiconductor devices with control electrode to produce the intermediate ac
    • H02M3/325—Conversion of dc power input into dc power output with intermediate conversion into ac by static converters using discharge tubes with control electrode or semiconductor devices with control electrode to produce the intermediate ac using devices of a triode or a transistor type requiring continuous application of a control signal
    • H02M3/335—Conversion of dc power input into dc power output with intermediate conversion into ac by static converters using discharge tubes with control electrode or semiconductor devices with control electrode to produce the intermediate ac using devices of a triode or a transistor type requiring continuous application of a control signal using semiconductor devices only
    • H02M3/338—Conversion of dc power input into dc power output with intermediate conversion into ac by static converters using discharge tubes with control electrode or semiconductor devices with control electrode to produce the intermediate ac using devices of a triode or a transistor type requiring continuous application of a control signal using semiconductor devices only in a self-oscillating arrangement
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01F—MAGNETS; INDUCTANCES; TRANSFORMERS; SELECTION OF MATERIALS FOR THEIR MAGNETIC PROPERTIES
    • H01F27/00—Details of transformers or inductances, in general
    • H01F27/24—Magnetic cores
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01F—MAGNETS; INDUCTANCES; TRANSFORMERS; SELECTION OF MATERIALS FOR THEIR MAGNETIC PROPERTIES
    • H01F30/00—Fixed transformers not covered by group H01F19/00
    • H01F30/06—Fixed transformers not covered by group H01F19/00 characterised by the structure
    • H01F30/16—Toroidal transformers

Definitions

  • the present invention relates to a magnetic core for a transformer, and more particularly to a transformer core for a power converter. Background technique
  • the existing self-excited push-pull converter and the transformer used therein have the circuit structure derived from the self-excited oscillation push-pull transistor single-transform DC converter invented by GH Royer in 1955, and also used as a Royer circuit, which is also The beginning of the high-frequency switching control circuit; part of the circuit comes from the self-excited push-pull dual transformer circuit invented in 1957 by Jen Sen (somewhere translated as "Jingsen”), which is called self-oscillating Jensen circuit or Jensen. Circuit; These two circuits are collectively referred to as self-excited push-pull converters.
  • the self-excited push-pull converter is described in Electronic Engineering Press, Principles and Designs of Switching Power Supplies, pages 67 to 70, ISBN ISBN 7-121-00211-6.
  • the main form of the circuit is the well-known Royer circuit and the self-oscillating Jensen circuit.
  • Figure 1-1 shows a common application of a self-excitation push-pull converter.
  • the circuit structure is a Royer circuit.
  • the circuit shown in Figure 1-2 is the famous self-oscillating Jensen circuit.
  • the Chinese constant transliteration is the "Jingsen" circuit.
  • the circuit is oscillated by the magnetic saturation characteristics of the transformer B1.
  • the self-oscillating frequency and driving function of the circuit are changed to the magnetically saturated transformer B1.
  • the main power transformer B2 can operate in an unsaturated state.
  • f is the oscillation frequency
  • w is the working magnetic induction ( ⁇ ), generally takes 50% ⁇ 70% magnetic saturation point £ m value
  • N is the number of turns of the coil
  • S is the effective cross-sectional area of the core
  • 1 ⁇ 2 is the working power supply voltage.
  • the circuit structure of Figure 1-1 is: The input filter capacitor C is connected between the voltage input terminal and the ground to filter the input voltage; the filtered input voltage is connected to the startup circuit, and the startup circuit is connected in parallel by the bias resistor R1 and the capacitor C1. Composition; two ends of the bias resistor R1 and the voltage input terminal and two push-pull transistors TR1
  • the TR2 base provides positive feedback to the center tap of the transformer B1 primary windings NBI and NB2; the emitters of the two push-pull transistors TR1, TR2 are common, and the two collectors are respectively connected to the transformer primary windings NPI and NP2
  • the terminal is connected to the two ends of the transformer primary windings NBI and NB2, and the center taps of the primary windings NP1 and NP2 are connected to the voltage input terminal; the secondary winding NS of the transformer B1 is connected to the output circuit to the voltage output terminal.
  • the parallel circuit of the bias resistor R1 and the capacitor C1 provides forward bias to the base and emitter of the transistors TR1 and TR2 through the windings NBI and NB2.
  • the two transistors TR1 and TR2 start to conduct. Since the characteristics of the two transistors are not exactly the same, one of the transistors will be turned on first. It is assumed that the transistor TR2 is turned on first, generating the collector current IC2, and the corresponding winding of the coil NP2.
  • the voltage is positive and negative, according to the same name end relationship, the base coil NB2 winding also appears up and down the negative induced voltage, this voltage increases the base current of the transistor TR2, which is a positive feedback process, so very Accordingly, the transistor TR2 is satisfactorily turned on; correspondingly, the voltage of the coil NBI winding corresponding to the transistor TR1 is up and down, and this voltage reduces the base current of the transistor TR1, and the transistor TR1 is quickly turned off completely.
  • the current in the coil NP2 corresponding to the transistor TR2 and the magnetic induction generated by this current increase linearly with time, but the inductance of the coil NP2 decreases rapidly when the magnetic induction increases to or near the saturation point B m of the core of the transformer B1. Therefore, the collector current of the triode TR2 switch tube is sharply increased, the rate of increase is much larger than the increase of the base current, the triode TR2 switch tube is out of saturation, and the collector-to-emitter voltage drop UCE of the triode TR2 switch tube is increased, correspondingly Ground, the voltage on the winding of the transformer NP2 is reduced by the same value, and the voltage induced by the winding of the coil NB2 is reduced.
  • the base voltage of the triode TR2 switch tube is also lowered, causing the triode TR2 switch tube to change in the cut-off direction.
  • the transformer B1 The voltage on the coil will be reversed, causing the other transistor TR1 to conduct, after which the process is repeated to form a push-pull oscillation.
  • the waveform of the output of the winding Ns is shown in Fig. 2. It can be seen that the operating waveform of the self-excited push-pull converter is close to a square wave except that the "collector resonance type Royer circuit" outputs a sine wave or an approximate sine wave.
  • CCFL inverter The collector resonant type Royer circuit is also called "CCFL inverter", which will also be referred to as CCFL inverter or CCFL converter.
  • CCFL converter is connected to the main power winding ten times inductance in the power supply circuit. The above inductance is used to obtain an output sine wave or an approximate sine wave.
  • Figure 3 is a square hysteresis loop of the core of the transformer B1, where +B m and B m are the two magnetic saturation points of the core, where +B m is called the first quadrant saturation point because the point +B m falls on In the first quadrant of the coordinates of Fig. 3, a B m is called the third quadrant saturation point, and the core of the transformer B1 works in the half cycle of Fig. 2.
  • the route of the point movement is ABCDE, and the movement route is EFGHA in the second half of the cycle.
  • the circuit performs push-pull conversion, that is, the other The triode is turned on, and the corresponding transistor is turned off. Since the triode has a storage time, that is, the base of the triode receives the turn-off signal, and the collector current is delayed until it is turned off, and the storage time is generated.
  • the movement point of the magnetic core working point moves from point D to point E.
  • the movement path of the core working point moves from point H to point A.
  • the hysteresis of the core causes the collector of the transistor. The current is increased unnecessarily and the loss is generated.
  • Push-pull oscillation using the core saturation characteristic the output waveform of the transformer is approximately square wave, and the conversion efficiency of the circuit is high.
  • the core is close to saturation at a specific time, so the core with the air gap cannot be used.
  • a self-excited push-pull converter must use a magnetically saturated core, and a core plus air gap is a well-known method of resisting magnetic saturation.
  • the magnetic core herein is the same as the other known documents, and the magnetic core means the same meaning, referring to the ferrite material, that is, by various oxidations.
  • a sintered magnetic metal oxide consisting of a mixture of irons, the core being used at high frequencies.
  • the Iron Core is a silicon-based material that is only suitable for low-frequency inductor lines and low transformers. It is generally used for low frequency and audio.
  • a similar structure shown in Figure 1-2 is the circuit in which the switch drive function is disconnected from the main power transformer, as shown in Figure 1-2.
  • the self-oscillating frequency and driving function of the circuit are realized by the magnetically saturated transformer B 1 , and therefore, the main power transformer B2 can operate in an unsaturated state.
  • B 1 exhibits magnetic saturation, the volume of B1 is small, and the energy consumed by magnetic saturation is small. Under the same conditions, the overall efficiency of the Jensen circuit is slightly higher.
  • the Royer circuit uses the core saturation characteristic for push-pull oscillation, its no-load operating current is not too small, as shown in Table 1 for the measured parameters of the Royer circuit. If the circuit of Figure 1-1 is used, it is a converter with input DC 5V, output DC 5V, and output current 200mA, that is, output power 1W.
  • the subsequent output of the transformer adopts the circuit structure of FIG. 4.
  • FIG. 4 is a well-known full-wave rectification circuit, and diodes D41 and D42 each adopt a Schottky diode of the type RB 160.
  • Capacitor C is a luF capacitor
  • resistor R1 is 1 ⁇
  • capacitor C1 is 0.047uF capacitor
  • transistors TR1 and TR2 are switching transistors with a magnification of about 200 times.
  • the maximum working current of the collector is 1A; Number of turns of the primary coil ⁇ and ⁇ 2 20 ⁇ , the feedback coils NBI and NB2 have a number of turns of 3 ⁇ , the secondary coils Nsi and NS2 have a number of turns of 23 ⁇ , respectively, and the core has a common ferrite with an outer diameter of 5 mm and a cross-sectional area of 1.5 mm 2 .
  • the toroidal core commonly known as the magnetic ring, is shown in Figure 5 in a perspective view of its shape.
  • the measured circuit has a no-load operating current of 18 mA, an operating frequency of 97.3 KHz, which is close to ⁇ .
  • the conversion efficiency test circuit uses the circuit of Figure 6, the VI voltmeter is the operating voltage Vin, ie the input voltage; the A1 current meter is the input current ⁇ , That is the working current; the V2 voltage meter is the output voltage Vout, and the A2 current meter is the output current lout; then the conversion efficiency can be calculated by the formula (2).
  • Vin is the operating voltage, ie the input voltage, Iin is the input current; Vout is the output voltage, lout is the output current.
  • RL is a variable load, which can effectively reduce the measurement error.
  • Both the ammeter and the voltmeter use the 200mA and 20V or 200V of the MY65 4-digit digital multimeter, and four and more multimeters are used at the same time.
  • the MY65 4-digit digital multimeter has an internal resistance of 10 ⁇ when measuring voltage and an internal resistance of 1 ⁇ for 200 mA current. When the current exceeds 200mA, two ammeters are used to measure in parallel at 200mA, and the current readings of the two meters are added to measure the value. Parallel measurement of ammeters is a mature technology in existing electronic engineering.
  • Jensen circuit see Figure 1-2, although a small transformer B1 is used to achieve magnetic saturation, and the main power transformer B2 operates in an unsaturated state to improve efficiency. In fact, due to the use of two transformers, loss occurs. One more device, the design of the small transformer B1 must take into account the output power of the entire circuit. Rate, after careful debugging, the output current of 5V, Jensen circuit output 5V/200mA is reduced to 16mA, when the output is only 5% of full load, the efficiency is improved by 40.91% than the above Royer circuit.
  • Self-excitation push-pull converter taking the Royer circuit as an example, wants to improve the conversion efficiency of the circuit. Since the circuit is in push-pull operation, each push-pull conversion is achieved by approaching or entering the magnetic saturation of the magnetic core. The energy consumed is lost in the form of heat. To improve the efficiency, it is necessary to reduce the operating frequency of the converter. It can be seen from equation (1) that under the premise of constant input voltage, the parameters in the denominator of the formula can only be increased. , increase the working magnetic induction Bw, or increase the number of turns of the coil N, or increase the effective cross-sectional area of the core 8.
  • the converter products have selected a magnetic core with a large magnetic induction intensity Bw, increasing the number of turns N of the coil, increasing the copper loss and increasing the effective cross-sectional area S of the core, and increasing the proximity or entry of the magnetic core each time.
  • the loss at saturation makes the conversion efficiency of the converter not increase and decrease. Therefore, when designing a self-excited push-pull converter, it is more difficult to choose between these parameters.
  • the effective cross-sectional area S of the small transformer B1 core is small, resulting in insufficient driving power, the switching transistor can not enter a good saturation conduction, and the voltage drop loss is also large.
  • the conversion efficiency of the converter is also low; the effective cross-sectional area S of the small transformer B1 core is large, and its own loss is also large; the problem can be solved by increasing the number of turns of the coil N, but the following process problems are brought about, : There are many turns N. Since the small transformer B1 must work under magnetic saturation, the air gap cannot be opened, which brings great difficulties to the winding.
  • Self-excitation push-pull converter taking the Royer circuit as an example, it can be seen from equation (1) that if the input voltage is raised, if the operating frequency of the self-excited push-pull converter remains unchanged, then the formula 1) The corresponding parameter of the denominator should be increased. For the same series of industrial-grade small module power supplies of the same power, the core of the same size is often used. At this time, only the number of turns of the coil N can be changed to solve the problem, as shown in Fig. 1. -1 The published circuit parameters, if the input is 24V, the number of turns of the primary side coil NPI and NP2 will be increased from 20 ⁇ at each input of 5V to 96 ⁇ , since transformer B1 in Figure 1-1 is required.
  • the number of turns can be reduced to 48 ⁇ , but the effective cross-sectional area S of the transformer B1 is doubled, and the loss is doubled at the same frequency.
  • the conversion efficiency of the converter is reduced.
  • each push-pull conversion is realized by the magnetic core approaching magnetic saturation or entering magnetic saturation, so after the operating frequency is increased, the loss is increased, and the conversion efficiency is lowered. .
  • the effective cross-sectional area S of the small transformer B1 core is smaller.
  • the primary side coil is only one set, two lines can be used. Wrap around, only 30 turns, and then serially get 60 ⁇ of the original side, but because the small transformer B1 has a smaller diameter, whether it is machine or hand-wound, there are processing difficulties.
  • the small transformer B1 can hardly be processed.
  • the effective cross-sectional area S of the core is doubled, the number of turns can be reduced, and at the same operating frequency, the loss itself is doubled, and the conversion efficiency of the converter is reduced.
  • the power supply is unacceptable; at 50KHz, the loss is relatively small.
  • the self-excitation push-pull converter is working.
  • the waveform is close to the square wave, and the rising edge of the square wave is a step signal.
  • the fundamental frequency refers to the fundamental frequency of the operating frequency of the self-excitation push-pull converter, that is, the frequency of the square wave in FIG.
  • Power frequency means 50Hz or 60Hz refers to the frequency of the AC power supply used in industry.
  • the principles of AC magnetic saturation regulators for household and industrial use that were not widely popular in the 1970s can be found in the "Electronic World” and “Radio” magazines of the early 1980s.
  • a magnetic core for a turbulent inductor (MAGNETIC CORE FOR CHOKE COIL) is shown in the patent of JP 60032312 A, published on Feb. 19, 1985, and the problem to be solved is shown in the patent of JP62165310A published on July 21, 1987.
  • the choke inductor (CHOKE COIL) is the same problem, obtaining a large inductance at a small current and a relatively small inductance at a large current, which is used as a freewheeling inductor in a switching power supply. It can improve the output ripple of the switching power supply in the discontinuous mode. In the case of outputting a small current, the switching power supply enters the discontinuous mode (DCM).
  • DCM discontinuous mode
  • the technical problem to be solved by the present invention is to overcome the shortcomings of the prior art magnetic core used in the self-excitation push-pull converter, so that the self-excitation push-pull converter has a self-excitation push-pull type when the load is lightly loaded.
  • the efficiency of the converter is significantly improved; at rated load, the efficiency is further improved; reducing the magnetic force in the self-excitation push-pull converter The number of turns of the coil on the saturation transformer; the operating frequency of the self-excited push-pull converter is increased while the loss remains low.
  • the present invention provides a magnetic core for a transformer, the magnetic core being composed of a closed thick portion and a thin portion, the fine portion being excited by the same small to large magnetic field than the thick portion.
  • the detail is in operation, approaching or reaching the first first quadrant saturation point or the third quadrant saturation point in the instant, at other times in the first quadrant saturation point inherent in the detail Between the third quadrant saturation points. There are one or more details of the details.
  • the length of the detail is 0.05 mm or more, the length of the total magnetic circuit is less than one eighth, and the cross-sectional area is less than 80% of the cross-sectional area of the rough portion, and 4% or more; more preferably, the cross-sectional area of the detail portion
  • the cross-sectional area of the thick portion is 50% or less and 6.25% or more.
  • the smaller the cross-sectional area of the thin portion the smaller the length of the thin portion is, the smaller the total magnetic path length of the magnetic ring is. To ensure the mechanical strength of the detail.
  • the thick portion is the same as the detail material.
  • transition body at the joint of the thick portion and the detail portion, so that the core is easily demolded after the magnetic powder is molded.
  • the thick portion has two or more bumps for preventing the winding on the thick portion from sliding to the detail, or to position the different windings to prevent the intersection between them.
  • the present invention also provides a transformer wound by the magnetic core, wherein the thick portion is wound around a coil, and the thin portion is not wound around the coil.
  • the working principle of the present invention is that the above technical solution is applied to a self-excitation push-pull converter, taking the Royer circuit of FIG. 1-1 as an example, and the transformer B1 in the self-excitation push-pull converter is the same as the above. According to the technical solution of the present invention, the push-pull operation principle of the circuit does not change. When the transistor TR2 is saturatingly turned on, correspondingly, the transistor TR1 is completely turned off, the current in the winding of the coil NP2 corresponding to the transistor TR2, and the magnetic induction intensity generated by the current are over time.
  • the linearity increases, but when the magnetic induction increases to the saturation point B m of the core of the transformer B1, the inductance of the coil decreases rapidly, so that the collector current of the triode TR2 switch tube increases sharply, and the rate of increase is much larger than the increase of the base current.
  • the triode TR2 switch tube is out of saturation, and the collector-to-emitter voltage drop UCE of the triode TR2 switch tube is increased. Accordingly, the voltage on the winding of the transformer NP2 is reduced by the same value, and the voltage induced by the winding of the coil NB2 is reduced, resulting in The base voltage of the triode TR2 switch tube is also reduced, causing the triode TR2 switch tube to change in the cut-off direction.
  • the transformer B The voltage on the coil 1 will be reversed, turning on the other transistor TR1. Thereafter, this process is repeated to form a push-pull oscillation.
  • the energy loss is mainly caused by the increase of the magnetic induction intensity to the saturation point B m of the core of the transformer B1.
  • the core having the same cross-sectional area is used, and the entire core is required to reach the saturation point B m , and more energy is required. Saturated.
  • the path of the transformer core moving in the upper half cycle is ABCDE, and in the lower half cycle, the motion path is EFGHA, that is, in a complete cycle, the working area of the prior art core is the entire area, such as The shading 71 in Fig. 7 is shown.
  • the core cross-sectional area is different, and the detail is magnetically saturated before the same small-to-large magnetic field excitation, but the length is short, so that the small section
  • the fine core reaches its own saturation point B m , which can also cause the push-pull oscillation of the circuit.
  • the working area of the core of the detail is the whole area, which is the same as the shadow 71 in Figure 7, but the length of the detail is very short, and the energy consumed.
  • the rough part is larger than the detail, and accommodates more magnetic lines of force.
  • the area is significantly reduced, so the energy consumed by the thick portion is very low, so the energy consumed by the core of the present invention is generally greatly reduced. That is, with the magnetic core of the present invention, the energy consumption per push-pull conversion can be reduced, and at the same operating frequency, the consumption of the self-excited push-pull converter circuit using the magnetic core of the present invention is reduced, which is manifested as no-load of the circuit.
  • the working current is reduced; for the same reason, the operating frequency of the self-excited push-pull converter can be increased without a large increase in loss, and the conversion efficiency can still be reduced at a high frequency.
  • the no-load operating current of the self-excited push-pull converter circuit is correspondingly reduced, that is, the no-load loss is reduced in proportion, and accordingly, the present invention is applied to the self-excitation push-pull type.
  • the efficiency of the converter is significantly improved when the load is lightly loaded.
  • the load is lightly loaded to full load, the corresponding efficiency of the converter will be significantly improved.
  • the cross-sectional area of the thick portion can be increased, so that the inductance per turn increases proportionally with the increase of the cross-sectional area, thus achieving the same
  • the inductance and the total number of turns are correspondingly reduced, thereby reducing the number of turns of the coil on the magnetic saturation transformer B1 in the self-excited push-pull converter, achieving the object of the present invention.
  • Detailed working principles and formula derivations will be presented in the examples.
  • the magnetic core with the same cross-sectional area is used, and the increase of the cross-sectional area leads to an increase in the magnetic saturation loss, resulting in an increase in the no-load operating current.
  • the conversion efficiency of the self-excited push-pull converter does not change at the full load or decreases. At the light load, the conversion efficiency is seriously degraded due to the large loss.
  • the patented solutions all have to operate all or part of the core and core in a stable magnetic saturation state. They achieve the purpose of the invention by using stable magnetic saturation. For example, an AC magnetic saturation regulator obtains a stable AC voltage output.
  • the choke inductor (CHOKE COIL) mentioned has a large inductance at a small current and a relatively small inductance at a large current.
  • the choke inductor uses a stable magnetic saturation over a large operating range.
  • the inductance decreases linearly with increasing current.
  • the losses caused by AC magnetic saturation regulators and choke inductors are large because they last for a long time in a cycle after magnetic saturation occurs during operation. Even account for more than one-half of a cycle.
  • the present invention only uses a very short length of fine core to achieve a small amount of magnetic saturation inducing circuit push-pull conversion, magnetic saturation occurs only in an instant, its duration can hardly be accurately measured, see Figure 7, because the invention only has details working in the figure In the shadow 71 area of 7, because the length of the detail is short, the working area of the core of the detail moves from point D to point E.
  • the hysteresis of the core causes the collector current of the transistor to increase without loss, but the length of the detail is small. In this process, the hysteresis is significantly reduced, so the loss is also reduced.
  • the hysteresis is relatively small, and the time taken for the working area of the core portion of the detail to move from point D to point E is shortened. That is, the time during which the detail in the core of the present invention approaches or reaches the saturation point instantaneously is shortened. That is, when the detail is in operation, the first first quadrant saturation point or the third quadrant saturation point of the detail is approached or reached in an instant, and the first quadrant saturation point and the third quadrant inherent in the detail are at other times. Between saturation points.
  • the present invention is intended to reduce the negative effects of magnetic saturation, which is an essential difference from the prior art.
  • the present invention has the following remarkable effects:
  • a high-efficiency micropower DC/DC module power supply of less than 100 mW can be realized.
  • Figure 1-1 shows the common application circuit diagram of the self-excited push-pull converter Royer circuit
  • Figure 1-2 shows the common application circuit diagram of the self-excitation push-pull converter Jensen circuit
  • Figure 2 is a waveform diagram of the output terminal of the Royer circuit winding Ns of Figure 1-1;
  • Figure 3 is a square hysteresis loop of the transformer B1 core in the ⁇ -lRoyer circuit;
  • Figure 4 is a known full-wave rectification circuit;
  • Figure 5 is a prior art magnetic ring core
  • FIG. 6 is a schematic diagram of the general conversion efficiency test in this paper.
  • Figure 7 is a diagram showing the working area of the core in the self-excitation push-pull converter
  • FIG. 8 is a working area diagram of a rough corresponding core in a self-excitation push-pull converter of the present invention
  • FIG. 9-1 is a front view of a magnetic core according to first to third embodiments of the present invention.
  • Figure 9-2 is a side view of a magnetic core according to first to third embodiments of the present invention.
  • Figure 9-3 is a plan view of a magnetic core according to first to third embodiments of the present invention.
  • Figure 9-4 is a perspective view of a magnetic core according to first to third embodiments of the present invention.
  • Figure 10-1 is a schematic view of the magnetic circuit in the thick portion of the first to third embodiments of the present invention
  • Figure 10-2 is a schematic view of the magnetic circuit l t in the detail of the first to third embodiments of the present invention
  • 1 is a magnetic core main view for comparison of prior art
  • Figure 11-2 is a side view of a magnetic core for comparison in the prior art
  • Figure 11-3 is a top view of a magnetic core for comparison in the prior art
  • FIG. 12 is a comparison diagram of efficiency after loading a circuit according to a fourth embodiment of the present invention
  • FIG. 13-1 is a front view of a magnetic core according to a fourth embodiment of the present invention
  • Figure 13-2 is a side view of a magnetic core according to a fourth embodiment of the present invention.
  • FIG. 13-3 is a top view of a magnetic core according to a fourth embodiment of the present invention.
  • Figure 14 is a perspective view showing the efficiency of the fourth embodiment of the present invention and the prior art magnetic core loading circuit;
  • Figure 15-1 is a front view of a magnetic core according to a fifth embodiment of the present invention;
  • Figure 15-2 is a view showing a magnetic iv of a fifth embodiment of the present invention;
  • Figure 15-3 is a view of a magnetic office according to a fifth embodiment of the present invention.
  • Figure 15-4 is a perspective view of a magnetic field according to a fifth embodiment of the present invention.
  • Figure 16-1 is a front view of a magnetic body according to a sixth embodiment of the present invention.
  • Figure 16-2 is a view showing a magnetic state according to a sixth embodiment of the present invention.
  • Figure 16-3 is a view showing a magnetic field of a sixth embodiment of the present invention.
  • Figure 16-4 is a perspective view of a magnetic body according to a sixth embodiment of the present invention.
  • Figure 17-1 is a magnetic front view showing a seventh embodiment of the present invention.
  • Figure 17-2 is a view showing a magnetic state of a seventh embodiment of the present invention.
  • Figure 17-3 is a view showing a magnetic field of a seventh embodiment of the present invention.
  • Figure 17-4 is a perspective view of a magnetic body according to a seventh embodiment of the present invention.
  • Figure 18-1 is a front view of a magnetic body according to an eighth embodiment of the present invention.
  • 18-2 is a view showing a magnetic state of an eighth embodiment of the present invention.
  • 18-3 is a view showing a magnetic field of an eighth embodiment of the present invention.
  • Figure 18-4 is a perspective view of a magnetic body according to an eighth embodiment of the present invention.
  • Figure 19-1 is a magnetic front view showing a ninth embodiment of the present invention.
  • Figure 19-2 is a view showing a magnetic view of a ninth embodiment of the present invention.
  • Figure 19-3 is a view showing a magnetic field of a ninth embodiment of the present invention.
  • Figure 19-4 is a perspective view of a magnetic field according to a ninth embodiment of the present invention. detailed description
  • the magnetic core is a closed magnetic ring without an air gap magnetic circuit
  • the ring body has a cylindrical shape and is composed of a thick portion and a thin portion of the same material, and the length of the detail portion is 0.05 mm or more.
  • the total magnetic circuit length is less than one eighth; and the sectional area of the detail is less than 80% and 4% or more of the cross-sectional area of the thick portion.
  • the core of the first embodiment of the present invention has the same core cross-sectional area as in the prior art, and the detail is smaller than the cross-sectional area of the core in the prior art.
  • the ratio of the cross-sectional area of the thick portion to the cross-sectional area of the detail portion which is the reciprocal of the percentage in the technical solution, and is recorded as a constant k, which is a reciprocal of "80% or less, 4% or more", that is, 1.25 times to 25 times.
  • Figure 5 shows a prior art magnetic ring core, as shown in Figure 5. As shown, the cross-sectional area is equal at 50, then, according to the prior art, the coil inductance wound around it
  • L formula (3) where: is the relative magnetic permeability of the core, and S-like in the formula (1), is the effective cross-sectional area of the core (cm 2 ), N is the number of turns of the coil, and t is the length of the magnetic circuit (cm)
  • the circumference of the circular dotted line 51 in FIG. 5 is the magnetic path length 4.
  • FIG. 9-1 to 9-4 are a structural view of a magnetic core of a first embodiment
  • Fig. 9-1 is a front view of a magnetic core according to a first embodiment of the present invention
  • Fig. 9-2 is a first embodiment of the present invention
  • FIG. 9-3 is a top view of a magnetic core according to a first embodiment of the present invention
  • FIG. 9-4 is a perspective view of a magnetic core according to a first embodiment of the present invention.
  • the circumferential dotted line 51 in Figure 9-1 is the geometric magnetic circuit length.
  • the geometric magnetic circuit is actually divided into two segments, one segment in the thick portion, and the length is recorded as l w , which is the inside of the thick portion 52 in Figure 9-1.
  • the equivalent length of the magnetic circuit / t is its inherent length ⁇ .
  • 11-1, 11-2, and 11-3 are respectively a front view, a side view, and a top view of a magnetic core for comparison in the prior art.
  • the core of the prior art is compared in the magnetic core 11-1.
  • the length of the geometric magnetic circuit 51 is equal to the circumferential imaginary of the magnetic core of the present invention in Figure 9-1.
  • the geometric magnetic path length of the line 51 since the effective cross-sectional area of the core for comparison in the prior art in Fig. 11-1 is equal, the length of the geometric magnetic circuit 51 is equal to the actual magnetic path length, and the core of Fig. 11-1
  • the effective sectional area is equal to the effective sectional area of the thick portion of the magnetic core of the present invention in FIG. 9-1, that is, kS 1 ; then the magnetic cores of FIGS. 11-1 to 11-3 are:
  • the transformer realized by the magnetic core of the invention has the inductance in the same number of turns: Formula ( 10) According to the formula (10), since the cross-sectional area of the thick portion is larger than the detail portion, that is, k is always greater than 1, the inductance of the N ⁇ coil of the magnetic core of the first embodiment of the present invention is smaller than that of the prior art magnetic core.
  • the magnetic core of the present invention is applied to a self-excitation push-pull converter, and can also induce a push-pull oscillation of the circuit. Since the inductance is close, the operating frequency changes weakly, since the magnetic saturation only appears in the detail, the energy loss is low, and the no-load input current On the small, the conversion efficiency of the self-excited push-pull converter is significantly improved.
  • the lower limit of the length of the detail is 0.05 mm, because 0.05 mm is the current magnetic core molding process or cutting process.
  • the limit, in fact, below 0.05 mm, can better achieve the object of the present invention.
  • the N ⁇ coil inductance of the core of the first embodiment of the present invention is smaller than the N ⁇ inductance L 2 of the prior art core, and a constant y is introduced. , is a percentage, here is easy to express with a decimal, requires y less than 0.1, can be infinitely close to zero, set:
  • Equation (10-5) is simplified to: — 1 Formula (10-6)
  • the triode Since the triode has a storage time, that is, the base of the triode receives the turn-off signal, and the collector current is delayed until it is turned off, the storage time is generated in Figure 3, and the magnetic core operating point moves from D point moves to point E, correspondingly, or the movement point of the core working point moves from point H to point A.
  • the magnetic flux in the core increases, which may cause Saturation occurs in the rough portion where the area is not much different. Therefore, when the cross-sectional area of the detail is less than 80% of the cross-sectional area of the thick portion, it is ensured that the movement path of the above-mentioned core working point does not occur from point D to point E.
  • the crude magnetic saturation caused when the movable portion. K above i.e. 80% of the reciprocal, 1.25.
  • X is the ratio of the length of the detail 53 to the total length of the magnetic circuit 51, 0.12 is approximately one-eighth, ⁇ , and the length of the detail is 0.05 mm or more and the length of the total magnetic circuit is less than one-eighth.
  • the geometry causes the inductance to decrease, and instead increases the number of turns, thereby reducing the effect of the implementation. Therefore, the value range of K is required to be 1.25 times to 25, and the sectional area of the detail is 80% or less and 4% or more of the cross-sectional area of the rough portion.
  • Capacitor C is a luF capacitor
  • resistor R1 is 1 ⁇
  • capacitor C1 is 0.047uF capacitor
  • transistors TR1 and TR2 are switching transistors with a magnification of about 200 times.
  • the maximum working current of the collector is 1A;
  • the number of turns of the primary coils ⁇ and ⁇ 2 is 20 ⁇ , the number of turns of the feedback coils ⁇ and ⁇ 2 is 3 ⁇ , the number of turns of the secondary coils Nsi and NS2 is 23 ⁇ , respectively, and the core is 5 mm in outer diameter.
  • the first embodiment of the present invention employs a magnetic core having an outer diameter of 5 mm and a rough cross-sectional area of 1.5 mm 2 , and a detailed cross-sectional area of 1.2 mm 2 , that is, a k value of 1.25 and a detail length of 1 mm.
  • the coil is not wound around the detail, and the transformer in FIG. 1-1 is produced by the same number of turns.
  • the no-load current of the self-excited push-pull converter is also reduced from the prior art 18.0 mA to 14.1 mA of the present invention, i.e., the no-load loss is reduced from the prior art 90 mW to the 70.5 mW of the present invention.
  • Embodiment 2
  • the core structural view, the core is a magnetic air-gap magnetic circuit is closed, cylindrical ring, by the thick portion of the same material composition and details, details length 0.05 mm or more and a total magnetic circuit length of one-eighth or less; and the cross-sectional area of the detail portion is less than 80% and 4% or more of the cross-sectional area of the thick portion.
  • the cross-sectional area of the detail is 4% of the cross-sectional area of the thick portion, that is, the second embodiment of the present invention is adopted: the core has an outer diameter of 5 mm, The cross-sectional area of the thick portion is 1.5 square millimeters, and the cross-sectional area of the detail is 0.06 square millimeter.
  • the detail 53 in Fig. 9-1 is actually cut out by a cutting process, and the diameter of the detail is 0.276 ⁇ 0.02 mm, that is, the k value is 25, The thickness of the machined cutter is limited, and the length of the detail is 0.15 mm, which cannot be further shortened.
  • the coil is not wound around the detail, and the transformer in FIG. 1-1 is produced by the same number of turns.
  • the transformer manufactured by the first embodiment of the present invention is loaded into the circuit, when the output current is 5% of the full load of 200 mA, that is, when 10 mA is output.
  • the circuit test efficiency of Figure 6 is used. After the measured data is combined with Table 1, as shown in Table 5:
  • the conversion efficiency of the self-excited push-pull converter is less than 100mW under light load, that is, the output is 20mA or less, and the efficiency is significantly improved. Since the value of k is too large, the effect is remarkable. However, the diameter of the detail is 0.276 ⁇ 0.02 mm, which is very large in processing, extremely easy to break in the experiment, and the yield is extremely low.
  • the no-load current of the self-excited push-pull converter is also reduced from the prior art 18.0 mA to the present invention of 4.8 mA, i.e., the no-load loss is reduced from the prior art 90 mW to the 24 mW of the present invention.
  • Embodiment 3
  • the core of the structure diagram of FIG. 9-1 to FIG. 9-4 is still used, and the core is a closed magnetic ring without an air gap magnetic circuit, and the ring body is cylindrical, and is composed of a thick portion and a detail of the same material, and the length of the detail is 0.05. More than a millimeter, the total magnetic circuit length is less than one-eighth; and the cross-sectional area of the detail is 80% or less and 4% or more of the cross-sectional area of the thick portion.
  • the core has an outer diameter of 5 mm, a rough cross-sectional area of 1.5 mm 2 , and a detailed cross-sectional area of 0.75 mm 2 , that is, a k value of 2 and a detail length of 1 mm.
  • the coil is not wound around the detail, and the transformer in FIG. 1-1 is produced by the same number of turns.
  • the conversion efficiency of the self-excited push-pull converter is obviously improved from light load to full load, and the comparison chart is made by software, see FIG. 12, wherein the 2# curve is adopted.
  • the conversion efficiency graph of the self-excited push-pull converter after the invention, wherein the 1# curve is a conversion efficiency graph of the prior art self-excitation push-pull converter.
  • the no-load current of the self-excited push-pull converter is also reduced from 18.0 mA of the prior art to 12.0 mA of the present invention, that is, the no-load loss is reduced from 90 mW of the prior art to 60 mW of the present invention, that is, every Only the product has been reduced by 30mW.
  • the self-excited push-pull converter currently in use has at least 1 billion micropower power modules, and if all of the technical solutions of the present invention are used, it can save more than 30 million kWh per hour.
  • the invention not only solves the technical problem to be solved from the working principle, but also the above-mentioned a large number of first-disclosed formula derivation also gives a strong theoretical support of the invention, and at the same time, through experimental demonstration, the invention can be fully used in industrial applications and achieves the expected. effect. Since the self-excited push-pull converter has been publicized since 1955, after more than half a century of development and innovation, no one can use the technical means of the present invention to solve the technical problem to be solved by the present invention. The theoretical level is not sufficiently deep enough to understand that the present invention can be brought to a significant effect by simple improvement.
  • Embodiment 4 is not sufficiently deep enough to understand that the present invention can be brought to a significant effect by simple improvement.
  • the transformer used in the self-excitation push-pull converter of the fourth embodiment of the present invention is slightly different from the first, second and third embodiments, but the essence is the same, and the magnetic core is also a closed magnetic ring without an air gap magnetic circuit, and the ring body is Cylindrical, consisting of thick and thin parts of the same material, the length of the detail is 0.05 mm or more, and the total magnetic path length is less than one-eighth; and the cross-sectional area of the detail is less than 50% of the cross-sectional area of the thick portion. , 6. 25% or more.
  • the core core portion and the prior art core core cross-sectional area are the same, and the cross-sectional area of the detail portion is smaller than that of the prior art, and their ratio is l/k.
  • the cross-sectional area of the detail is equal to the cross-sectional area of the prior art, that is, the cross-sectional area of the thick portion is the cross-sectional area of the prior art. Times.
  • 11-1, 11-2, and 11-3 are respectively a front view, a side view, and a top view of a magnetic core used for comparison with a magnetic core according to a fourth embodiment of the present invention.
  • the cross-sectional area of the magnetic core of the prior art is S. 2 , then substituting into the formula (3), and obtaining the magnetic inductance of the prior art of the magnetic core of Figure 11-1, the coil inductance L 3 of the N winding is also:
  • FIG. 13-1 is a front view of a magnetic core according to a fourth embodiment of the present invention
  • Fig. 13-2 is a fourth embodiment of the present invention
  • FIG. 13-3 is a top view of a magnetic core according to a fourth embodiment of the present invention
  • a magnetic core has a small section and a core cut for comparison in the prior art.
  • the cross-sectional area of the detail 53 having the same area, that is, the detail 53 of the core in Fig. 13-1 is equal to the above S 2 , but the length thereof is short; accordingly, the cross-sectional area of the thick portion 52 in Fig.
  • the technical core cross-sectional area is large, equal to kS 2 , and its ratio is the reciprocal of the percentage in the technical solution, which is recorded as a constant k, which is reciprocal according to the above technical solution, that is, 1.25 times to 25 times, correspondingly, wound in the thick part
  • the inductance of the single coil of the upper coil is increased.
  • the inductance of the coil which is also wound by N ⁇ is 1 ⁇ 4 :
  • S 2 is the effective cross-sectional area (cm 2 ) of the detail of the core
  • gPkS 2 is the effective cross-sectional area of the thick portion
  • N is the coil
  • the denominator is the total equivalent length (cm) of the magnetic circuit, which is the sum of the magnetic path in the thick portion and the equivalent length of the magnetic circuit l t in the detail, and the effective sectional area of the detail is the effective sectional area of the thick portion.
  • l / k the magnetic circuit l t in the detail is equivalent to the effective cross-sectional area of the thick part, it is multiplied by k, the length is equivalent to kl t;
  • the inductance is
  • the magnetic path length ⁇ in the detail is sufficiently short, such as toward 0.05 mm, the sum of the product of 1 ⁇ 2 and the sum of the coarse magnetic circuits is closer to the magnetic core of the magnetic core of the prior art of Fig. 11-1.
  • the path length t that is, the inductance L 4 kL 3 of the N turns of the core of the present invention.
  • the inductance is increased by about k times, that is, the number of turns can be considerably reduced, and the same inductance as in the background art can be achieved; that is, the number of turns of the present invention can be Quite a reduction, the number of turns in the prior art and the number of turns in the present invention, the ratio of which is:
  • the sectional length of the detail portion is less than 80% of the cross-sectional area of the thick portion, and the origin of 4% or more.
  • the first embodiment corresponds to the foregoing mentioned: After the value exceeds 25, The "window" at the center of the magnetic ring is too small, and it is often necessary to extend the length of the geometric magnetic circuit 51 in Fig. 9-1. As can be seen from the formula (3), the extension of the geometric magnetic circuit 51 causes the inductance to decrease, instead To increase the number of turns, the implementation effect is reduced.
  • the magnetic core of the present invention is applied to a self-excitation push-pull converter, and the push-pull oscillation of the circuit can also be induced. Since the magnetic saturation occurs only in the thin portion 53, the length is short, and the energy is short. The loss is low, that is, the no-load input current of the self-excitation push-pull converter circuit is small, and the conversion efficiency of the self-excited push-pull converter is obviously improved. Since the energy loss is low, the self-excitation push-pull converter can be operated. The operating frequency is further increased, which brings about a good advantage: The number of turns wound on the core of the fourth embodiment of the present invention can be further reduced. The actual effect of the core of the fourth embodiment is embodied by a set of measured data.
  • Capacitor C is a luF capacitor
  • resistor R1 is 1 ⁇
  • capacitor C1 is 0.047uF capacitor
  • transistors TR1 and TR2 are switching transistors with a magnification of about 200 times.
  • the maximum working current of the collector is 1A;
  • the number of turns of the primary coils ⁇ and ⁇ 2 is 20 ⁇ , the number of turns of the feedback coils ⁇ and ⁇ 2 is 3 ⁇ , the number of turns of the secondary coils Nsi and NS2 is 23 ⁇ , respectively, and the core is 5 mm in outer diameter.
  • the fourth embodiment of the present invention employs: the core has an outer diameter of 5 mm, a rough cross-sectional area of 3 mm 2 , and a detail cross-sectional area of 1.5 mm 2 , that is, a k value of 2 and a detail length of 0.5 mm.
  • the coil is not wound on the detail, the number of turns of the primary coils NPI and NP2 is 7 ⁇ , the number of turns of the feedback coils NB1 and NB2 is 2 ⁇ , and the number of turns of the secondary coils Nsi and NS2 is 8 ⁇ , respectively.
  • the operating frequency of the measured circuit is 139 KHz, and the no-load input current is 6.9 mA.
  • the no-load current of the self-excited push-pull converter is also reduced from the prior art 18.0 mA to the present invention.
  • Figure 15-1 is a fifth embodiment of the present invention
  • Figure 15-1 is a front view of a magnetic core according to a fifth embodiment of the present invention
  • Figure 15-2 is a side view of a magnetic core according to a fifth embodiment of the present invention
  • 15-3 is a top view of a magnetic core according to a fifth embodiment of the present invention
  • FIG. 15-4 is a perspective view of a magnetic core according to a fifth embodiment of the present invention; similarly, there is a small section 53 having a small core cross-sectional area, which is symmetrically removed on a cylindrical magnetic ring.
  • a sheet-like detail 53 is formed, the length of the detail is 0.05 mm or more, and the total magnetic path length is one-eighth or less; the cross-sectional area of the detail is 80% or less and 4% or more of the cross-sectional area of the thick portion.
  • the working principle is the same as the working principle in the above invention and the working principles of the first embodiment to the fourth embodiment, and details are not described herein again.
  • FIG. 16-1 to 16-4 are a sixth embodiment of the present invention
  • FIG. 16-1 is a front view of a magnetic core according to a sixth embodiment of the present invention
  • FIG. 16-2 is a side view of a magnetic core according to a sixth embodiment of the present invention
  • 16-3 is a top view of a magnetic core according to a sixth embodiment of the present invention
  • 16-4 is a perspective view of a magnetic core according to a sixth embodiment of the present invention; similarly, there is a small section 53 having a small core cross-sectional area, a thick portion 52, and a sixth embodiment Further improved features: There is a transition body 54 between the thick portion and the detail portion, and the transition body 54 can be equivalent to a part of the detail portion, which corresponds to three details in the embodiment, and the cross-sectional area of the transition body 54 is from large to small, from differential From the point of view, there are innumerable details.
  • the transition body 54 is provided for the magnetic core to be demolded after the magnetic powder is molded. In fact, this is shown in Figures 9-1 to 9-4. A further improvement of the first embodiment.
  • the cross-sectional areas of the two symmetrical thin portions 54, the thin portion 53 and the two symmetrical thin portions 54 are not equal, then the thin portion 53 having the smallest cross-sectional area acts, and the cross-sectional area of the thin portion 53 having the smallest area is between
  • the cross-sectional area of the thick portion is 80% or less and 4% or more.
  • the inside of the core corresponding to the detail 54 is not magnetically saturated, and does not participate in the magnetic saturation operation.
  • the length of the detail, the transition body 54 is required to be short, and the sum of the length of the transition body and the detail is 0.05 mm or more and the total magnetic path length is less than one eighth.
  • the working principle is the same as the working principle in the above invention and the working principles of the first embodiment to the fourth embodiment, and details are not described herein again. Due to the presence of the transition body 54, the length of the detail 53 can be zero, and at the same time there is a portion having a minimum cross-sectional area, and the portion having the smallest cross-sectional area can achieve magnetic saturation first, and the same can be achieved.
  • FIG. 17-1 to 17-4 are a seventh embodiment of the present invention
  • FIG. 17-1 is a front view of a magnetic core according to a seventh embodiment of the present invention
  • FIG. 17-2 is a side view of a magnetic core according to a seventh embodiment of the present invention
  • 17-3 is a top view of a magnetic core according to a seventh embodiment of the present invention
  • FIG. 17-4 is a perspective view of a magnetic core according to a seventh embodiment of the present invention
  • An improvement feature of the seventh embodiment of the present invention is: On the basis of the sixth embodiment, two or more bumps 55 are added to the thick portion to prevent the winding on the thick portion from sliding to the detail, and the bump 55 can be anywhere on the rough. Another effect of the bump 55 is: It is possible to position different windings in order to prevent crossover between them.
  • the transition body 54 can be equivalent to a part of the detail, and the transition body 54 is provided for the magnetic core to be molded in the magnetic powder. Post demolding is convenient, in fact this is a further improvement of the sixth embodiment of Figs. 16-1 to 16-4.
  • the length of the detail and the transition body 54 are required to be short.
  • the working principle is the same as the working principle in the above invention and the working principles of the first embodiment to the fourth embodiment, and details are not described herein again. Since the length of the transition body 54, the length 53 can be zero, the same can be achieved.
  • FIG. 18-1 to 18-4 are eighth embodiment of the present invention
  • Fig. 18-1 is a front view of a magnetic core according to an eighth embodiment of the present invention
  • Fig. 18-2 is a side view of a magnetic core according to an eighth embodiment of the present invention
  • 18-3 is a top view of a magnetic core according to an eighth embodiment of the present invention
  • FIG. 18-4 is a perspective view of a magnetic core according to an eighth embodiment of the present invention
  • the magnetic ring without closed air gap is composed of a flat rough portion 52 and a thin portion 53 of the same material.
  • the length of the detail portion 53 is 0.05 mm or more and the total magnetic circuit length is one eighth or less.
  • the sectional area of the detail portion 53 is 80% or less and 4% or more of the sectional area of the thick portion 52.
  • the working principle is the same as the working principle in the above-mentioned invention and the working principles of the first embodiment to the fourth embodiment, and details are not described herein again.
  • FIG. 19-1 to 19-4 are a ninth embodiment of the present invention
  • FIG. 19-1 is a front view of a magnetic core according to a ninth embodiment of the present invention
  • FIG. 19-2 is a side view of a magnetic core according to a ninth embodiment of the present invention
  • 19-3 is a top view of a magnetic core according to a ninth embodiment of the present invention
  • FIG. 19-4 is a perspective view of a magnetic core according to a ninth embodiment of the present invention; likewise, a small portion 53 having a small core cross-sectional area and a thick portion 52 are present.
  • transition body 54 between the thick portion and the detail portion of the ninth embodiment, and the transition body 54 can be equivalent to a part of the detail portion.
  • the transition body 54 is provided for the purpose of demolding the magnetic core after the magnetic powder is molded, in fact, this is the figure 18- A further improvement of the eighth embodiment of 1 to 18-4. Since the length of the transition body 54, the detail 53 can be zero, the same can be achieved.
  • the working principle of the ninth embodiment is the same as the working principle of the above-mentioned invention and the working principles of the first embodiment to the fourth embodiment, and details are not described herein again.

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Chemical & Material Sciences (AREA)
  • Composite Materials (AREA)
  • Dc-Dc Converters (AREA)
  • Coils Or Transformers For Communication (AREA)

Abstract

本发明公开了一种变压器用磁心，所述的磁心由闭合的粗部和细部组成，所述的细部在相同的由小到大的磁场激励下比所述的粗部先达到磁饱和；所述的细部在工作时，只在瞬间接近或达到所述的细部第一象限饱和点或第三象限饱和点，其它时间都在所述的细部固有的第一象限饱和点和第三象限饱和点之间；本发明克服现有的磁心用在自激推挽式变换器存在的缺点，使得自激推挽式变换器在负载轻载时，自激推挽式变换器的效率得到显著提升；在额定负载时，效率进一步提升；降低自激推挽式变换器中磁饱和变压器上线圈的匝数；提高自激推挽式变换器的工作频率而损耗仍维持较低水平。

Description

一种变压器用磁心
技术领域
本发明涉及变压器用磁心， 特别涉及电源变换器用的变压器磁心。 背景技术
现有的自激推挽式变换器及其使用的变压器， 其电路结构来自 1955年美国罗 耳 ( G.H.Royer)发明的自激振荡推挽晶体管单变压器直流变换器，也作 Royer电路， 这也是实现高频转换控制电路的开端； 部分电路来自 1957年美国查赛 （Jen Sen, 有的地方译作 "井森")发明的自激式推挽双变压器电路， 后被称为自振荡 Jensen电 路或 Jensen电路； 这两种电路， 后人统称为自激推挽式变换器。 自激推挽式变换 器在电子工业出版社的 《开关电源的原理与设计》 第 67页至 70页有描述， 该书 ISBN号 7-121-00211-6。电路的主要形式为上述著名的 Royer电路和自振荡 Jensen 电路。
图 1-1示出的为自激推挽式变换器常见应用， 电路结构为 Royer电路； 图 1-2 示出的电路就是著名的自振荡 Jensen电路， 中文常音译为"井森"电路， 在图 1-1 和图 1-2中，电路都要利用变压器 B1的磁心饱和特性进行振荡，在图 1-2的 Jensen 电路中， 电路的自振荡频率和驱动功能， 改由磁饱和的变压器 B1来实现， 因此， 主功率变压器 B2能工作在不饱和状态。
Royer电路的振荡频率是电源电压的函数， 在电子工业出版社的《开关电源的 原理与设计》 第 68页第 18行有描述， 该书 ISBN号 7-121-00211-6。 这里引用如 下： f = χ \ 4 Ηζ 公式 (l) BwSN
式中： f为振荡频率； w为工作磁感应强度 (Τ)， 一般取 50%〜70%磁饱和点 £m 值； N为线圈匝数； S为磁心有效截面积； ½为工作电源电压。
为了方便理解 Royer电路的工作原理，特别是磁心饱和特性进行振荡这一点， 这里以图 1-1为例， 说明其工作原理。
图 1-1的电路结构为： 输入滤波电容 C连接于电压输入端与地之间， 对输入电 压进行滤波； 滤波后的输入电压接入启动电路， 启动电路由偏置电阻 R1 和电容 C1并联组成； 偏置电阻 R1的两端分别与电压输入端以及为两个推挽晶体管 TR1、 TR2基极提供正反馈的变压器 B1原边线圈 NBI和 NB2的中心抽头连接；两个推挽 晶体管 TR1、 TR2的发射极共地， 两个集电极分别连接变压器原边线圈 NPI和 NP2 的两个端头， 基极连接变压器原边线圈 NBI和 NB2的两个端头， 原边线圈 NP1和 NP2中的中心抽头连接电压输入端； 变压器 B1的副边线圈 NS连接输出电路至电 压输出端。
其工作原理简述为： 参见图 1-1， 接通电源瞬间， 偏置电阻 R1和电容 C1并联 回路通过线圈 NBI和 NB2绕组为三极管 TR1和 TR2的基极、发射极提供了正向偏 压， 两只三极管 TR1和 TR2开始导通， 由于两个三极管特性不可能完全一样， 因 此， 其中一只三极管会先导通， 假设三极管 TR2先导通， 产生集电极电流 IC2， 其 对应的线圈 NP2绕组的电压为上正下负， 根据同名端关系， 其基极线圈 NB2绕组 也出现上正下负的感应电压， 这个电压增大了三极管 TR2的基极电流， 这是一个 正反馈的过程， 因而很快使三极管 TR2饱和导通； 相应地， 三极管 TR1对应的线 圈 NBI绕组的电压为上正下负， 这个电压减小了三极管 TR1的基极电流， 三极管 TR1很快完全截止。
三极管 TR2对应的线圈 NP2绕组里的电流， 以及这个电流产生的磁感应强度随 时间而线性增加， 但磁感应强度增加到接近或达到变压器 B1磁心的饱和点 Bm时， 线圈 NP2的电感量迅速减小， 从而使三极管 TR2开关管的集电极电流急剧增加， 增 加的速率远大于基极电流的增加， 三极管 TR2开关管脱离饱和， 三极管 TR2开关 管的集电极到发射极的压降 UCE增大， 相应地， 变压器 NP2绕组上的电压就减小同 一数值，线圈 NB2绕组感应的电压减小，结果使三极管 TR2开关管基极电压也降低， 造成三极管 TR2开关管向截止方向变化， 此时， 变压器 B1线圈上的电压将反向， 使另一只三极管 TR1导通， 此后， 重复进行这一过程， 形成推挽振荡。 绕组 Ns的 输出端的波形如图 2所示， 可见， 除了"集极谐振型 Royer电路"输出正弦波或近似 正弦波外， 自激推挽式变换器的工作波形是接近方波的。 集极谐振型 Royer电路又 称"冷阴极灯管逆变器 (CCFL inverter)", 也会简称为 CCFL逆变器或 CCFL变换器， CCFL变换器在供电回路串入主功率绕组十倍电感量以上的电感， 以获得输出正弦 波或近似正弦波。 图 3为变压器 B1磁心的方形磁滞回线， 其中 +Bm、 一 Bm为磁心 的两个磁饱和点， 其中 +Bm称为第一象限饱和点， 因为该点 +Bm落在图 3坐标中第 一象限， 一 Bm称为第三象限饱和点， 在图 2的半个周期内， 变压器 B1磁心的工作 点运动的路线为 ABCDE,在下半个周期内运动路线为 EFGHA。事实上，三极管 TR2 或 TR1对应的线圈绕组里的电流， 以及这个电流产生的磁感应强度随时间而线性 增加到图 3中 D点或 H点时， 电路就会进行推挽转换， 即另一只三极管导通， 而对 应的三极管会截止， 由于三极管存在一个存储时间 (storage time), 即三极管基极接 收到关断信号， 而集电极电流要延时才能下降直到关断， 存储时间会产生在图 3 中， 磁心工作点运动路线从 D点向 E点移动， 对应地， 或磁心工作点运动路线从 H 点向 A点移动， 在这个移动过程中， 磁心的磁滞作用会导致三极管集电极电流无谓 增加而产生损耗。
其特点为： 利用磁心饱和特性进行推挽振荡， 变压器输出波形为近似方波， 电路的变换效率较高。 磁心要在特定的时间瞬间接近饱和状态， 所以无法采用存 在气隙的磁心。 自激推挽式变换器必需使用磁饱和式磁心， 而磁心加气隙是公知 的抗磁饱和的手段。
鉴于各种文献不区分磁心与磁芯和铁心以及铁芯， 本文中的磁心和公知的其 它文献一样， 和磁芯 (Magnetic Core)表示相同的意思， 指铁氧体材料， 即由各种氧 化铁混合物组成的一种烧结磁性金属氧化物，磁心多用在高频。而铁心、铁芯 (Iron Core)是硅片型材料， 只适合低频率的电感线、 低变压器， 一般用于低频和音频。
图 1-2示出的相似结构，就是开关驱动功能与主功率变压器脱离的电路， 如图 1-2所示。 前文描述过， 电路的自振荡频率和驱动功能， 改由磁饱和的变压器 B 1 来实现， 因此， 主功率变压器 B2能工作在不饱和状态。 虽然 B 1出现磁饱和， 因 为 B1的体积小， 磁饱和消耗的能量小， 在相同条件下， Jensen电路的总体效率略 高。
上述的自激推挽式变换器因为磁心存在磁饱和， 而存在以下缺点：
1、 负载轻载时， 变换器的效率低
由于 Royer电路是利用磁心饱和特性进行推挽振荡， 所以其空载工作电流不 会太小， 如表一为 Royer电路实测参数。 如使用图 1-1的电路， 做成输入直流 5V， 输出直流 5V， 输出电流为 200mA的变换器， 即输出功率 1W。 变压器的后续输出 采用图 4的电路结构， 图 4为公知的全波整流电路， 二极管 D41和 D42均采用型 号为 RB 160的肖特基二极管。 电路的主要参数为： 电容 C为 luF电容， 电阻 R1 为 1ΚΩ， 电容 C1为 0.047uF电容， 三极管 TR1和 TR2为放大倍数在 200倍左右 的开关三极管， 其集电极最大工作电流为 1A; 其中， 原边线圈 ΝΡΙ和 ΝΡ2的圈数 分别为 20匝， 反馈线圈 NBI和 NB2的圈数分别为 3匝， 副边线圈 Nsi和 NS2的圈 数分别为 23匝， 磁心采用外直径 5毫米， 横截面积 1.5平方毫米的常见铁氧体环 形磁心， 俗称磁环， 其外形立体图参见图 5。
实测电路的空载工作电流为 18mA， 工作频率为 97.3KHz， 接近 ΙΟΟΚΗζ, 变 换效率测试电路采用图 6的电路， VI电压表头为工作电压 Vin， 即输入电压； A1 电流表头为输入电流 Πη， 即为工作电流； V2电压表头为输出电压 Vout， A2电流 表头为输出电流 lout; 那么变换效率可以用公式 (2)计算得出。
电路的变换效率为： χ 100% 公式 (2)
Figure imgf000006_0001
式中： Vin为工作电压， 即输入电压， Iin为输入电流； Vout为输出电压， lout 为输出电流。 测试时都采用图 6的接线方式， RL为可变负载， 可以有效地减小测 量误差。电流表和电压表均使用 MY65型 4位半数字万用表的 200mA档和 20V档 或 200V档， 同时使用了四块及四块以上的万用表。
MY65型 4位半数字万用表在测电压时， 内阻为 10ΜΩ, 200mA电流档的内 阻为 1Ω。 当电流超过 200mA时， 采用了两块电流表置于 200mA档并联测量， 把 两块表的电流读数相加， 即为测量值。 电流表并联测量是现有电子工程的成熟技 术。
用图 1-1的电路， 按上述参数制作好后， 当输出电流在 200mA的 5%时， 即输 出 10mA时， 工作频率为 97.3KHz， 实测参数如下表一：
表一
Figure imgf000006_0002
从上表可以看出， 当输出只有满载的 5%时， 效率只有 38.03%， 这个效率目前 在工业界的小功率模块电源中极具代表性。
Jensen电路， 参见图 1-2， 尽管采用了一个小变压器 B1来实现磁饱和， 而主 功率变压器 B2工作在不饱和状态， 以此来提高效率， 事实上， 由于采用了两只变 压器， 产生损耗的器件多了一个， 小变压器 B1的设计时要兼顾整个电路的输出功 率，精心调试后，输出 5V，输出 5V/200mA的 Jensen电路的空载电流下降至 16mA， 当输出只有满载的 5%时， 效率比上述的 Royer电路有所提升， 达 40.91%。
2、 额定负载时， 效率无法进一步提升
自激推挽式变换器， 以 Royer电路为例， 想提升电路的变换效率， 由于电路 在推挽式工作时， 每次推挽转换， 都是靠接近或进入磁心磁饱和实现的， 磁饱和 消耗的能量， 以热量形式损耗掉， 想提高效率， 需降低变换器的工作频率， 从公 式 (1)可以看到， 在输入电压不变的前提下， 只能加大公式中分母中的参数， 加大 工作磁感应强度 Bw， 或加大线圈匝数 N， 或加大磁心有效截面积8。 目前的变换 器产品都已选用工作磁感应强度 Bw极大的磁心， 加大线圈匝数 N， 带来铜损耗增 力 而磁心有效截面积 S的加大， 同时增加了每次接近或进入磁心磁饱和时的损 耗， 使得变换器的变换效率不增反降。 因此， 在设计自激推挽式变换器时， 在这 些参数之间取舍是一件较为困难的事。
对于 Jensen电路， 想提升电路的变换效率， 同样相似的原因， 小变压器 B1 磁心有效截面积 S取小了， 导致推动功率不足， 开关三极管不能进入良好的饱和 导通， 引起的压降损失也大， 变换器的变换效率也低； 小变压器 B1磁心有效截面 积 S取大了， 自身的损耗也大； 可以通过加大线圈匝数 N来解决问题， 但又带来 下述的工艺问题， 即： 匝数 N多， 由于小变压器 B1必需工作在磁饱和状态下， 不 能开气隙， 这给绕制带来很大的困难。
3、 输入电压比较高时， 变压器 B1上的匝数过多， 加工困难
自激推挽式变换器， 以 Royer电路为例， 从公式 (1)可以看到， 在输入电压升 高的前提下， 若自激推挽式变换器工作频率保持不变， 那么， 公式 (1)中的分母对 应参数要加大， 对于同一系列、 同功率的工业级小模块电源， 经常采用同一尺寸 的磁心， 这时， 只能改变加大线圈匝数 N来解决问题， 如图 1-1公布的电路参数， 若做成输入 24V的产品，原边线圈 NPI和 NP2的匝数就要由 5V输入时的各 20匝， 上升至各 96匝， 由于图 1-1中变压器 B1必需工作在磁饱和状态下， 不能开气隙， 这给绕制带来很大的困难， 目前在不到 10mm直径的小磁环上绕制这么多匝漆包 线， 无论是机器绕制还是手工绕制， 都存在加工困难。 采用机器绕制， 当第一层 绕好， 绕第二层时， 第二层线很难叠在第一层上， 会破坏第一层的线序， 结果越 绕越差、 越乱； 而手工绕制， 这么多圈数， 全靠操作工人用心记忆， 很难保证不 多绕一两匝， 或少绕一两匝， 匝数一旦改变， 则输出电压就出现偏差， 严重时变 压器装机后， 不能实现原有功能。
若把磁心有效截面积 S加大一倍， 则匝数可以降低为 48匝， 但这时变压器 B1磁心有效截面积 S取大了一倍， 同样的频率下， 自身的损耗也大了一倍， 变换 器的转换效率降低。
所以， 目前工业领域及市面上较难见到工作在 48V及以上电压的自激推挽式 变换器模块， 也是这个原因， 只能降低效率来换取较少匝数。
4、 工作频率难以提高
由于自激推挽式变换器电路是利用磁心饱和特性进行推挽振荡， 每次推挽转 换都是靠磁心接近磁饱和或进入磁饱和实现的， 所以工作频率提升后， 损耗增加， 变换效率下降。
对于 Jensen电路， 同样相似的原因， 小变压器 B1磁心有效截面积 S更小了， 在 24V输入电压下， 经常需要绕制 60匝的原边， 由于原边线圈仅为一组， 可以用 双线并绕，只绕 30匝，然后串接得到 60匝的原边，但由于小变压器 B1直径更小， 无论是机器还是手工绕制， 都存在加工困难。 而对于 48V输入电压下， 小变压器 B1几乎无法加工。 同样， 若把磁心有效截面积 S加大一倍， 则匝数可以降低， 同 样的工作频率下， 自身的损耗也大了一倍， 变换器的转换效率降低。 在 CN101290828的专利中示出了用于工作在磁饱和区的不等截面铁心结构， 无论在工作段绕上输入或输出绕组， 和上世纪 70年代未广为流行的家用、 工业用 交流磁饱和稳压器一样， 只能工作在正弦波或波形失真不大的正弦波下， 都无法 克服自激推挽式变换器存在的上述缺点。 该专利存在的不足， 请参见电子工业出 版社的 《开关电源设计》 第 174页：
该书 ISBN号 7-121-01755-5， 该页最后一段有详细描述，原文为 "需要指出是 的， 绝大多数由带状迭片叠成的铁心表面辐射面积很小， 因而具有较高的热阻， 约为 40〜100°C/W。 除非固定在散热片上， 它们的总损耗必须维持在 1W以下"； 该书 174页的表 6.3同时表明了铁心的损耗很大， 在 ΙΟΟΚΗζ工作频率下损耗最小 的是 Toshiba MB铁心材料， 其铁心损耗为 1.54W/cm3， 即每立方厘米该种铁心在 ΙΟΟΚΗζ的工作频率下， 其固有损耗达 1.54W, 这在工业界的小功率模块电源中是 无法接受的； 而在 50KHz下损耗相对较小， 前文提到， 自激推挽式变换器是工作 波形是接近方波的， 方波的上升沿属阶跃信号， 按傅立叶变换展开后， 其频率可 以是基频的 20倍以上， 即为 50KHzx20=lMHz， 这时这种铁心的损耗是很大的， 基频指自激推挽式变换器的工作频率基频， 即图 2中方波的频率。
事实上， 在该书 174页第三段开始已说明： "材料通常被加工成薄带状， 卷成 柱形"。 这个卷带技术在工频用环形变压器中大量使用， 目的也是为了获得无气隙 的铁心。使用带状迭片叠成直径小于 10mm的环形铁心， 本身的加工难度极大， 这 种情况下， 都会选择磁粉模压后烧结成的磁心， 在 ΙΟΟΚΗζ工作频率下， 磁心的铁 损相对都在几十至几百 mW/cm3级左右，相对低得多，请参见电子工业出版社的《开 关电源设计》 第 184页的表 7.1对应 ΙΟΟΚΗζ部分的参数。
说明： 工频指 50Hz或 60Hz指工业上用的交流电源的频率。上世纪 70年代未 广为流行的家用、工业用交流磁饱和稳压器原理可以参阅上世纪 80年代初期的《电 子世界》 和 《无线电》 杂志。
在 1985年 2月 19日公开的 JP60032312A的专利中示出了一种扼流电感用磁 心 (MAGNETIC CORE FOR CHOKE COIL), 其要解决的问题和 1987年 7月 21 日 公开的 JP62165310A的专利中示出的扼流电感 (； CHOKE COIL)是同一问题，在小电 流时获得较大的电感量， 而在大电流下获得相对较小的电感量， 这样用于开关电 源中作为续流电感使用时， 可以改善在断续模式下开关电源的输出纹波， 在输出 小电流情况下， 开关电源进入断续模式 (DCM)下工作， 这时， 续流电感使用这两 项专利， 可以得到较大的电感量， 这样， 开关电源的工作模式会向连续模式 (CCM) 转移， 流过电感的电流变小， 但流过电感的电流因为变小而持续的时间延长， 这 样改善输出纹波， 这也是业界目前公知的技术， 从这两分公开的文件中附图也可 以看出，请见 JP60032312A的公开文件中第 5图、 JP62165310A的公开文件中第 2 图。 这两项专利技术也无法克服自激推挽式变换器在前文描述的缺点， 这些缺点 都由自激推挽式变换器采用的变压器磁心存在磁饱和引起的。 发明内容
有鉴如此， 本发明要解决的技术问题是， 克服现有的磁心用在自激推挽式变 换器存在的缺点， 使得自激推挽式变换器在负载轻载时， 自激推挽式变换器的效 率得到显著提升； 在额定负载时， 效率进一步提升； 降低自激推挽式变换器中磁 饱和变压器上线圈的匝数； 提高自激推挽式变换器的工作频率而损耗仍维持较低 水平。
为解决上述技术问题， 本发明提供一种变压器用磁心， 所述的磁心由闭合的 粗部和细部组成， 所述的细部在相同的由小到大的磁场激励下比所述的粗部先达 到磁饱和； 所述的细部在工作时， 只在瞬间接近或达到所述的细部第一象限饱和 点或第三象限饱和点， 其它时间都在所述的细部固有的第一象限饱和点和第三象 限饱和点之间。 所述的细部有一个或一个以上。
优选地， 所述的细部长度为 0. 05毫米以上、 总磁路长度八分之一以下， 且截 面积介于粗部截面积 80%以下、 4%以上； 更优地， 细部的截面积介于粗部的截面积 50%以下、 6. 25%以上， 与粗部的截面积相比， 细部的截面积越小， 则细部的长度 占磁环总磁路长度也越小为好， 以保证细部的机械强度。
优选地， 所述的粗部与细部材质相同。
优选地， 所述的粗部和细部连接处存在过渡体， 从而使得磁心在磁粉模压后 脱模方便。
优选地， 所述的粗部上有两个或两个以上的凸点， 用于防止粗部上的绕线滑 到细部， 或对不同的绕组进行区域定位， 以防它们之间交叉等。
本发明还提供一种利用上述磁心绕制的变压器， 所述的粗部绕制线圈， 所述 的细部不绕线圈。 本发明的工作原理为， 上述技术方案应用于自激推挽式变换器中， 以图 1-1 的 Royer电路为例， 图 1-1 自激推挽式变换器中的变压器 B1采用上述本发明技术方 案， 电路的推挽工作原理并没有变化， 当三极管 TR2饱和导通， 相应地， 三极管 TR1完全截止， 三极管 TR2对应的线圈 NP2绕组里的电流， 以及这个电流产生的磁 感应强度随时间而线性增加， 但磁感应强度增加到变压器 B1磁心的饱和点 Bm时， 线圈的电感量迅速减小， 从而使三极管 TR2开关管的集电极电流急剧增加， 增加 的速率远大于基极电流的增加， 三极管 TR2开关管脱离饱和， 三极管 TR2开关管 的集电极到发射极的压降 UCE增大， 相应地， 变压器 NP2绕组上的电压就减小同一 数值， 线圈 NB2绕组感应的电压减小， 结果使三极管 TR2开关管基极电压也降低， 造成三极管 TR2开关管向截止方向变化， 此时， 变压器 B1线圈上的电压将反向， 使另一只三极管 TR1导通， 此后， 重复进行这一过程， 形成推挽振荡。 能量损耗主要是由于磁感应强度增加到变压器 Bl磁心的饱和点 Bm时产生的， 现有技术使用截面积相同的磁心， 基本上要让整个磁心达到饱和点 Bm， 而要较多 的能量才能达到饱和。 如图 3所示， 变压器磁心在上半个周期内运动的路线为 ABCDE, 在下半个个周期内运动路线为 EFGHA, 即在一个完整周期内， 现有技术 磁心的工作区域为整个区域， 如图 7中阴影 71所示。
而本发明存在一小段和现有技术中磁心截面积不相同的细部， 细部在相同的 由小到大的磁场激励下比粗部先达到磁饱和， 但其长度却很短， 让这一小段细部 磁心达到其本身的饱和点 Bm， 同样可以引起电路的推挽振荡， 细部这一部分的磁 心的工作区域为整个区域， 和图 7中阴影 71相同， 但细部的长度很短， 消耗的能 量大幅降低； 粗部因为面积比细部大， 容纳较多的磁力线， 当细部出现接近磁饱 和的瞬间， 粗部并没有工作在饱和状态， 其对应的磁心工作区域为图 8中阴影 81 所示， 面积明显减小， 所以粗部消耗的能量很低， 故本发明的磁心消耗的能量总 体降低很大。 即采用本发明的磁心， 可以降低每次推挽变换时的能量消耗， 在同 样的工作频率下， 采用本发明磁心的自激推挽式变换器电路的消耗会降低， 表现 为电路的空载工作电流下降； 同样原因， 可以实现自激推挽式变换器工作频率提 升而损耗上升并不大， 实现在高频率下变换效率仍可不下降。
如上所述， 在相同的工作频率下， 自激推挽式变换器电路的空载工作电流就 会相应降低， 即空载损耗就会同比例降低， 相应地， 本发明应用于自激推挽式变 换器中， 在负载轻载时， 变换器的效率会明显提升。 同样， 在负载轻载至满载整 个区间工作时， 变换器的对应效率都会明显提升。
基于上述原理， 由于粗部并不需要工作在两个饱和点之间， 所以可以加大粗 部的截面积， 这样每匝电感量会随截面积的增加而成比例增加， 这样， 实现相同 的电感量， 总匝数就会相应下降， 从而降低自激推挽式变换器中磁饱和变压器 B1 上线圈的匝数， 实现本发明的目的。 详细的工作原理和公式推导将在实施例中展 现。
而现有技术使用截面积相同的磁心， 加大截面积会导致因磁饱和损耗加大， 导致空载工作电流增加， 自激推挽式变换器变换效率在满载时不变或反而有所下 降， 而在轻载时由于损耗大， 变换效率下降严重。 背景技术提及的专利方案都必 须让整个或部分磁心、 铁心工作在稳定的磁饱和状态下， 它们是利用稳定的磁饱 和实现发明目的， 比如， 交流磁饱和稳压器获得稳定的交流电压输出， 背景技术 提及的扼流电感 (CHOKE COIL)在小电流时获得较大的电感量，而在大电流下获得 相对较小的电感量， 扼流电感要在很大的工作范围内利用稳定的磁饱和实现电感 量随电流增加而线性降低， 交流磁饱和稳压器和扼流电感带来的损耗都很大， 因 为它们在工作时出现磁饱和后， 磁饱和持续的时间占一个周期中较长时间， 甚至 占一个周期中二分之一以上。 而本发明仅利用长度极短的细部磁心实现少量的磁 饱和引发电路的推挽转换， 磁饱和只在瞬间出现， 其持续时间几乎无法精确测量， 参见图 7， 由于本发明只有细部工作在图 7中阴影 71区域， 因细部长度短， 细部 这一部分的磁心的工作区域从 D点运动至 E点， 磁心的磁滞作用会导致三极管集 电极电流无谓增加而产生损耗， 但因细部长度小， 在这一过程中， 磁滞明显降低， 因此产生的损耗也降低； 由于细部长度短， 磁滞相对很小， 细部这一部分的磁心 的工作区域从 D点运动至 E点所花的时间会縮短， 即本发明磁心中细部瞬间接近 或达到饱和点所持续的时间縮短。 即所述的细部在工作时， 只在瞬间接近或达到 所述的细部第一象限饱和点或第三象限饱和点， 其它时间都在所述的细部固有的 第一象限饱和点和第三象限饱和点之间。
如上所述， 本发明是要降低磁饱和带来的负面影响， 这是和现有技术的本质 区别。
也正因为如此， 当本发明磁心存在两个或两个以上细部时， 若多个细部的截 面积相等， 它们相当于串联关系， 等效于一个总细部。 其工作原理等同于上述原 理。
当本发明磁心存在两个或两个以上细部时， 若多个细部的截面积不等， 那么， 截面积最小的细部起作用， 其它细部对应的磁心内部不会磁饱和， 从而不参与工 作， 由于其它细部的截面积较粗部小， 会减小整个磁心上的线圈的电感量。 其工 作原理等同于上述原理。
与现有技术相比， 本发明具有如下显著的效果：
(1)使得自激推挽式变换器在负载轻载时， 自激推挽式变换器的效率得到显著 提升；
(2)在额定负载时以及从空载至满载的整个工作区间， 变换效率进一步提升； 即降低了输入端的功率消耗。
(3)降低自激推挽式变换器中磁饱和变压器上线圈的匝数。
(4)使得自激推挽式变换器的变压器绕制效率提高， 单件生产工时縮短。 (5)使得 48V及以上输入电压的自激推挽式变换器成为可能， 且工艺简单。
(6)使得自激推挽式变换器在空载时， 工作电流下降。
(7)提高自激推挽式变换器的工作频率。
(8)可以实现 lOOmW以下的高效率微功率 DC/DC模块电源。 附图说明
图 1-1 为自激推挽式变换器 Royer电路常见应用电路图；
图 1-2 为自激推挽式变换器 Jensen电路常见应用电路图；
图 2 为图 1-1的 Royer电路绕组 Ns的输出端的波形图；
图 3 为图 Ι-lRoyer电路中变压器 B1磁心的方形磁滞回线； 图 4 为公知的全波整流电路；
图 5 为现有技术的磁环式磁心；
图 6 为本文中通用的变换效率测试原理图；
图 7 为磁心在自激推挽式变换器中工作区域图；
图 8 为本发明粗部对应磁心在自激推挽式变换器中工作区域图； 图 9-1 为本发明第一至第三实施例的磁心主视图；
图 9-2 为本发明第一至第三实施例的磁心侧视图；
图 9-3 为本发明第一至第三实施例的磁心俯视图；
图 9-4 为本发明第一至第三实施例的磁心立体图；
图 10-1 为本发明第一至第三实施例的粗部内的磁路 ^示意图； 图 10-2 为本发明第一至第三实施例的细部内的磁路 lt示意图； 图 11-1 为现有技术对比用的磁心主视图；
图 11-2 为现有技术对比用的磁心侧视图；
图 11-3 为现有技术对比用的磁心俯视图；
图 12 为本发明第四实施例和现有技术磁心装入电路后效率对比图； 图 13-1 为本发明第四实施例的磁心主视图；
图 13-2 为本发明第四实施例的磁心侧视图；
图 13-3 为本发明第四实施例的磁心俯视图；
图 14 为本发明第四实施例和现有技术磁心装入电路后效率对比图； 图 15-1 为本发明第五实施例的磁心主视图； 图 15-2 为本发明第五实施例的磁 ι 则视图;
图 15-3 为本发明第五实施例的磁 ι 府视图；
图 15-4 为本发明第五实施例的磁 立体图;
图 16-1 为本发明第六实施例的磁 主视图;
图 16-2 为本发明第六实施例的磁 ι 则视图;
图 16-3 为本发明第六实施例的磁 ι 府视图；
图 16-4 为本发明第六实施例的磁 立体图;
图 17-1 为本发明第七实施例的磁 主视图;
图 17-2 为本发明第七实施例的磁 ι 则视图;
图 17-3 为本发明第七实施例的磁 ι 府视图；
图 17-4 为本发明第七实施例的磁 立体图;
图 18-1 为本发明第八实施例的磁 主视图;
图 18-2 为本发明第八实施例的磁 ι 则视图;
图 18-3 为本发明第八实施例的磁 ι 府视图；
图 18-4 为本发明第八实施例的磁 立体图;
图 19-1 为本发明第九实施例的磁 主视图;
图 19-2 为本发明第九实施例的磁 ι 则视图;
图 19-3 为本发明第九实施例的磁 ι 府视图；
图 19-4 为本发明第九实施例的磁 立体图。 具体实施方式
实施例一
图 94至图 9-4为本发明第一实施例磁心， 磁心为无气隙磁路闭合磁环， 环体 呈圆柱状， 由同材质的粗部和细部组成， 细部长度为 0.05毫米以上、 总磁路长度 八分之一以下； 且细部的截面积介于所述的粗部的截面积 80%以下、 4%以上。
为了清晰地展现本实施例的效果， 本发明第一实施例粗部和现有技术中磁心 截面积相同， 细部比现有技术中磁心截面积小。 粗部的截面积和细部的截面积的 比值， 该比值是技术方案中的百分比的倒数， 记作常数 k， 按上述 "80%以下、 4% 以上"取倒数， 即 1.25倍至 25倍， 图 5示出的是现有技术的磁环式磁心， 如图 5 所示， 其截面积 50处处相等， 那么， 按现有的公知技术， 绕在它上面的线圈电感
Figure imgf000015_0001
L 公式 (3) 式中： 为磁心相对磁导率， 和公式 (1)中的 S—样， 为磁心有效截面积 (cm2) ， N为线圈匝数， t为磁路长度 (cm), 图 5中的圆周虚线 51的周长即为磁 路长度 4。
图 9-1至图 9-4系列示出的是第一实施例磁心的结构图， 图 9-1为本发明第一 实施例的磁心主视图； 图 9-2为本发明第一实施例的磁心侧视图； 图 9-3为本发明 第一实施例的磁心俯视图； 图 9-4为本发明第一实施例的磁心立体图。 其中图 9-1 中的圆周虚线 51为几何磁路长度，该几何磁路事实上被分为两段，一段在粗部内， 其长度记作 lw， 为图 9-1中粗部 52内部的那段磁路； 几何磁路的另一段在细部内， 其长度记作 lt， 为图 9-1中细部 53内部的那段磁路。 图 10-1和图 10-2分别示出了 本发明中磁路 ^和1,的示意图， 图 10-1中， 虚线 61表示在粗部内的磁路长度 lw， 图 10-2中， 虚线 62表示在细部内的磁路长度 lt， 那么， 本发明第一实施例中， 绕在 粗部上线圈的电感量计算方法， 就可以根椐法拉第定律推导出， 设细部的有效截 面积为 S1 ; 那么粗部的有效截面积为 kS1 ; 代入公式 (3)， 那么粗部上绕制 N匝的线 圈电感量1^为：
_ 4π χ μί χ kS{ x N2
h = ^ ^it ^ 公式 (4) 式中： 为磁心相对磁导率； Si为磁心的细部的有效截面积（cm2)， ^81为 粗部的有效截面积， 和公式 (1)中的 S—样； N为线圈匝数； 分母为磁路总等效长度 (cm), 为粗部内的磁路 /^和细部内的磁路 /t的等效长度之和，细部由于其有效截面积 是粗部的有效截面积的 l/k， 细部内的磁路 ^若按粗部的有效截面积等效， 按公知 理论， 就要乘以 fc， 长度等效为 ¾， 艮
1等效长度― Klt
磁路 /t的等效长度， 是其固有长度 ^的^咅。 图 11-1、 11-2、 11-3分别为现有技 术对比用的磁心主视图、 侧视图、 俯视图， 为了方便说明本发明的原理， 现有技 术对比用的磁心图 11-1中的几何磁路 51的长度等于图 9-1中本发明磁心的圆周虚 线 51的几何磁路长度，由于图 11-1中现有技术对比用的磁心的有效截面积处处相 等， 其几何磁路 51长度就等于实际磁路长度， 且图 11-1中的磁心的有效截面积等 于图 9-1中本发明磁心的粗部的有效截面积， 即为 kS1 ; 那么图 11-1至图 11-3的磁 心有：
le=lw+lt 公式 (6)
Ae= kSj 公式 (7) 那么， 把上式代入公式 (3)， 图 11-1现有技术对比用的磁心同样绕制 N匝的线 圈电感量 L2为：
_ 4π χ μί χ kS1 χ Ν1
L2 = 公式 (8) 用公式 (4)和公式 (8)相比， 得：
Lx _
~ = 公式 (9)
Figure imgf000016_0001
艮卩， 本发明磁心实现的变压器， 在同匝数下， 其电感量为：
Figure imgf000016_0002
公式 (10) 根据公式 (10)， 由于粗部截面积比细部大， 即 k恒大于 1， 所以本发明第一实施 例磁心的 N匝线圈电感量 小于现有技术磁心的 N匝电感量 L2， 只要取得不太大， 而细部内的磁路长度 It足够短， 那么本发明磁心的 N匝电感量1^可以很接近 L2， 如 L!=0.99L2, 这时， 由于存在细部， 本发明的磁心应用于自激推挽式变换器中， 同 样可以引发电路推挽振荡， 由于电感接近， 工作频率变化微弱， 由于磁饱和只出 现在细部内， 能量损耗低， 空载输入电流就小， 自激推挽式变换器的变换效率明 显提升。
由于细部磁路 It越小越好， 从而磁环需要达到磁饱和的部分也越少， 损耗降低 也越明显， 细部的长度下限取 0.05mm， 是因为 0.05mm是目前磁心模压成型工艺 或切割工艺的极限， 事实上， 0.05mm以下可以更好地实现本发明目的。
以下是对权利要求中的几个尺寸限定进行证明的过程， 至公式 (10-10)， 本证 明的方法、 过程是未公开过的学术首创：
参考公式 (6)， 设 X为细部长度 53占总磁路长度 51的比例， 那么有： lt =x le 公式（10- 1) lw =(l-x) le 公式（10-2) 根据公式 (10)， 本发明第一实施例磁心的 N匝线圈电感量 小于现有技术磁心 的 N匝电感量 L2，引入一个常数 y，为百分比，这里用小数方便表达，要求 y小于 0.1， 可以无限地接近零， 设：
L l-yj 公式（10-3) 把公式 (6)、 公式（10-1)、 公式（10-2)、 公式（10-3)代入公式（10)得：
( 、
- y)L2 公式（10- 4)
Figure imgf000017_0001
对公式（10-4)进行简化， 左右约去 L2，等式右边分子分母约去 ， 得：
1 1
\ -x + kx 1 + x(k― 1)
公式（10-5)简化为： — 1 公式（10- 6)
Figure imgf000017_0002
因为 y小于等于 0.1， 那么有下述工程计算中的等式：
Λ 2 Λ 公式（10-7)
1- _y ¾ 1
当 y=0. 1时， 公式（10-7)出现 0. 99« 1， 误差是 1%， 已满足工程计算的精度， 因为 y小于等于 0.1， 从公式（10-7)得出：
即：
1 Λ
- ^ ~ + y 公式 do- 9) 公式（10-9)中， y取最大值 0. 1，则出现：
艮 1.1的循环小数约等于 1.1的情况， 误差为 1%， 而当 y下降至 0.05即 5% 时，出现 1.05263 « 1.05，误差为 0.25%，已非常精确。把公式（10-9)代入公式（10-6) , 则有： x(k - 1) =― 1 1 + y - 1 = y
1 - y 即：
1) -公式（10-10) y - x(k - 公式（10-3)可以看到， 本发明希望 y值越小越好， 这样， 本发明的磁心的电 感量越接近期望值， 这样不用增加本发明第一实施例线圈匝数， 通过实验实测发 现， 细部的截面积介于所述的粗部的截面积 80%以下时， 才保证当细部出现磁饱和 时， 粗部不出现磁饱和。 由于三极管存在一个存储时间 (storage time), 即三极管基 极接收到关断信号， 而集电极电流要延时才能下降直到关断， 存储时间会产生在 图 3中， 磁心工作点运动路线从 D点向 E点移动， 对应地， 或磁心工作点运动路 线从 H点向 A点移动。磁心工作点运动路线从 D点向 E点移动时， 会引发磁心中 磁通增加， 这时会引发面积相差不大的粗部内出现饱和， 所以， 细部的截面积介 于所述的粗部的截面积 80%以下时，才确保不会出现上述磁心工作点运动路线从 D 点向 E点移动时引发粗部出现磁饱和。 k即上述的 80%的倒数， 为 1.25。
当 k取 1.25时， 公式 (10-10)中的 y若不大于 3%， 那么：
x=y/(k-l)=0.03/(1.25-l)=0.12
X为细部长度 53占总磁路长度 51的比例， 0.12近似为八分之一， δΡ， 所述的 细部长度为 0. 05毫米以上、 总磁路长度八分之一以下。
当然， 这里仅以 y为 3%为例说明的， 下表二给出了电子工程中常用可以容忍 的偏差值， 都是依据公式 (10-10)计算所得， 如下：
表二：
Figure imgf000018_0001
通过实验实测验证， 本发明 y取 3%以下时， 才能保证较好的实施效果。 在实际应用中， 值超过 25以后， 即粗部截面积是细部截面积的 25倍后， 细 部的强度不好保证， 容易断裂， 断裂后反而产生气隙， 无法应用于自激推挽式变 换器， 值超过 25以后， 由于磁环中心的可以绕线的 "窗口面积"过小， 经常需 要延长图 9-1中几何磁路 51的长度来实施， 从公式 (3)可以看到， 几何磁路 51延 长， 会使得电感量降低， 反而要增加匝数， 从而降低了实施效果。 因此， K的取值范围要求为 1.25倍至 25，此时细部截面积为粗部截面积的 80% 以下、 4%以上。
下面以一组实测数据来体现第一实施例磁心的实际效果。 以图 1-1为例，现有 技术采用和背景技术中介绍的方案相同：
使用图 1-1的电路， 做成输入直流 5V， 输出直流 5V， 输出电流为 200mA的 变换器， 即输出功率 1W。 变压器的后续输出采用图 4的电路结构， 图 4为公知的 全波整流电路。 电路的主要参数为： 电容 C为 luF电容， 电阻 R1为 1ΚΩ， 电容 C1为 0.047uF电容， 三极管 TR1和 TR2为放大倍数在 200倍左右的开关三极管， 其集电极最大工作电流为 1A; 其中， 原边线圈 ΝΡΙ和 ΝΡ2的圈数分别为 20匝， 反 馈线圈 ΝΒΙ和 ΝΒ2的圈数分别为 3匝， 副边线圈 Nsi和 NS2的圈数分别为 23匝， 磁心采用外直径 5毫米， 横截面积 1.5平方毫米的常见铁氧体环形磁心。
按上述参数制作好后， 当输出电流在 200mA满载的 5%时， 即输出 10mA时， 实测参数参见背景技术表一， 效率为 38.03%。
本发明第一实施例采用：磁心采用外直径 5毫米，粗部横截面积 1.5平方毫米， 而细部横截面积为 1.2平方毫米， 即 k值为 1.25， 细部的长度为 1毫米。 细部上不 绕线圈，采用同样的匝数制作图 1-1中的变压器，采用本发明第一实施例制作的变 压器装入电路后， 当输出电流在 200mA满载的 5%时， 即输出 10mA时， 同样采 用图 6的电路测试效率， 实测数据结合表一后， 如表三所示：
Figure imgf000019_0001
明显地， 采用本发明后， 自激推挽式变换器的变换效率在负载轻载时明显提 升， 上升了 (42.05%-38.03%)=4.02%。
进一步地， 对负载从轻载、 满载进行了测试； 记录在表四：
表四：
Figure imgf000019_0002
本发明 25.6 5.066 9.95 5.482 42.05% 现有技术 242.0 5.024 198.00 5.094 82.96%
100% 1.54% 本发明 239.1 5.025 199.00 5.102 84.50% 明显地， 采用本发明后， 自激推挽式变换器的变换效率在轻载、 满载， 都有 提升， 由于 k取值较小， 所以效果较为一般。
自激推挽式变换器的空载电流也由现有技术的 18.0mA下降至本发明的 14.1mA, 即空载损耗由现有技术的 90毫瓦下降至本发明的 70.5毫瓦。 实施例二
图 9-i至图 9-4为本发明第二实施例磁心结构图，磁心为无气隙磁路闭合磁环， 环体呈圆柱状， 由同材质的粗部和细部组成， 细部长度为 0.05毫米以上、 总磁路 长度八分之一以下；且细部的截面积介于所述的粗部的截面积 80%以下、 4%以上。
为了清晰地展现本实施例的效果， 在本发明第二实施例磁心中， 细部的截面 积为粗部的截面积的 4%， 即本发明第二实施例采用： 磁心采用外直径 5毫米， 粗 部横截面积 1.5平方毫米， 而细部横截面积为 0.06平方毫米， 实际用切割工艺把 图 9-1中的细部 53切出来， 细部直径为 0.276 ± 0.02mm， 即 k值为 25， 受加工的 环切机刀片厚度限制， 细部的长度为 0.15毫米， 无法进一步縮短。 细部上不绕线 圈，采用同样的匝数制作图 1-1中的变压器，采用本发明第一实施例制作的变压器 装入电路后， 当输出电流在 200mA满载的 5%时， 即输出 10mA时， 同样采用图 6 的电路测试效率， 实测数据结合表一后， 如表五所示：
表五:
Figure imgf000020_0001
明显地， 采用本发明后， 自激推挽式变换器的变换效率在负载轻载时明显提 升， 上升了 (69.49%-38.03%)=31.46%。 这是在输出 10mA电流时测出来的， 即输出 50mW时测出的效率。 进一步地， 对输出 20mA电流进行了测试， 对应的负载率为 (20÷200) X 100%, 记录在表六：
表六：
Figure imgf000021_0001
明显地， 采用本发明后， 自激推挽式变换器的变换效率在轻载 lOOmW以下， 即输出 20mA以下， 效率都有显著提升， 由于 k取值过大， 所以效果显著。 但细 部直径为 0.276 ± 0.02mm， 加工很大， 在实验中极容易断裂， 成品率极低。
自激推挽式变换器的空载电流也由现有技术的 18.0mA下降至本发明的 4.8mA, 即空载损耗由现有技术的 90毫瓦下降至本发明的 24毫瓦。 实施例三
第三实施例仍采用图 9-1至图 9-4结构图磁心， 磁心为无气隙磁路闭合磁环， 环体呈圆柱状， 由同材质的粗部和细部组成， 细部长度为 0.05毫米以上、 总磁路 长度八分之一以下；且细部的截面积介于所述的粗部的截面积 80%以下、 4%以上。
上述两个实施例中， k用两个极限值展示了实施效果， 为了清晰地展现本实施 例的效果， 第三实施例中常数 k取中间值 2来展示实施效果。 本发明第一实施例 应用， 磁心采用外直径 5毫米， 粗部横截面积 1.5平方毫米， 而细部横截面积为 0.75平方毫米， 即 k值为 2， 细部的长度为 1毫米。 细部上不绕线圈， 采用同样的 匝数制作图 1-1中的变压器，采用本发明第一实施例制作的变压器装入电路后， 当 输出电流在 200mA满载的 5%时， 即输出 10mA时， 同样采用图 6的电路测试效 率， 实测数据结合表一后， 如表七所示：
表七:
Figure imgf000021_0002
明显地， 采用本发明后， 自激推挽式变换器的变换效率在负载轻载时明显提 升， 上升了 (45.80%-38.03%)=7.77%。
进一步地， 对负载从轻载至满载整个区间， 按 5%步长全部进行了测试； 当负 载率超过 40%后， 按 10%步长进行了测试， 记录在表八：
表八:
Figure imgf000022_0001
明显地， 采用本发明后， 自激推挽式变换器的变换效率从轻载至满载整个区 间， 都有明显提升， 用软件做成对比图表， 参见图 12， 其中的 2#曲线为采用本发 明后自激推挽式变换器的变换效率曲线图， 其中的 1#曲线为现有技术的自激推挽 式变换器的变换效率曲线图。 自激推挽式变换器的空载电流也由现有技术的 18.0mA下降至本发明的 12.0mA, 即空载损耗由现有技术的 90毫瓦下降至本发明的 60毫瓦， 即每只产品 降低了 30mW。
据不完全统计， 目前在使用中的自激推挽式变换器所制成微功率电源模块， 至少有 10亿只， 若全部使用本发明的技术方案， 每小时可以节约 3000万度电以 上。
本发明不仅从工作原理上解决所要解决的技术问题， 而且上述大量首次公开 的公式推导也给予了本发明有力的理论支持， 同时通过实验论证， 本发明完全可 以在工业应用中使用并达到预期的效果。自激推挽式变换器自 1955年被公开以来， 经过半个多世纪的发展创新， 至今无人能够利用本发明的技术手段来解决本发明 所要解决的技术问题， 正是由于对该电路的理论层面理解不够深刻充分， 从而在 此之前无法意识到本发明可以通过简单的改进从而带来显著的效果。 实施例四
本发明第四实施例自激推挽式变换器采用的变压器和第一、 二、 三实施例略 有区别， 但本质是相同的， 磁心也为无气隙磁路闭合磁环， 环体呈圆柱状， 由同 材质的粗部和细部组成， 细部长度为 0. 05毫米以上、 总磁路长度八分之一以下； 且细部的截面积介于所述的粗部的截面积 50%以下、 6. 25%以上。
在第一实施例中， 磁心粗部和现有技术的磁心截面积相同， 而细部的截面积 比现有技术的截面积小， 它们的比值为 l/k。 为了充分地展现本实施例的效果， 在 本发明第四实施例采用的变压器磁心中， 细部的截面积等于现有技术的截面积， 即粗部的截面积是现有技术的截面积的 k倍。
图 11-1、 11-2、 11-3分别为现有技术用于和本发明第四实施例磁心对比用的 磁心主视图、侧视图、俯视图，设现有技术的磁心的截面积为 S2，那么代入公式 (3)， 得到图 11-1现有技术对比用的磁心同样绕制 N匝的线圈电感量 L3为：
4π χ μί x S2 x N2
½ = j 公式 (11) 本发明第四实施例磁心参见图 13-1至图 13-3，图 13-1为本发明第四实施例的 磁心主视图； 图 13-2为本发明第四实施例的磁心侧视图； 图 13-3为本发明第四实 施例的磁心俯视图； 第四实施例磁心存在一小段和现有技术中用于对比的磁心截 面积相同的细部 53， 即图 13-1中磁心的细部 53的截面积等于上述的 S2， 但其长 度却很短；那么相应地，图 13-1中粗部 52的截面积就比现有技术的磁心截面积大， 等于 kS2， 其比值是技术方案中的百分比的倒数， 记作常数 k， 按前文的技术方案取 倒数， 即 1.25倍至 25倍， 相应地， 绕在粗部上线圈的单匝电感量增加， 本发明第 四实施例磁心， 同样绕制 N匝的线圈电感量1^4为：
4π X μί χ kS2 x N
L -公式 (12)
+ klt
式中： 为磁心相对磁导率； S2为磁心的细部的有效截面积（cm2)， gPkS2为 粗部的有效截面积， 和公式 (1)中的 S—样； N为线圈匝数； 分母为磁路总等效长度 (cm), 为粗部内的磁路 ^和细部内的磁路 lt的等效长度之和，细部由于其有效截面积 是粗部的有效截面积的 l/k， 细部内的磁路 lt若按粗部的有效截面积等效， 就要乘 以 k， 长度等效为 klt;
(12)和公式 (11)相比， 得： 公式 (13)
Figure imgf000024_0001
即， 本发明第四实施例磁心实现的变压器， 在同匝数下， 其电感量为
LA = k .公式 (14)
Figure imgf000024_0002
根据公式 (14)， 若细部内的磁路长度 ^足够短， 如趋向 0.05mm， ½的乘积和粗 部磁路 /^之和就越接近图 11-1现有技术对比用的磁心的磁路长度 t， 即本发明磁心 的 N匝线圈的电感量 L4 kL3。
绕同样的匝数， 但电感量却加大了约 k倍， 即所绕的匝数可以相当地降低， 即可同样实现和背景技术中相同的电感量； 即本发明所绕的匝数可以相当地降低， 现有技术的匝数和本发明的匝数， 其匝比《为：
Π = - 公式 (15) k
即当 取 25时， 《=1/5=0.2， 对于背景技术中提到的输入电压 24V的产品， 需 要绕制 96匝才能获得较好的工作效率， 而本发明可以用 96 X 0.2=19.2匝， 取整用
20匝来实现。 这也是技术要求中， 所述的细部长度截面积介于所述的粗部的截面 积 80%以下、 4%以上的来历。 实施例一对应的前文提到过： 值超过 25以后， 由 于磁环中心的 "窗口"过小， 经常需要延长图 9-1中几何磁路 51的长度来实施， 从公式 (3)可以看到， 几何磁路 51延长， 会使得电感量降低， 反而要增加匝数， 从 而降低了实施效果。
同样， 当 取 16时， 《=1/4=0.25， 可以让匝数降低为原有匝数的 1/4， 方便绕 制。
在图 13-1中由于存在细部 53， 本发明的磁心应用于自激推挽式变换器中， 同 样可以引发电路推挽振荡， 由于磁饱和只出现细部 53内， 而其长度较短， 能量损 耗低， 即自激推挽式变换器电路的空载输入电流就小， 自激推挽式变换器的变换 效率明显提升， 由于能量损耗低， 就可以把工作自激推挽式变换器的工作频率进 一步提高， 这样带来好的好处是： 本发明第四实施例磁心上绕制的匝数可以进一 步降低。 下面以一组实测数据来体现第四实施例磁心的实际效果。
现有技术用于对比的方案， 采用和背景技术中、 实施例一中介绍的方案相同， 为了方便对比， 这里引用如下：
使用图 1-1的电路， 做成输入直流 5V， 输出直流 5V， 输出电流为 200mA的 变换器， 即输出功率 1W。 变压器的后续输出采用图 4的电路结构， 图 4为公知的 全波整流电路。 电路的主要参数为： 电容 C为 luF电容， 电阻 R1为 1ΚΩ， 电容 C1为 0.047uF电容， 三极管 TR1和 TR2为放大倍数在 200倍左右的开关三极管， 其集电极最大工作电流为 1A; 其中， 原边线圈 ΝΡΙ和 ΝΡ2的圈数分别为 20匝， 反 馈线圈 ΝΒΙ和 ΝΒ2的圈数分别为 3匝， 副边线圈 Nsi和 NS2的圈数分别为 23匝， 磁心采用外直径 5毫米， 横截面积 1.5平方毫米的常见铁氧体环形磁心。
按上述参数制作好后， 当输出电流在 200mA满载的 5%时， 即输出 10mA时， 实测参数如背景技术表一， 效率为 38.03%。 其它参数见表七、 表八中对应的现有 技术部分。
本发明第四实施例采用： 磁心采用外直径 5毫米， 粗部横截面积 3平方毫米， 而细部横截面积为 1.5平方毫米， 即 k值为 2， 细部的长度为 0.5毫米。 细部上不 绕线圈， 原边线圈 NPI和 NP2的圈数分别为 7匝， 反馈线圈 NB1和 NB2的圈数分 别为 2匝，副边线圈 Nsi和 NS2的圈数分别为 8匝，采用本发明第四实施例制作的 变压器装入电路后， 实测电路的工作频率为 139KHz， 空载输入电流为 6.9mA。
当输出电流在 200mA满载的 5%时， 即输出 10mA时， 同样采用图 6的电路 测试效率， 实测数据结合表一后， 如表九所示： 表九:
Figure imgf000026_0001
明显地， 采用本发明后， 自激推挽式变换器的变换效率在负载轻载时明显提 升， 上升了 (61.48%-38.03%)=23.45%。
进一步地， 对负载从轻载至满载整个区间， 按 5%步长全部进行了测试； 当负 载率超过 40%后， 按 10%步长进行了测试， 记录在表十：
表十:
口 Iin Vin lout Vout
腳 广口口 效率
(mA) (V) (mA) (V)
现有技术 28. 4 5. 060 9. 96 5. 487 38. 03%
5% 23. 45% 本发明 17. 5 5. 066 9. 95 5. 478 61. 48% 现有技术 40. 0 5. 045 20. 00 5. 424 53. 76%
10% 20. 58% 本发明 28. 9 5. 052 20. 00 5. 427 74. 34% 现有技术 51. 1 5. 031 30. 10 5. 381 63. 00%
15% 16. 83% 本发明 40. 3 5. 037 30. 10 5. 384 79. 84% 现有技术 62. 7 5. 016 40. 00 5. 341 67. 93%
20% 15. 12% 本发明 51. 4 5. 022 40. 10 5. 346 83. 05% 现有技术 73. 7 5. 002 50. 00 5. 305 71. 95%
25% 12. 70% 本发明 62. 8 5. 008 50. 10 5. 314 84. 65% 现有技术 85. 1 4. 987 59. 90 5. 269 74. 37%
30% 11. 67% 本发明 73. 7 4. 994 60. 00 5. 278 86. 04% 现有技术 96. 3 4. 973 69. 90 5. 236 76. 42%
35% 10. 31% 本发明 85. 0 4. 978 70. 00 5. 243 86. 74% 现有技术 107. 6 4. 960 79. 70 5. 202 77. 68%
40% 9. 58% 本发明 96. 1 4. 965 79. 90 5. 211 87. 26% 现有技术 130. 1 4. 931 99. 70 5. 138 79. 85%
50% 8. 02% 本发明 118. 5 4. 936 99. 90 5. 145 87. 87% 现有技术 153. 8 4. 900 120. 50 5. 073 81. 11%
60% 7. 21% 本发明 141. 5 4. 907 120. 70 5. 081 88. 33% 现有技术 174. 9 4. 873 139. 40 5. 013 81. 99%
70% 6. 31% 本发明 162. 7 4. 879 139. 60 5. 021 88. 30% 现有技术 199. 8 4. 847 161. 80 4. 950 82. 70%
80% 5. 60% 本发明 187. 4 4. 845 161. 90 4. 952 88. 30%
90% 现有技术 220. 0 5. 030 180. 00 5. 131 83. 46% 4. 84% 本发明 208. 2 5. 032 180. 20 5. 134 88. 31%
现有技术 242. 0 5. 024 198. 00 5. 094 82. 96%
100% 5. 04% 本发明 229. 1 5. 025 199. 00 5. 091 88. 00% 明显地， 采用本发明后， 自激推挽式变换器的变换效率从轻载至满载整个区 间， 都有明显提升， 用软件做成对比图表， 参见图 14， 其中的 2#为采用本发明后 自激推挽式变换器的变换效率曲线图， 其中的 1#曲线为现有技术的自激推挽式变 换器的变换效率曲线图。
自激推挽式变换器的空载电流也由现有技术的 18.0mA下降至本发明的
6.9mA, 即空载损耗由现有技术的 90毫瓦下降本发明的 34.5毫瓦。 同时工作频率 也由现有技术的 97.3KHZ上升至本发明实施例二的 139KHz。带来的好处， 就是把 原边线圈 NPI和 NP2的圈数分别为现有技术的 20匝下降为 7匝， 节约绕制工时， 也不容易出错。
从表六可以看到， 当负载为 10%时， 即输出 20mA的电流时， 本发明的效率 仍有 74%， 若减小磁心， 专门设计成微功率的 DC/DC变换器， 即可实现效率进一 步提升。 综上所述， 第四实施例的综合实施效果是比较好的。
实施例五
图 15-1至图 15-4为本发明第五实施例， 图 15-1 为本发明第五实施例的磁心主视 图； 图 15-2 为本发明第五实施例的磁心侧视图； 图 15-3 为本发明第五实施例的 磁心俯视图； 图 15-4 为本发明第五实施例的磁心立体图； 同样存在一小段磁心截 面积小的细部 53， 在圆柱状的磁环上对称切除形成一个片状的细部 53， 细部长度 为 0.05毫米以上、总磁路长度八分之一以下；细部的截面积介于粗部的截面积 80% 以下、 4%以上。 工作原理同上述的发明内容中的工作原理以及实施例一至实施例 四的工作原理， 这里不再赘述。
实施例六
图 16-1至图 16-4为本发明第六实施例， 图 16-1 为本发明第六实施例的磁心 主视图； 图 16-2 为本发明第六实施例的磁心侧视图； 图 16-3 为本发明第六实施 例的磁心俯视图； 图 16-4 为本发明第六实施例的磁心立体图； 同样存在一小段磁 心截面积小的细部 53， 粗部 52， 以及第六实施例的进一步改进特征： 粗部和细部 之间存在过渡体 54， 过渡体 54可以等效为细部的一部分， 相当于本实施例存在三 个细部， 过渡体 54的截面积从大到小， 从微分的观点来看， 其实是存在无数个细 部，设置过渡体 54是为了磁心在磁粉模压后脱模方便，事实上这是对图 9-1至 9-4 的第一实施例的进一步改进。 当本实施例存在一个细部 53， 两个对称细部 54， 细 部 53和两个对称细部 54的截面积不等， 那么， 截面积最小的细部 53起作用， 面 积最小的细部 53的截面积介于粗部的截面积 80%以下、 4%以上。 细部 54对应的 磁心内部不会磁饱和， 从而不参与磁饱和工作。
同样， 细部的长度、 过渡体 54要求短， 过渡体和所述的细部长度之和为 0.05 毫米以上、 总磁路长度八分之一以下。 工作原理同上述的发明内容中的工作原理 以及实施例一至实施例四的工作原理， 这里不再赘述。 由于存在过渡体 54， 细部 53的长度可以为零， 这时一样存在一个最小截面积的部分， 最小截面积的部分可 以实现先达到磁饱和， 一样可以实现发明目的。
实施例七
图 17-1至图 17-4为本发明第七实施例， 图 17-1 为本发明第七实施例的磁心 主视图； 图 17-2 为本发明第七实施例的磁心侧视图； 图 17-3 为本发明第七实施 例的磁心俯视图； 图 17-4 为本发明第七实施例的磁心立体图；
本发明第七实施例的改进特征是： 在第六实施例的基础上， 在粗部上增加了 两个或两个以上凸点 55， 以防止粗部上的绕线滑到细部， 凸点 55可以在粗部上任 意地方。 凸点 55另一作用是： 可以对不同的绕组进行区域定位， 以防它们之间交 叉等。
由于同样存在一小段磁心截面积小的细部 53， 粗部 52， 以及粗部和细部之间 的过渡体 54， 过渡体 54可以等效为细部的一部分， 设置过渡体 54是为了磁心在 磁粉模压后脱模方便， 事实上这是对图 16-1至 16-4的第六实施例的进一步改进。
同样细部的长度、 过渡体 54要求短。 工作原理同上述的发明内容中的工作原 理以及实施例一至实施例四的工作原理， 这里不再赘述。 由于存在过渡体 54， 细 部 53的长度可以为零， 一样可以实现发明目的。
实施例八
图 18-1至图 18-4为本发明第八实施例，图 18-1 为本发明第八实施例的磁心 主视图； 图 18-2 为本发明第八实施例的磁心侧视图； 图 18-3 为本发明第八实施 例的磁心俯视图； 图 18-4 为本发明第八实施例的磁心立体图； 无气隙磁路闭合磁 环由同材质的扁平状粗部 52和细部 53组成， 细部 53长度为 0. 05毫米以上、 总 磁路长度八分之一以下； 细部 53的截面积介于所述的粗部 52的截面积 80%以下、 4%以上。 工作原理同上述的发明内容中的工作原理以及实施例一至实施例四的工作原 理， 这里不再赘述。
实施例九
图 19-1至图 19-4为本发明第九实施例， 图 19-1 为本发明第九实施例的磁心 主视图； 图 19-2 为本发明第九实施例的磁心侧视图； 图 19-3 为本发明第九实施 例的磁心俯视图； 图 19-4 为本发明第九实施例的磁心立体图； 同样存在一小段磁 心截面积小的细部 53， 粗部 52。
第九实施例的粗部和细部之间存在过渡体 54，过渡体 54可以等效为细部的一 部分， 设置过渡体 54是为了磁心在磁粉模压后脱模方便， 事实上这是对图 18-1 至 18-4的第八实施例的进一步改进。 由于存在过渡体 54， 细部 53的长度可以为 零， 一样可以实现发明目的。
第九实施例的工作原理同上述的发明内容中的工作原理以及实施例一至实施 例四的工作原理， 这里不再赘述。
以上仅是本发明的优选实施方式， 应当指出的是， 上述优选实施方式不应视 为对本发明的限制， 本发明的保护范围应当以权利要求所限定的范围为准。 对于 本技术领域的普通技术人员来说， 在不脱离本发明的精神和范围内， 还可以做出 若干改进和润饰， 这些改进和润饰也应视为本发明的保护范围。 比如， 采用各种 几何形状截面积的磁环实施上述细部、 粗部， 或采用方形、 椭圆形的磁环实施上 述整个磁心的外形。

Claims

权利要求
1、 一种变压器用磁心， 其特征在于： 所述的磁心由闭合的粗部和细部组成， 所述 的细部在相同的由小到大的磁场激励下比所述的粗部先达到磁饱和； 所述的细部 在工作时， 只在瞬间接近或达到所述的细部第一象限饱和点或第三象限饱和点， 其它时间都在所述的细部固有的第一象限饱和点和第三象限饱和点之间。
2、 根据权利要求 1所述的磁心， 其特征在于所述的细部为一个或一个以上。
3、 根据权利要求 1所述的磁心， 其特征在于： 所述的细部长度为 0.05毫米以上、 总磁路长度八分之一以下； 且所述的细部的截面积介于所述的粗部的截面积 80% 以下、 4%以上。
4、 根据权利要求 3所述的磁心， 其特征在于： 所述的细部截面积介于所述的粗部 的截面积 50%以下、 6.25%以上。
5、根据权利要求 1所述的磁心， 其特征在于： 所述的粗部和所述的细部材质相同。
6、 根据权利要求 1所述的磁心， 其特征在于： 所述的粗部和所述的细部连接处存 在方便脱模的过渡体。
7、 根据权利要求 1所述的磁心， 其特征在于： 所述的粗部上有两个或两个以上的 用于防止绕线滑到细部， 或对不同的绕组进行区域定位的凸点。
8、一种用权利要求 1所述磁心绕制的变压器， 其特征在于： 所述的粗部绕制线圈， 所述的细部不绕线圈。
PCT/CN2012/072656 2011-12-22 2012-03-20 一种变压器用磁心 WO2013091315A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US14/356,886 US10008312B2 (en) 2011-12-22 2012-03-20 Magnetic core for transformer

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201110436359.XA CN102436907B (zh) 2011-12-22 2011-12-22 一种变压器用磁心
CN201110436359.X 2011-12-22

Publications (1)

Publication Number Publication Date
WO2013091315A1 true WO2013091315A1 (zh) 2013-06-27

Family

ID=45984915

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2012/072656 WO2013091315A1 (zh) 2011-12-22 2012-03-20 一种变压器用磁心

Country Status (3)

Country Link
US (1) US10008312B2 (zh)
CN (1) CN102436907B (zh)
WO (1) WO2013091315A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102437767B (zh) * 2011-12-22 2014-01-29 广州金升阳科技有限公司 一种自激推挽式变换器
CN103384121A (zh) * 2013-07-03 2013-11-06 苏州倍加福防爆电气有限公司 一种磁饱和本安电源
PL226676B1 (pl) 2015-06-29 2017-08-31 Akademia Górniczo Hutnicza Im Stanisława Staszica W Krakowie Przetwornica izolacyjna

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5117214A (en) * 1991-01-28 1992-05-26 Powercube Corporation Integrated magnetic power converter core
CN1622234A (zh) * 2003-11-28 2005-06-01 三垦电气株式会社 电抗器装置
CN1788410A (zh) * 2004-03-26 2006-06-14 三垦电气株式会社 开关电源装置
CN102074331A (zh) * 2010-12-13 2011-05-25 山东大学 一种磁饱和电抗器
CN201936722U (zh) * 2010-11-11 2011-08-17 河南逐鹿电器有限公司 一种磁阀式可控电抗器铁芯

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US2149634A (en) * 1936-09-10 1939-03-07 Jr Edmund O Schweitzer Transformer fault indicating means
US3143686A (en) * 1962-08-14 1964-08-04 Rca Corp Flyback transformer and transistorized deflection circuit
JPS5574111A (en) * 1978-11-29 1980-06-04 Hitachi Ltd Transformer
EP0303994B1 (en) * 1987-08-21 1993-05-05 Nippon Telegraph And Telephone Corporation Push-pull current-fed dc-dc converter
US5506764A (en) * 1994-01-31 1996-04-09 Astec International, Ltd. Electrical power converter with step-gapped transformer
JP2004111528A (ja) * 2002-09-17 2004-04-08 Matsushita Electric Ind Co Ltd マグネトロン駆動用昇圧トランス
DE202005017998U1 (de) * 2004-11-16 2006-07-20 JUNG FONG ELECTRONICS CO., LTD., Shen Ken Hsiang Elektrisches Bauelement mit dem Effekt eines variablen Luftspalts
JP4751266B2 (ja) * 2006-02-09 2011-08-17 株式会社タムラ製作所 リアクトル部品
JP5521792B2 (ja) * 2010-06-03 2014-06-18 トヨタ自動車株式会社 リアクトル
US8791782B2 (en) * 2011-01-28 2014-07-29 Uses, Inc. AC power conditioning circuit
CN102226976A (zh) * 2011-04-07 2011-10-26 常熟开关制造有限公司(原常熟开关厂） 一种电流互感器

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5117214A (en) * 1991-01-28 1992-05-26 Powercube Corporation Integrated magnetic power converter core
CN1622234A (zh) * 2003-11-28 2005-06-01 三垦电气株式会社 电抗器装置
CN1788410A (zh) * 2004-03-26 2006-06-14 三垦电气株式会社 开关电源装置
CN201936722U (zh) * 2010-11-11 2011-08-17 河南逐鹿电器有限公司 一种磁阀式可控电抗器铁芯
CN102074331A (zh) * 2010-12-13 2011-05-25 山东大学 一种磁饱和电抗器

Also Published As

Publication number Publication date
CN102436907B (zh) 2014-01-01
US10008312B2 (en) 2018-06-26
US20140354384A1 (en) 2014-12-04
CN102436907A (zh) 2012-05-02

Similar Documents

Publication Publication Date Title
WO2013091323A1 (zh) 一种自激推挽式变换器
CN202167993U (zh) 具有无损缓冲电路的移相全桥开关电源变换器
Liu et al. A phase-shift soft-switching control strategy for dual active wireless power transfer system
WO2013091315A1 (zh) 一种变压器用磁心
CN202523539U (zh) 一种变流变压器
CN214674878U (zh) 一种电源适配器
CN210429528U (zh) 一种基于二次侧串联型分数阶电路的无铁芯变压器
CN103997217A (zh) 一种基于平板型电感的定频串联谐振变换器
CN101020272A (zh) 环形高效直流弧焊机
CN210429529U (zh) 一种基于二次侧并联型分数阶电路的无铁芯变压器
CN202282744U (zh) 一种大功率高频高压开关电源
CN203813669U (zh) 一种单相和三相交流斩控调压补偿式交流稳压器
CN105991044B (zh) 一种非接触供电副边整流电路及方法
CN201167013Y (zh) 数码发电机逆变器电感
CN207638564U (zh) 一种应用新型变压器的推挽电路
Ke et al. Optimal Design of Electromagnetic Induction High-Efficiency Wireless Power Transmission System
CN206471179U (zh) 多段气隙式磁性组件
CN203225660U (zh) 一种用于太阳能发电系统的直流变换器
Yu et al. Design Optimization of High Frequency Transformer for Resonant DC/DC Converter
CN220962989U (zh) 磁芯结构、变压器、微型逆变器和光伏系统
CN203444962U (zh) 一种平面变压器
CN201450448U (zh) 一种新型开关电源
Wu et al. Design and research on tapped inductor of large step-up ratio DC/DC converter
CN115360922A (zh) 一种推挽变换器的磁集成设计方法
CN101742789A (zh) 一种恒功率电子镇流器

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12859015

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 14356886

Country of ref document: US

122 Ep: pct application non-entry in european phase

Ref document number: 12859015

Country of ref document: EP

Kind code of ref document: A1