GB2033163A

GB2033163A - Variable leakage transformers

Info

Publication number: GB2033163A
Application number: GB7935340A
Authority: GB
Original assignee: TDK Corp
Current assignee: TDK Corp
Priority date: 1978-10-14
Filing date: 1979-10-11
Publication date: 1980-05-14
Also published as: GB2033163B

Description

SPECIFICATION Variable leakage transformers The present invention relates to variable leakage transformers.

A transformer has, in general, a magnetic core having a closed magnetic path, and a primary and a secondary winding wound on said magnetic core. All the input power applied to the primary winding is available at the output of the secondary winding except for a small amount of loss in the transformer. The output voltage V2 across the secondary winding is given by: V2 = (n2/n) x Va, where is the voltage across the primary winding and n1 and n2 are the number of turns of the primary and secondary windings, respectively.

If it is required to control the output power of the transformer, a controllable switching device such as a SCRJSilicon Controlled Rectifier) or a transistor may be employed at the output of the transformer. In such controllable switching device, the pulse width during each cycle is varied by controlling the conducting time of the device. However, a prior controllable switching device has the disadvantage that the circuit is very complicated and the cost of the device is rather high.

Another known arrangement for controlling an AC power source is a magnetic amplifier, in which a saturable reactor is inserted between the power source and the load, and by controlling the reactor, the power transferred from the source to the load is controlled. However, a magnetic amplifier has the disadvantage in that the voltage across the load must be the same as that of the power source, and the saturable reactor does not function as a variable-voltage transformer.

The present invention seeks to overcome the disadvantages and limitations of the prior art by providing a new and improved controllable transformer which operates on the principle of leakage flux control.

In accordance with the invention there is provided a variable leakage transformer comprising of a core defining a closed main magnetic path having a gap of non-magnetic material in the same and a closed sub-magnetic path, the main magnetic path having at least one common path with the sub-magnetic path, a primary winding wound on the common path of the core, a secondary winding wound on the main magnetic path of the core, and a control winding wound on said sub-magnetic path of the core for controlling the magnetic flux in said sub-magnetic path.

Preferably, the magnetic flux in said sub-magnetic path is controlled by flow of direct current in a control winding wound on said sub-magnetic path.

In the particular application of the present transformer in a stabilized power source, the direct current in said control winding reflects the change of the output voltage across the secondary winding.

In order that the invention may be better understood, several embodiments thereof will now be described by way of example only and with reference to the accompanying drawings in which: Figure 1 is a cross sectional view of a known transformer; Figures 2(at, 2(8) and 2(C) show the principle operational waveforms of the transformer in Figure 1; Figures 3(A) and 3(B) show the actual operational waveforms of the transformer in Figure 1; Figure 4 is a cross sectional view of another known transformer; Figure 5 is a perspective view of still another known transformer; Figure 6 shows the other structure of the core for the use of the transformer in Figure 5; Figure 7shows the equivalent circuit of an embodiment of the transformer of the present invention;; Figure 8 shows the structure of an embodiment of the transformer of the present invention, and relates to the equivalent circuit of Figure 7; Figure 9 is the structure of another embodiment of the transformer of the present invention, and relates to the equivalent circuit of Figure 7; Figure 10 shows the characteristics of the transformer shown in Figure 8 or Figure 9; Figure 11 shows the characteristics of the core utilized in the still another embodiment of the transformer of the present invention; Figures 12(A) and 12(B) show the characteristics of the transformer which utilizes the core shown in Figure 11; Figure 13 is the circuitry of the stabilized power source using the transformer of the present invention; Figure 14 is the embodiment of the circuitry of the control circuit of the power source in Figure 13;; Figure 15 shows the operational waveforms of the circuit shown in Figure 13; Figure 16 shows the other operational waveforms of the circuit shown in Figure 13; and Figure 17 is the circuit of still another stabilized power source using the transformer of the present invention.

Figure 1 shows a known transformer. The transformer comprises a magnetic core 1, composed of the combination of either an E shaped magnetic core portion and I shaped magnetic core portions or a pair of E shaped magnetic core portions, is a three legged structure defining 2 magnetic paths, a sub-magnetic path AA'B'B and a main magnetic path AA'C'C. The middle leg AA' and the side leg CC' of the magnetic core 1 are provided with a primary coil 2 and a secondary coil 3, respectively. The number of turns of the primary coil 2 is assumed to be N, and that of the secondary coil N2.

A load resistor RL is connected to the secondary coil 3 and a AC input voltage Bin is applied to the primary coil 2. The magnetic flux + generated thereby is divided into magnetic flux X, flowing in the sub-magnetic path AA'B'B, and magnetic flux 4)2 flowing in the main magnetic path AA'C'C. That is, the formula 4) = 4)i + 4)2 is satisfied.

Of these two fluxes, the one that supplies power to the load resistor RL is the one that interlinks with the secondary coil 3, i.e. 2. The flux 1 does not contribute in the supply of power to the load resistor RL. At this stage, if the magnetic reluctance is increased by application of a control magnetic field Hc to the sub-magnetic path AA'B'B from outside, the magnetic flux X, decreases with resultant proportionate increase in the magnetic flux 4)2. Therefore, the output voltage Vout of the secondary coil 3, i.e. the power supplied to the load resistor RL, increases.It can thus be seen that power transferred from the primary to the secondary can be controlled by changing the strength of the control magnetic field Hc applied to the sub-magnetic path AA'B'B from outside.

The control magnetic field Hc can be obtained, for instance, by means of a winding wound on the sub-magnetic path, for instance on the leg BB' and, by controlling the D.C. current in such winding, the strength of magnetic field Hc is controlled.

Suppose that a rectangular wave voltage is applied to the primary coil 2, as input voltage Vjn such as shown in Figure 2(A). If the sub-magnetic path AA'B'B is in a saturated condition due to the control magnetic field Hcthen, because this sub-magnetic path is considered to be equivalent to non-existent as a magnetic circuit, the output wave form of the output voltage Vout thus obtained is identical with that of the input voltage Vjn shown in Figure 2(B).

The amplitude of Vout is determined by the ratio between the number of turns N1 on the primary coil 2 and the number of turns N2 on the secondary coil 3. That is, Vout = Vin (N2/N1).

Suppose that the sub-magnetic path AA'B'B is put in a non-saturated condition by weakening the control magnetic field Hc. Some of the magnetic flux o generated by input voltage Vjn flows round the sub-magnetic path during the time t1, i.e. from the time that input voltage Vjn is applied until the time the sub-magnetic path AA'B'B is saturated by the input voltage Vin, and, the output voltage Vout in effect becomes almost zero.

At the end of time period t1, the sub-magnetic path AA'B'B becomes saturated. This saturated condition is maintained until inversion of the input voltage Vin. Therefore, as shown in Figure 2(C) the output voltage Vout (= Vin N2/N1) appears only during the time (T/2 - tl). At this stage, the length of time period to can be changed by changing the magnetized condition of the sub-magnetic path AA'B'B, or in other words by the magnitude of the control magnetic field H, applied to the sub-magnetic path AA'B'B.

Thus, by increasing or decreasing the magnitude of the control magnetic field Hc, the pulse width of output voltage Vout can be controlled. During the period t1, power consumption in the primary winding is zero, because the whole magnetic flux o flows in the sub-magnetic path AA'B'B and does not interlink with the secondary coil 3, and therfore, the load as viewed from the primary coil 2 is in a condition equivalent to an open circuit. In other words, the control does not affect power efficiency while the operation is always highly efficient.

Performance of this transformer shows that the pulse width of the output voltage Vout can be efficiently controlled by increasing or decreasing the magnitude of the control magnetic field Hc.

It should be appreciated of course that an input voltage of sinusoidal waveform is also possible and provides the same effect as a rectangular waveform, although the drawing shows a rectangular waveform for simplicity of explanation.

As explained above, the leakage of the magnetic flux from the main magnetic path to the sub-magnetic path can be controlled by varying the magnetic flux in the sub-magnetic path, and thus, the coupling between the primary and secondary windings and the conduction period in each cycle of the input voltage are controlled. And it should be noted that the control of the conduction period provides the control of the power transmission from the primary winding to the secondary winding.

In the above-described transformer, the primary coil 2 was wound around the middle leg of the magnetic core 1 and the secondary coil 3 was wound around one of the side legs. As a result of this arrangement, the magnetic coupling between the primary coil 2 and the secondary coil 3 has a tendency to be insufficient. If the coupling is insufficient, even if the rectangular wave input voltage Vjn as given in Figure 3(A) is applied, the wave form of the output voltage Vout becomes deformed as shown in Figure 3(B), and the efficiency of the transformer is decreased. This results from insufficient coupling between the primary coil 2 and the secondary coil 3.

Figure 4 depicts a second known transformer which provides a configuration with improvements on the above mentioned drawback. In this configuration, the primary coil is divided into a first primary coil 2(A) and a second primary coil 2(B). The first primary coil 2(A) surrounds the middle leg AA' of the magnetic core 1 and the second primary coil 2(B) surrounds the side leg CC'. The second coil 3 is would over the second primary coil 2(B). With this arrangement, the coupling between the primary and the second coils becomes greater and the wave form of the output voltage Vout and that of the input voltage Vin can be made almost identical.

The configuration of Figure 4 with the divided primary windings is advantageous in particular when the magnetic core constituting the main magnetic path and the sub-magnetic path is made of ferrite, whose magnetic permeability is rather small. When the magnetic permeability of the core is small, the coupling between the primary winding and the secondary winding is insufficient, and the output waveform is deformed. This insufficiency is overcome by dividing the primary winding into a first primary winding and a second primary winding, one of which is closely coupled with the secondary winding, as shown in Figure 4.

Since the operation of a ferrite core is excellent at high frequency, a transfomer utilizing a ferrite core operating in high frequency can be made of small size.

Preferably, a ferrite core with the magnetic permeability being approximate 300~1000 is utilized, the frequency of the input signal is in the range 20-30 kHz, and 50 per cent of the turns of the primary winding is separated to form the second primary winding.

The preferred ferrite core is obtained in the commerical market under the tradename H7C1 manufactured by TDK Electronics, Co. Ltd., Tokyo, Japan. The magnetic permeability of the H7C1 core is approximate 5,000 in a ring shaped core, and is approximate 300~1000 in assembled E and I cores. The permeability in an assembled E and I core is considerably decreased compared with that of a ring shaped core, because of the inevitable air gap at the coupling portion between the E shaped core portion and the I shaped core portion.

Figure 5 illustrates another known transformer. In Figure 5, the magnetic core 1Aforming the main magnetic path is made up of the combination of a U shaped core portion 1A(U) and an I shaped core portion iA(I), and the magnetic core 1 B forming the sub-magnetic path is made up of a combination of an E shaped core portion 1 B(E) and an I shaped core portion 1 B(l). The primary coil 2 is wound in common around one of the legs of the magnetic core portions 1A and the middle leg of the magnetic core portion 1 B while the secondary coil 3 is wound around the other leg of the magnetic core portion 1A. Both the side legs of the magnetic core portion 1 B are provided with respective control coils 4A and 4B.The control coils 4A and 4B are connected in series so that the voltages induced in these coils will cancel each other when the input voltage Vjn is applied to the primary coil 2.

In the above arrangement, the magnetized condition of the magnetic core portion 1 B forming the sub-magnetic path can be changed by the control current lc fed to the control coils 4A and 4B. For example, when the rectangular wave such as shown in Figure 2(A) is applied as input voltage Vjnl the magnetization strength of the sub-magnetic path becomes large is control current lc is large, while the time period tin Figure 2(C) becomes short. On the other hand, if the control current lc is small, the strength of magnetization of the sub-magnetic path falls while the time period t1 becomes long. As a result, the pulse width of output voltage Vout can be controlled.

The magnetic core is made up of the E-l magnetic core portions and the U-l magnetic core portions combined. However, it should be appreciated that a magnetic core having the same effect may be formed by a combination of a four-legged magnetic core portion IC and a T shaped magnetic core portions ID as depicted in Figure 6.

Although Figure 5 illustrates the use of a combined U-l core and E-i core, it should be appreciated that many modifications of this particular arrangement, for instance U-U core and E-E core, are possible.

In the transformer of Figure 5, the waveform of the output voltage Vout can be improved by dividing the primary coil 2 as described above in relation to the embodiment of Figure 4 and by closely coupling a part of the primary coil to the second coil 3.

The characteristics of the transformer are improved by providing a narrow air gap in the main magnetic path. This air gap improves the characteristics of the transformer when the load of the transformer is small, this is to say, when the load is small. The magnetic reluctance of the main magnetic path is inherently small, and so the presence of the sub-magnetic path does not greatly affect the magnetic flux in the main magnetic path, and thus, the output voltage on the secondary winding can not be controlled when the load is small.

The improvement for overcoming this problem is described with reference to Figure 7 to 10.

Figure 7 shows the equivalent circuit of the present transformer, in which an ordinary transformer T, having a primary winding 2B and a secondary winding 3 coupled closely with the primary winding 2B, and a saturable reactor T2 having winding 2A connected in series with the primary winding 2B are provided.

Assuming that the inductance of the saturable reactor T2 is L1, and the inductance of the primary winding 2B of the transformer T1 is L2, the voltages V11 and V72 across the reactor winding 2A and the primary winding 2B respectively are proportional to the inductances L1 and L2 respectively, and are given below:

It should be noted that the inductance L1 of the saturable reactor becomes small when the control current in the windings 4A and 4B in Figure 5 is large and, in this event, the voltage V11 is small and the voltage V12 is large and the output voltage is large. When the control current is small, the inductance L1 is large, the voltage V11 is large, the voltage V12 is small and the output voltage is small.

Figure 8 shows an improvement over Figure 1. The feature of Figure 8 is the presence of a narrow air gap lOin the main magnetic path. The presence of the air gap 10 increases the magnetic reluctance of the main magnetic path and so the output voltage can be controlled even when the load is small.

Figure 9 is an improvement of the structure of Figure 5. The feature of Figure 9 is the presence of spacers 10a and lOb of non-magnetic material, for example copper, at the top of the U shaped core portion 1A(U).

The thickness of the spacers is preferably 100-1 50rim. Due to the presence of those spacers, the magnetic reluctance in the main magnetic path is increased and the same effect as that described above in relation to Figure 8 is obtained.

Figure 10 shows the effect of the air gap or a spacer in the main magnetic path. In Figure 10 the horizontal axis shows the output current (12) (logarithmic scale) which corresponds to the load, and the vertical axis shows the output voltage (V2). The curve (A) shows the characteristics when no air gap nor spacers is provided, and the curve (B) shows the characteristics when an air gap or spacers of non-magnetic material is provided when the control current is constant. It should be appreciated in Figure 10 that the presence of an air gap or spacers improves the characteristics for a small load, and the deviation of the output voltage of the transformer is decreased by the presence of an air gap or spacers.In Figure 10, the curve (A) shows that the output voltage (V2) is constant for the range from Imjn(A) to 1max of the output current (12), and the curve (B) shows that the output voltage (V2) is constant for the range from Imjn(B) to Imax of the output current (12).

Apparently, lmjn(B) is smaller than Imin(A), and so the substantial operational range of the curve (B) is wider than that of the curve (A).

Figure 11, shows the replacement of said air gap or spacers. Figure 11 shows the characteristics of the magnetic permeability (t) of the core material utilized for the main magnetic path and/orthe sub-magnetic path. The curve (A) in Figure 11 shows the H-ll characteristics of X013 type manganese-zinc ferrite material maufactured by TDK Electronics Co. Ltd., in Tokyo, Japan, and the curve (B) in Figure 11 shows the characteristics of H7C1 type ferrite material.In Figure 11, the horizontal axis shows the magnetic flux (J;oersted) in the core, and the vertical axis shows the magnetic permeability (cm), and it should be appreciated from Figure 11 that the material of curve (A) is preferable to the material of curve (B), since the ratio of the permeability when the magnetic flux is large to that when the magnetic flux is small, is larger for the material (A). Actually, the material (A) shows that the permeability for H = 0.35 oersted is 7,100 and the permeability for H = 0.04 oersted is 2,700, so the ratio is 7.100/2,700 = 14,000, while the material (B) shows that the permeability for H = 0.35 oersted is 5,600 and the permeability for H = 0.04 oersted is 2,550, so the ratio is 5,600/2,550 = 9,800. Preferably, the value of that ratio is larger than 10,000.

Some experimental conditions for the curves of Figure 11 are that a sample core is a toroidal shape with an inner diameter of 19 mm, the frequency is 25 KHz and the temperature is 25"C. That characteristic, in particular the characteristic of the curve A is equivalent to the air gap or spacers. When the core of the characteristics of the curve (A) in Figure 11 is utilized, the characteristics of the inductances L1 and L2 of the windings 2A and 2B respectively for the change of the load current (12) are shown in Figures 12(a) and 12(b) respectively. It should be noted from Figures 12(A) and 12(B) that the inductance L1 is comparatively large when the load current (12) is small, and the output voltage of the transformer is well controlled even when the load is small.

It should be appreciated that the combination of the features mentioned in Figure 4, which is the division of the primary winding when a ferrite core is utilized, and those in Figures 8 or 9 which show the presence of an air gap or spacers, and/or Figures 11 to 12(B) will provide a more improved transformer.

Figure 13 depicts an embodiment of a stabilized D.C. power source for use with the transformer of the present invention. In the drawing, the primary winding N1 of a pulse width control transformer 21 is a so-called bifilar coil with a centre tap. To this primary winding N1 is connected a self-exciting push-pull converter 23 incorporating transistors Q1 and Q2 and an oscillation transformer 22. A direct current power source 24 is connected between the centre tap of the primary winding N, and the emitter of the transistors Q1 and Q2 SO that D.C power is fed to the converter 23.

The secondary winding N2 of the transformer 21 is also provided with a centre tap. A full-wave rectification circuit 25 composed of rectifier diodes D1, D2, choke Land capacitor C1, is connected to the secondary winding N2. The DC output voltage Vout of full-wave rectification circuit 25 is fed to a load 26. The transformer 21 is equipped with an auxiliary supply winding N3 on the main magnetic path and a control winding N4 on the sub-magnetic path. A DC control voltage Vc, produced by rectification by diode D3 and smoothed by capacitor C2 is applied together with the Ac output of the auxiliary supply winding N2, to a control circuit 27.

The control circuit 27 detects changes in the output voltage Vout and controls the value of the control current lc to be fed to the control winding N4.

It should be appreciated in Figure 13 that the transformer 21 is a variable leakage transformer of the type described above. The push-pull convertor 23 provides the input voltage of the rectangular waveform to the primary winding N1 of the transformer 21.

In Figure 13, the battery 24 can be replaced by the combination of a commercial alternative current power supply and a rectifying circuit connected to the same, and in this case, a stabilized D.C. power supply from a commercial A.C. power supply is obtained. An example of the design of such a stabilized D.C. power supply is shown below.

1) Input A.C. power voltage; 85~135volts 2) Output D.C. voltage; 24 volts 3) Output current; 0.1~6.0 amperes 4) Core; H7C1 type ferrite core manufactured by TDK Electronics Co. Ltd., Tokyo, Japan, the cross sectional area of El portion in Figure 9 is 2.4 cm2, and the cross section area of Ul portion in Figure 9 is 1.2 cm2, and the permeability of the core is approximately 300-1000 in an assembled form as shown in Figure 9.

5) Thickness of the spacers; 1001lm (plastics film) (for instance polyethylene-terephthalate film) 6) Numberofturns ofthe primary winding; First primary winding; 25x 2 Second primary winding; 25x2 7) Number of turns of the secondary winding; 24 x 2 It should be appreciated from the above explanation that although the voltage of the commercial power source depends upon the location (for instance, 100 volts in Japan, and 114 volts in other countries) a single power supply can provide stabilized D.C. output voltage by controlling the control current in the transformer.

During experiments with the above design, we obtained a ratio of output current (lmax/lmin(B); see Figure 10 of more than 30, while the ratio (lmax/lmin(A); see Figure 10) is less than 15 if no air gap is provided. Hence the presence of the air gap provides twice as wide a range of the output current as that with no air gap.

As illustrated in Figure 14, the control circuit 27 comprises a series circuit consisting of, a transistor Q3 which feeds control current lc to the control winding N4, upon receiving the control voltage Vc. The control circuit also comprises a zener diode ZD that detects the output voltage Vout of the rectification circu it 5, and a transistor 04 controlled by zener diode ZD.

The transistor Q4 is switched on upon breakdown of the zener diode ZD when the output voltage Vout exceeds the zener voltage of the zener diode ZD, and thus reduces the emitter current of the transistor Q3, i.e.

control current 1c In the circuit of the Figures 13 and 14, if the sub-magnetic path of the transformer 21, is sufficiently saturated by having the prescribed control current lc passing through the control winding N4, the sub-magnetic path effectively becomes non-existent.The P - P' voltage of the self-exciting push-pull type converter 23 assumes a rectangular waveform as indicated in Figure 15(A). The collector current 11 of the transistor Q1 and the collector current 12 of the transistor 02 assume waveforms as depicted in Figures 15(B) and 15(C) respectively. Therefore, the voltage induced in the secondary winding N2, as in the case of an ordinary transformer, becomes a rectangular waveform similar to that induced in the primary winding as shown in Figure 15(D).

In this situation, transfer of power from the primary to the secondary is maximised and the output DC voltage Vout also reaches its maximum value.

When the terminal voltage of the DC power source 24 rises, or the power required by the load 26 decreases resulting in an increase in the output DC voltage Volt, current flows through the zener diode ZD in the control circuit 27 and the base bias current of the transistor Q3 is divided by the transistor 04. As a result, emitter current of the transistor 03, i.e. control current lc decreases and the magnetic reluctance of the sub-magnetic path of the transformer 21 drops. Under these conditions most of the magnetic flux generated by the primary winding N1 flows through the sub-magnetic path and reduces the coupling with the secondary winding N2.

Therefore, even if the P - P' voltage of the self-exciting push-pull type converter 23 assumes the rectangular waveform indicated in Figure 16(A), the collector current Ii of the transistor Q1 and the collector current 12 of the transistor Q2 assume waveforms as shown in Figures 16(B) and 16(C) respectively. That is, until the end of the time period t1 when the sub-magnetic path has been magnetically saturated by the magnetic flux established by the primary winding N1, only very little current corresponding to the exciting current for exciting the magnetic core of the transformer 21, flows as collector current of each transistor. At the end of time period t1, the sub-magnetic path becomes saturated and the magnetic flux of the primary winding N1 couples with the secondary winding N2.Therefore, the voltage appearing in the secondary winding N2 becomes a bipolar type pulse such as shown in Figure 16(D). That is, compared with Figure 15(D) the pulse width becomes shorter by as much as the time duration t1 and therefore, in response thereto, the output DC voltage Vout drops.

Thus, in the circuit of Figure 13, the output DC voltage Veut can be constantly maintained at a given level if the control current lc and the characteristics of the zener diode ZD are appropriately set in dependence upon the output DC voltage Vout desired. In this situation, the collector currents of the transistors Q1 and Q2 of the aforementioned converter become only exciting current at the time t1, and the output end of the converter 23 reaches an open circuit or no-load condition. Therefore, the control does not cause a power loss, while conversion efficiency of the converter 23 scarcely decreases. In the conventional pulse control system, in order to control operation of a switching element, a feed back circuit from the secondary to the primary of the transformer is necessary.However, in this circuit pulse width control can be accomplished on the secondary side of the transformer 21, and the primary is irrelevant to the control. Therefore, a high dielectric strength as prescribed by various safety standards can be readily provided between the primary and the secondary. Further, since a complex circuit arrangement is not required, costs can be reduced.

Figure 17 is another embodiment of this invention with a configuration appropriate to the above situation.

In this embodiment, the output of the converter 23 is applied to the primary winding L1 of an output transformer 30 provided with the required number of output windings. And, to each of a plurality of output windings N11, N12 and N13 respectively of the control transformers 21X, 21Y and 21Z. Each of the control windings N41, N42 and N43 of the control transformers 21X, 21Y and 21Z are provided with a respective control circuit 27X, 27Y and 27Z. With these control circuits, the control of the output pulse width of the secondary windings N21, N22, N23 of the control transformers 21 X, 21 Y and 21 Z respectively is effected: i.e.

regulation of output voltage is accomplished.

In the configuration described in Figure 17 above, a single converter 23 can be used for each stabilized output in common. Compared with a situation where a converter is provided for each regulated output, this arrangement greatly simplifies the circuit configuration and, therefore, is less costly. Further, the transformer 30 can be removed, in which case, the primary windings N11, N12, N13 are connected in parallel directly to the output of the converter 23.

As described above, with these embodiments, output pulse width is controlled by the control transformer.

Therefore, the oscillation frequency of the converter can be arbitrarily set to meet the current demand. This facilitates the design of the converter. Further, the use of the frequency higher than 10 KHz is possible by utilizing a ferrite core, and such high frequency allows the small size of the apparatus. At the same time, control of the operation of the switching element in the converter is unnecessary. Since the feedback circuit from the secondary to the primary of the transformer is unnecessary, insulation between the primary and the secondary is facilitated. Special contrivances for a feedback circuit are not necessary either. The circuit arrangement is simple while affording economical improvement. Further, the presence of an air gap 10 or spacers is the important feature of the apparatus, and due to this air gap, the output voltage is stabilized even when the load is small.

Claims

1. A variable leakage transformer comprising a core defining a closed main magnetic path having a gap of non-magnetic material in the same and a closed sub-magnetic path, the main magnetic path having at least one common path with the sub-magnetic path, a primary winding wound on the common path of the core, a secondary winding wound on the main magnetic path of the core, and a control winding wound on said sub-magnetic path of the core for controlling the magnetic flux in said sub-magnetic path.

2. A variable leakage transformer according to Claim 1, wherein said gap is filled with a spacer made of non-magnetic material.

3. A variable leakage transformer according to Claim 1 wherein said gap is an air gap.

4. A variable leakage transformer according to any one of Claims 1, 2 or 3 wherein the ratio of the magnetic permeability of the core when the magnetic flux in the core is large to that when the magnetic flux in the core is small, is large.

5. A variable leakage transformer according to any one of the preceding Claims, wherein said core is made of ferrite, and said primary winding is divided into a first primary winding portion and a second primary winding portion connected in series with each other, the first primary winding portion being wound on the common magnetic path, the second primary winding portion being wound on the main magnetic path and wherein the second primary winding portion is closely coupled magnetically with the secondary winding.

6. A variable leakage transformer as claimed in any one of Claims 1 to 4 wherein the core comprises a U shaped core portion and an I shaped core portion together defining said main magnetic path, an E shaped core portion and an I shaped core portion together defining said sub-magnetic path, the centre leg of said E shaped core portion being connected to one leg of said U shaped core portion, and wherein a pair of spacers of non-magnetic material are inserted between the top ofthe legs of said U shaped core portion and the related I shaped core portion, and wherein the primary winding is wound common to the connected centre leg of said E shaped core portion and said one leg of the U shaped core portion, and the secondary winding is wound on the other leg of said U shaped core portion, and said control windings are wound on the pair of side legs of said E shaped core portion and are connected in series with each other.

7. A variable leakage transformer according to Claim 6, wherein said cores are made of ferrite, and at least a part of said primary winding is wound on the same leg of said U shaped core portion as that on which the secondary winding is provided.

8. A variable leakage transformer substantially as hereinbefore described with reference to Figures 7 to 12 of the accompanying drawings.