RU2402149C1 - Key power amplifier - Google Patents

Abstract

FIELD: radio engineering.

SUBSTANCE: key power amplifier (KPA) comprises active key element (AKE), serially connected the first inductance L₁ and the second capacitor C₂, which form a serial generating L₁C₂ - circuit, parallel connected the first capacitor C₂, the second inductance L₂ and active load R, forming parallel L₂C₁R - circuit, and source of supply voltage V_ds with throttle (Th) and blocking capacitor C_bl for isolation of HF circuits by supply. Inlet of AKE is device inlet. AKE inlet is connected to the first inductance L₁, the second output of which, at one side, via the second capacitor C₂ is connected to output of KPA, and at the other side - via throttle Th is connected simultaneously to the first terminal of supply source V_ds and to the first output of blocking capacitor C_bl. The second terminal of supply source V_ds and the second output of blocking capacitor C_bl are connected both to common point of AKE, being simultaneously a common bus of KPA, and to common point of parallel L₂C₁R - circuit, other common point of which is output of KPA. Values of main elements satisfy the following ratios

where Q_L - quality of loaded circuit, ω_s - angle frequency of input signal, V_ds - voltage of supply source, P - level of output signal capacity. Using values of main elements calculated in compliance with given formulas in amplifier fully realises mode of its operation in amplifier in class "E", when it is possible to produce efficiency factor, which is theoretically equal to 100%. Such optimal mode of class "E" in amplifier may be provided at higher frequencies compared to available devices.

EFFECT: increased maximum frequency of operation with maximum possible efficiency.

8 dwg

Description

The proposed device relates to the field of radio engineering and can be used in highly economical radio transmitters operating in frequency ranges up to the microwave range.

Known power amplifier operating in key mode, which is called class "E". (See Sokal NO, Sokal AD Class E - a new class of high-efficiency tuned single-ended switching power amplifiers // IEEE Jour, of Solid-State Circuits, Vol.SC-10, No. 3. June 1975, P . 168-176). This mode of operation provides the amplifier with a theoretical efficiency of 100%. The amplifier (see Fig. 1) contains an active key element, a capacitor shunting it, a series LC circuit, an active load, and a power supply with a choke for decoupling the RF circuits from the power supply.

. The disadvantage of this class “E” amplifier is the low maximum frequency of its operation at the highest possible efficiency. When this amplifier is operating in the optimum mode of efficiency, its maximum frequency f _{max is} calculated by the formula:

.

Hereinafter, V _ds is the voltage of the power source of the amplifier, I _max is the maximum value of the key current, C _ds is the linear output capacity of the key.

Known power amplifier operating in class "E" mode with a shunt key inductance. (See Kazimierczuk M. Class E tuned power amplifier with shunt inductor // IEEE Jour. Of Solid-State Circuits, Vol. SC-16, No. 1. February 1981. P.2-7). The amplifier (see Fig. 2) contains an active key element, shunting its inductance L, series L _r C _f - circuit, active load and power supply source.

. A disadvantage of the known device is the low maximum frequency of its operation at the highest possible efficiency. When this amplifier is operating in the optimum mode of efficiency, its maximum frequency f _{max is} calculated by the formula:

.

Hereinafter, I _ds is the constant component of the amplifier current, V _max is the maximum value of the key voltage.

Also known is a power amplifier operating in class "E" with a shunt key LC circuit. (See Grebennikov AV, Jaeger H. Class E with parallel circuit - a new challenge for high-efficiency RF and microwave power amplifiers // IEEE MTT-S Int. Microwave Symp. Dig., 2002, TJ2D-1, P.1627 -1630). The amplifier (see Fig. 3) contains an active key element, shunting its LC circuit, a serial L ₀ C ₀ circuit, an active load and a power supply source.

. The disadvantage of this device is the low maximum frequency of its operation at the highest possible efficiency. When this amplifier is operating in the optimum mode of efficiency, its maximum frequency f _{max is} calculated by the formula:

.

Here the output power is P = I _ds V _ds (with efficiency = 100%), and I _max = 2,647I _ds .

In addition, a class “E” power amplifier with a shunt key capacitance is known. (See Fig. 2.19 b) in the book: Improving the Efficiency of Powerful Radio Transmitting Devices / A.D. Artym, A.E. Bakhmutsky, E.V. Kozin et al. Ed. A.D. Artyma. - M.: Radio and Communications, 1987. - 176 p.). The amplifier (see Fig. 4) contains an active key element, a capacitor C ₀ shunting the key, which forms a forming circuit with inductance L ₀ , a blocking (isolation) capacitor C _bl filtering parallel L _K C _K circuit, active load R _N and power supply with a choke for decoupling the RF circuits for power.

, которая получена, используя выражение (1.55) и Таблицу 2.1 в указанной выше книге.The disadvantage of this device is the low maximum frequency of its operation at the highest possible efficiency. When this amplifier is operating in the optimum mode of efficiency, its maximum frequency f _{max is} calculated by the formula:

, which is obtained using the expression (1.55) and Table 2.1 in the above book.

Closest to the proposed technical solution is a power amplifier (See RF patent for the invention No. 2306667 with priority dated 12/22/2005, H03F 3/193. Key power amplifier / A. Baranov). This amplifier also works in class "E", and its circuit is dual to the classical amplifier circuit Sokal NO and Sokal AD. The amplifier (see Fig. 5) contains an active key element, a series inductance L ₁ isolation capacitor C _bl , parallel to L ₂ CR-circuit with active load R and a power supply voltage V _ds with a choke Dr for decoupling the RF circuits for power.

. The disadvantage of the prototype device is the low maximum frequency of its operation at the highest possible efficiency. When this amplifier is operating in the optimum mode of efficiency, its maximum frequency f _{max is} calculated by the formula:

.

The technical effect to which the proposed solution is aimed is to increase the maximum frequency of its operation at the highest possible efficiency.

где Q_L - добротность нагруженного контура, ω_S - угловая частота входного сигнала, согласно изобретению между выходом усилителя и общей точкой первой индуктивности L₁ и дросселя введен второй конденсатор С₂, который вместе с первой индуктивностью L₁ образует последовательный формирующий L₁C₂-контур, блокировочный конденсатор C_бл подключен между общей шиной усилителя и первой клеммой источника питания, а величины основных элементов удовлетворяют следующим соотношениям:This effect is achieved by the fact that in a key power amplifier containing an active key element, the first point of which is its output, the second is its input and simultaneously the input of the amplifier, and the third is its common point and the common bus of the amplifier as a whole, the first inductance L ₁ , one terminal of which is connected to the output of the active core element and the other through a throttle - to the first power supply terminal, the second terminal of which is connected to the common bus amplifier, C _plaque blocking capacitor and the parallel connection of the second I inductance L _2, the first capacitor C _1, resistive load R, which form a parallel L ₂ C ₁ R-circuit, a common output which is the output of the amplifier, and the second - is connected to the common bus amplifier, the inductance value L ₂ is defined by:

where Q _L is the Q factor of the loaded circuit, ω _S is the angular frequency of the input signal, according to the invention, a second capacitor C _{2 is} introduced between the output of the amplifier and the common point of the first inductance L ₁ and the inductor, which together with the first inductance L ₁ forms a series forming L ₁ C ₂ -contour, blocking capacitor C _bl connected between the common bus of the amplifier and the first terminal of the power source, and the values of the main elements satisfy the following relationships:

,

,

,

,

,

,

,

,

where V _ds is the voltage of the power source, P is the power level of the output signal.

A schematic diagram of the proposed device is presented in Fig.6. The amplifier contains an active key element, serial to the key L ₁ C ₂ circuit, a parallel L ₂ C ₁ R circuit with an active load R and a power supply voltage V _ds with a choke Dr and a blocking capacitor C _bl for decoupling the RF circuits for power.

, где A₀ вычисляется из начального условия: V_S(θ)=0 при θ=0. В результате при 0≤θ<π

. Во включенном состоянии (π≤θ<2π) напряжение на ключе V_S(θ)=0, тогда напряжение на индуктивности L₁-V_L(θ)=V_C(θ)+V_R(θ) можно записать в виде:

. Здесь

вычисляется из начального условия V_L(θ)=V_C(θ) при θ=π. Полученное уравнение может быть представлено в форме дифференциального уравнения второго порядка:

с общим решением вида:The proposed amplifier operates as follows. The principle of operation is explained by the waveforms shown in Fig.7. These stress plots are obtained as follows. Let the instantaneous voltage V _R (θ) on the resistor R be of the form: V _R (θ) = a ₀ V _ds sin (θ + φ), where θ = ω _S t, ω _S is the angular frequency of the input signal, t is the current time , and a ₀ and φ are the quantities defined below. When the switch is open (during 0≤θ <π), the current through the switch I _S (θ) = I _ds -I _C (θ) = 0, therefore, the current I _C (θ) = I _ds flows through the capacitor C ₂ . Since at 0≤θ <π the current does not flow through the switch, there is no voltage across the inductance L ₁ : V _L (θ) = 0. Then the voltage on the key is V _S (θ) = V _C (θ) + V _R (θ), and the voltage on the capacitor

, where A _{0 is} calculated from the initial condition: V _S (θ) = 0 for θ = 0. As a result, for 0≤θ <π

. In the on state (π≤θ <2π), the key voltage V _S (θ) = 0, then the voltage across the inductance L ₁ -V _L (θ) = V _C (θ) + V _R (θ) can be written in the form:

. Here

is calculated from the initial condition V _L (θ) = V _C (θ) for θ = π. The resulting equation can be represented in the form of a second-order differential equation:

with a general solution of the form:

,

,

, A₁=a₁I_ds, A₂=a₂I_ds, а a₁ и a₂ - некоторые константы, определяемые ниже. Для того чтобы найти неизвестные константы: q, a₀, а₁, a₂ и φ, необходимы пять уравнений:Where

, A ₁ = a ₁ I _ds , A ₂ = a ₂ I _ds , and a ₁ and a ₂ are some constants defined below. In order to find the unknown constants: q, a ₀ , and ₁ , a ₂ and φ, five equations are needed:

,

,

,

,

,

,

,

,

.

.

Equations (2) and (3) are the initial conditions. Equation (3) is the condition for the constant component of the current I _ds expressed in terms of the Fourier series. And expressions (5) and (6) are the main conditions for the optimal operation of this amplifier in class "E". The last two conditions are necessary to achieve the maximum possible efficiency of the device - 100%, that is, in order for the voltage V _S (θ) and the current flowing through the switch, I _S (θ) not to have significant values at the same time, and their the product I _S (θ) V _S (θ) was equal to zero.

As a result, the numerical solution of a system of five equations with five unknowns gives the following values:

, то можно вычислить a₀:If we add to equation (11) the condition for the constant component of the voltage V _ds , expressed through the Fourier series -

, then we can calculate a ₀ :

It is with these values of the constants (7) - (10), (12) in Fig. 7 that the voltage and current plots are plotted, normalized to the values of V _ds and I _ds . Formulas for the main waveforms can be rewritten in the form:

The obtained equations (14), (15) are dual with respect to the wave forms for the amplifier in the class “E” with a forming circuit parallel to the key. (See generator: Kozyrev VB. Single-cycle key generator with a filter circuit. - In the book: Semiconductor devices in the telecommunication technique. Edited by I.F. Nikolayevsky. - M.: Communication, 1971 - Issue 8. - S.152-166 and a similar device: Grebennikov AV, Jaeger N. Class E with parallel circuit - a new challenge for high-efficiency RF and microwave power amplifiers // IEEE MTT-S Int. Microwave Symp. Dig., 2002, TJ2D-1. P.1627-1630). Therefore, the proposed scheme is dual with respect to both schemes.

If we expand the voltage V _S (θ) in a Fourier series, then its active V _R and reactive V _X amplitudes can be found from the expressions:

Then the phase angle between the current of the fundamental frequency I _S1 (θ) and voltage V _S1 (θ) is equal to:

, можно получить оптимальные величины элементов L₁, C₂, R формирующего L₁C₂R-контура:Given the found values q = 1.412, a ₀ = 1.211 and the fact that the phase angle can be represented as a function of the elements of the output circuit -

, you can get the optimal values of the elements L ₁ , C ₂ , R forming the L ₁ C ₂ R-circuit:

, то величину С₁ можно найти по формуле:Given that the inductance value L _{2 is} determined through the quality factor Q _{L of the} loaded filter L ₂ C ₁ circuit in the form:

, then the value of C ₁ can be found by the formula:

Thus, if the elements in the amplifier are calculated according to formulas (17) - (20), then the class “E” mode of operation will be fully implemented in the amplifier. If at the same time we consider the active key element to be ideal, then in such an amplifier 100% efficiency can be obtained, a P = I _ds V _ds . Using equations (14) and (15), calculating the peak values of current I _max and voltage V _max : I _max = 3,647I _ds V _max = 2,647V _ds , the maximum frequency of the amplifier operation can be determined from equations (17) - (19) in such an optimal mode in two equivalent forms:

и

- для известных устройств и

- для предлагаемого усилителя. Следовательно, даже при такой грубой оценке, когда величина С₂ выбрана равной С_ds, максимальная частота предлагаемого усилителя выше в 1,45 и 2,47 раза по сравнению с f_max рассмотренных аналогов (схемы Козырева В.Б. и схемы Sokal N.O. и Sokal A.D.). Очевидно, что, если величину C₂ выбрать меньшей С_ds, то отличие f_max этих устройств будет еще больше.Since the output impedance of all transistor switches is capacitive in nature, it is the value of the output capacitance of the key C _ds in the parallel (or serial) output circuit that gives the main limitation on the operating frequency of most class “E” amplifiers. For example, to increase f _max in the circuits of amplifiers of class "E" Kozyreva VB and Sokal NO, Sokal AD, it is not possible to set the output capacitance value lower than C _ds . In the proposed device, the output capacity of the key C _ds does not increase, but decreases the total capacity of the forming circuit. In fact, the value of C ₂ in the formula (21) decreases, and f _max additionally increases. The output capacitance of the key With _ds here can also be fully compensated with the inductance L ₁ . Thus, in the proposed amplifier, the output capacitance of the key C _ds (in the serial output type circuit) is not involved in the operation of the forming circuit, af _{max is} not limited to it. Here, the frequency boundary for the optimal implementation of an amplifier of class “E” f _max can be moved to a higher frequency region, since there is no practical limit to reduce the values of C ₂ and L ₁ in formulas (21) and (22). The decrease in C ₂ and L _{1 is} limited only by the possibility of their practical feasibility. Although the value of the capacitance C ₂ can be chosen arbitrarily small, for a rough comparison of f _{max of the} proposed amplifier and f _{max of the} known devices (Kozyrev VB schemes and Sokal NO, Sokal AD), we choose it equal to C _ds . If, in addition, the supply voltage and the maximum current of the switch in the proposed device are the same as V _ds , I _max in the schemes of V. Kozyrev and Sokal NO, Sokal AD, then by the numerical coefficients in the formulas for f _max you can judge the difference f _{max of} these devices. These ratios are as follows:

and

- for known devices and

- for the proposed amplifier. Therefore, even with such a rough estimate, when the value of C _{2 is} chosen equal to C _ds , the maximum frequency of the proposed amplifier is 1.45 and 2.47 times higher than the f _{max of the} considered analogues (V. Kozyrev’s scheme and Sokal NO and Sokal AD). Obviously, if the value of C _{2 is} chosen less than C _ds , then the difference f _{max of} these devices will be even greater.

и

) в формулах для f_max можно также судить об отличии максимальных частот этих устройств. Следовательно, даже при такой грубой оценке, когда величина С₂ выбрана равной С_ds, f_max предлагаемого усилителя выше в 1,33 раза. На практике же величина С₂ может быть выбрана меньшей С_ds, тогда отличие f_max этих устройств будет еще больше.Compare the f _{max of the} proposed amplifier and f _max device (See the book: Improving the efficiency of powerful radio transmitting devices / A.D. Artym, A.E. Bakhmutsky, E.V. Kozin, etc. Edited by A.D. Artym. - M .: Radio and communications, 1987. - 176 p.). In this analogue, it is also impossible to set the value of the output capacitance C ₀ (See figure 4), less C _ds . However, in the proposed device, the output capacity of the key C _ds does not increase, but decreases the total capacity C _{2 of the} forming circuit. This means that the frequency boundary for the optimal implementation of the class “E” amplifier f _max can here be moved to a higher frequency region, since the values of C ₂ and L ₁ in formulas (21) and (22) can be reduced to those values that In principle, they can be implemented in practice. Although the value of the capacitance C ₂ can be chosen arbitrarily small, for a rough comparison of the maximum frequencies of the proposed amplifier and the circuit Artyma A.D. we choose it equal to C _ds . If, in addition, in both amplifiers V _ds and I _{max are} chosen the same, then by numerical coefficients (

and

) in the formulas for f _max you can also judge the difference in the maximum frequencies of these devices. Therefore, even with such a rough estimate, when the value of C _{2 is} chosen equal to C _ds , f _{max of the} proposed amplifier is 1.33 times higher. In practice, the value of C ₂ can be chosen less than C _ds , then the difference f _{max of} these devices will be even greater.

и

) в формулах для f_max также можно судить об отличии f_max этих устройств. Следовательно, по сравнению с максимальными частотами рассмотренного аналога и прототипа f_max предлагаемого усилителя здесь выше в 1,71 раза.In order to compare the f _{max of the} proposed amplifier with the maximum frequencies of other known devices (See Kazimierczuk M. Class E tuned power amplifier with shunt inductor // IEEE Jour. Of Solid-State Circuits, Vol.SC-16, No. 1. February 1981, P.2-7 and RF patent for invention No. 2306667 with priority dated December 22, 2005, H03F 3/193. Key power amplifier / Baranov A.V.), we use the equivalent (21) formula (22) for f _max . We choose the values of L (See figure 2) and L ₁ (See figure 5 and 6) the same and at the same time arbitrarily small, since they are all equally bounded from below. If in addition the current consumption and the maximum voltage on the key in the proposed device are the same as I _ds , V _max in the prototype and amplifier Kazimierczuk M., then by the numerical coefficients (

and

) in the formulas for f _max you can also judge the difference f _{max of} these devices. Therefore, compared with the maximum frequencies of the considered analogue and prototype f _{max of the} proposed amplifier here is 1.71 times higher.

Thus, the optimal mode of class "E" in the proposed amplifier in comparison with known devices can be achieved at higher frequencies.

An example of a specific implementation of the device.

,

. Здесь f=ω_S/2π,

,

и

,

- электрические длины (в градусах) и волновые сопротивления микрополосковых линий передач, которые соответствуют индуктивности L₁ емкости С₁. Если для пересчета L₁ и C₁ выбрать

=50 Омам и

=40 Омам, то

=23.2° и

=15.9°. Отсюда нетрудно найти длины микрополосков

и

, где

,

- длины волн в соответствующих микрополосковых линиях передач. Индуктивность L₂=25,4 нГн, емкость С₂=4,3 пФ, C_бл=1000 пФ и дроссель Др с индуктивностью 250 нГн в макете усилителя удобно выполнить в виде сосредоточенных элементов. Следует отметить, что здесь величина С₂ выбрана большей 4,3 пФ (С₂=10 пФ) для того, чтобы скомпенсировать выходную емкость ключа C_ds (в выходной цепи последовательного типа). Выходная емкость ключа C_ds (в цепи последовательного типа) составляет 10-12 пФ при той настройке входной цепи, которая использована в предлагаемом макете. Для оценки C_ds здесь использована методика, предложенная в работе Баранова А.В. «Дуальные СВЧ-усилители мощности в классе «Е»». - Радиотехника, 2007, №3, с.71-78. Все элементы были реализованы в макете усилителя (См. фиг.8). Напряжение на затворе транзисторного ключа выбрано приблизительно на (0,3-0.6) В меньше, чем напряжение отпирания транзистора. Кроме того, уровень входного сигнала был выбран таким, чтобы транзистор работал в режиме насыщения. Эти условия необходимы для того, чтобы транзистор ближе соответствовал идеальному ключу.The power amplifier is made on an LDMOSFET transistor type MRF282S from Motorola (now Freescale Semiconductor). The amplifier operates at a frequency of 920 MHz. If you choose V _ds = 20.0 V, P = 10 W and Q _L = 5, then the calculated values of the amplifier elements are as follows: L ₁ = 3.5 nH, C ₂ = 4.3 pF, C ₁ = 1.2 pF, L ₂ = 25.4 nH; R = 29.3 Ohm. The inductance L ₁ and capacitance C _{1 were} converted into segments of microstrip transmission lines implemented on a substrate of Al ₂ O ₃ ceramic with a dielectric constant of 9.8. Recalculation was carried out using well-known formulas:

,

. Here f = ω _S / 2π,

,

and

,

- the electrical lengths (in degrees) and wave impedances of the microstrip transmission lines, which correspond to the inductance L _{1 of the} capacitance C ₁ . If for the conversion of L ₁ and C ₁ choose

= 50 ohms and

= 40 ohms then

= 23.2 ° and

= 15.9 °. From here it is not difficult to find the lengths of microstrips

and

where

,

- wavelengths in the respective microstrip transmission lines. Inductance L ₂ = 25.4 nH, capacitance C ₂ = 4.3 pF, C _bl = 1000 pF and the inductor Dr with an inductance of 250 nH in the amplifier layout is conveniently performed in the form of lumped elements. It should be noted that here the value of C _{2 is} chosen to be greater than 4.3 pF (C ₂ = 10 pF) in order to compensate for the output capacitance of the key C _ds (in the output circuit of a sequential type). The output capacitance of the key C _ds (in the serial type circuit) is 10-12 pF with the input circuit setting used in the proposed layout. To evaluate C _ds, we used the technique proposed in the work of A. Baranov. "Dual microwave power amplifiers in the class" E "." - Radio engineering, 2007, No. 3, p. 71-78. All elements were implemented in the layout of the amplifier (See Fig. 8). The voltage at the gate of the transistor switch is selected approximately (0.3-0.6) V less than the unlock voltage of the transistor. In addition, the input signal level was chosen so that the transistor operates in saturation mode. These conditions are necessary in order for the transistor to more closely match the ideal key.

The main characteristics of the developed amplifier were measured. When the input power P _I = 0.25 W, the supply voltage V _ds = 20.0 V and the current consumption I _ds = 0.51 A, the maximum value of the efficiency for the added power was obtained, the efficiency = 71.1%. Accordingly, the stock efficiency of this amplifier is slightly higher and equals 73.5%. The output power level P was 7.5 watts.

Thus, an example of a specific implementation of the proposed device confirms the possibility of obtaining high values of efficiency (more than 70%) in amplifiers of increased power in the microwave range. A comparison of the numerical coefficients in the formulas for the maximum frequency of operation of the proposed and similar devices in the “E” class modes proves that f _max in the proposed amplifier is 1.71 times higher than the highest frequency analogs and the prototype, in particular.

Claims (1)

A key power amplifier containing an active key element, the first point of which is its output, the second is its input and simultaneously the input of the amplifier, and the third is its common point and the common bus of the amplifier as a whole, the first inductance L ₁ , which is connected to the output of the active one output the key element, and through the inductor to the first terminal of the power source, the second terminal of which is connected to the common bus of the amplifier, the blocking capacitor C _bl and the second inductance L ₂ connected in parallel, the first condenser sator C ₁ , the active load R, which form a parallel L ₂ C ₁ R - circuit, one common output of which is the output of the amplifier, and the second is connected to the common bus of the amplifier, the inductance value L _{2 is} determined by the formula

,
where Q _L is the quality factor of the loaded circuit;
ω _S is the angular frequency of the input signal,
characterized in that between the output of the amplifier and the common point of the first inductance L ₁ and the inductor, a second capacitor C _{2 is} introduced, which together with the first inductance L ₁ forms a series circuit L ₁ C ₂ - circuit, a blocking capacitor C _bl connected between the common bus of the amplifier and the first power supply terminal, and the values of the main elements satisfy the following relationships:

,

,

,
where V _ds is the voltage of the power source, P is the power level of the output signal.

Images