DE2949070A1 - CLOCKED POWER SUPPLY WITH SEVERAL OUTPUTS - Google Patents

Description

. € ΪΚ €

N.V. Philips'Gloeilampenfabrieken PHN 9302

Clocked power supply unit with several outputs

The invention relates to a pulsed power supply of the flow converter type having a primary circuit with two Input terminals that can be connected to a DC voltage supply source and secondary circuits with output terminals for connecting a load, the primary circuit between the input terminals at least one controllable circuit element, which is periodically controlled by a control circuit, in series with a primary winding of a transformer, while each secondary circuit contains a secondary winding of the transformer, a rectifier diode, a freewheeling diode and a choke coil contains, wherein at least one secondary circuit with setting and stabilizing means for a size on the Output terminals is provided.

Such a power supply unit is known from DE-OS 26 08 167. There is a flow converter with two galvanic separate outputs described, of which one output by means of a controlled clocked series transistor in the Output circuit an adjustable and stabilized output voltage independently of the other parameters of the converter supplies. For this purpose, a special electronic control circuit is provided for the clocked series transistor. Part of the switching period does not need to build up for setting and stabilizing the output voltage and maintaining this voltage is used, whereby a pulse width control takes place per output circuit.

A power supply unit is also described in DE-OS 26 34 193, which corresponds to the type mentioned at the beginning. The pulse width regulation per secondary output circuit is here from

030026/0669

PHN 9302

an electronic circuit that controls a thyristor, which also acts as a rectifier in the secondary circuit serves.

Fig. 1 of this patent also indicates that not all secondary output circuits need to be provided with the setting and stabilizing means and that the primary circuit can hold two circuit elements arranged in series with the primary winding and bridged by a freewheeling diode. Both circuit elements are periodically brought into the conductive state by a pulse generator.

Both known power supply units should be provided with control units for each regulated secondary output circuit. These control units are rather intricate in structure and contain many electronic components and mostly control transformers in order to be able to properly control the switching transistors or the switching thyristors.

The invention has the object of providing a great simplification of the switching means with the associated control unit , and is characterized in that the setting and stabilizing means contain a saturable choke coil, which is arranged with a power winding in series with the secondary winding and the rectifier diode and further at least contains a control winding for setting a bias in the core of the choke coil, and that the means further contain a control unit which is connected to the control winding and is provided with an input for the variable to be set and stabilized.

The pulse width regulation per secondary output circuit is obtained here by the fact that the leading edge of each square-wave voltage pulse is applied to the secondary winding of the transformer

030 0 26/0669

p PHN 9302

fonnator is delayed due to the voltage-time Integral representing the difference between the saturation flux and the bias flux in the core of the saturable Choke coil is proportional. By regulating the bias is in this way at a certain secondary Transformer voltage an adjustable time delay can be achieved. The premagnetization can be with a a simple control unit, whereby a parameter to be regulated, such as the output voltage or the output current, is converted into a measured variable which can be compared with a reference variable, after which the difference signal obtained the premagnetization by means of a current through the control winding or by impressing a voltage which ensures demagnetization up to a desired induction value.

The advantages obtained with the control circuit according to the invention are:

- Low cost price of the used items;

expensive switching transistors or thyristors and a complicated one Control circuit are avoided.

- Low power dissipation; the power winding of the choke coil has a very small resistance, while the losses in the control winding, the magnetic circuit and the Control unit are very small, in contrast to the losses in the switching transistors or thyristors and in the control circuit of the known power supply units.

- A simple linear control circuit can be applied.

An improvement in efficiency can still be obtained if, according to the invention, the core of the saturable Choke coil is made of magnetic material with low remanent magnetism.

030 0 26/0 6 69

/ * PHN 9302

An additional control current, possibly via a special one Control winding to pre-magnetize the core of the choke coil in such a way that a steep area of the B-H curve is coated is not necessary. This shift of the B-H curve is necessary e.g. with magnetic Material with a rectangular loop in which the remanent magnetism is close to the saturation value of the flux is equivalent to.

An embodiment of a power supply unit according to the invention is characterized in that the control unit contains a diode and a control voltage source, which are arranged in series with the control winding, such that during the first period of time the diode is blocked by the control voltage of the source and during part of the second duration is conductive due to the induction voltage across the control winding, so that the bias is determined by the voltage-time integral of the control voltage over the part of the second duration, the control circuit makes the controllable circuit element conductive during a first period and this element during a the second period of time.

It is advantageous that the magnetic energy stored in the choke coil is optimally used. The voltage-time integral in the first time period is equal to the voltage-time integral in the second time period. There is no need for a continuous preset current to flow in the control windings and the control voltage source can easily be obtained, for example according to another embodiment of the invention, which is characterized by Is that the control voltage source is the output of an operational amplifier, this amplifier amplifies the difference signal between a reference signal and a signal proportional to said variable.

030026/0669 COPY

'PHN 9302

In another embodiment of the invention is the diode the base-emitter diode of a transistor. It is advantageous that the demagnetization current is greatest Part bypasses the control voltage source, which is especially important when the output of an operational amplifier from the outside cannot draw any or only a little electricity. Usually such an output supplies current. This embodiment is characterized in that the Diode is the base-emitter diode of a transistor with its collector on one end and its emitter on the other End of the control winding is connected.

The invention can also be used very advantageously in a clocked power supply in which the control circuit for the circuit elements in the primary circuit one in the Includes pulse width controlled oscillator and a comparison circuit that has an input to the output terminals one of the secondary circuits is connected so that the voltage is applied to these terminals by applying the pulse width control is constant.

So there is already a stabilization of the different output voltages. A mutual influence through load changes can be avoided and a fine adjustment, each output voltage is thereby possible that the input of the control unit, which belongs to a different secondary circuit, to the associated output terminals connected. This increases the output voltage, independent of load changes, kept constant over a wide range and fine adjustment of this output voltage causes.

For the saturable choke coil, small dimensions can be chosen here, while a simple amplifier circuit can be used to regulate the output voltage. The control circuit consumes practically no energy,

030026/0669 COPY

& PHN 9302

namely less than, for example, 3 % of the nominal output performance; the setting range of each output voltage is, for example, + and - 5 % _t while the stabilization is, for example, much better than 1 % . It should be noted that the use of a saturable inductor for control purposes is well known. US Pat. No. 3,087,107 describes a DC voltage supply source which is fed from an AC voltage network via a transformer. A saturable choke coil is arranged between the secondary winding and the rectifier diode, which with the help of two control windings receives a premagnetization that is determined on the one hand by the load current and on the other hand by a current that comes from an amplifier circuit that compares the output voltage with a reference value. The existing rectifier circuit is a half-wave rectifier circuit. The full travel design is described in U.S. Patent 3,105,184. Both known supply circuits work with a sinusoidal input voltage and peak rectification, so that the regulation only becomes effective after the peak value of the sinusoidal voltage has been passed. Since it is also indicated that the sinusoidal voltage comes from a supply network with a frequency of 50 or 60 Hz, it follows from this and from the above that a long exposure time is required and thus a large choke coil must be used. The losses in the circuit will also be relatively large, which is due, inter alia, to the use of a magnetic material with a rectangular loop, which requires an additional setting current.

The invention relates to forward converters with multiple outputs, the switching frequency of which is known to be high, wherein a square wave voltage is emitted by the secondary windings. The regulation can here for each output circuit from the

Be effective from the leading edge.

030026/0669

PHN 9302

US Pat. No. 3,445,775 also describes the use of saturable inductors for pulse width control for a control circuit for a push-pull converter. The square wave voltage an oscillator is in the pulse width of a magnetic amplifier using core material Controlled rectangular loop, which has an adjustment control winding for giving a fixed bias and a Includes diode control voltage source circuit for obtaining a voltage-time integral. The present invention however, uses one choke coil for each secondary output of a forward converter which is to be considered several secondary outputs, namely for the power control of this output and not for receiving an in the pulse width of the controlled oscillator.

Some embodiments of the invention are shown in the drawing and are described in more detail below. It demonstrate

1 shows a general block diagram with measures according to the invention,

2 shows a B-H curve with low remanent magnetism, 3 shows a B-H curve with a rectangular loop; FIGS. 4A and 4B show a series of time diagrams for the course of currents, voltages and flux, FIGS. 5A and 5B show a detail of a current-time diagram, 6 shows a comparison circuit for the output voltage, 7 shows a comparison circuit for the output current, 8, 9 and 10 modifications of the control for the control winding, and

11 is a circuit diagram of a regulated secondary circuit according to the invention.

Fig. 1 shows a circuit diagram of a forward converter, the has four galvanically isolated outputs. A transformer 1 with secondary windings 2, 3, 4 and 5 is used for this purpose Mistake. These windings can of course

030026/0669

βί / fO ^PHN 9302

also overlap or even be combined with the primary winding 6. A degaussing winding 7 is the in the transformer stored magnetic energy via the diode 8 back to the DC voltage supply source, the is connected to terminals 9 and 10. A control circuit 11 supplies control pulses to a controllable one Circuit element 12, which may for example be a transistor, and which connects winding 6 to the for a first period of time DC voltage source connects, and disconnects therefrom for a second period of time.

Each secondary circuit contains a rectifier diode 13 »which because of the character of the forward converter during the first Time is conductive, a choke coil 14, which can store magnetic energy, some of which is transferred to the buffer con-

is capacitor 15 and to the load connected to the output 16, 17, 18 or 19 are emitted, and further a freewheeling diode 20, which the current through the coil 14 during the time during which the rectifier diode 13 can not conduct this current.

The secondary circuits 2-16 and 3-17 are provided with adjustment and stabilization means according to the invention. It is stated that there may also be secondary circuits for which a stabilization of the output variables is not necessary is like circuit 4-18, and that a secondary circuit can be controlled by influencing the primary circuit. In addition becomes the output voltage of circuit 5-19 of a comparison circuit 21, which emits a control signal, which is sent via the connection 22 to a control input 23 of the control circuit 11 is supplied so that, for example, the pulse width of the first time period is changed in a compensatory manner. 1 that stabilization can also be obtained by an additional secondary winding 24 which is connected to a comparison circuit 25 which is connected to the line 22.

030026/0669 Copy ^

29A9Q70.V

PHN 9302

For the output circuits 2-16 and 3-17, a saturable choke coil 26 is arranged between the secondary transformer winding 2 or 3 and the diode 13, of which a power winding 27 carries the circulating current and a control winding 28 causes the premagnetization of the core material. This control winding is connected to a control unit 29 which has an input 30 via which data of a variable to be set and stabilized are fed to the output terminals. This size can be the

Ό input voltage, output current or output power . This is indicated symbolically in FIG. 1 with the connection 31.

In FIG. 2 a BH curve of the magnetic Kernmäterial of the saturable inductor 26 of FIG 1 shown _this. Material has a low remanent magnetism, which means that the induction value B _r at a magnetic field strength of zero is small compared to the Sittigungsinduktionswert B _e that occurs at a field strength H.

The control winding 28 can set premagnetization values B ₁ with a field strength H ₁ or -B ₂ with a field strength -H ₂ . The values H ₁ and H ₂ can be obtained with the aid of a controllable current source circuit in the control unit 29 of FIG. 1 or by a current or ip which flows in the control winding at the point in time at which the secondary winding 2, 3 receives voltage delivery begins, thus at the beginning of the first period of time. It is also possible to route the currents I ₁ and i _{2 of} constant value through a special control winding and to use the control winding 28 to demagnetize the value B ₁ to the value B 1 or B ₂ . The forward magnetization, which is brought about by the secondary voltage, which is completely over the winding 27 during the first part of the first period, then runs in the opposite direction, for example from B ₉ to B or from B ₁ to B ". Reason-

£ _ S Io

In addition, it can also run from B ₁ ^ to B ". After this

030026/0669

copy

y6 PHN 9302

Ah

Induction law, the voltage across a coil is equal to the change in flux in the coil per unit of time, or the voltage-time integral is equal to the total change in flux. I _n Fig. 2, an arrow 40, the change in induction B ₁ -B _3, an arrow 41 B ₃ -B _1, and an arrow B 42 _s + B ₂ and is known that the flux change these values through the factor n * A proportional , where η is the number of turns of a winding and A is the surface area in the cross section of the magnetic core material.

If the voltage in the voltage-time integral associated with the aforementioned flux changes is the secondary square-wave voltage during the forward magnetization, a response time proportional to the changes can be achieved. It can now also be seen that the demagnetization can take place during the second period in that the field is rapidly reduced within this period, so that a set field belonging to H ₁ or H ₂ remains, or by the fact that during the second period a voltage of a source is applied to a (control) winding of the coil, whereby the field is reduced during part of the second period, but to a desired induction _{value, for example B 1} or a value close to B _r . A preset value for the magnetization is not required, which means that a winding and constant control current are superfluous.

It is clear, however, that only the area between B_ and B_ can be covered with it. For a larger area, the BH curve must be shifted _{with an auxiliary field such as H 2.}

In Fig. 3, a B-H curve for magnetic material having a rectangular loop is shown. It can be seen from this that the value B for the remanent magnetism is almost the same the saturation value B "is. In order to still have a useful one

030026/0669

PHN 9302

To get a change in flow, an auxiliary field, such as H ₁ or Hp, will always be necessary here.

In Fig. 4a and Fig. 4B some timing diagrams are shown, of which the diagram V___ represents the secondary transformer voltage. At time t ₁ in FIG. 4A, the circuit element 12 of FIG. 1 becomes conductive, as a result of which the square-wave voltage 43 arises on the secondary windings. At time t ^ the first period ends and the second period begins. For the demagnetization of the transformer 1, the diode 8 becomes conductive and the supply voltage is across the winding 7. The voltage 46 is thus present across the secondary windings. Due to the saturable choke coil, which represents a high impedance at the beginning of the first period, the voltage V___ is available for the rectifier circuit at time t ~ or t. This is indicated in diagram V__ with lines 45 and 44, respectively.

The diagram I___ shows the current through coil 14, lake

At constant loading, line 53 is obtained in the event that the full secondary voltage would be available. In the event that the saturable coil brings _{about a response time of tp-t 1} , the streamline 55 is obtained, while the streamline 54 is obtained _{for the response time t ^ -t 1.}

The diagram I represents the current through the power winding 27 of the saturable choke coil 26, the current _{values between the times t 2} -t _Zf and t ^ -t ^ being equal to the values shown in the diagram I _OQ _. At time t /. the current I_ “becomes zero and flows in the

sp

Control winding 28 a demagnetization current.

At times t _g , t _1Q , t ₁₁ and t ₁₂ indicated in FIG. 4B, the same events take place, however

030026/0669

PHN 9302

with the difference that, in the event that the premagnetization is set by a current source control from the control unit 29, the demagnetization can only take place by means of a current in the power winding 27. This is indicated at times t ₁₂ , t ₁₃ and t ^. At the time Χ _Λ0 , the voltage V _PQ reverses its direction. The freewheeling diode 20 is conductive and carries the currents 47 or 48, as shown in the diagram I ₀₀ ^. The coil 26 now generates an induction voltage across the winding 27 »which is almost equal to the negative voltage V___ (line 46), whereby the rectifier diode 13 can be conductive for a demagnetizing current that is 49 and 50 im

Diagram I is indicated.
sp

ifi In diagram B of both figures is the course of the induction or the flux in the saturable reactor.

During the time periods t ^ -t ^ or ^ ₁ ^ t ₁₂ , the coil is saturated (value B ₁ ,). The forward magnetization by the voltage V takes place during the period Ϊ £ ~ ^ 1 ^or ^ 1Ο ~ * 9 from the pre-magnetization value B ^ and during t, -t ^ or t ₁₁ -tq from the value B, instead.

The demagnetization by means of a control voltage source takes place during the period t -ti, the values B ₁ or B 1 being reached again as the final value at time t. The point in time t is the point in time at which the demagnetization of the core of the transformer 1 has ended. The voltage 46 becomes zero, whereby the blocking of the diode 13 is removed and the voltage across the winding also becomes zero. As a result, the control voltage source can no longer be effective. The field present in the core of the coil 26 at this point in time is now maintained by a current in the winding 27, as _is indicated _{in diagram I with the current values i b} and i d. For the other demagnetization process, the values B-.

03002 6/0669

COPY

PHN 9302

and B, in Fig. 4B at times t .., and t,., reached, these values being recorded externally in the control unit 29 during the remaining part of the period of time Funds are maintained.

In FIGS. 5A and 5B, the transient current range between times t ^ and t, or times tg and t ₁ ^ from diagram I in FIGS. 4A and 4B is shown in detail. The straight rising streamlines can be derived from the law of induction according to the formula i = V _._ / Lt

OCv

clarify, where L is the self-induction of the coil, referred to the winding 27, and t is the time course. aside from that it can be seen from this that the current i corresponds to the voltage

V is just proportional.
lake

At time t ^ or tg the secondary voltage is corresponding the rectangle 43 in FIG. 4 is present. This voltage is across the power winding 27 of the coil 26 and must be equal to a certain change in flux per unit of time, which over the B-H curve of the magnetic material in a Field strength change per unit of time can be converted, which again is proportional to a change in current per unit of time is. The diagrams of FIGS. 5A and 5B can thus be explained with the aid of FIG. 2 or 3. FIG.

Fig. 5A shows the course of the current I for the times t, t _A , t ~ and t 1. At time t_, the current I through the winding 27 assumes the value i ^ or i, via the field strength and BH curve belonging _{to the inductions B 1 and B.} At time t ₁ , the current increases by a value i ^, which corresponds to the flat part of the BH curve in FIGS. 2 and 3, which is indicated by the points χ and y. The current then runs according to line 59 or 58 or according to lines 60 or 61 when the input voltage, and thus also the secondary voltage, has increased.

Ü 3 O O? B / O 8 6 9 COPY

PHN 9302

As is known, the relationship between field strength and current can be written in simplified form as:

H.l = n «i,

where 1 is the length of the magnetic circuit and η is the number of turns of the winding through which the current i flows. With the help of this relationship it can be seen from FIG. 2, for example, that the field strength H ₁ , which belongs to the current i _b , is already present as a premagnetization, and that then the BH curve up to the field strength H_ belonging to the saturation induction B_ is run through. From this point in the diagram on, the induction changes slightly, but the field strength and thus the current through the winding 27 increases considerably. In FIG. 5A, this is indicated for the associated current by the arrows 56 and 57. It should be noted that the coil 26 in the secondary circuit represents a high impedance in the region of the time-varying field strength between H ₁ and H _s , but a low impedance with regard to the other electrical circuit parameters at a field strength greater than Hp. For the sake of clarity, however, in diagram I of FIGS. 4A and 4B, the transient current range is shown enlarged in relation to the current on lines 54 and 55. If voltage regulation is desired for the secondary circuit and the associated control unit is set up accordingly, the regulation to constant output voltage with increasing input voltage can be explained with reference to FIGS. 4A and 5A. At a higher secondary voltage, the line and the line 60 change and the switching point is reached at time t. The control unit measures an increasing output voltage and thereby increases the voltage of the control voltage source. This results in a greater demagnetization in the period t _z -t ^ in Fig. 4A, diagram B, B ₁ changes to B- ,, while in Fig. 5A the associated

Current i, to i ·, changes, so that in this figure the switching point t “is reached via the line. Since t _y -t _{1 is} greater than t_-t ₁

030026/0669

29A9Ü70

PHN 9302

is, of the remaining part of the switching period in which V___ is effective, a smaller part remains, which thus forms the compensation for the higher voltage.

In Fig. 5B the Einschwingstrombereich is shown, which corresponds to the diagram of FIG I_. AB. An external

sp

The impressed field H ₁ or H ,, which _{corresponds to the induction values B 1} or B , is present on both ts at time tg. From this point in time on, the current I runs according to the line 51 or, in the case of a higher secondary voltage, according to the line 52, which is comparable to the line 59 or 60 or the line 58 or 61 in FIG. 5A. At times tt ₁₀ , t and t ₁₁ , the core of the coil is saturated . As far as the currents are concerned, it can be said that the current i _{a is} equal to 1 / n · (H ₅ -H ₁ ) and the current i _{c is} equal to 1 / n · (H -H ₅ ), where H, is the field strength at of induction is.

With voltage regulation in the manner already described with reference to FIG. 5A, the control unit will reduce the premagnetization from B ₁ to B, or from H ₁ to H, so that the current i at the time t - is converted into the current i.

cL ■ "· ^ **

with the switching point at time t.

In Fig. 6 a part of the control unit 29 is shown. The output voltage of a secondary circuit is selected as the variable to be controlled. For this purpose, the connection 31 is attached between the input 30 and the output terminal 32 and the common connection of the circuit is connected to the terminal 33.

An operational amplifier 62 is connected to a voltage divider 64, 65 via a non-inverting input 63. The inverting input 66 is connected to a Zener diode 67 , which is set with a resistor 68. The amplifier and the zener diode can be

are fed, e.g. by rectifying V___. above

lake

the winding 2 is obtained. But both can also

03002B / 0669

yS PHN 9302

the manner shown in Fig. 6 are fed. At the output 69-70 of the amplifier 62 a voltage is obtained, the difference between the Zener reference voltage and the output voltage divided by the divider 64, 65 is proportional to terminals 32 and 33. The output voltage of the amplifier is used to control the control winding 28 via an intermediate circuit. The latter is in the 7, 8, 9 and 10 are described in more detail.

A comparison circuit is shown in FIG. 7 , the current through the load being selected as the variable to be controlled at the output terminals 32, 33. For this purpose, this current is passed through a measuring resistor 71 via connection 31 and input 30. The voltage across this resistor is compared with a Zener voltage of the Zener diode 72. The voltages are fed to input 63 and input 66, respectively, by means of the same dividers 73, 74. The output 69-70 supplies the amplified differential voltage which is used, for example, to set a controllable current source circuit which consists of a transistor 75 and an emitter resistor 76. The collector current obtained is passed through the control winding 28 in order to set a bias.

The circuits of Figs. 8, 9 and 10 indicate how the output voltage of the amplifier 62 itself can be used to obtain the demagnetization. The control winding 28 is arranged in series with a diode 77, which can also be the base-emitter diode of a transistor 78. the The series circuit is connected to the output of the amplifier in such a way that the current 79, which is used for demagnetization ensures, flows through the diode 77 or the collector-emitter circuit, the induction voltage across the winding being the same the control voltage across terminals 69, 70 plus the

Diode voltage drop is. From the stress-time integral

030026/0669

PHN 9302 follows:

where V _{t is} the control voltage across the coil and η is the number of turns of the control winding. In Fig. The induction B is shown after the previous equality in the course of time between t- and t_. Both the field strength H and the associated current i have almost the same profile, it being assumed that the relative permeability of the material is almost constant. By a suitable choice of the turns ratio between the power winding and the control winding, magnetization and demagnetization values can be matched to one another.

8, 9 and 10, the output of the operational amplifier with terminals 69 and 70 is shown floating. It is readily possible for terminal 70 to be connected to the common connection of terminal 33 or for terminal 69 to be connected to terminal 32 with the feed line. Another possible circuit according to FIG. 9 is shown in FIG.

In Fig. 11, a secondary output circuit is shown, which is also shown in Fig. 1, the control unit 29 of Fig. 6 and Fig. 8 or 9 being used. The terminal of the amplifier 62 is connected to the common connection to the terminal 33 and the terminal 69 supplies a control voltage to an emitter follower 80 with the emitter resistor 81. The diode 77 can be omitted if the base-emitter reverse voltage of the transistor 80 is sufficiently high . The control voltage for winding 28 is determined here by V_ "-V / -Q, where V _e " _ represents the output voltage across terminals 32 and 33 and VgQ represents the output voltage of amplifier 62. Since this tension is in the

If subtraction mentioned has a negative sign, the functions at input terminals 63 and 66 are related to

030026/0669

PHN 9302

Fig. 6 reversed. Again, the stabilization is the Voltage clearly visible at terminals 32 and 33. For example, if this voltage tends to increase, because the supply voltage of the converter increases or the load current decreases, the voltage at the input increases 66 assumes the differential voltage between the inputs 63 and 66 and the voltage at the terminal decreases. In the "out of phase" part of the converter period, namely in the second time period, so the voltage across the winding 28 will be greater. This means in this "out of phase" "Part a faster demagnetization to a lower induction value, so that in the" in-phase "part, and the first time period, possibly despite the larger secondary transformer voltage, the response

¹ S time increases, because now a greater change in induction or flux has to be bridged. The effective "on" time decreases as a result, or the transformer delivers power for a shorter time. This counteracts the original increase in tension. This is dependent on the control loop gain, which is essentially determined by the gain factor of the amplifier 62. The regulation works very quickly because the change in the bias magnetization is already effective in the following period of the control oscillator.

030026/0669

Claims (6)

-49- PHN 9302 PATENT CLAIMS; 1. J Pulsed power supply of the flow converter type with a * in the primary circuit with two input terminals connected to a DC voltage supply source can be connected, and secondary circuits with output terminals for connecting a Load, the primary circuit between the input terminals at least one controllable circuit element that periodically controlled by a control circuit, contains in series with a primary winding of a transformer, during each secondary circuit a secondary winding of the transformer, includes a rectifier diode, a freewheeling diode and a choke coil, with at least one secondary circuit is provided with setting and stabilizing means for a size at the output terminals, characterized in that that the adjustment and stabilization means contain a saturable choke coil (26) with a power winding (27) is arranged in series with the secondary winding (2, 3) and the rectifier diode (13) and fathers at least a control winding (28) for setting a bias in the core of the choke coil, and that further the means contain a control unit (29) connected to the control winding (28) and having an input (30) is provided for the size to be set and stabilized. 2. Pulsed power supply according to claim 1, characterized in that the core of the saturable choke coil is made of magnetic There is material with low remanent magnetism (Fig. 2). 3. Clocked power supply according to claim 1 or 2, wherein the control circuit during a first period of time 30026/0669 ORIGINAL INSPECTED 2 * Γ PHN 9302 makes controllable switching element conductive and blocked during a second time period, characterized in that the control unit includes a diode (77) and a control voltage source (69, 70), which ihe in R _e are arranged with the control winding such that, during the first time period the diode (77) is blocked by the control voltage of the source and is conductive during a part of the second period due to the induction voltage across the control winding (28), so that the bias is determined by the voltage-time integral of the control voltage over the part of the second period will. 4. Pulsed power supply according to claim 3, characterized in that the control voltage source is the output of an operational amplifier (62) which amplifies the difference signal between a reference signal (67) and a signal (30) proportional to said variable. 5. Pulsed power supply according to claim 3 or 4, characterized in that the diode (77) is the base-emitter diode of a transistor (78) whose collector with one end and its emitter with the other end of the control winding (28) is connected. 6. Pulsed power supply according to one of the preceding claims, in which the control circuit for the circuit elements in the primary circuit contains an oscillator controlled in the pulse width and a comparison circuit which is connected to an input is connected to the output terminals of one of the secondary circuits, so that the voltage is applied to these Clamping with the help of the pulse width control is constant, characterized in that the input (30) of the control unit (29), which belongs to another secondary circuit (2, 3), to the associated output terminals (32, 33; 34, 35) connected. 030 0 26/0669