Power amplifier
Patent Information
- Authority / Receiving Office
- EP · EP
- Patent Type
- Applications
- Current Assignee / Owner
- ZODIAC DATA SYSTEMS
- Filing Date
- 2024-07-03
- Publication Date
- 2026-05-13
AI Technical Summary
Existing multi-channel semiconductor power amplifiers, such as SSPA, face limitations in output power due to saturation effects and decreasing maximum achievable output power with increasing input signal frequency, particularly in applications like ground-to-satellite communication, and suffer from significant combination losses in combiners which degrade efficiency and introduce isolation issues between channels.
A power amplifier system with a radial type combiner using a configuration of sub-combiners, input couplers, and an output coupler, where each input coupler divides the amplified signal into two phase-shifted signals to be combined in separate sub-combiners, allowing for complete channel isolation while minimizing losses by using waveguides to propagate signals without dielectric losses.
The solution enables efficient combination of multiple channels with reduced power losses and ensures isolation between channels, enhancing the overall efficiency and power handling capacity of the amplifier system, particularly in high-frequency applications like satellite communication.
Smart Images

Figure FR2024050896_09012025_PF_FP_ABST
Abstract
Description
[0001] Power amplifier
[0002] TECHNICAL FIELD
[0003] The invention relates to a power amplifier, in particular a multi-channel semiconductor type power amplifier (known as "SSPA") suitable in particular for communication between a satellite and the ground.
[0004] STATE OF THE ART
[0005] Power amplification systems allow an initial input signal to be used to deliver an output signal that is amplified relative to the initial input signal.
[0006] These systems use, for example, semiconductor power amplifiers, known to those skilled in the art by the acronym SSPA for "Solid State Power Amplifier". However, the output power achievable by such SSPA amplifiers is limited by a saturation effect when the power of the input signal is too high. In addition, the maximum achievable output power decreases with increasing input signal frequency. As a result, with existing SSPA amplifiers, a single amplification path may not be sufficient to achieve the output powers required by certain applications, for example in ground-to-satellite communication.
[0007] Power amplification systems therefore generally comprise a divider for dividing the input signal into different amplifier channels, at least one elementary power amplifier per amplifier channel for amplifying the signal transmitted in the amplifier channel and a combiner for recombining the amplified signals leaving the amplifier channels so as to form the amplified output signal.
[0008] A combiner can be evaluated by its combining efficiency:
[0009] P s
[0010] T = — — x 100
[0011] PE
[0012] Where P s is the power at the output of the combiner of the amplified output signal and P Eis the total power of the amplified signals in each amplifier channel at the input of the combiner. Power losses between the input and output of the combiner, otherwise known as combination losses, have multiple causes and are evaluated by the general formula, in dB: For example, a combination loss of 1 dB corresponds to a combination efficiency of 80%.
[0013] There are three main combiner technologies: radial combiners, tree combiners and spatial combiners.
[0014] Tree combiners generally include several stages of binary adders allowing the different amplified signals coming out of the amplifier channels to be recombined two by two.
[0015] Tree amplifiers allow efficient combining of a small number of channels. However, combining losses increase rapidly with the number of combined channels and the length of the combining lines. The combining efficiency T of an n-channel tree power combiner with stage losses a dB , the value of a dB being fixed by the length of the lines and the ohmic losses due to the presence of resistances, is given by the following relation:
[0016] T = 100 X 10“ ln * a dB /
[0017] For example, a Wilkinson type power combiner has stage losses a dB around 0.7 dB in Ka band, which means that the combining efficiency of such a combiner cannot exceed 62% when it includes 8 combining channels.
[0018] Spatial amplifiers generally comprise several elementary amplifiers placed in parallel on a panel perpendicular to the direction of propagation of the input signal, or placed on several plates parallel to the direction of propagation of the input signal. The propagation of the input signal before and after the elementary amplifiers can be done via waveguides or beams generated by antennas. At the input of each elementary amplifier, transitions allow switching from these guided or beam propagation modes to planar propagation. After each elementary amplifier, a new transition allows switching back to a guided or beam propagation mode.
[0019] The combining efficiency of spatial amplifiers is independent of the number of combined channels. However, the power level admitted in these amplifiers is limited because of the difficulty of evacuating the heat generated by the elementary amplifiers. Furthermore, the losses by combining spatial amplifiers are generally significant because of the dispersion in phase and amplitude of the signals circulating in the different amplifier channels before their recombination. For example, the relationship between the combination losses by unbalance P (which is one of the causes of combination losses) and the unbalances in phase 0 and in amplitude A of an isolated combiner with two combination channels is given by the following formula: [ cos 6
[0020] P = 10 x log(0.5 + 1 + )
[0021] Thus, if the signals propagated in the two channels are 180° out of phase, no power is delivered to the output of the combiner.
[0022] Radial amplifiers generally comprise a plurality of combining and / or dividing paths that extend radially from an axis in a plane perpendicular to said axis.
[0023] Due to this radial geometry, the signals circulating in the different combination and / or division channels have the same phase and the same amplitude, so that the unbalance losses P are very low. In addition, the combination efficiency T is advantageously independent of the number of channels so that it is possible to combine a large number of channels from a large number of amplifiers and thus achieve very high powers.
[0024] A major problem with these radial combiners remains the lack of isolation between the combining channels: the presence of a short circuit or an open circuit on one of the combining channels results in a loss and oscillations on the signal transmitted to the output of the combiner.
[0025] One possible solution is to arrange an isolator on each combination channel. However, this solution is expensive and bulky and results in losses of the order of 0.2 dB.
[0026] Another solution is to add resistive septa in the central part of the combiner at the level of the recombination of the channels. However, a part of the amplified signal dissipates in the septa instead of being transmitted to the output by coupling between said amplified signal and these resistive septa. Such dissipation represents a loss of power to the combination so that the combination efficiency is degraded. BRIEF DESCRIPTION OF THE INVENTION
[0027] An aim of the invention is to have a power amplification system, in particular of the multi-channel semiconductor type (called "SSPA"), comprising a divider, amplifier channels and at least one combiner making it possible to combine a large number of channels. The desired amplifier preferably comprises a radial type combiner. The desired amplifier must further comprise a device making it possible to isolate the combination channels from each other, so that a degradation present on a given combination channel does not disturb the propagation and recombination of waves circulating in other combination channels remaining functional. A solution is desired that ensures the isolation of the combination channels of the amplifier without causing significant combination losses.
[0028] For this purpose, the invention proposes a power amplifier comprising:
[0029] - a power divider configured to receive an initial input electromagnetic signal and divide it into a plurality of elementary signals, the power of the initial input electromagnetic signal being distributed over the elementary signals,
[0030] - an amplifier comprising a plurality of amplifier channels, each amplifier channel being configured to amplify an elementary signal of the plurality of elementary signals,
[0031] - a power combiner configured to recombine the plurality of amplified elementary signals into an amplified initial electromagnetic signal, the power combiner comprising a first sub-combiner, a second sub-combiner, a plurality of input couplers and an output coupler, the first sub-combiner comprising:
[0032] - a first cylindrical cavity along an axis of symmetry (Z) forming a first radial waveguide, the first cylindrical cavity comprising a first plurality of inputs arranged in a plane (P) perpendicular to the axis of symmetry (Z),
[0033] - and a first plurality of rectangular waveguides, each waveguide of the first plurality of waveguides extending radially with respect to the axis of symmetry (Z) from an input of the first plurality of inputs, the second sub-combiner comprising:
[0034] - a second cylindrical cavity along the same axis of symmetry (Z) forming a second radial waveguide, the second cylindrical cavity comprising a second plurality of inputs arranged in a plane (P') perpendicular to the axis of symmetry (Z), the second plurality of inputs being superimposed with the first plurality of inputs,
[0035] - and a second plurality of rectangular waveguides, each waveguide of the second plurality of waveguides extending radially with respect to the axis of symmetry (Z) from an input of the second plurality of inputs, each input coupler being configured to receive an amplified elementary signal from an amplifying path and to divide said amplified elementary signal into a first signal intended to be transmitted in a waveguide of the first plurality of waveguides and into a second signal intended to be transmitted in a waveguide of the second plurality of waveguides, the first signal being in phase with the amplified elementary signal and the second signal being phase-shifted by a quarter of a wavelength with respect to the amplified elementary signal,the first sub-combiner being configured to combine the first signals transmitted in each waveguide of the first plurality of waveguides into a first combined signal and the second sub-combiner being configured to combine the second signals transmitted in each waveguide of the second plurality of waveguides into a second combined signal, the output coupler being configured to combine the first combined signal and the second combined signal into an output electromagnetic signal by inducing a phase shift of the first combined signal by a quarter wavelength, so that the output electromagnetic signal has an amplified power compared to the initial input electromagnetic signal.,
[0036] The system of input couplers and output couplers in waveguide advantageously allows the combination paths to be completely isolated from each other, while limiting combination losses. Indeed, an implementation using waveguides makes it possible to limit losses at the power combiner as much as possible. The electromagnetic signal propagates, in the nominal case, only in guiding structures filled with air and enclosed by walls. No dielectric loss is therefore introduced into the transmitted signal.
[0037] According to other optional features of the invention taken alone or in combination when technically possible:
[0038] - each input coupler and the output coupler are “branchline” type couplers;
[0039] - at least one input coupler is a rectangular waveguide coupler extending, radially with respect to the axis of symmetry (Z), from a waveguide of the first plurality of waveguides and a waveguide of the second plurality of waveguides;
[0040] - the power amplifier further comprises at least a first transition between an input coupler of the plurality of input couplers and the waveguide of the first plurality of waveguides, the first transition preferably comprising a rectangular waveguide impedance transformer of variable section, and / or further comprises at least a second transition between an input coupler of the plurality of input couplers and the waveguide of the second plurality of waveguides, the second transition preferably comprising a rectangular waveguide impedance transformer of variable section;
[0041] - the first combined signal is output by a first output of the first sub-combiner, and the second combined signal is output by a second output of the second sub-combiner, the power amplifier further comprising a third transition between the first output and the output coupler and / or a fourth transition between the output coupler and the second output;
[0042] - each waveguide of the first plurality of waveguides and of the second plurality of waveguides is a rectangular waveguide comprising a capacitive element positioned inside the waveguide so as to form staircase steps of a depth measured in the radial direction in which the waveguide extends equal to a quarter of the wavelength of the initial input electromagnetic signal;
[0043] - the number of input couplers, the number of waveguides of the first plurality of waveguides and the number of waveguides of the second plurality of waveguides are equal;
[0044] - each amplifier channel comprises a phase shifter of the elementary signal comprising a microstrip line of variable length;
[0045] - the variable length microstrip line is formed by a first microstrip line on a first printed circuit and by a second microstrip line and a third microstrip line on a second printed circuit, the third microstrip line being parallel to the second microstrip line, a first end of the first microstrip line being in contact with the second microstrip line and a second end of the first microstrip line being in contact with the third microstrip line, the first printed circuit being movable relative to the second printed circuit;
[0046] - the power divider comprises: a third cylindrical cavity along the same axis of symmetry (Z), the third cylindrical cavity comprising an input of the initial electromagnetic input signal on the axis of symmetry (Z) and a plurality of outputs of the third cylindrical cavity on the same plane (P”) perpendicular to the axis of symmetry (Z), a plurality of air-suspended strip-plate lines, each air-suspended strip-plate line extending radially with respect to the axis of symmetry (Z) in the plane (P”) from the interior of the third cylindrical cavity through an output of the plurality of outputs of the third cylindrical cavity, each air-suspended strip-plate line being configured to pick up an elementary signal of the plurality of elementary signals;
[0047] - the power amplifier comprises a coaxial input extending axially along the axis of symmetry (Z) and opening into the third cylindrical cavity via the input of said third cylindrical cavity, the coaxial input comprising a conductive core which extends inside the third cylindrical cavity to the plane (P”); - each amplifier channel comprises a printed circuit extending radially in the plane (P”) from an air-suspended triplate line of the plurality of air-suspended triplate lines, each amplifier channel further comprising an elementary amplifier placed on the printed circuit.
[0048] BRIEF DESCRIPTION OF THE FIGURES
[0049] Other characteristics and advantages of the invention will emerge from the detailed description which follows, with reference to the appended drawings, in which:
[0050] - Figures 1A and 1B represent a power amplifier according to an embodiment of the invention comprising a power combiner, a power divider and amplifier channels, Figure 1A being a three-dimensional representation and Figure 1B a sectional view,
[0051] - figure 2 represents a block view of the connections between the first sub-combiner, the second sub-combiner, the input couplers and the output coupler of a power combiner according to one embodiment of the invention,
[0052] - Figure 3 represents a three-dimensional view of a power combiner according to an embodiment of the invention of radial symmetry with respect to the axis of symmetry (Z) comprising a first sub-combiner, a second sub-combiner, input couplers and an output coupler,
[0053] - Figures 4A and 4B represent two sectional views of a power combiner according to an embodiment of the invention without the external metal parts. Figure 4A represents a section along a first plane comprising the axis of symmetry (Z) of the power combiner and two input couplers, and Figure 4B represents a section along a second plane comprising the axis of symmetry (Z) of the power combiner, an input combiner and the output coupler,
[0054] - Figure 5 represents a three-dimensional view of an exemplary implementation of a power combiner according to an embodiment of the invention comprising a first sub-combiner, a second sub-combiner, a set of input couplers and an output coupler comprising a lower metal plate, an intermediate metal plate and an upper metal plate,
[0055] - Figure 6 represents a sectional view of the first sub-combiner and the second sub-combiner according to an embodiment of the invention with the external metal parts,
[0056] - Figures 7A and 7B represent a power divider according to an embodiment of the invention according to which the power divider is of radial and quasi-planar architecture, Figure 7A being a three-dimensional representation and Figure 7B a sectional view, - Figure 8 represents a three-dimensional view of a power combiner according to the embodiment of Figures 7A and 7B, mounted on amplifier channels according to one example,
[0057] - figure 9 represents a sectional view of the power combiner and the amplifier channels according to the embodiment of figures 7A, 7B and 8,
[0058] - Figures 10A and 10B represent an embodiment of a phase compensation device according to an embodiment of the invention, comprising a first printed circuit and a second printed circuit which form a microstrip line of variable length depending on the positioning of the first printed circuit relative to the second printed circuit, Figure 10A representing the microstrip line in open circuit and Figure 10B representing the microstrip line producing a phase shift of a quarter of a wavelength.
[0059] Throughout the figures, similar elements have identical references.
[0060] DETAILED DESCRIPTION OF EMBODIMENTS
[0061] The invention relates to an amplifier of a signal comprising a power divider, a plurality of amplifying channels and a power combiner. The power divider is configured to divide an initial input electromagnetic signal into a plurality of elementary signals, the power of the initial input electromagnetic signal being distributed over the elementary signals. Each amplifying channel is configured to amplify an elementary signal of the plurality of elementary signals and the power combiner is configured to recombine the plurality of amplified elementary signals into an output electromagnetic signal, such that the output electromagnetic signal has an amplified power relative to the initial input electromagnetic signal.
[0062] The initial input electromagnetic signal, for example, has a frequency preferably between 27 GHz and 31 GHz.
[0063] The initial power of the initial input electromagnetic signal is for example between 37 and 39 dBm and the final power of the output electromagnetic signal is for example between 51 dBm and 53 dBm.
[0064] Figures 1A and 1B show an embodiment of the power amplifier according to an embodiment of the invention comprising a power combiner 1, a power divider 2 and amplifier paths 3, Figure 1A being a three-dimensional representation and Figure 1B a sectional view. Power combiner
[0065] The power combiner includes a first sub-combiner, a second sub-combiner, a plurality of input couplers, and an output coupler.
[0066] The first sub-combiner comprises a first plurality of combining paths and the second sub-combiner comprises a second plurality of combining paths.
[0067] Each input coupler of the plurality of input couplers is configured to capture the amplified elementary signal from an amplifying channel and divide said amplified elementary signal into a first signal transmitted in a combination channel of the first plurality of combination channels and into a second signal transmitted in a combination channel of the second plurality of combination channels.
[0068] The second sub-combiner therefore comprises as many combination channels as the first sub-combiner and the number of combination channels of each sub-combiner is equal to the number of input couplers of the plurality of input couplers and to the number of amplifier channels of the plurality of amplifier channels.
[0069] The first sub-combiner is configured to combine the first signals transmitted in each combination path of the first plurality of waveguides into a first combined signal and the second sub-combiner is configured to combine the second signals transmitted in each combination path of the second plurality of waveguides into a second combined signal.
[0070] Finally, the output coupler is configured to combine the first combined signal and the second combined signal into the output electromagnetic signal.
[0071] The power combiner according to the invention has a radial architecture. More specifically, the first sub-combiner and the second sub-combiner each have radial symmetry: the combination paths of the first, respectively the second, sub-combiner are similar and extend regularly from a central axis. Furthermore, the first sub-combiner and the second sub-combiner are similar and in particular all the dimensions of the first sub-combiner are equal to the dimensions of the second sub-combiner. Finally, the combination paths of the first sub-combiner extend from the same central axis as the paths of the second sub-combiner and each combination path of the first sub-combiner is superimposable on a combination path of the second sub-combiner. The radial architecture makes it possible to combine a large number of combination paths.For example, each sub-combiner among the first sub-combiner and the second sub-combiner comprises two combination paths, or a number of combination paths strictly greater than two.
[0072] The inventors consider that if power losses are tolerable at the power divider level since the signal has not yet been amplified, the power preamplification stages can be configured to compensate for these losses, it is important to limit these losses as much as possible at the power combiner level.
[0073] Indeed, any loss at the combiner is definitive. In order to minimize power losses, the power combiner according to the invention is implemented in a waveguide. In the following, an embodiment of the power combiner according to the invention is described in more detail.
[0074] With reference to Figures 3 and 4A and 4B, the first sub-combiner 4 comprises a first cylindrical cavity 8 along an axis of symmetry (Z) forming a first radial waveguide. The first cylindrical cavity comprises a first plurality of inputs at the periphery of the first cylindrical cavity arranged in a plane (P) perpendicular to the axis of symmetry (Z). More precisely, the intersection of the first cylindrical cavity with the plane P defines a first circle (C1) and the inputs of the first plurality of inputs are distributed along the circle (C1).
[0075] The first sub-combiner 4 further comprises a first plurality of waveguides 9, each waveguide 9 of the first plurality of waveguides extending radially in the plane (P) relative to the axis of symmetry (Z) from an input of the first plurality of inputs.
[0076] In other words, each combination path of the first plurality of combination paths extends radially with respect to the axis of symmetry (Z) and comprises a waveguide 9 of the first plurality of waveguides.
[0077] Due to the radial symmetry of the first sub-combiner 4, all the waveguides 9 of the first plurality of waveguides are equally distributed around the axis of symmetry (Z) and have the same dimensions.
[0078] The second sub-combiner 5 comprises a second cylindrical cavity 10 forming a second radial waveguide having the same axis of symmetry (Z) as the first cylindrical cavity. The second cylindrical cavity 10 comprises a second plurality of inputs arranged on the periphery of the second cylindrical cavity in a plane (P') perpendicular to the axis of symmetry (Z). More precisely, the intersection of the second cylindrical cavity with the plane (P') defines a second circle C2 and the inputs of the second plurality of inputs are distributed along the circle C2.
[0079] The second sub-combiner 5 further comprises a second plurality of waveguides 11, each waveguide 11 of the second plurality of waveguides extending radially in the plane (P') relative to the axis of symmetry (Z) from an input of the second plurality of inputs.
[0080] In other words, each combination path of the second plurality of combination paths extends radially with respect to the axis of symmetry (Z) and comprises a waveguide of the second plurality of waveguides.
[0081] Due to the radial symmetry of the second sub-combiner 5, all the waveguides 11 of the second plurality of waveguides are equally distributed around the axis of symmetry (Z) and have the same dimensions.
[0082] Further, each input of the second plurality of inputs is aligned with an input of the first plurality of inputs. That is, each input of the first plurality of inputs is the orthogonal projection of an input of the second plurality of inputs onto the plane (P). In this way, each combination path of the first plurality of combination paths is aligned with a combination path of the second plurality of combination paths.
[0083] As previously mentioned, the first sub-combiner 4 and the second sub-combiner 5 are similar. In particular, all dimensions of the first cylindrical cavity 8 are equal to the dimensions of the second cylindrical cavity 10 and all dimensions of a waveguide 9 of the first plurality of waveguides are equal to the dimensions of a waveguide 11 of the second plurality of waveguides.
[0084] The radial symmetry of the first sub-combiner 4 and the second sub-combiner 5, their superposition and the fact that the first sub-combiner 4 is similar to the second sub-combiner 5 advantageously allow good operation of the channel isolation system described below and prevents phase dispersion of the amplified elementary signals circulating in the different combination channels which would be a source of combination losses. Each waveguide 9, 11 of the first plurality of waveguides and of the second plurality of waveguides is preferably a rectangular waveguide. In other words, a section of the waveguide in a plane perpendicular to the direction in which the waveguide 9, 11 extends is rectangular and has a height H parallel to the axis of symmetry (Z) and a width I perpendicular to said axis of symmetry (Z).
[0085] The first signal transmitted by each input coupler 6 of the plurality of input couplers - respectively the second transmitted signal - propagates in a waveguide 11 of the first plurality of waveguides - respectively of the second plurality of waveguides - towards the first radial waveguide - respectively towards the second radial waveguide with the fundamental mode TE10 of the rectangular waveguide.
[0086] In a local coordinate system defined for each rectangular waveguide, having an axis (Oz) parallel to the direction of propagation in said waveguide and axes (Ox) and (Oy) perpendicular to each other and perpendicular to the axis (Oz), the different components of the electric field and the magnetic field are given by the following formulas:
[0087] H y= 0 where I is the width of the rectangular waveguide section, co is the operating pulsation, A is the field amplitude constant, / 3 is the phase constant and / c is the magnetic permeability.
[0088] The first signals transmitted by the set of input couplers 6 of the plurality of couplers in the set of waveguides 9 of the first plurality of waveguides - respectively the second signals transmitted by the set of input couplers 6 of the plurality of couplers in the set of waveguides 11 of the second plurality of waveguides - are then emitted in the first radial waveguide - respectively in the second radial waveguide - where they combine into the first combined signal - into the second combined signal. The first combined signal and the second combined signal propagate respectively in the first radial waveguide and in the second radial waveguide according to the E00 mode whose components of the electric field and the magnetic field are given by the following formulas:
[0089] Eg = E r = 0
[0090] H z = H r= 0 where h is the height along the axis of symmetry (Z) of the radial waveguide, r the radius of the radial waveguide, V(r) the voltage wave and l(r) the current wave.
[0091] A width I constituting the largest dimension of the waveguide section, and a height H parallel to the axis of symmetry (Z) and less than the width I, allow a better transition without reflection between the TE10 mode of the rectangular waveguide and the E00 mode in the first or second radial waveguide in which the electric field is parallel to the axis of symmetry Z.
[0092] Furthermore, the first sub-combiner 4 and the second sub-combiner 5 may advantageously comprise at least one capacitive element 12 per waveguide 9, 11 of the first plurality of waveguides and of the second plurality of waveguides, so as to allow a reflection-free transition of the electromagnetic wave coming from the waveguide 9, 11 towards the radial waveguide from which said waveguide 9, 11 extends.
[0093] More specifically, each capacitive element 12 is positioned in the waveguide 9, 11 of the first plurality of waveguides or of the second plurality of waveguides at the junction with the corresponding radial waveguide.
[0094] For example, in the case where each waveguide 9, 11 of the first plurality of waveguides and of the second plurality of waveguides is a rectangular waveguide, each capacitive element 12 can form inside the waveguide 9, 11 in which said capacitive element 12 is positioned staircase steps of a depth measured in the radial direction in which the waveguide 9, 11 extends equal to a quarter of the wavelength of the initial input electromagnetic signal, so that the section of the waveguide measured perpendicular to the radial direction in which said waveguide extends gradually decreases in the direction of propagation of the first transmitted signal or the second transmitted signal at each staircase step of the capacitive element.
[0095] The reduction in section by the capacitive elements 12 can be caused alternatively or in combination by a reduction in the width I and / or the length L of the waveguide.
[0096] As previously mentioned, each input coupler 6 of the plurality of input couplers is configured to capture the amplified elementary signal coming from an amplifying channel and divide said amplified elementary signal into a first signal transmitted in a waveguide 9 of the first plurality of waveguides and into a second signal transmitted in a waveguide 11 of the second plurality of waveguides, the first transmitted signal being in phase with the amplified elementary signal and the second transmitted signal being phase-shifted by a quarter of a wavelength relative to the amplified elementary signal.
[0097] More specifically and with reference to Figure 2, each input coupler 6 of the plurality of input couplers is preferably a branch-line type coupler. Each input coupler 6 comprises an input channel IN 62, a first output channel otherwise called THRU channel not inducing a phase shift, a second output channel otherwise called COUP channel inducing a phase shift of a quarter of a wavelength, and an isolation channel 63. The signal entering via the input channel of such a coupler is divided into an output signal emitted by the THRU channel and into a signal emitted by the COUP channel, the two output signals being of the same amplitude at 3 dB. The output signal emitted by the COUP channel is phase shifted by a quarter of a wavelength relative to the signal entered via the input channel and relative to the output signal emitted by the THRU channel.The phase shift between the two THRU and COUP channels is induced by a difference between the length of the electrical path between the input channel and the COUP channel on the one hand, and the length of the electrical path between the input channel and the THRU channel on the other hand. This difference in length is preferably equal to a quarter of the wavelength.
[0098] In the combiner according to one embodiment of the invention, for each input coupler of the plurality of input couplers:
[0099] - the input channel IN 62 is connected with an amplifier channel 3 of the plurality of amplifier channels,
[0100] - the first output path or THRU path is connected with a waveguide 9 of the first plurality of waveguides, and, - the second output path or COUP path is connected with a waveguide 11 of the second plurality of waveguides.
[0101] Thus, the amplified signal transmitted by an amplifying channel is picked up by the input coupler 6 of the plurality of input couplers to which the amplifying channel is connected. Said amplified signal is divided into a first signal emitted by the first output channel or THRU channel and transmitted in the waveguide 9 of the first plurality of waveguides to which said THRU channel is connected and into a second signal of the same amplitude at 3 dB as the first signal, the second signal being emitted by the second output channel or COUP channel and transmitted in the waveguide 11 of the second plurality of waveguides to which said COUP channel is connected.
[0102] Such connections between the input couplers 6 of the plurality of input couplers, the amplifier paths and the waveguides 9, 11 of the first and second pluralities of waveguides advantageously make it possible to isolate the combination paths from each other.
[0103] Indeed, the amplified elementary signal arriving on the input channel of a first input coupler of the plurality of input couplers is divided into a first signal transmitted in the first sub-combiner by the THRU channel and into a second signal transmitted in the second sub-combiner by the COUP channel phase-shifted from the first signal and the amplified elementary signal by a quarter wavelength. Through the second sub-combiner, the second signal can reach the input channel of a second input coupler of the plurality of input couplers. Given the previously described connections of the second coupler of the plurality of couplers with the first and second sub-combiner, the second signal arrives on the input channel of the second input coupler of the plurality of input couplers via the COUP channel of said second coupler.Said passage through the COUP path of the second input coupler of the plurality of input couplers induces an additional phase shift in said second signal of a quarter of a wavelength. The overall phase shift of the second signal is therefore half a wavelength on the input path of the second input coupler of the plurality of input couplers. As for the first signal transmitted in the first sub-combiner, it can also reach, through the first sub-combiner, the input path of the second sub-combiner via the THRU path of said second input coupler of the plurality of input couplers without any phase shift being induced therein.When the first signal and the second signal arrive at the input channel of the second input coupler of the plurality of input couplers, the first signal and the second signal are of the same amplitude and in opposite direction: the second coupler adds them together so that the first signal and the second signal cancel each other out.
[0104] Preferably, each input coupler 6 of the plurality of input couplers is a rectangular waveguide coupler which extends radially relative to the axis of symmetry (Z) in a plane (P3) perpendicular to the axis of symmetry (Z) from the waveguide 9 of the first plurality of waveguides and the waveguide 11 of the second plurality of waveguides in which the first signal and the second signal are respectively transmitted by the at least one input coupler 6.
[0105] In particular, the isolation path 63 of each input coupler 6 may comprise a waveguide extending radially in the plane (P') from the input coupler 6 connected to a waveguide load 64 extending in a direction parallel to the axis of symmetry (Z).
[0106] The input path IN 62 of each input coupler 6 may comprise a waveguide extending radially in the plane (P) from the input coupler 6.
[0107] The power amplifier according to one embodiment of the invention may further comprise at least one first transition 13 between an input coupler 6 of the plurality of input couplers and the waveguide 9 of the first plurality of waveguides in which the first signal is transmitted by the at least one input coupler 6 and / or at least one second transition 14 between an input coupler 6 of the plurality of input couplers and the waveguide 11 of the second plurality of waveguides in which the second signal is transmitted by the at least one input coupler 6.
[0108] The first transition 13 and the second transition 14 are preferably rectangular waveguide impedance transformers of variable section. More specifically, the section of the rectangular waveguide impedance transformer in a plane perpendicular to the direction in which the waveguide of the first plurality of waveguides or the waveguide of the second plurality of waveguides extends may have a rectangular shape with a height H along the axis of symmetry (Z) and a depth P perpendicular to the axis of symmetry (Z) and the dimensions of the height H and / or the depth P may vary along the impedance transformer in the radial direction in which the waveguide 9 of the first plurality of waveguides or the waveguide 11 of the second plurality of waveguides extends.Preferably, the height H and / or the depth P vary periodically, defining portions of the impedance transformer of length equal to a quarter of the wavelength of the initial input electromagnetic signal.
[0109] A variable depth P advantageously makes it possible to obtain the desired impedance, more precisely to pass from the impedance of each channel of the input coupler to the impedance of the waveguides of the first and second plurality of waveguides. A variable height H advantageously makes it possible to obtain a transition between coupler and waveguide without reflection.
[0110] Furthermore, the first sub-combiner 4 may comprise a first output 15 through which the first combined signal is output in the first sub-combiner 4 and the second sub-combiner 5 may comprise a second output 16 through which the second combined signal is output in the second sub-combiner 5.
[0111] As previously mentioned, the output coupler 7 is arranged to combine the first combined signal and the second combined signal by inducing a phase shift on the first combined signal of a quarter wavelength without inducing a phase shift on the second combined signal, so as to form the output electromagnetic signal.
[0112] More precisely, the output coupler 7 is a “branch-line” type coupler comprising an output channel OUT, a first input channel otherwise called THRU channel not inducing a phase shift and a second input channel otherwise called COUP channel inducing a phase shift of a quarter of a wavelength and an isolation channel. In other words, the output coupler 7 is a coupler of the same type as the input couplers 6, but used in the reciprocal direction of said coupler: when two signals enter respectively via the THRU and COUP channels, the coupler combines them on its OUT channel (the IN channel in the forward direction) while phase shifting by a quarter of a wavelength the signal entering via the COUP channel.
[0113] In the power combiner 1 according to one embodiment of the invention:
[0114] - the first input channel of the output coupler 7 or THRU channel is connected with the second output 16 of the second sub-combiner 5,
[0115] - the second input channel of the output coupler 7 or COUP channel is connected with the first output 15 of the first sub-combiner 4,
[0116] - the output channel IN of the output coupler 7 is connected with a global output 17 of the power combiner 1 through which the output electromagnetic signal is emitted.
[0117] Thus, the first combined signal in the first sub-combiner 4 is transmitted by the first output 15 of the first sub-combiner 4 and picked up by the COUP channel of the output coupler to which the first output 15 of the first sub-combiner 4 is connected which induces a phase shift of a quarter of a wavelength to said first combined signal. Furthermore, the second combined signal in the second sub-combiner 5 is transmitted by the second output 16 of the second sub-combiner 5 and picked up by the THRU channel of the output coupler 7 to which the second output 16 of the second sub-combiner 5 is connected. The output coupler 7 combines the first combined signal, phase shifted by a quarter of a wavelength, with the second combined signal, so as to form the output electromagnetic signal which is emitted by the input channel of the output coupler 7 into the overall output 17 of the combiner 1.
[0118] The phase shift of a quarter wavelength of the first combined signal by the COUP channel of the output coupler 7 advantageously allows the first combined signal and the second combined signal to have the same phase when they are combined so as to form the output electromagnetic signal and thus to limit the combination losses caused by the phase imbalances whose impact on the combination losses is preponderant compared to the amplitude imbalances.
[0119] The power combiner 1 may comprise a third transition 18 between the first output 15 of the first sub-combiner 4 and the output coupler 7 and a fourth transition 19 between the second output 16 of the second sub-combiner 5 and the output coupler 7.
[0120] Preferably, the first output 15, the second output 16, the output coupler 7, the third transition 18 and the fourth transition 19 are made of rectangular waveguides, so as to minimize the losses in the combination of the first combined signal and the second combined signal. For example, the rectangular waveguide of the first output 15 and the rectangular waveguide of the second output 16 both extend from the axis of symmetry (Z) in the same direction perpendicular to the axis of symmetry (Z). The rectangular waveguide of the first output 15 may be included in a plane (P1) and the rectangular waveguide of the second output 16 in a plane (P2), the planes (P1) and (P2) being perpendicular to the axis of symmetry (Z) but disjoint from each other and disjoint from the planes (P) and (P').The output coupler 7 may extend radially relative to the axis of symmetry (Z) in a plane perpendicular to the axis of symmetry (Z) between the planes (P1) and (P2), for example in the same plane (P3) as the input couplers (6), from the rectangular waveguide of the first output 15 and the rectangular waveguide of the second output 16.
[0121] The third transition 18 and the fourth transition 19 may each comprise a rectangular waveguide impedance transformer of variable cross-section. More specifically, the rectangular waveguide cross-section of the impedance transformer of the third, respectively fourth, transition in a plane perpendicular to the direction in which the waveguide of the impedance transformer extends may have a rectangular shape with a height H along the axis of symmetry (Z) and a depth P perpendicular to the axis in which the waveguide of the impedance transformer extends and the dimensions of the height H and / or the depth P may vary along the waveguide of the impedance transformer. Preferably, the height H and / or the depth P vary periodically, defining portions of the impedance transformer of length equal to a quarter of the wavelength of the initial input electromagnetic signal.
[0122] The third, respectively fourth, transition may further comprise a rectangular waveguide bend making it possible to bring the first, respectively second, output of the first, respectively second, sub-combiner into the plane of the output coupler.
[0123] In order to transmit the first combined signal from the first radial waveguide to the first output 15 of the first sub-combiner 4, the first cylindrical cavity 8 may comprise a first output on the axis of symmetry (Z) and the first sub-combiner 4 may comprise a first coaxial output 20 in a coaxial waveguide extending axially along the axis of symmetry (Z) from the output of the first cylindrical cavity 8 to the first output 15 of the first sub-combiner 4.
[0124] Similarly, in order to transmit the second combined signal from the second radial waveguide to the second output 16 of the second sub-combiner 5, the second cylindrical cavity 10 may comprise a second output on the axis of symmetry (Z) and the second sub-combiner 5 may comprise a second coaxial output 21 in a coaxial waveguide extending axially along the axis of symmetry (Z) from the second output of the second cylindrical cavity 10 to the second output 16 of the second sub-combiner 5.
[0125] In the first coaxial input 20 and in the second coaxial input 21, the first combined signal and the second combined signal propagate according to the fundamental TEM mode of the coaxial guide, the components of the electric field and the magnetic field of which are given by the following formulas:
[0126] E g = E Z = Q
[0127] H r = Hz = 0 where R1 is the radius of the inner conductor of the coaxial waveguide, R2 is the radius of the outer conductor of the coaxial waveguide, r is the coordinate of a cylindrical coordinate system defined with respect to the axis of symmetry (Z), V is the voltage wave and I is the current wave.
[0128] Preferably, the first sub-combiner 4 and the second sub-combiner 5 comprise capacitive elements 22, for example metal rings arranged in the first radial waveguide and the second radial waveguide, said metal rings extending in a plane perpendicular to the axis of symmetry Z and being centered on the axis of symmetry Z.
[0129] The capacitive elements 22 between the first radial waveguide and the first coaxial input 20 on the one hand and between the second radial waveguide and the second coaxial input 21 on the other hand advantageously allow a reflection-free transition from the propagation mode E00 of the radial waveguide to the fundamental mode of the coaxial TEM waveguide.
[0130] Preferably again, the internal conductor 23 of the coaxial waveguide of the first, respectively of the second, coaxial input 20, 21 extends inside the first, respectively of the second, radial waveguide to the face of the first, respectively of the second, cylindrical cavity 8, 10 opposite the first, respectively of the second, output of said first, respectively of the second, cylindrical cavity 8, 10. Such an extension of the internal conductor 23 of the coaxial waveguide advantageously makes it possible to avoid having a floating central conductor and thus to improve the passage of the wave from the radial waveguide to the coaxial waveguide.
[0131] Particularly advantageously, the internal conductor 23 of the coaxial waveguide is integral with the metal plate forming the part of the radial waveguide opposite the output of the radial waveguide. An example of an embodiment of such an embodiment will be given below.
[0132] Such an embodiment of the internal conductor 23 of the coaxial waveguide advantageously allows for better centering of the internal conductor 23 in the sheath formed by the external conductor as well as a good evacuation path for the power dissipated by the internal conductor 23 of the coaxial waveguide by thermal conduction towards the metal plate. A conductor machined directly with the metal plate forming the radial waveguide also makes it possible to dispense with the use of glue or foam as could be the case if the conductor were simply inserted into a slot in the metal plate, to plug said slot.
[0133] The use of a coaxial waveguide between the radial waveguide and the rectangular waveguide advantageously allows the output signal to be transferred without reflection between the radial waveguide and the coaxial waveguide on the one hand and then between the coaxial waveguide and the rectangular waveguide on the other hand. Since the fundamental mode of a rectangular waveguide does not have radial symmetry, such a transfer without reflection would not be possible directly between the radial waveguide and the rectangular waveguide.
[0134] If the first output 15 of the first sub-combiner 4 and the second output 16 of the second sub-combiner 5 are rectangular waveguides extending from the axis of symmetry (Z) in the same direction perpendicular to the axis of symmetry (Z), a transition between the first, respectively second, coaxial output 20, 21 and the first, respectively second output 15, 16, of the first, respectively second, sub-combiner 4, 5 may comprise an extension of the internal conductor 23 of the coaxial waveguide inside the rectangular waveguide along the axis of symmetry (Z).
[0135] Said transition between the first, respectively second, coaxial output 20, 21 and the first, respectively second, output 15, 16, of the first, respectively second, sub-combiner 4, 5 may further comprise a capacitive pad 24 arranged inside the rectangular waveguide forming the first output 15 of the first sub-combiner 4 or the second output 16 of the second sub-combiner 5, at a distance from the end of the rectangular waveguide by which the signal is transmitted to the output coupler equal to a quarter of the wavelength of the initial electromagnetic input signal.
[0136] Such a capacitive pad 24 advantageously makes it possible to adjust the imaginary value of the impedance at the output of the rectangular waveguide.
[0137] In a particular embodiment of the power combiner 1 and with reference to Figures 5 and 6, said power combiner 1 comprises, from a rear face to a front face, a lower metal part 25, an intermediate metal part 26 and an upper metal part 27 arranged against each other and perpendicular to the axis of symmetry (Z). The lower metal part 25 may comprise on its front face recesses closed by the intermediate metal part 26, so that said recesses of the lower metal part 25 closed by the intermediate metal part 26 form the waveguides 9 of the first plurality of waveguides and the first radial waveguide.
[0138] Symmetrically, the upper metal part 27 may comprise on its rear face recesses closed by the intermediate metal part 26, so that said recesses of the upper metal part 27 closed by the intermediate metal part 26 form the waveguides 11 of the second plurality of waveguides and the second radial waveguide.
[0139] The intermediate metal part 26 may comprise the input couplers 6 of the plurality of rectangular waveguide input couplers. More specifically, the intermediate metal part 26 comprises at each input coupler 6 openings of length equal to a quarter of the wavelength of the initial input electromagnetic signal configured so as to connect the input path of the coupler, the COUP path and the THRU path.
[0140] The arrangement of the lower metal piece 25, the intermediate metal piece 26 and the upper metal piece 27 can be made such that each input coupler 6 of the intermediate metal piece 26 is aligned with a recess of the lower metal piece 25 and a recess of the upper metal piece 27 so as to be connected with a waveguide 9 of the first plurality of waveguides and a waveguide 11 of the second plurality of waveguides.
[0141] Such an embodiment of the combiner has the advantage of being easily manufacturable by machining. Indeed, the use makes it possible to obtain parts with a more controlled surface roughness than parts resulting from other types of manufacturing process, for example additive manufacturing. A well-controlled surface roughness makes it possible to limit metal losses.
[0142] In the case where the input path IN 62 and the isolation path 63 of each input coupler 6 comprise a rectangular waveguide extending respectively in the plane (P) and in the plane (P'), the lower metal part 25 and the upper metal part 26 may further comprise extension recesses, so that said extension recesses closed by the intermediate metal part 26 form the rectangular waveguides respectively of each input path IN 62 and of the isolation path 63 which extend radially from the input coupler.The first, respectively second, outlet 15, 16 of the first, respectively second, sub-combiner 4, 5 can be formed by a first, respectively second, metal outlet part 28, 29 fixed on the rear, respectively front, face of the lower metal plate 25, respectively upper 27, and a first, respectively second, cover 30, 31 arranged against the first, respectively second, metal outlet part 28, 29.
[0143] For example, the first, respectively second, metal output part 28, 29 comprises a recess which is closed by the first, respectively second, cover 30, 31, so as to form a rectangular metal waveguide which extends perpendicular to the axis of symmetry (Z).
[0144] In the case where the first sub-combiner 4 and the second sub-combiner 5 respectively comprise a first coaxial output 20 and a second coaxial output 21 which respectively connect the first radial waveguide to the first output 15 of the first sub-combiner 4 and the second radial waveguide to the second output 16 of the second sub-combiner 5, the intermediate metal part 26 may comprise a first flange and a second flange 32 made of material which extend axially on the axis of symmetry (Z) on either side of said intermediate metal part 26. The lower metal part 25 and the upper metal part 26 may furthermore each comprise a through cylindrical opening which extends axially on the axis of symmetry (Z) from the rear face to the front face of said metal part.The arrangement of the lower metal part 25, the intermediate metal part 26 and the metal part 27 is made so that the first flange of the intermediate metal part 26 is engaged in the cylindrical opening of the lower metal part 25 and the second flange 32 of the intermediate metal part 26 is engaged in the upper metal part 27. The first, respectively second, flange 32 forms the internal conductor 23 of the coaxial waveguide of the first, respectively second, coaxial output 20, 21.
[0145] In the case where the transition between the first, respectively second, coaxial output 20, 21 and the first, respectively second, output 15, 16 comprises a capacitive pad 24, said capacitive pad 24 may be an adjustment screw 33 engaged in an opening of the first, respectively second, cover. The screw advantageously makes it possible to adjust the imaginary part of the impedance and undesirable effects due to manufacturing defects. Alternatively, the capacitive pad 24 may comprise one or more metal pads machined directly in the first, respectively second, cover or in the first, respectively second, metal output part.Finally, in the case where the power combiner 1 comprises a third transition 18 between the first output 15 of the first sub-combiner 4 and the output coupler 7 and a fourth transition 19 between the second output 16 of the second sub-combiner 5 and the output coupler 7 and said third transition 18 and fourth transition 19 comprise elbows, the output coupler 7 may be included in the intermediate metal plate 26. The lower metal plate 25 and the upper metal plate 27 may each further comprise an additional recess which forms the COUP path and the THRU path of the output coupler.
[0146] In an alternative embodiment of the power combiner not shown, each sub-combiner includes a lower metal piece and an upper metal piece.
[0147] Power divider
[0148] The power divider is configured to receive the initial input electromagnetic signal and divide it into a plurality of elementary signals, the power of the initial input electromagnetic signal being distributed over the elementary signals.
[0149] In the following, a preferred embodiment of the power divider 2 is described in which said power divider 2 is radial and quasi-planar. The radial symmetry of the power divider implies that the power divider 2 comprises a plurality of similar division paths which extend regularly from a central axis. Said central axis is the same axis as the axis of symmetry (Z) of the power combiner.
[0150] In this embodiment and with reference to Figures 7A and 7B, the power divider 2 comprises a third cylindrical cavity 34 with the same axis of symmetry (Z) as the first cylindrical cavity and the second cylindrical cavity, the third cylindrical cavity 34 comprising an input 35 of the initial electromagnetic input signal on the axis of symmetry (Z) and a plurality of outputs 36 of the third cylindrical cavity 34 on the same plane (P”) perpendicular to the axis of symmetry (Z).
[0151] The power divider 2 further comprises a plurality of air-suspended strip-plate lines, each air-suspended strip-plate line 37 extending radially with respect to the axis of symmetry (Z) in the plane (P”) from an outlet 36 of the plurality of outlets of the third cylindrical cavity 34. Each air-suspended strip-plate line 37 comprises successively along the axis of symmetry (Z), from a rear face to a front face, a first metal plate 38, a first air thickness 39, a dielectric substrate plate 40, a second metal plate 41 deposited on the dielectric substrate plate 40, a second air thickness 42 and a third metal plate 43.
[0152] More specifically, the second metal plate 41 deposited on the dielectric substrate plate 40 can be produced by a copper deposition with a method for manufacturing a printed circuit or any other method for manufacturing a printed circuit. The printed circuit is preferably suspended in a metal channel of rectangular section extending radially from the entrance of the third cylindrical cavity, a first face of the metal channel parallel to the printed circuit forming the first metal plate 38 and a second face of the metal channel parallel to the printed circuit forming the third metal plate 43. The use of a metal channel advantageously makes it possible to eliminate radiation losses.
[0153] More specifically still and with reference to Figure 9, the third cylindrical cavity 34 and the metal channels may be formed by a first metal part 44 or lower metal part and by a second metal part 45 or upper metal part arranged against each other perpendicular to the axis of symmetry (Z). In other words, when the first metal part 44 and the second metal part 45 are arranged against each other, recesses in the first metal part 44 coincide with recesses in the second metal part 45, so as to form the third cylindrical cavity 34 and the metal channels.
[0154] Each air-suspended stripline 37 is configured to pick up an elementary signal from the plurality of elementary signals. For this purpose, the second metal plate 41 of each air-suspended stripline 37 may comprise a linear extension 46 inside the third cylindrical cavity 34, the first air thickness 39 and the second air thickness 42 opening into the third cylindrical cavity 34. Preferably, the third cylindrical cavity 34 further comprises a dielectric substrate disc 47 in the plane (P”) which extends the dielectric substrate plates inside the third cylindrical cavity and the linear extension 46 of each second conductive plate 41 inside the third cylindrical cavity 34 is deposited on said dielectric substrate disc 47. The dielectric substrate disc 47 and the dielectric substrate plates 40 of the air-suspended stripline 37 are made in one piece.In other words, the dielectric substrate disc 47 and the dielectric substrate plates 40 form a single piece. The dielectric substrate disc 47 advantageously makes it possible to fix the central core of a coaxial input of the divider as will be described later.
[0155] The air-suspended triplate lines 37 are all identical and regularly distributed around the axis of symmetry (Z). In other words, all their dimensions are equal.
[0156] The initial electromagnetic input signal to be amplified is emitted by the input 35 of the third cylindrical cavity 34. The power divider 2 makes it possible to divide said initial electromagnetic input signal into a plurality of elementary signals of the same phase and the same amplitude, each elementary signal being captured by a suspended air triplate line 37.
[0157] If the printed circuit of the air-suspended strip lines 37 necessarily induces dielectric and conductive losses, the first air thickness 39 and the second air thickness 42 of the air-suspended strip lines 37 advantageously allow a transfer between input signal and elementary signal mainly by air, so that the dielectric and conductive power losses are limited to the maximum. For example, each air-suspended strip line 37 may comprise a first air thickness 39 of 0.25 mm, a dielectric substrate plate 40 of thickness 0.5 mm and a second air thickness 42 of 0.5 mm. Preferably, the air thickness along the axis of symmetry (Z) represents 75% of the total thickness of the air-suspended strip line 37.The printed circuit allows for simple integration of the elementary amplifiers on each amplifier channel so as to allow efficient transfer of the heat emitted by said elementary amplifiers and the implementation of a simple solution for phase shifting the elementary signal transmitted on each amplifier channel. Embodiments will be described later.
[0158] The power divider 2 may further comprise a coaxial input 48 extending axially along the axis of symmetry (Z) and opening into the third cylindrical cavity 34 via the input 35 of said third cylindrical cavity 34, the coaxial input 48 comprising a conductive core 49 which extends inside the third cylindrical cavity 34 to the plane (P”). A coaxial input is advantageously used at the input of the divider where the power level is still low and where losses are still acceptable due to its low cost.
[0159] Preferably, if the third cylindrical cavity 34 comprises a dielectric substrate disc 47, said dielectric substrate disc 47 is pierced in its center and the conductive core 49 of the coaxial input 48 is fixed on the face of the dielectric substrate disc 47 opposite the input of the third cylindrical cavity 34. The fixing of the central core advantageously makes it possible not to leave the conductive core 49 floating, which would less effectively constrain the positioning of said conductive core 49 on the axis of symmetry (Z) and would risk breaking the radial symmetry.
[0160] Preferably, each output 36 of the third cylindrical cavity 34 is aligned with an input of the first plurality of inputs of the first cylindrical cavity 8. Such a configuration of the outputs 36 of the third cylindrical cavity 34, of the inputs of the first cylindrical cavity and of the inputs of the second plurality of inputs advantageously makes it possible to have a path to be traveled of the same length for each elementary signal from an output of the third cavity to an input of the first plurality of inputs and which corresponds to the shortest path.
[0161] Amplifying pathways
[0162] Each amplifier channel 3 comprises at least one elementary amplifier 50 SSPA, so that an elementary signal transmitted by the power divider 2 is amplified.
[0163] Preferably and with reference to Figures 8 and 9, each amplifier channel 3 comprises a printed circuit extending radially in the plane (P”) from an air-suspended stripline 37 of the plurality of air-suspended striplines, and the elementary amplifier 50 of each amplifier channel 3 is placed on the printed circuit. For example, the printed circuit of each amplifier channel is an extension of the printed circuit of the air-suspended striplines 37. More precisely, the transition between the air-suspended striplines 37 and the amplifier channels 3 corresponds to a closure of the metal channel with an extension only of the printed circuit of the air-suspended striplines 37.
[0164] A configuration of the coplanar amplifier paths 3 and elementary amplifiers 50 advantageously makes it possible to provide a transfer path for the heat emitted by the elementary amplifiers which is efficient since a single flat metal plate can serve as a support for the amplifier paths and the elementary amplifiers. In a preferred embodiment, said single metal plate is the first metal part
[0165] 44 of the power divider 2 which extends radially beyond the second metal part
[0166] 45 as shown in Figure 9 so as to support the amplifier channels and thus facilitate heat dissipation.
[0167] The radial geometry of the divider theoretically allows the transmission of elementary signals that are all in phase with each other. However, elementary amplifiers, and to a lesser extent, assembly imperfections, can induce a significant phase shift between the amplified elementary signals (up to 60°). Indeed, each elementary amplifier will potentially induce a phase shift in the elementary signal it amplifies and the value of the induced phase shift can vary from one elementary amplifier to another, even if all the elementary amplifiers are part of the same batch supplied by the same manufacturer. However, to obtain the best possible combination efficiency, it is particularly advantageous for all the amplified elementary signals to have the same phase.
[0168] For this purpose, each amplifier channel 3 may comprise a phase shifter of the elementary signal transmitted in said amplifier channel 3, so that the phase of the elementary signal is adjusted independently of the phase of the other elementary signals transmitted in the other amplifier channels.
[0169] If the amplifier channel 3 is a printed circuit, the phase shifter may comprise a microstrip line portion of adjustable length. Adjustable means that an operator can set the length of the microstrip line portion, so as to increase or decrease the overall length of the electrical path of the amplifier channel on which the adjustable microstrip line portion is mounted.
[0170] With reference to Figures 10A and 10B, the phase shifter comprises for example a first microstrip line 51 on a first dielectric substrate 52 and a second microstrip line 53 and a third microstrip line 54 on a second dielectric substrate 55. The third microstrip line 54 is parallel to the second microstrip line 52. Furthermore, in a closed circuit (example of Figure 10B), a first end 56 of the first microstrip line 51 is in contact with the second microstrip line 53 and a second end 57 of the first microstrip line 51 is in contact with the third microstrip line 54.
[0171] For example, the second dielectric substrate 55 is the dielectric substrate of the amplifier channel 3 and the amplifier channel 3 is in open circuit at the level of the phase shifter with contact of a first microstrip portion 58 of the amplifier channel 3 with the second microstrip line 53 and contact of a second microstrip portion 59 of the amplifier channel 3 with the third microstrip line 54.
[0172] More precisely, the first microstrip portion 58 of the amplifier path can extend radially with respect to the axis of symmetry (Z) from the power divider 2 to one end of the second microstrip line 53 perpendicular to said second microstrip line 53, and the second microstrip portion 59 of the amplifier path 3 can extend along the same radial axis as the first microstrip portion 58 of the amplifier path 3 from one end of the third microstrip line 54 to the end of the amplifier path 3.
[0173] The contact of the first microstrip line 51 with respectively the second microstrip line 53 and the third microstrip line 54 makes it possible to electrically connect the first microstrip portion 58 of the amplifier channel to the second microstrip portion 59 of the amplifier channel 3. Furthermore, the positioning of the first dielectric substrate 52 relative to the second dielectric substrate 55 makes it possible to adjust the length of the microstrip line portion between the first microstrip portion 58 of the amplifier channel 3 and the second microstrip portion 59 of the amplifier channel 3.
[0174] The transition from a 36 air-suspended stripline to a microstrip line allows easy access to the 50 elementary amplifiers and the optional phase shift system.
[0175] Each amplifier channel 3 may further comprise, at the output of the elementary amplifier 50, a transition from a microstrip line to an air-suspended stripline extending radially along the axis of symmetry (Z) in the direction of the amplifier channel 3 then a transition 60 from an air-suspended stripline to a metal cavity 61 of a rectangular waveguide oriented parallel to the axis of symmetry (Z). The amplified elementary signal then propagates in quasi-TEM mode in the air-suspended stripline then in TE10 mode in the metal cavity of the rectangular waveguide.In particular, the transition 60 between the air-suspended stripline and the metal cavity of the rectangular waveguide may comprise a "Patch" type probe and a short circuit positioned on the axis parallel to the axis of symmetry (Z) passing through the probe in the opposite direction to the power combiner 1 at a distance from the probe equal to the wavelength of the initial input electromagnetic signal: the short circuit is seen by the probe as an open circuit, so that the propagation of the amplified elementary signal is in the direction parallel to the axis of symmetry (Z) towards the power combiner 1. The amplified elementary signal thus propagates in the metal cavity 61 of the rectangular waveguide to the input path IN 62 of an input coupler 6 of the plurality of input couplers. The length of the metal cavities 61 may be set to any length.
Claims
CLAIMS 1. Power amplifier comprising: - a power divider (2) configured to receive an initial input electromagnetic signal and divide it into a plurality of elementary signals, the power of the initial input electromagnetic signal being distributed over the elementary signals, - an amplifier comprising a plurality of amplifier channels (3), each amplifier channel (3) being configured to amplify an elementary signal of the plurality of elementary signals, - a power combiner (1) configured to recombine the plurality of amplified elementary signals into an amplified initial electromagnetic signal, the power combiner (1) comprising a first sub-combiner (4), a second sub-combiner (5), a plurality of input couplers (6) and an output coupler (7), the first sub-combiner (4) comprising: - a first cylindrical cavity (8) along an axis of symmetry (Z) forming a first radial waveguide, the first cylindrical cavity (8) comprising a first plurality of inputs arranged in a plane (P) perpendicular to the axis of symmetry (Z), - and a first plurality of rectangular waveguides (9), each waveguide (9) of the first plurality of waveguides extending radially with respect to the axis of symmetry (Z) from an input of the first plurality of inputs, the second sub-combiner (5) comprising: - a second cylindrical cavity (10) along the same axis of symmetry (Z) forming a second radial waveguide, the second cylindrical cavity (10) comprising a second plurality of inputs arranged in a plane (P') perpendicular to the axis of symmetry (Z), the second plurality of inputs being superimposed with the first plurality of inputs, - and a second plurality of rectangular waveguides (11), each waveguide (11) of the second plurality of waveguides extending radially with respect to the axis of symmetry (Z) from an input of the second plurality of inputs, each input coupler (6) being configured to receive an amplified elementary signal from an amplifying channel (3) and to divide said amplified elementary signal into a first signal intended to be transmitted in a waveguide (9) of the first plurality of waveguides and into a second signal intended to be transmitted in a waveguide (11) of the second plurality of waveguides, the first signal being in phase with the amplified elementary signal and the second signal being phase-shifted by a quarter of a wavelength with respect to the amplified elementary signal,the first sub-combiner (4) being configured to combine the first signals transmitted in each waveguide (9) of the first plurality of waveguides into a first combined signal and the second sub-combiner (5) being configured to combine the second, signals transmitted in each waveguide (11) of the second plurality of waveguides into a second combined signal, the output coupler (7) being configured to combine the first combined signal and the second combined signal into an output electromagnetic signal by inducing a phase shift of the first combined signal by a quarter wavelength, so that the output electromagnetic signal has an amplified power compared to the initial input electromagnetic signal.
2. Power amplifier according to the preceding claim, in which each input coupler (6) and the output coupler (7) are “branch-line” type couplers.
3. Power amplifier according to one of the preceding claims, wherein at least one input coupler (6) is a rectangular waveguide coupler extending, radially with respect to the axis of symmetry (Z), from a waveguide (9) of the first plurality of waveguides and a waveguide (11) of the second plurality of waveguides.
4. Power amplifier according to one of the preceding claims, further comprising at least one first transition (13) between an input coupler (6) of the plurality of input couplers and the waveguide (9) of the first plurality of waveguides, the first transition (13) preferably comprising a rectangular waveguide impedance transformer of variable section, and / or further comprising at least one second transition (14) between an input coupler (6) of the plurality of input couplers and the waveguide (11) of the second plurality of waveguides, the second transition (14) preferably comprising a rectangular waveguide impedance transformer of variable section.
5. Power amplifier according to one of the preceding claims, wherein the first combined signal is output by a first output (15) of the first sub-combiner (4), and the second combined signal is output by a second output (16) of the second sub-combiner (5), the power amplifier further comprising a third transition (18) between the first output (15) and the output coupler (7) and / or a fourth transition (19) between the output coupler (7) and the second output (16).
6. Power amplifier according to one of the preceding claims, wherein each waveguide (9,11) of the first plurality of waveguides and of the second plurality of waveguides is a rectangular waveguide comprising a capacitive element (12) positioned inside the waveguide (9,11) so as to form staircase steps of a depth measured in the radial direction in which the waveguide (9,11) extends equal to a quarter of the wavelength of the initial input electromagnetic signal.
7. Power amplifier according to one of the preceding claims, wherein the number of input couplers (6), the number of waveguides (9) of the first plurality of waveguides and the number of waveguides (11) of the second plurality of waveguides are equal.
8. Power amplifier according to one of the preceding claims, in which each amplifier channel (3) comprises a phase shifter of the elementary signal comprising a microstrip line of variable length.
9. Power amplifier according to the preceding claim, wherein the variable length microstrip line is formed by a first microstrip line (51) on a first printed circuit (52) and by a second microstrip line (53) and a third microstrip line (54) on a second printed circuit (55), the third microstrip line (54) being parallel to the second microstrip line (52), a first end (56) of the first microstrip line (51) being in contact with the second microstrip line (53) and a second end (57) of the first microstrip line (51) being in contact with the third microstrip line (54), the first printed circuit (52) being movable relative to the second printed circuit (55).
10. Power amplifier according to one of the preceding claims, the power divider (2) comprising: - a third cylindrical cavity (34) along the same axis of symmetry (Z), the third cylindrical cavity (34) comprising an input (35) of the initial electromagnetic input signal on the axis of symmetry (Z) and a plurality of outputs (36) of the third cylindrical cavity (34) on the same plane (P”) perpendicular to the axis of symmetry (Z), - a plurality of air-suspended strip-plate lines (37), each air-suspended strip-plate line (37) extending radially with respect to the axis of symmetry (Z) in the plane (P”) from inside the third cylindrical cavity (34) through an outlet (36) of the plurality of outlets of the third cylindrical cavity, each air-suspended strip-plate line (37) being configured to pick up an elementary signal of the plurality of elementary signals.
11. Power amplifier according to claim 9, comprising a coaxial input (48) extending axially along the axis of symmetry (Z) and opening into the third cylindrical cavity (34) via the input (35) of said third cylindrical cavity (34), the coaxial input (48) comprising a conductive core (49) which extends inside the third cylindrical cavity (34) up to the plane (P”).