WO1999014844A1 - Process for optimising the performance of switching power packs and switching power pack for carrying out this process - Google Patents

Abstract

In a process for optimising the performance of switching power packs, the input voltage Ue or input current Ie is periodically switched on and off with a switching frequency f, and an output voltage Ua or output current Ia are obtained by means of a transformer or choke, at least one parameter of the input voltage Ue or input current Ie being variable. Also disclosed is a switching power pack for carrying out this process. In order to increase the power gain of switching power packs fitted with transformers and to avoid oversized transformers or chokes, the actual magnetic field is measured in the transformer or choke and at least one parameter of the input voltage Ue or current Ie is varied depending on the actual magnetic field to regulate the magnetic field and avoid saturation of the transformer or choke. This active core magnetisation control also allows dispensing with the air gap in the core of the transformer without any risk of saturation.

Description

Method for optimizing the performance of switching power supplies and switching power supply for carrying out this method

The invention relates to a method for optimizing the performance of switched-mode power supplies according to the preamble of patent claim 1, and to a switched-mode power supply for carrying out this method according to the preamble of patent claim 7.

Power supplies are necessary for the supply of electronic devices which derive one or more DC or AC voltages of the appropriate size from the AC network. In conventional power supplies, the voltage translation and the mostly required galvanic isolation from the network are taken over by a transformer. However, such network transformers can only be reduced in volume and weight to a limited extent, which is why such power supplies do not meet today's miniaturization requirements. In addition, in the case of voltage or current regulation, the relatively high power loss requires the use of large heat sinks, which further increases the total volume and weight of the power supplies.

This has been remedied by so-called switched-mode power supplies, in which the mains voltage is rectified and screened and then "chopped" with a relatively high frequency. By increasing the operating frequency, the disadvantages of conventional power supplies with 50 Hz transformers can be greatly reduced, since the higher frequencies Voltages can be translated with much higher efficiency, which results in switching power supplies that have a significantly lower volume and weight than conventional power supplies.

When dimensioning the transformers of switched-mode power supplies, the working area is usually chosen to be so small for safety reasons that it is highly unlikely that saturation will occur. In the area of core saturation, the current in the primary winding of the transformer would in fact have impermissibly high values. Furthermore, a certain residual magnetism (remanence) can build up when the control is not completely symmetrical, which is equivalent to a zero point shift. Because of this shift, the distance decreases on one side of the magnetization characteristic for core saturation, which is why the safety margin is chosen even larger. Due to the core saturation on the one hand and the remanence on the other hand, only a part of the possible work area is used for safety reasons. The transformer is therefore not optimally used for the desired output power, but rather oversized, or the power yield achieved in a switched-mode power supply with a transformer of a certain size will be very low.

Various circuit arrangements attempt to prevent the zero point shift and, consequently, saturation of the transformer. For example, the circuit according to US Pat. No. 4,553,198 A aims to control the transformer as symmetrically as possible by detecting the primary-side current. No. 4,939,633 A also shows the regulation of a symmetrical current load on the primary winding of the transformer, the primary-side current of the transformer being detected with the aid of a magnetic sensor in its own magnetic core. With the self-oscillating circuits according to US 4,395,751 A and US 4,519,023 A, in which the transistors are controlled via their own load current via an auxiliary winding, good control symmetry of the transformer can be achieved. In these known circuits, however, component tolerances and offset errors also lead to zero point shifts in the magnetic field of the transformer. By controlling the control symmetry, only one cause of the undesired effect of an asymmetrical permanent magnetization is combated. Other influences on the nuclear magnetization are not taken into account. The problem of asymmetrical control due to component tolerances is only shifted from the power level to the control electronics. Undefined states of the electronics in the switch-on and switch-on phase due to fluctuations in the supply voltage of the control and the power section can also result in asymmetrical controls. Loads on the secondary side, for example when connecting or terminating, can also cause short-term asymmetrical currents and thus nuclear magnetizations. It is therefore necessary to overdimension the core in order to achieve the necessary safety distances for core saturation. Dispensing with the air gap in the core would only be conceivable with an increased saturation risk with correspondingly large safety distances. The object of the invention is to provide measures by means of which the power yield of switching power supplies with transformers can be increased. The usual over-dimensioning of the transformers or chokes in switching power supplies is to be avoided in this way.

The object of the invention is achieved in that the current magnetic field in the transformer or in the choke is measured, and in that at least one parameter of the input voltage U _e or of the input current I is used to regulate the magnetic field for the purpose of preventing saturation of the transformer or the choke _{e is} changed depending on the current magnetic field. By detecting the magnetic field in the transformer or in the choke, appropriate control of the parameters of the input variables can prevent the transformer from saturating and subsequently being overloaded. This makes it possible to significantly increase the power yield compared to conventional switching power supplies of the same size, since the entire working range is used in accordance with the magnetization characteristic of the transformer or the choke. The active regulation of nuclear magnetization eliminates the risk of saturation. At the same time, the air gap in the core of the transformer or the choke, which is necessary in particular in the case of forward converters and which serves to counteract possible residual magnetization, can be eliminated. This reduces the leakage losses and the magnetic resistance, which means that increased performance can be achieved.

The method according to the invention can be improved in that the primary current I _{p of} the transformer or the inductor is measured and that the primary current I _{p of} the transformer or the inductor is used to regulate the magnetic field of the transformer or the inductor. Since the relationship between the primary current I _p and the magnetic field changes due to the non-linear relationship outside of specified conditions, the core saturation can be recognized more accurately, faster and more reliably by recording both parameters.

According to a variant of the method according to the invention, the switching times ti, t _{2 of} the input voltage U _e or the input current l _e are dependent on the current one Magnetic field in the transformer or the choke changed. The control of the switching times is a simple and quick way.

Alternatively or in combination with the change in the switching times, the switching frequency f of the input voltage U _e or the input current L can also be changed as a function of the current magnetic field.

In addition, as an alternative or in addition to the abovementioned variation possibilities, the amplitude of the input voltage U _e or the input current L can also be changed as a function of the current magnetic field in the transformer or the inductor and thus also the output voltage U _a or the output current I _a independently of the respective load conditions at the output of the switching power supply and without the risk of core saturation.

According to a further feature of the invention it is provided that when a possible remanence (residual magnetism) of the transformer or the choke is measured, a corresponding counter-magnetization in the transformer or the choke to eliminate the remanence is caused by a corresponding change in the parameters of the input voltage Ue or the input current le. This is particularly necessary in the case of switched-mode power supplies which operate on the principle of the single-ended converter in order to prevent residual magnetism (zero point shift) and consequently saturation. In the case of a switched-mode power supply, which works on the principle of a push-pull converter, the residual magnetism can be reduced simply by changing the duty cycle of the switching times tj, t _{2 of} the input voltage U _e or the input current L. Thus, the switch-on time of that voltage or that current which causes magnetization in the opposite direction of remanence is correspondingly increased compared to the switch-on time of the voltage or current in opposite polarity, in order to produce a corresponding counter-magnetization.

Another object of the invention is to provide a switching power supply for carrying out the above-mentioned method for performance optimization. This object is achieved in that a sensor, for example a Hall sensor or a magnetoresistive sensor for measuring the current magnetic field, is provided in the transformer or in the choke, and in that the sensor is connected to a control unit, the output of which is connected to the switching stage . It follows that with a certain nominal output power of the power supply, this can be made much smaller, since the efficiency is higher and the losses in the transformer or in the choke are lower, so that the transformer or the choke does not have to be oversized as usual, but rather can be made much smaller.

If, in addition, the control unit is connected to the primary side of the transformer or the choke, so that the primary current I _{p of} the transformer or the choke can be used as an input variable for the control, the control can be carried out faster and more accurately, since the core saturation can be detected more quickly and reliably can.

If the control unit is formed by a microcontroller, there are all the associated advantages, such as miniaturization, flexibility in programming or rapid control. Instead of a microcontroller, however, there can also be an analog circuit.

The invention is explained in more detail with reference to the attached figures and compared with known methods and devices.

Show in it

Fig. La-lb the working area of a conventional switching power supply transformer in

Comparison of the working range of the transformer of a switching power supply according to the inventive method, Fig. 2 shows a switching power supply according to the principle of a bridge push-pull converter in a schematic representation, Fig. 3a-3c, the control voltages and currents of the switching transistors and the

Primary voltage of the transformer from the circuit of FIG. 2 in time

Course, 4 shows the block diagram of a switched-mode power supply according to the invention with active regulation of the nuclear magnetization, and FIGS. 5a-5d the temporal profiles of the primary voltage of the transformer to illustrate some possibilities of the regulation according to the invention.

Figures la and lb schematically show the magnetization characteristic of a transformer. The magnetic induction B as a function of the magnetic field strength H or the magnetic flux Φ as a function of the flooding Θ is shown. The area of the core saturation is characterized in that the magnetic flux Φ or the induction B cannot be significantly increased despite the increase in the current in the primary winding of the transformer or the flux Θ. Almost all elementary magnets of the core material are aligned in the area of saturation. The inductive resistance of the winding decreases in the area of saturation, as a result of which only the undesirable ohmic component of the resistance limits the current in the winding and this reaches inadmissibly high values. In order not to reach the core saturation range, safety distances in the magnetization characteristic are maintained, whereby the maximum possible magnetic flux Φ is not used. Furthermore, a certain residual magnetism, which is referred to as remanence, can build up in the case of incompletely symmetrical control of switched-mode power supplies which operate on the principle of push-pull converters, and in the case of switched-mode power supplies which work according to the single-ended principle. This is equivalent to a zero point shift of the magnetic flux Φ. This asymmetry reduces the distance to the core saturation on the side of the magnetization curve on which the residual magnetism occurs, which is why the safety distance is chosen to be even larger. The usual work area is indicated in Fig. La by the solid line.

In Fig. Lb, on the other hand, the working range of a switching power supply according to the invention is shown, which is achieved in that the magnetic flux Φ or the magnetic field strength H is measured and the input variables of the switching power supply are regulated depending on the flux Fluss so that no core saturation and none Zero shift occurs. The zero point shift is prevented, for example, in such a way that the magnetic core is magnetized longer or more in the opposite direction to the shift. When the core saturation is detected, the magnetic flux Φ is not increased further, ie it is blocked off. So can the entire working area can be used without running the risk that the working area runs into the saturation of the magnetization characteristic. An overload occurring in spite of the active regulation cannot lead to any damage, since the core saturation does not occur suddenly and the transition to core saturation is therefore "round" and therefore the inductance does not drop abruptly. It is therefore even possible to slightly over the kink of the work area Magnetization line to expand beyond saturation.

Fig. 2 shows the circuit diagram of a switched-mode power supply based on the principle of a bridge push-pull converter, in which the input voltage U _e or the input current L is periodically switched on and off with the aid of four transistors T-T in a bridge circuit. The currents Ii and I ₂ flow in phase opposition over the primary winding of the transformer and cause a primary voltage Ui there, which is transformed into a desired output voltage U _a . To stabilize the output voltage U _a , it is known to regulate the duty cycle and the frequency of the switching operations of the transistors. A bridge push-pull converter has the advantage that a lower number of turns on the transformer is necessary, and the disadvantage that four transistors are required. In principle, the present invention can also be applied to other types of switching power supplies.

3a and 3b show the control voltages and control currents of the transistors as a function of time. The transistors Ti and T ₃ are simultaneously the

Period ti turned on, whereupon a current through the primary winding of the

Transformer flows. During the blocking phase of the transistors Ti and T ₃ , the

Transistors T ₂ and T _{4 turned on} simultaneously for a duration t ₂ , so that a current I ₂ in

The opposite direction to the current L flows through the primary winding of the transformer. This switching sequence is repeated with the period T corresponding to an operating frequency f.

The times tι = t ₂ are normally selected. The duration tj or t ₂ per period T is set in accordance with the power desired at the output. At the beginning of the diagrams, the time period is selected according to a 50% power yield at the output, and on

End of the diagrams for a 100% power yield. 3c shows the resulting voltage U [on the primary side of the transformer. 4 schematically shows a block diagram of a switching power supply according to the invention with active control of the nuclear magnetization. After any transformers and rectifiers (not shown), there is an input voltage U _e which is fed to a switching stage 1. In switching stage 1, which is usually constructed from switching transistors, the input voltage U _{e is} periodically switched on and off at a relatively high frequency f, that is to say "chopped up" to a certain extent. The resulting square-wave voltage U | is translated with the aid of a transformer 2. Because of the higher frequency f, a smaller transformer 2 is required compared to a 50 Hz transformer The transformer 2 also takes on the frequently required galvanic isolation between the primary and secondary side.If such galvanic isolation is not necessary, the squarewave voltage Uj can also be converted by a choke Subsequently, the secondary voltage U _{2 is} rectified and sieved in a subsequent stage 3, so that a stable output voltage U _a is present, according to the invention a sensor 4 is provided for measuring the magnetic field strength in the transformer 2. This sensor 4 can be designed, for example, as a Hall sensor be the s I use the Hall effect. A semiconductor wafer arranged in a magnetic field and through which a current flows delivers a voltage proportional to the magnetic flux density. Likewise, the sensor 4 can be designed as a magnetoresistive sensor which shows a resistance dependent on the magnetic field. The sensor 4 can, for example, be glued into a slot in the core material of the transformer 2, the adhesive advantageously being mixed with ferrite particles. The slot should be made as small as possible so that the homogeneity of the magnetic field is not significantly disturbed and the possible scatter losses are kept small. The signal originating from the sensor 4 is fed to a control unit 5. The control unit 5 prepares the sensor signal for further processing. The control unit 5 delivers a corresponding control signal with which the switching stage 1 is controlled. For example, the frequency of the oscillator in the switching stage 1, which generates the frequency f with which the input voltage U _{e is} periodically switched on and off, can be changed as a function of the control signal, ie as a function of the magnetic field in the core of the transformer 2. With the help of a pulse width modulator, the switch-on time of the input voltage U _e can also be changed depending on the control signal or the magnetic field. Usually, the output voltage U _a or the output current I _a or the secondary voltage U _{2 is also fed} into the control circuit. This However, feedback could also be omitted for constant load conditions and applications that are independent of network fluctuations. To improve the method, the primary current I _{p of} the transformer 2 or the choke can also be included in the control. This measure detects core saturation faster and more reliably than when only the magnetic field is detected.

For clarification, FIGS. 5a-5d schematically show the time profiles of the primary voltage Uj on the transformer in different control cases for a switched-mode power supply with a push-pull converter. 5a shows the course for symmetrical operation, ie the same residual magnetism in both directions. The transistors of the switching stage are switched on for a certain time t in accordance with the required output power. In this case tι = t. If residual magnetism builds up in one direction of the magnetization characteristic, this situation is immediately recognized by the sensor and appropriate control is carried out. This case is shown in FIG. 5b, where the switch-on time of those transistors which produce a magnetic field in the opposite direction of the remanence is increased by an amount Δt and the switch-on time of the other transistors is reduced by the amount Δt. This asymmetrical operation reduces the remanence and corrects the zero point shift. This prevents the work area from reaching core saturation. 5c shows the case in which higher power is required at the output of the switching power supply. Under constant control of the current magnetic field, the switch-on times ti and t ₂ in the switching stage are increased to a value t 'in order to achieve the required output power. The switched-mode power supply is dimensioned in such a way that, with maximum power requirement at the output, the working area in the magnetization just reaches the areas of the core saturation and thus the full working area is used. 5d finally shows the case of lower power requirements, the switch-on times ti and t ₂ in the switching stage being reduced to a corresponding value t ". In addition, the frequency f and the amplitude of the input voltage U _e or the input current L can be regulated (not shown).

The low volume and low weight of such power supplies result in extreme advantages in a wide variety of applications. Think of them Lighting technology, where the power supplies required for gas discharge or halogen lamps no longer have to be housed in an additional housing, as before, but can be built into the housing of the lamps themselves due to their small size. Likewise, by using the power supplies according to the invention, the volume and weight of electric welding devices can be reduced to such an extent that the power supplies can even be installed in the handle of the welding device or worn comfortably on the belt. As a result, more complicated welding work, such as on a scaffold or the like, can be carried out much more easily and conveniently.

The power supplies according to the invention can also bring extreme advantages in traffic engineering, since, for example, electric traction vehicles require much smaller and lighter converters with the same drive power. The application possibilities are almost unlimited.

Claims

Claims: 1. Method for optimizing the power of switching power supplies, in which the input voltage U _e or the input current L is periodically switched on and off with a switching frequency f and a with a transformer or a choke Output voltage U _a or an output current I _{a is} obtained, in which at least one parameter of the input voltage U _e or the input current l _{e can be} changed, characterized in that the current magnetic field is measured in the transformer or in the inductor, and that for regulation of the magnetic field for the purpose of preventing saturation of the transformer or the choke at least one Parameters of the input voltage U _e or the input current L is changed depending on the current magnetic field. 2. The method according to claim 1, characterized in that the primary current I _{p of} the transformer or the inductor is measured, and that the primary current I _p of Transformer or the choke is used to regulate the magnetic field of the transformer or the choke. 3. The method according to claim 1 or 2, characterized in that the switching times ti, t _{2 of} the input voltage U _e or the input current l _e depending on the current Magnetic field can be changed. 4. The method according to at least one of claims 1 to 3, characterized in that the switching frequency f of the input voltage U _e or the input current L is changed depending on the current magnetic field. 5. The method according to at least one of claims 1 to 4, characterized in that the amplitude of the input voltage U _e or the input current L is changed depending on the current magnetic field. 6. The method according to at least one of claims 1 to 5, characterized in that when measuring a possible remanence of the transformer or the choke by appropriate change in the parameters of the input voltage U _e or the input current l _e a corresponding counter-magnetization in the transformer or the choke to eliminate remanence. 7. Switching power supply for performing the method according to at least one of claims 1 to 6, with a switching stage (1) for periodically switching on and off an input voltage U _e or an input current l _e , and a transformer (2) or a choke for extraction an output voltage U _a or an output current I _a , the switching frequency (f) and / or the switching times (ti, t ₂ ) and / or the amplitude of the input voltage U _e or the input current L can be changed, characterized in that in the transformer (2) or in the choke a sensor (4), for example a Hall sensor or a magnetoresistive sensor for measuring the current magnetic field, and that the sensor (4) is connected to a control unit (5), the output of which is connected to the switching stage (1). 8. Switched-mode power supply according to Claim 7, characterized in that the control unit (5) is connected to the primary side of the transformer (2) or the choke so that the primary current I _{p of} the transformer (2) or the choke can be used as an input variable for the control. 9. Switching power supply according to claim 7 or 8, characterized in that the control unit (5) is formed by a microcontroller.

IE

Fig. 2 2.3

3.3