FR2625400A1 - MICROWAVE ENERGY GENERATING SYSTEM - Google Patents

Abstract

A power supply for a magnetron is disclosed. A power supply for a magnetron includes a full-wave bridge inverter 22 which has four solid-state switching elements 46F, 46S, 46T, 46R which are switched by a pulse-width modulated control circuit 24 to provide power. energy to the primary winding of a power transformer 26, the secondary of which feeds the magnetron 32. The power of the magnetron is controlled by applying to the switching elements control pulses having a width proportional to a input signal which is applied to a transconductance amplifier. Application to microwave ovens.

Description

The present invention relates generally to a feeding system

for a cooking magnetron, and

  it is more particularly concerned with such a system

  a double-wave bridge inverter.

  Most microwave ovens found

  currently on the market use a food

  50 or 60 Hz LC, which is generally in accordance with

  the description made in the U.S. Patent No. 3,396,342,

  delivered to Feinberg on August 6, 1968. More than twenty years ago there is this type of power supply which is used in microwave cooking appliances from low-power appliances called "subcompacts" to appliances.

  combination of the electric oven / microwave oven type.

  Among the advantages of the Feinberg feeder system are the simplicity that results from the use of only four components, and a good control of the power factor. The disadvantages include the size

  and the high weight that is achieved by commanding

  only by the duty cycle, the operation of the filament in a non-continuous power regime at power levels other than 100%, a current draw

  high and the noise of the laminated magnetic circuit. The incon-

  The size and weight of the Feinberg system is the result of the need to use a 50 or 60 Hz transformer with a nominal power of about 1.2 kVA. Iron and

  copper of such a transformer typically weigh

  about 700 g, and they occupy a volume of 1710 cm3. In addition, it is necessary to use a component consisting of

  capacitor of large physical size, when

  that such a transformer is used, so as to achieve a constant current regulation for the magnetron power

  vis-à-vis variations of the mains voltage.

  A symmetrical system has been used or proposed in connection with the supply of a cooking magnetron. Although the symmetric system avoids some of the disadvantages

  Feinberg's feeder configuration, such a system

  The symmetrical system has drawbacks such as high cost, complex logic, power transistors to

  high voltage in Darlington mounting, damping networks

  reactive (that is, dissipate power),

  an imbalance inherent in the volt-

  seconds for each half-cycle of operation, resulting from uncontrolled blocking characteristics of switching transistors, a bad input power factor

  (eg 0.6), a high level of disturbance generation

  electromagnetic interference, a poor conversion efficiency

  and a higher cost of magnetic components. The

  magnetic components with a higher cost corresponding to

  dent to a structure employing a leakage transformer

  variables as a power control means.

  Another prior structure is described in U.S. No 4 281 372, issued at Kornrumpf on 28

  July 1981. Although it is generally used

  This structure has several disadvantages. In particular, this structure is more complex and expensive than the Feinberg system. A peak current of more than 4 amps is required to provide 700 watts of microwave power. This peak current tends to show an initial operation of the magnetron on undesirable modes, and finally a short-term failure. In addition, this structure can cause excessively high electromagnetic disturbances. This

  structure also has the disadvantage of requiring

  switching sistors that must accept currents from

  ridge high and which must hold very high tensions.

  A further disadvantage of this structure is that complex logic is required to control the

  operation of the switching elements.

  A principle goal of the invention is therefore to

  a microwave energy generation system that includes

  a diet which is economical and which has a

  space-saving and low weight.

  A further object of the invention is to provide

  a microwave energy signal generation system having a power supply circuit that has a high efficiency,

  a minimum level of electromagnetic interference emissions

  and eliminates or minimizes the inconveniences

  from the power circuits envisaged above.

  The invention provides a structure for the food

  of a cooking magnetron by the use of a corrugated

  their bridging double alternation. The inverter uses four

  switching elements to alternate the flow of the

  in a primary winding of a power transformer. A secondary winding of the power transformer connects the power supply to the magnetron. In one form of the invention, the switching elements are

  semiconductor switching elements, more particularly

  MOS field effect transistors (or MOSFETs), and a diode is connected in parallel to each of the semiconductor switching elements. A bridge rectifier is used to provide power to the inverter circuit. The energy that the bridge rectifier provides to the inverter circuit is continuous energy having a minimum voltage during operation, and a ripple that is produced by

  the alternative energy to the rectifier input, this corrugation

  having an amplitude at least as large as the minimum voltage. The inverter circuit comprises a control circuit

  to control the operation of the switching elements.

  In one form of the invention, the control circuit generates

  control pulses or gate attack pulses for

  the semiconductor switching elements at frequencies of the order of 20-30 kHz, so that the frequency of the current in the primary winding is similarly of the order of 20-30 kHz. One form of the invention uses a control circuit comprising a pulse width modulator. The primary winding is preferably the only one

  primary winding of the transformer.

  In a more specific aspect of the invention, the control terminals of two of the switching elements are

  floating relative to the other two switching elements

  and an isolated driver for applying signals to the floating control terminals. In a form

  of the invention, the isolated driving circuit comprises a trans-

  insulation former having a primary winding connected to the control circuit and having two secondary windings,

  with each secondary winding connected to one of the terminals

  Floating command names. In another form of the invention

  the isolated driving circuit comprises two capacitors,

  each capacitor corresponding to one of the elements of

  mutation having a floating control terminal. Each capacitor can be operated to use its load

  to close the corresponding switching element.

  In another specific aspect of the invention, the

  control circuit receives a control input signal re-

  present in the magnetron, and the control circuit can be operated to switch pairs of the switching elements to an open state in

  function of the value of the command input signal. A disc

  positive power adjustment which is incorporated in the

  control circuit makes it possible to change the power of the magnet.

  tron by changing the length of intervals during which the switching elements are closed. More specifically, the power adjusting device can be a variable power control resistor, and the input signal of

  control also depends on the setting of the variable resistance

power adjustment.

  The rest of the description refers to the drawings

  annexed which represent respectively:

  Figure 1: a functional synoptic diagram of the system

the subject of the invention;

  Figure 2: a simplified diagram of the system of the in-

  vention; Figure 3: a sequential diagram of signals that

  are generated in different parts of the system of the invention.

  tion; Figure 4: details of a control circuit that is used with the system of the invention;

  Figure 5A: an isolated driver circuit that can be

  instead of using an isolation transformer of Figure 4; Figure 5B: A current absorbing circuit that could be used in the driving circuit of Figure 5A. Figure 6: a structure usable for regulating the filament voltage;

  Figure 7: A configuration that could be used

with the invention; and

  Figures 8, 9 and 10: representations respectively

  front, side and section of a radiator.

General structure

  In the block diagram of Figure 1, the system

  of microwave energy generation 14 of the invention

  takes a protection filter against electrical disturbances

  tromagnetic, 16, which is connected to a line of a conventional electrical power distribution network. The filter 16 prevents the system 14 from sending harmful signals to the AC line. The protection filter against

  the electromagnetic disturbances 16 is connected to a

  As shown, the output signal of the rectifier / filter 18 on the line 20 is a voltage signal.

  unregulated, which means that it has a negative

  a significant result of the 50 or 60 Hz frequency that is applied to the system 14. As will be described hereinafter in detail, the system uses the unregulated DC voltage on the line 20 because this permits the use of in the rectifier / filter 18) a filter capacitor smaller than that which would otherwise be necessary. The smallest capacitor absorbs from the line of the AC mains a current significantly lower than

  which would otherwise be necessary.

  It should of course be noted that the block diagram of the system 14 is somewhat simplified, insofar as it does not include safety switches,

  user control, an input fuse and

  similar products that are well known in the context of a

  microwave energy generation system.

  The unregulated DC voltage present on line 20 is applied to an inverter / driver circuit 22 with full-wave bridge. The inverter / driver 22, which is under the control of a control circuit 24, provides a power transformer 26 with a high AC voltage at a frequency of about 20 kHz. The control circuit 24 can receive input signals from the user,

  as regards the power setting, the operating time

  and other conditions usually

  users when operating a microwave oven

  waves. As shown, the control circuit 24 is connected

  As will be discussed hereinafter in more detail, the control circuit 24 receives a feedback signal from the power transformer.

power 26.

  The power transformer 26 supplies power to a voltage doubler 30 which in turn supplies the magnetron 32. The magnetron 32 also receives power

  for its filament from the power transformer 26.

  Inverter and associated circuits As shown in Figure 2, the protection filter

  against electromagnetic disturbances 16 of the

  me of microwave energy generation 14 receives a voltage

  120 volts through relay contacts

  34 and a fuse 36. The protection filter con-

  be the electromagnetic disturbances 16 is a filter in

  double pi which includes capacitors 38 and inductan-

  40. The signal from filter 16 is applied to the

  18R bridge trainer which applies a rectified signal to the

  42 and 44. The signal is filtered by a filter capacitor 18C so that the signal between the lines 42 and 44 is an unregulated DC voltage signal. If we

  uses a filter capacitor of 30 pF, 250 V =, the

  The signal between lines 42 and 44 will vary between 30 and volts. The ripple or level variation that results from the AC input signal is therefore at least as large as the normal minimum voltage during operation, i.e., 30 volts. By operating the system 14

  from an unregulated DC voltage, the

  to use a capacitor with a capacitance value

  cited high for the filter capacitor 18C. The utilisa-

  a capacitor of sufficiently high value for the filter would reduce the ripple of the AC signal between lines 42 and 44, but would result in the initial absorption

  of a very high current, likely to melt the fusi-

  36 and / or cause the tripping of a circuit breaker

  in the home electrical installation of the user.

  The inverter / driver 22 includes first

  first, second, third and fourth transistors 46F, 46S,

  46T and 46R. Transistors fulfill the function of

  Semiconductor switching elements for switching to the output lines 48 and 50 of the inverter the unregulated DC voltage which is present between the input lines 42 and 44 of the inverter. The switching elements 46F, 46S, 46T and 46R are blocked and released by the control circuit 24. In the embodiment considered, the elements of FIG.

  switching are field effect transistors of

  270 V, 18 A, which are marketed by Interna-

  Rectify as IRF 640.

  The control circuit 24 is directly connected

  at the control terminals of the semiconductor switching elements

  46S and 46R conductors. (For the MOS field effect transisoters which are shown, the control terminal will obviously be the gate.) In addition, the control circuit 24 controls the switching elements with effect transistors.

  MOS field 46F and 46T through a transforma-

  insulation motor having a primary winding 52 and secondary windings 54F and 54S. The isolation transformer performs the function of an insulated driver to enable the control terminals (more precisely, the gates) of the switching elements 46F and 46T to float with respect to the switching elements 46S and 46R. The

  insulation transformer, which includes the primary winding

  Mayor 52 and the secondary windings 54F and 54S, is a simple 1: 1: 1 pulse transformer. The drains of the semiconductor switching elements 46F and 46T are directly connected to the input line 42 of the inverter, and can therefore be considered as input terminals of these switching elements, while the source terminals

  switching elements 46F and 46T may be considered

  as exit points, since they are directly

  connected to the respective output lines 50 and 48 of the inverter. On the other hand, the sources of transistors 46S and 46R operate as input terminals to the extent that they receive via resistor 56 the

  input signal from the input line 44 of the

  while the drains of the switching elements 46S and 46R operate as output terminals in so far as they are respectively connected to the output lines 50 and

48 of the inverter.

  5. A diode 58 is connected in parallel with each of the switching elements 46F, 46S, 46T and 46R. The diodes 58 prevent the destruction of the transistor switching elements by excessive heating during the momentary dead interval which occurs between the blocking of a pair of switching elements and the unblocking of the other pair.

switching elements.

  The output lines 48 and 50 of the inverter are connected to a primary winding 60 of a transformer

  26. The ratio of transformation between the coil

  primary element 60 and the high-voltage secondary winding

  sion 64 is established to produce across a winding

  secondary wave of 2000 volts when this winding

  It is charged to output an average current of 540 mA. This voltage is doubled at half-wave by a diode 66 and a charge-holding capacitor 68. The wave

  resulting square, corresponding to a nega-

  4000 volts, is applied to the cathode of the cooking magnetron 70. The power transformer 26 can have

  typically a primary winding 60 comprising

  both 24 turns and a secondary high voltage winding 64 having 440 turns. In addition, a secondary low voltage winding 72, having a single turn, provides the 3 volts and 14 amps (in rms value) that are needed

  for the magnetron filament 70, and a secondary winding

  2-turn, 74, provides low-voltage energy

  for supplying the control circuit 24.

  Fundamental Operation of the Inverter and the Control Circuit The fundamental operation of the inverter 22 will be explained by continuing to consider FIG. 2, but also considering the sequence diagram of FIG.

  3. The control circuit 24, which is envisaged in detail below,

  after, may be used to provide different power levels to magnetron 70. Parts (a) - (f) of Figure 3 relate to low power operation (20%), while parts (g) (1) ) concern a high-power magnetron operation (100%) Considering first of all the operation at

  low power, it is noted that the control circuit 24 pro-

  duit a command or gate attack pulse that is

  represented in part (a) of figure 3, and this impulse

  The control pulse appears at the output A of the control circuit 24 in FIG. 2. The control pulse unblocks or closes.

  transistor switching element 46S and, via the

  1.5 of primary 52 and secondary 54S, it closes the

  46T switchgear. The control circuit 24 controls the power that is supplied to the magnetron 70 by controlling the width of the pulse. In part (a) of Figure 3, the pulse has a width of 5 microseconds. The frequency of the pulses is constant, and the control pulses A

  are produced repetitively during a series of pre-

  first time intervals beginning every 40 microse-

  onds. A series of control pulses B (of which only one

  is shown in part (b) to facilitate the illustra-

  tion), interspersed between the first intervals of time,

  are produced on the output B of the control circuit 24.

  The control pulse B closes the switching elements 46F and 46R. As shown in part (c) of FIG. 3, the alternating closure of the pairs of switching elements (46S and 46T together, and 46F and 46R together) applies the unregulated DC voltage (whose peak value rises

  up to 160 volts) in alternating directions to the winding

  primary power transformer 60. 26. The

  rant flowing in the primary winding 60 is

  FIG. 3, while the voltage across the secondary winding 64 is shown in part (e). The current resulting from the magnetron, about

  800 mA, is shown in part (f) of Figure 3.

  The operation of the circuit shown in parts (g) - (1l of FIG. 3 is substantially identical to

  than that of parts (a) - (f) of Figure 3, with the excep-

  This is because the control pulses A and B have a larger width, which increases the width of all associated signal pulses. This corresponds to a higher value of the time during which the 800 mA current is applied to the magnetron, as shown in part L of Fig. 3. Therefore, an average current of 270 mA is applied during high power operation, while the average current is only 50 mA for operation

at low power.

  It should be noted that even in the high power operation of parts (g) - (1) of Fig. 3, there must be a short, inactive area between the end of one of the control pulses on output A or exit B, and the

  start of the control pulse on the other output. The exis-

  this "inactive zone" appears best in the area.

  (i) of Figure 3, and it should be noted that this

  tive represents the delay since the end of the pulse of

  command B until the beginning of the control pulse A. This

  could be typically 2.5 microseconds

  for a total idle zone of 5 microseconds, in

  also staggering the corresponding delay between the end of

  the control pulse A and the start of the compression pulse

  B. Details of the control circuit The control circuit 24 provides the control functions with respect to the AC input voltage and the output power control. It detects the current of the magnetron and it uses this current information in

  a feedback system (closed-loop system) to

  the pulse width of the control signals which are applied to the inverter 22. The control circuit 24, which is shown in greater detail in FIG.

  in pulse width modulator circuit. For the convenience

  In the illustration, Figure 2 does not contain some

  components that appear in the more detailed representation

cut from Figure 4.

  The control circuit 24 comprises a generator chip

  24C driver signal driver of a suitable type, such as the SG3526J, which is a width modulator integrated circuit

  impulse made by Silicon General. We set the pace

  this pulse repetition, which is 20 kHz in the preferred embodiment, using a capacitor 76 and a

  resistance 78 for setting the temporal characteristics

  the, and these components could have respective values

  0.005 pF and 7.5 kQL. A resistor 80, which establishes the inactive area, could for example have a value of 91L. A slow start capacitor 82 is used which

  could have a characteristic value of 100 pF, for, in-.

  to produce a delay with linear variation from start up to obtaining the pulse width corresponding to the

normal running.

  When the control circuit 24 is in a state

  waiting time (that is, when it does not produce impulse

  control), the quiescent current is approximately 12 mA at a voltage of 15 V. The quiescent current is transmitted by a limiting resistor 84 connected to the control line.

  unregulated DC voltage 42, and is smoothed by the

  smoothing densifier 86, to remove the ripple that comes from the AC grid. Better control could be achieved by placing a 15 V zener diode (not shown) in parallel with the capacitor 86. During the generation of control pulses, the higher operating current for the control circuit is provided by the controller. low voltage winding 74 of the power transformer 26 (only the winding 74 is shown in FIG. 4). This supply current is rectified by the diode 8 and is filtered by the capacitor 86. A current limiting resistor 90, which may have a value of 33S-I, is used to provide voltage to the capacitors.

  common drives of the control circuit chip 24C.

  The regulation function is performed with respect to the voltage of the alternating electric network and the control of the output power, by detecting the current of the magnetron

  and using this signal to control the pulse width

  control signals produced by the control circuit

  of 24. A proportional control of the pulse width is carried out using an error amplifier 106 which

  is internal to the control circuit chip 24C.

  A toroidal transformer 94 completes the function of

  current transformer for detecting the current of the magnet

  tron. The secondary winding of this transformer comprises

  turns and the primary winding of the transformer 94 -

  has 1 turn, to provide a properly transformed current. The transformed current produces a voltage at

  terminals of the resistor 96, connected in series with a resistor

  variable power control 98, and this voltage is rectified by a diode 100 to apply a signal to the

* common node between the resistor 96 and the diode 100. This si-

  gnal is proportional to the current of the magnetron for a

  Given the variable power control resistor 98, the user sets the variable power control resistor 98 to select the desired output power for the magnetron. This proportional voltage signal is applied by a diode 102 to the positive signal input of a transconductance amplifier 106, in the chip of

  24C command. The signal ground line 44 floats by

  port to ground 92 relative to the alternative electric network, because of the conduction characteristics in the direct direction of

  diodes which form part of the bridge rectifier 18R (the

  18R is shown only in Figure 2). An iso-

  Therefore, it is necessary between the node of the magnetron 70 which is connected to the ground of the AC grid, and the signal ground line or input line of the inverter, 44. The transformer 94 also provides the necessary isolation. sary.

  The internal structure of the circuit board chip

  24C pulse width modulated control, provides a

  output signal having a proportional pulse width

  the input error, as determined by

  the error amplifier 106. During normal operation, the pulse width is determined by the voltage at the

  resistor terminals 96 and 98, and this voltage is

  even determined by the current of the magnetron. An example may be useful to better explain the reaction operation, and first consider the regulation with respect to the input voltage. If the circuit constants are set to obtain the full output power of the magnetron, ie 700 watts, from a low value, of 110 volts, of the alternating electric grid voltage, when the width modulator chip of pulse 24C produces pulse widths corresponding to a duty cycle of 75%, an increase in the AC mains voltage will cause an increase in the magnetron current which has the effect of increasing the input error which is applied to the Error amplifier 106. This increase in the input error in turn reduces the pulse width and reduces

  the magnetron current at the set point. For the com

  power, the user's control of the resistance

  variable power regulation 98 changes the voltage across the resistors 96 and 98, and therefore changes the signal

  which is applied to the positive input of the amplifier 106.

  This changes the pulse width of the control signal that is applied to the inverter 22, which in turn changes the

  power that is applied to the magnetron.

  Amplifier 106 is also used for

  Protect against excessive peak currents in switching elements 46F, 46S, 46T and 46R. For this purpose, the current detection resistor 56 is used to detect the value of the switching current. Voltage

  which is developed across the resistance 56 is applied

  by diode 104 to the positive input of the error amplifier 106. The diodes 104 and 102 are connected in a logical OR configuration, to allow the use of the amplifier 106 for both current control. magnetron and for limiting the switching current

in the inverter circuit 22.

  The gain and phase control of the amplifier

  an error or transconductance amplifier 106 is accom-

  folded by an external frequency compensation network, com-

  taking a capacitor 108 and a resistor 110. By way of example, using a capacitor 108 of 100 pF and a

  resistance 110 of 5 kIL, compensation is made for

  the stability for a gain equal to unity, placing a

pole at 400 Hz.

  The start and stop operations of the pulse width modulator control circuit chip 24C are performed by the control of the relay contact 112 which connects to the signal ground line 44, and disconnects

  this is an off entry (SD).

  Temperature probe isolation As shown in Figure 4, one can use a

  temperature probe 114 in association with the chip

  24C. Specifically, the temperature probe 114

  comprises a resistive element 116 which is a function of the time

  one terminal is connected to ground for the AC mains, 92. Electrical isolation is therefore required between the temperature probe 114 and the

  floating signal mass 44 of the control circuit chip

  24C. A toroidal transformer 118 is used to provide the necessary insulation. You can use a BNC connector or

  other (not shown) to connect the probe 114 to the trans-

  trainer 118 and disconnect it from the latter. The temperature probe 114 can be attacked by the pulses that are emitted on the output B of the chip 24C. These pulses are applied to the primary winding of the transformer 118 via a resistor 120. Moreover, these

  output pulses are applied to a resistance network.

  these temperature set point, which includes

  sistances 122 and 124 and a variable resistor 126. The common node between the resistor 120 and the primary winding of the

  transformer 128 is connected to the positive input of an amplifier

  comparator 128, internal to the 24C chip. When this built-in comparator amplifier 128 is active, it suppresses the control pulses which are emitted in A and

  B, which turns off the inverter 22. When the temperature

  As a result, the resistance of the element 116 decreases, which causes more current to flow in the primary winding of the isolation transformer 118. This current variation is detected in the form of a variable voltage across the resistor. 120. When this temperature dependent voltage is equal to the set point which is determined by the voltage generated by the resistors 122, 124 and 126, it activates the comparator amplifier 128 and switches the inverter to a mode. off. Therefore, when the temperature detected by the element 116 reaches the predetermined temperature which is

  established by the setting of the variable resistor 126, the

Microwave is interrupted.

   Floating capacitor isolated drive circuit Referring now to FIGS. 2 and 5A, an alternative solution to the driver circuit will be described.

  insulation comprising the primary winding 52 and the coils

  Secondary 54F and 54S (Figure 2). Instead of using an isolation transformer as the isolation driver of the embodiment of FIG. 2, FIG. A shows how a control circuit 24 can be connected to operate the upper transistor 46F by means of FIG. an insulated floating capacitor driver 152. In practice, the isolated driver would include

  two copies of the flow capacitor driver

  152, since it is easy to understand that a circuit 152

  milaire would be connected between the output B circuit of

  control 24 and the gate of transistor 46T (Figure 2).

  The floating capacitor driver 152 is based on the fact that power field effect transistors, such as MOSFET 46F, are voltage-controlled and absorb negligible gate current in the state.

  driver. The driver 152 therefore uses a condense

  154 to store the drive voltage. The condensation

  The value of the source of the 46F field effect transistor switch element and the drain

of the switching element 46S.

  When the transistor 46S is unlocked (the output B produces a control pulse), the capacitor 154 charges at 15 volts via the resistor

  156 and diode 158. As a result of the blocking or opening

  46S switching element and a short inactive zone (the control chip 24 operates in the same manner as described above), the control chip produces a pulse on the output A. This pulse of

  the control unit unlocks the switching element 46R and de-

  also cleaves, through the resistance

  , the circulation of a current in the absorption circuit

  The flow of current to the current absorbing circuit 162 activates the gate 164 which was previously inactive. The output signal of

  the door 164 then quickly unblocks the communication element.

  46F, via complementary switching elements 166 and 168 and associated resistors 170 and 172. A resistor 174 and a diode 176 are connected.

  at an entrance to the gate 164 to provide signals of

appropriate.

  At the moment when the pulse present at the output A of the pulse width modulator control circuit 24 terminates, the signal applied to the lower input of the gate 164 changes. The output signal of the gate 164 in turn reduces to a value almost equal to zero (the element of

  transistor switching 158 is conducting) the voltage of

  gate-source gate of transistor 46F, which has the effect of

  block the switching element 46F.

  An example of a circuit suitable for the circuit

  current absorption 162 may consist of a simple circuit

  current absorbing cooker such as that shown in FIG. 5B, in which a transistor 250 is used to make a current mirror circuit conductive

  formed by transistors 252 and 254, in which case no

  connection to the voltage of 15 volts is necessary. It should be noted that other circuits could similarly be used

current absorption.

  Power compensation for the filament In conventional LC power systems for

  furnace magnetrons, the feed winding of the

  lament of the magnetron is wound on the same magnetic circuit

  than the high voltage winding. Therefore, roll up

  the filament is energized when the primary winding

  mayor of the power transformer LC is energized,

  and it is off when the primary is off.

  The filament can therefore cool during the shutdown intervals of the operating cycle, which can have a duration

  on the order of 15 to 30 seconds with feed circuits

  50 or 60 Hz. Three seconds are approximately

  necessary to heat the cathode up to the time

  normal operating temperature when starting with a cold cathode. During this period, the magnetron may oscillate in an incorrect mode, or it may repeatedly jump between abnormal modes, particularly in the case of starting with a cold cathode. When the magnetron jumps to or out of abnormal mode, the oscillation often stops, which usually causes very high voltage transients (so

characteristic 12 to 14 kV).

  The present system avoids stopping intervals

  potentially long which are characteristic of

  control ratio by cyclic ratio. The cathode of the

  gnetron therefore remains at a nearly constant temperature,

  which improves the life of the magnetron and eliminates the

  waning of the periodic generation of high voltage transients.

  A technique commonly used in the prior art

  using a separate filament transformer which is placed in parallel with the power transformer and provides a constant filament voltage (with the exception of variations in the filament voltage that are produced

  by variations in the voltage of the electricity grid

  ternatif). This technique could be used with the cir-

  cooked inverter of the invention. However, in the embodiment considered, a secondary winding of low

  voltage on the power transformer provides the voltage

  filament pressure. In this system, the filament winding 72 (Fig. 2) can cause changes in the power supplied to the magnetron filament 70 when the pulse width is varied to adjust the power of the magnetron.

  as previously described in connection with FIG.

  FIG. 6 shows an exemplary embodiment of a structure designed to regulate the power of the filament in order to suppress variations which would otherwise appear

  under the effect of the variation of the pulse width.

  The primary winding 60 of the power transformer 26 is connected to the output lines 48 and 50 of the inverter, exactly as in FIG. 2. A control circuit for the magnetron filament is added to the secondary side of the transformer circuit. power, to provide the desired regulation for the filament. It should be noted that the additional secondary winding 74 of FIG.

  been omitted in Figure 6, for the sake of clarity.

  In Figure 6, it is easy to see that the condensa-

  182 and the diode 184 have the function of doubling the voltage, with a simple alternation, essentially of the same

  way that the capacitor 68 and the diode 66 of Figure 2.

  A controlled variable impedance device, in the form of a small saturable magnetic circuit reactor 186, which comprises a control winding 188 and two controlled windings 190 and 192, is operably connected to the filament supply circuit, between the secondary low voltage filament winding

  72 of the main power transformer 26, and the filament

  magnetron, to stabilize the power of the filament.

  The impedance of the variable impedance device changes under

  the effect of a control signal which is produced by a

  positive detection, which detects a change in the

  which is supplied to the primary winding of the transformer

power meter.

  The magnetic circuit of reactance 186 comprises

  two outer magnetic branches and one central branch

  the. The ordered windings 190 and 192 are wound in

  opposite directions on respective outer branches.

  The control winding 188 is wound on the in-

  térieure. One could use in the circuit of the embodiment considered a reactance having an inductance

  controlled winding which varies from about 150 pH in the abs-

  current in the control winding, up to about 50 pH with mA control winding current, ie a 3: 1 control ratio. (The filament is essentially a resistive load, so the

  control of the current basically also controls the

  detection device intended to detect

  variations in the power that is applied to the winding

  60 of the transformer 26 in the example of

  considered, is realized in the form of a

  detection resistor 194 which is connected in series between

  the cathode of the diode 184 and the mass 92 of the anode of the

  gnétron. Diodes 196 and 198 are connected so as to establish a current path through the resistor 194, respectively for the charging current and the charging current.

  capacitor discharge 182. The voltage at the point of con-

  diode 184 and resistance 194 is proportionally

  the magnetron current, which is a di-

  the power that is applied to the winding

  60. A voltage representative of the power

  plicated to the primary winding 60 therefore appears at the connection point of the diode 184 and the resistor 194. This voltage is used as control voltage for

  reactance 186. One end of the control winding

  of 188 of reactance 186, which is called the terminal of com-

  Mande, is connected to the connection point of the diode 184 and the resistor 194. The other end or the other-terminal

  is connected to a power supply voltage circuit of

  ference. The reference supply circuit - is powered by a low voltage secondary winding 202 of the power transformer 26. A rectifier circuit with

  double alternation 204, which is connected to the terminals of the en-

  secondary bearing 202, provides at 206 a continuous voltage

  pulsed naked which is filtered by a filter capacitor 207. This voltage is applied to a terminal of a diode

  zener 208 by a current limiting resistor 210.

  The other terminal of the zener diode 208 is connected to the mas-

  to provide a practical reference voltage

  constant force that is applied to the other end of

the control winding 188.

  With this configuration, the voltage across the control winding 188, which determines the inductance of the controlled windings 190 and 192, is the difference between the control voltage at the connection point of the

  diode 184 and the detection resistance 194, and the

  reference in 212. The value of the reference is selected.

  194 and the level of the zener voltage so as to

  miter control voltage at 212 to a range of values

  not exceeding the reference voltage, on the desired range

  of the magnetron current. Since the voltage applied to the control terminal of winding 188 is always less than or equal to the reference voltage, and

  because it varies according to the direct current of the

  gnetron, the voltage across the control winding 188, and thus the current in the winding 188, varies in

  inverse function of the magnetron current.

  As described in connection with FIG.

  vary the power that is applied to the primary winding

  mayor 60 by varying the pulse width that the

  Inverter circuit 22 (FIG. 2) applies to the primary winding

  mayor. When the pulse width increases, the current

  magnetron increases. When the magnetron current increases

  mente, the voltage at the connection point of the resistance 194

  and diode 184 increases proportionally. This decreases

  the difference in voltage across the winding

  188, which proportionally increases the impedance

  this controlled windings 190 and 192, connected in series

  with the magnetron filament, which has the effect of reducing

  re the current of the filament. Since the filament is es-

  sensitively a resistive element, reducing the current

  proportionally reduces the power of the filament. By way of

  Similarly, a decrease in the power that is applied

  the primary winding, resulting for example from reducing the width of the pulses which are applied to the primary winding, reduces the magnetron current. The control voltage is reduced proportionally, which

  increases the voltage difference across the windings

  188. This increase in the voltage at the terminals of the control winding decreases the impedance of the

  controlled windings 190 and 192, which increases the power

  which is applied to the filament. The characteristics of reactance 186, when employed in this manner, help to stabilize the filament tension against variations that would otherwise appear

  when the pulse width that is applied is varied

to the primary winding 60.

  Physical configuration of the circuit Figure 7 shows a characteristic structure

  using the double-wave bridge inverter on the inside

  a microwave oven. A printed circuit board 216 carries the electronic components for switching,

  comprising the four semiconductor switching elements

  46F, 46S, 46T and 46R. Radiators associated with the switching transistors, the modulator control chip

  pulse width and associated control components.

  The input filtering capacitor of the inverter, the isolation transformer for gate attack, and a power relay, which are not separately identified in FIG. 10, would also be mounted on the board.

circuit 216.

  To reduce to the lowest level possible

  operating temperatures of the switching transistors

  The radiators of the transistors are mounted in offset positions and are cooled by forced circulation.

  of air, being placed in the air circulation of the ventila-

  cooling fan 218 of the magnetron. Room 220 of

  microwave oven guides the forced circulation of air from

  cooling so that this air passes over all

  radiating faces, which are simply

  shown in FIG. 10 by the blocks located above the circuit circuit 216. Part of the forced air flow is directed to the power transformer 26. As shown, the power transformer 26 is placed very close to the power transformer 26. printed circuit board 216 to reduce the risk of

parasitic ductance.

  Figures 8, 9 and 10 show detailed views of

  radiator 219 that can be used in the configuration

  Figure 7. This radiator is used to establish

  appropriate operating conditions for

  switching sistors. Since the temperature of

  threshold for particular switching transistors

  appears at the geometric center of the chip, it is visible

  to evacuate this heat. It is also necessary to

  provide a sufficient thermal mass to avoid

  thermal cursors while the transistors go through

  conduction and blocking cycles. A mount plate

  ge 220 for a TO 220 is in thermal contact with a main thermal path 222 from which extend

  a series of cooling fins 224.

  It should be noted that various specific values of circuit elements, various specific structures and

  other details have been indicated only for the purpose of illustrating

tration.

Claims (20)

  1. Microwave energy generation system   characterized in that it comprises: a magnetron (32) capable of producing microwave energy for cooking; a power transformer (26) having a primary winding (60) and a secondary power supply winding   magnetron (64) which is connected to provide.   magnetron energy (32); a full wave bridge inverter circuit (22) having first and second input lines (42, 44) and first and second output lines (48, 50), with the first output line (48) connected at a first end of the primary winding   (60), and the second output line (50) connected to a second   primary end of the primary winding (60), said inverter circuit (22) comprising first, second, third and fourth semiconductor switching elements (46F,   46S, 46T, 46R), each of these switching elements to se-   comprising an input terminal, an output terminal and a control terminal, each control terminal being operable to close and open its element   associated semiconductor switching circuit, this   their (22) further comprising a control circuit (24) for controlling all the semiconductor switching elements (46F, 46S, 46T, 46R) by their respective terminals.   order so that during a first series of inter-   time: the first semiconductor switching element   conductors (46F) is closed to connect functional-   the first line of entry (42) to the second line of   output (50) to the fourth semicon-   (46R) is closed to functionally connect the second input line (44) to the first output line (48), and the second and third semiconductor switching elements (46S, 46T) are open,   first and fourth semiconductor switching elements   (46F, 46R) circulating a current in a first   first direction in the primary winding (60); and so   that during a series of second intervals of time,   with the first time slots: the second semiconductor switching element (46S) is closed to functionally connect the second input line (44) to the second output line (50), the third element   semiconductor switching circuit (46T) is closed to   functionally connecting the first input line (42) to the first output line (48), and the first and fourth semiconductor switching elements (46F, 46R) are open, closing the second and third semiconductor switching elements (46S, 46T) flowing a current in a second direction, opposite the first direction, into the primary winding (60); and in that the first and second semiconductor switching elements   (46F, 46S) extend in series between the first line of   gate (42) and the second input line (44), and the third and fourth semiconductor switching elements (46T, 46R) extend in series between the first input line   (42) and the second input line (44), and that the circuit   cooked control (24) receives a control input signal   representative of the current in the magnetron (32), and the circuit   control furnace (24) switches to pairs of the semiconductor switching elements (46F, 46S,   46T, 46R) at a time which depends on the input signal of mande.   2. Microwave energy generation system se-   lon claim 1, characterized in that it further comprises a diode (58) in parallel on each of the elements of   semiconductor switching (46F, 46S, 46T, 46R).   3. Microwave energy generation system se-   according to claim 1, characterized in that it further comprises a bridge rectifier (18R) having AC voltage input lines and capable of applying to the first and second input lines (42, 44) of the circuit inverter (22)   a DC voltage having a minimum level during operation   tion, and a ripple which is produced by the AC voltage on the input lines, this ripple having an amplitude at least as large as the minimum voltage level. 4. Microwave power generation system according to   Claim 3, characterized in that the connecting terminals   the first and third switching elements to se-   (46F, 46T) are floating in relation to se-   cond and fourth semiconductor switching elements   (46S, 46R), this system further comprising an attack circuit   than isolated (52, 54F, 54S; 152), and in that the circuit of   control (24) is able to control the first and third   th semiconductor switching elements (46F, 46T),   in that the isolated driver circuit applies   at the control terminals of the first and third   semiconductor switching elements (46F, 46T).   5. Microwave energy generation system according to claim 3, characterized in that the transformer   (26) further comprises a secondary winding of   filament (72) which is connected to a filament of the gnetron (32).   6. Microwave power generation system according to   claim 1, characterized in that the communication circuit   mande (24) generates control pulses for switching the semiconductor switching elements (46F, 46S, 46T, 46R) at a frequency greater than 20000 Hz, such that the frequency of the current in the primary winding (60) is greater than 20000 Hz.   7. Microwave power generation system according to   claim 6, characterized in that the communication circuit   Mande (24) comprises a pulse width modulator (24C).   8. Microwave power generation system according to   claim 7, characterized in that the communication circuit   Mande (24) is capable of closing pairs of the semiconductor switching elements (46F, 46S, 46T, 46R) during   a duration that depends on the command input signal.   9. Microwave energy generation system se-   according to claim 8, characterized in that the control circuit (24) comprises a variable power control resistor (98), and the control input signal depends on   setting of this variable resistance resistor (98).   10. Microwave energy generation system se-   1, characterized in that each of the semiconductor switching elements (46F, 46S, 46T, 46R) 'is a field effect transistor, and the winding   the primary winding (60) is the only primary winding of the power meter (26).   11. Microwave energy generation system   characterized in that it comprises: a magnetron (32) capable of producing microwave energy for cooking; a   power transformer (26) having a primary winding   mayor (60) and a magnetron secondary supply winding (64) which is connected to provide power to the magnetron (32); a full wave bridge inverter circuit (22) having first and second input lines (42, 44) and first and second output lines (48, 50), with the first output line (48) connected at a first end of the primary winding   (50), and the second output line (50) connected to a second   end of the primary winding (60), said inverter circuit (22) comprising first, second, third and   fourth switching elements (46F, 46S, 46T, 46R), each   one of these switching elements comprising a terminal   input terminal, an output terminal and a control terminal, each   that control terminal can be activated to close   and open its associated switching element, this circuit on-   converter (22) further comprising a control circuit (24) for controlling all the switching elements (46F, 46S, 46T, 46R) by their control terminals so that during a first series of switching intervals time: the   first switching element (46F) is closed for connection   functionally the first input line (42) to the   second output line (60), the fourth communication element   (46R) is closed to functionally connect the second input line (44) to the first output line (48), and the second and third switching elements.   (46S, 46T) be opened, the closure of the first and fourth   third switching elements (46F, 46R) flowing a current in a first direction in the primary winding (60); and so that during a series of seconds   intervals of time, alternating with the first intervals   the time: the second switching element (46S) is closed to functionally connect the second line   input (44) to the second output line (50), the third   sth switching element (46T) is closed to connect   functionally the first input line (42) to the first   the first line of exit (48), and the first and fourth   switching elements (46F, 46R) are open, closing second and third switching elements (46S, 46T)   circulating a current in a second direction, op-   placed in the first direction, in the primary winding   (60); and in that the first and second   tation (46F, 46S) extend in series between the first   input pin (42) and the second input line (44), and the third and fourth switching elements (46T, 46R) extend in series between the first input line (42) and the second line input (44); and in that the control terminals of the first and third switching elements   (46F, 46T) float relative to the second and fourth   switching devices (46S, 46R), the system further comprising an isolated driver (52, 54F, 54S; 152); and in that the control circuit (24) is capable of controlling the first and third switching elements (46F, 46T)   in that the isolated driver circuit applies   at the control terminals of the first and third switching elements (46F, 46T).   Microwave power generation system according to claim 11, characterized in that the isolated driving circuit comprises an isolation transformer (52, 54F, 54S) having a primary winding (52) connected to the   control circuit (24), and having two secondary windings   (54F, 54S), with each secondary winding of the isolation transformer connected to the control terminal of a corresponding one of the first and third elements of switching (46F, 46T).   13. Microwave energy generation system se-   according to claim 11, characterized in that the insulated driver (152) comprises a first capacitor (154) which charges when the second switching element (46S) is closed, and the load of the first capacitor (154).   closes the first switching element (46F) under the   control circuit dance (24); and in that the circuit   of isolated attack includes a second capacitor   when the fourth switching element (46R) is   closed, and the charge of this second capacitor closes the third   second switching element (46T) dependent on the control circuit.   14. Microwave energy generation system se-   Claim 11, characterized in that the control circuit (24) comprises a pulse width modulator (24C).   15. Microwave energy generation system se-   according to claim 14, characterized in that the control circuit (24) receives a control input signal which is   representative of the current in the magnetron (32), and this circuit   control furnace (24) is capable of switching the pairs of switching elements (46F, 46S, 46T, 46R) to an open state when the control input signal reaches a   predetermined level; and in that each of the elements of   mutation (46F, 46S, 46T, 46R) is a switching element semiconductor.   16. A microwave energy generation system characterized by comprising: a magnetron (32) capable of producing microwave energy for cooking; a   power transformer (26) having a primary winding   re (60) and a secondary magnet power winding   tron (64) which is connected to provide power to the   magnetron (32); a double-bridge bridge inverter circuit   (22), having first and second input lines (42, 44), and first and second output lines (48, 50)   with the first output line (48) connected to a first   first end of the primary winding (60), and the second output line (50) connected to a second end of   the primary winding (60), this inverter circuit (22) comprises   first, second, third and fourth switching elements (46F, 46S, 46T, 46R), each of these elements of   switching having an input terminal, a output terminal   and a control terminal, each control terminal can   being activated to close and open its associated switching element, said inverter circuit (22) further comprising a control circuit (24) for controlling all the switching elements (46F, 46S, 46T, 46R) by their control terminals so that during a first   series of time intervals: the first element of   (46F) is closed to functionally connect the first input line (42) to the second output line (50), the fourth switching element (46R) is closed to functionally connect the second input line (44). at the first output line (48), and the second and third switching elements (46S, 46T) are open, closing the first and fourth switching elements   (46F, 46R) circulating a current in a first direction   rection in the primary winding (60); and so that during a series of second time intervals, alternating with the first time intervals: the second element   switch (46S) is closed to connect functional-   the second input line (44) to the second output line (50), the third switching element (46T) is closed to functionally connect the first line   (42) to the first output line (48), and the first   first and fourth switching elements (46F, 46R) are open, closing the second and third switching elements (46S, 46T) to circulate a current in a second direction, opposite to the first direction, in   the primary winding (60); and in that the first and second   cond switching elements (46F, 46S) extend in series between the first input line (42) and the second line   (44), and the third and fourth elements of   mutations (46T, 46R) extend in series between the first   input signal (42) and the second input line (44), and in that the control circuit (24) receives a control input signal representative of the current in the magnetron (32), and the circuit controller (24) switches to an open state the pairs of switching elements (46F, 46S, 46T, 46R) at a   moment that depends on the command input signal.   17. Microwave energy generation system se-   according to claim 16, characterized in that the control circuit (24) comprises a variable power control resistor (98), and the control input signal depends on the setting of said variable power control resistor (98). .   18. Microwave energy generation system se-   according to claim 16, characterized in that the control circuit (24) comprises a pulse width modulator (24C).   19. Microwave energy generation system se-   according to claim 18, characterized in that the control circuit (24) further comprises a power adjusting device (98) which can be set to change the power of the magnetron (32), changing the length of the first and second time intervals.   20. Microwave power generation system according to claim 16, characterized in that the control terminals of the first and third switching elements (46F, 46T) float relative to the second and fourth switching elements (46S). , 46R), this system further comprising an isolated driver (52, 54F, 54S; 152); and in that the control circuit (24) is capable of controlling the first and third switching elements (46F, 46T)   in that the isolated driving circuit applies   at the control terminals of the first and third   switching elements (46F, 46T); and in that each of the switching elements (46F, 46S, 46T, 46R) is an element semiconductor switching.