A wall current based beam position detector

Abstract

The application provides a wall current based beam position monitor, which comprises a vacuum tube, strip electrodes extending along the length direction of the tube and spaced apart from the inner surface of the tube, the electrode starting end of each strip electrode is connected with a coaxial lead terminal on the electrode mounting section, and then connected with a signal acquisition system, the signal acquisition system comprises an extremely low input impedance high resistance amplifier connected with the terminal, a radio frequency front end and an ADC module, a soft magnetic material magnetic core is arranged between the electrode starting end and the inner surface of the vacuum tube, and the inner surface of the tube is plated with gold. The application introduces soft magnetic material to improve the characteristic impedance of the strip electrode, the inner surface of the vacuum tube is plated with gold, the low frequency cutoff frequency of the output signal is reduced by reducing the annular current; at the same time, the amplifier with extremely low input impedance is connected with the output end, the resistance value of the coaxial lead terminal is greatly reduced, the electrode coupling is reduced, and then the low frequency cutoff frequency of the differential mode signal is reduced.

Description

Technical Field

[0001] This invention relates to a beam position measurement device for accelerators, specifically to a strip beam position detector, which is used to reduce the low-frequency cutoff frequency and reduce inter-electrode coupling interference. Background Technology

[0002] Existing unobstructed beam position detectors (BPMs) generally employ electromagnetic induction detection, with their detection efficiency directly proportional to frequency; that is, the higher the detector's operating frequency, the better the system performance. However, since the frequency components of the beam within an accelerator are inversely proportional to the bundle length, and the slow-extraction beam in proton therapy has an extremely long beam length with very weak high-frequency components, conventional beam position detectors (BPMs) operating at higher frequencies are unsuitable for this type of beam condition due to their excessively high low-frequency cutoff frequencies.

[0003] Synchrotron-based tumor therapy devices typically employ a third-order resonant slow extraction structure. The longitudinal structure of the extracted beam is generally a drift beam of approximately ~s in length, but it includes kHz and MHz frequency components due to the operating frequency of the extraction devices; the corresponding time scale is on the order of ms and μs. Currently, there are no corresponding unobstructed online measurement methods. Taking the high-energy line of the Shanghai SAPT proton therapy project as an example, after slow extraction, there is only a profile target for beam tuning, but no unobstructed online measurement method for normal accelerator operation. During operation, end-point calibration relies on the position ionization chamber within the treatment head. Because of the lack of online position measurement methods within the high-energy line and its branches over a distance of tens of meters, monitoring and feedback of extraction and transmission efficiency during this period is impossible, significantly increasing the difficulty of maintaining dose stability in the treatment room.

[0004] Furthermore, the development of low-energy heavy particle beam application technology, taking a 2MeV H-particle application device of the China Changfeng Electromechanical Technology Research and Design Institute as an example, has an accelerating frequency of 425MHz, a micropulse length of 35mm, and a macropulse length of 50-300μs. Since its γ is approximately 1.002, its corresponding β is 0.063. Therefore, for a macropulse with a repetition frequency of 425MHz, the spacing between micropulses is approximately 45mm. The beam length and pulse spacing of the micropulses are almost equivalent. Since the longitudinal angle of the beam electromagnetic field is inversely proportional to γ, the lower beam energy worsens the beam length effect. Ultimately, the electromagnetic fields of the micropulses within the same macropulse are almost connected end-to-end, resulting in extremely weak high-frequency components of the electromagnetic field excited by the beam cluster. The main electromagnetic field components are concentrated in the ~kHz frequency band and its lower harmonics, determined by the macropulse length.

[0005] For ultra-long beams that are not focused or have poor focusing under low energy conditions, there is currently a lack of in-situ, high-precision, non-destructive measurement methods worldwide.

[0006] Currently, for focused beams, inductive beam position detectors such as button-type BPMs, strip BPMs (SBPMs), or triangular BPMs are generally used to perform non-destructive measurements of the beam position; alternatively, resonant cavity BPMs (CBPMs) can be used to perform non-destructive position measurements of the high-frequency components of the electromagnetic field deposited in the cavity structure.

[0007] However, for proton or heavy ion accelerators, poor beam focusing or even beam drift often occurs. In such cases, the output signal of an inductive detector typically exhibits the following characteristics:

[0008]

[0009] Among them, I b R is the spectral intensity of the beam, C is the resistance of the output circuit, and K is the detector capacitance, typically about 3 to 40 pF. K is a coefficient related to the detector size.

[0010] k=a×L / c0=a / ω0(2)

[0011] Where a is the angle subtended by the electrode to the beam, L is the electrode length, and c0 is the speed of light. Correspondingly, ω0 is the characteristic frequency of the detector.

[0012] Combining formulas (1) and (2), and assuming the beam length distribution is Gaussian, we can obtain:

[0013]

[0014] Where σ is the beam length and I0 is the average beam current intensity. It can be seen that for long beam clusters, due to the beam current spectral intensity I... b The exponential term in the spectrum decreases rapidly with increasing beam length, which in turn leads to a corresponding decrease in the selectable detector operating frequency band.

[0015] To ensure that the output signal U(ω) of formula (3) is strong enough, the exponent term must be kept out of the way, that is, the product of ω×σ in the exponent term must be kept out of the way. When the beam length σ increases, the working frequency band ω can only be reduced to keep the ratio of the product of the two to the speed of light in a suitable range.

[0016] After the operating frequency band is reduced, if the detector characteristic frequency remains unchanged, the ratio term of ω / ω0 in formula (3) will decrease linearly, thereby reducing the amplitude of the output signal.

[0017] However, the characteristic frequency of the detector cannot be reduced accordingly due to limited space. This results in a rapid decrease in the signal-to-noise ratio of the output signal of the inductive detector in the case of long bundles, making it difficult to apply to longer bundles. Generally, button-type BPMs can operate at frequencies up to 3 GHz, suitable for bundle lengths below 100 ps. Strip BPMs, on the other hand, can have detector lengths on the order of hundreds of millimeters, and their operating frequencies can be as low as 100 MHz, suitable for bundle lengths of 1 ns to 10 ns. The operating frequency of a strip BPM can be as low as 100 MHz.

[0018] The basic working principle of a triangular BPM is similar to that of a button BPM. Due to the use of a very large external resistor, its output signal amplitude and bandwidth are mainly determined by the electrode capacitance. Generally, the bandwidth is directly proportional to the capacitance, and the signal amplitude is inversely proportional to the capacitance. Its output signal response has the following form:

[0019]

[0020] By appropriately increasing the electrode length and selecting a suitable capacitor, its operating bandwidth can be as low as tens of kHz to 10 MHz, and the corresponding beam length can be as long as about 10 ns to 20 μs.

[0021] As for CBPM, since it picks up the high-frequency components of the beam bundle, the wavelength of the operating frequency band is linearly dependent on the transverse dimension of the detector; correspondingly, to ensure the Q value of the detector output signal, the longitudinal length of the detector is also linearly dependent on the transverse dimension. To control system costs and save beam longitudinal space, the transverse dimension of the detector generally needs to be controlled within tens of centimeters, corresponding to an operating frequency of several hundred MHz or higher, and an appropriate beam length of at least 1 ns.

[0022] For proton and heavy particle macropulse or drift beams, when the beam length exceeds 100 μs, the effective electromagnetic field component of the bundle electromagnetic field, even at higher energies, is only around kHz. Unfortunately, longer bundle lengths generally correspond to lower beam energies. For example, in a 2 MeV proton beam, γ is only around 1.002 (relative energy), while the longitudinal angle of the beam electromagnetic field is inversely proportional to γ. In other words, lower beam energy worsens the beam length effect, making other more conventional inductive beam position measurement methods unsuitable for this situation.

[0023] For low-energy bundles in the 100 μs range, the aforementioned non-destructive beam position measurement methods are largely unusable because the beam electromagnetic field components are concentrated below kHz. While scraping-type beam position detectors can be chosen, the large beam spot diameter and the corona effect at low energies mean that the amount of bundle charge scraped between electrodes may not have a simple linear dependence on the beam centroid position, making them unsuitable for this application. As for blocking-type beam position detectors, they can meet the beam position measurement requirements at this stage, but are not suitable for applications requiring beam feedback.

[0024] Foreign beam measurement researchers, during their research on wall current detectors, discovered that the differential-mode signal of the wall current detector could be used for beam position detection (see references [1, 2]). Based on the wall current detector, CERN researchers developed a wall current detector with position measurement capabilities. The common-mode signal of this detector can be used for beam intensity measurement; the differential-mode signal can be used for beam position measurement. Its common-mode signal low-frequency cutoff frequency meets theoretical expectations at 2.45 kHz; however, its differential-mode signal cutoff frequency exhibits an anomalous rise to 282 kHz. Based on this theoretical guidance, the CERN scheme made targeted improvements to address this anomalous cutoff frequency rise, raising its low-frequency cutoff frequency to 10 kHz. After introducing a separate amplifier specifically for low frequencies, the apparent cutoff frequency of its differential-mode signal was processed to 800 Hz (see references [3, 4]). However, since the signal-to-noise ratio was not truly improved, its applicable beam length is still determined by the true low-frequency cutoff frequency of its differential-mode signal at 10 kHz.

[0025] Since the low-frequency cutoff frequencies of the common-mode signal and differential-mode signal in the CERN scheme are 150Hz and 10kHz respectively, and based on the basic measurement principle, the common-mode signal is the sum of the electrode signals, while the differential-mode signal is the difference between the electrode signals. Without additional interference, the spectral distributions of the electrode signals, common-mode signal, and differential-mode signal should be consistent. The most likely reason for the severe degradation of the low-frequency cutoff frequency of the differential-mode signal is the very strong signal coupling between the electrodes in the low-frequency band; and the coupling is caused by the imbalance of the induced voltage amplitude between the electrodes when the beam deviates from the center. This is not the reason, as suggested in references [3][4], where the differential-mode signal inductance is much smaller than the common-mode signal inductance.

[0026] The overall structure of a SBPM (Strip Beam Position Detector) includes a vacuum assembly, electrodes, and a feedthrough for signal extraction. This device is a commonly used detector in the beam measurement field. One of the most important tasks in SBPM development is matching the characteristic impedance of the strip electrodes to achieve a good match with the 50-ohm characteristic impedance of the output path. However, what is theoretically very simple becomes extremely complex in practice due to the introduction of a multi-electrode structure. This is because coaxial transmission line theory posits that the number of independent internal conductor structures within a closed channel determines the number of intrinsic TEM modes. Each independent TEM mode corresponds to a specific electrode coupling mode and characteristic impedance distribution. Poor electrode design will result in good common-mode signal extraction but significant attenuation of differential-mode signals. From this perspective, the situation demonstrated at CERN closely matches the common-mode signal issues frequently encountered in SBPM design. Furthermore, the literature lacks specific content and considerations regarding the characteristic impedance matching of the electrode structure.

[0027] Furthermore, another explanation for low-frequency interference is that low-frequency signals, due to their longer wavelengths, are more easily coupled between electrodes. In SBPM design, an equivalent ground effect is typically introduced between electrodes to mitigate crosstalk.

[0028] References:

[0029] [1]Brian Fellenz,Jim Crisp,An improved Resistive Wall monitor

[0030] [2]P.odier,Geneva,Anew wide band wall current monitor,DIPAC.2003

[0031] [3]M.Gasior,GeneVa,An inductive Pick-up for beam position and current measurement,Proc.DIPAC(2003),P.53-55

[0032] [4] R.Corsini, et.al. Precision measurement of beam current, position and phase for an e+e-Linear collider, 1st workshop of ELAN. Summary of the Invention

[0033] The purpose of this invention is to provide a beam position detector based on wall current to reduce the low-frequency cutoff frequency and reduce inter-electrode coupling interference.

[0034] To achieve the above objectives, the present invention provides a beam position detector based on wall current, comprising a vacuum tube, the vacuum tube including an electrode mounting section and a conventional section other than the electrode mounting section; the inner diameter of the vacuum tube at the electrode mounting section is larger than the inner diameter at the conventional section, thereby forming an electrode groove; four strip electrodes extending along the length direction of the vacuum tube and spaced apart from the inner surface of the vacuum tube at the electrode mounting section are installed at the electrode groove, each strip electrode having an electrode start end and an electrode end at both ends along the length direction of the vacuum tube, the electrode start end being connected to the inner conductor of a coaxial lead-out terminal on an electrode mounting section, such that the electrode start end is connected to a signal acquisition system outside the vacuum tube through the coaxial lead-out terminal, the signal acquisition system including four ultra-low input impedance high-impedance amplifiers connected one-to-one with the four coaxial lead-out terminals, a radio frequency front-end connected to all the ultra-low input impedance high-impedance amplifiers, and an ADC module connected to the radio frequency front-end; a soft magnetic core is provided between the electrode start end and the inner surface of the vacuum tube at the electrode mounting section, and the inner surface of the vacuum tube is gold-plated.

[0035] The electrode is 200 mm long and the total length of the vacuum tube is 250 mm, such that the angle between the strip electrode and the center of the beam position detector based on wall current is 15°.

[0036] Each strip electrode has an electrode start and an electrode end at both ends along the length of the vacuum pipe. The electrode start is spaced apart from the electrode mounting section and the conventional section, and the electrode end is fixed on the end face facing the electrode mounting section.

[0037] Each strip electrode has a thickness of 2 mm, and the gap between each strip electrode and the inner surface of the vacuum pipe at the electrode mounting section is 5 mm. The conventional section includes a first conventional section upstream of the electrode mounting section and a second conventional section downstream of the electrode mounting section. The distance between the electrode start end and the first conventional section is 3 mm.

[0038] The soft magnetic material core is made of manganese-zinc ferrite, with a permeability better than 10,000, and its applicable frequency range includes at least DC to 5MHz.

[0039] The soft magnetic material core includes nine magnetic ring structures arranged sequentially along the length of the vacuum pipe. Each magnetic ring structure has a length of 21.5 mm, and its inner and outer diameters are 33.55 mm and 38.45 mm, respectively.

[0040] The surface of the strip electrode is plated with gold.

[0041] The negative terminal of the input of each ultra-low input impedance high-impedance amplifier is connected to the output terminal of the coaxial lead, and the positive terminal of the input of the ultra-low input impedance high-impedance amplifier is grounded.

[0042] The inner core and inner wall of the coaxial lead-out terminal are gold-plated, and the inner core of the wires of the ultra-low input impedance high-impedance amplifier are all gold-plated. The input impedance of the ultra-low input impedance high-impedance amplifier is less than 10μΩ.

[0043] The ADC module is an FPGA, and the coaxial output terminal is an SMA type coaxial feed-through.

[0044] 1) This invention improves the characteristic impedance of the strip electrode by introducing soft magnetic material, and at the same time, the inner surface of the vacuum tube is plated with gold to improve the conductivity of the inner surface of the vacuum tube outside the detector electrode, so that the resistance of the tube wall of the vacuum tube is small enough, thereby reducing the low frequency cutoff frequency of the output signal by reducing the loop current.

[0045] 2) The output terminal of the coaxial lead of the present invention is connected to an amplifier with extremely low input impedance, which greatly reduces the resistance value across the coaxial lead, reduces the ring current, and at the same time reduces the electrode voltage of the beam position detector of the present invention when it is working, avoids destroying the deviation of its 0 potential boundary condition, reduces the coupling between electrodes, and thus reduces the low frequency cutoff frequency of the differential mode signal.

[0046] 3) The present invention reduces the center angle of the electrode pair to the detector, thereby reducing the low-frequency coupling between the electrodes, reducing the main peak frequency on the spectral response curve, and thus increasing the low-frequency response amplitude of the beam position detector based on wall current of the present invention. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the overall structure of a beam position detector based on wall current according to an embodiment of the present invention;

[0048] Figure 2A It is along Figure 1 A schematic diagram of the cross-section of line AA in the diagram;

[0049] Figure 2B It is along Figure 1 A schematic diagram of the cross-section of the BB line in the diagram;

[0050] Figure 2C It is along Figure 1 A schematic diagram of the cross-section of the CC line in the diagram;

[0051] Figure 3A and Figure 3B This is a schematic diagram illustrating the working principle of a strip BPM and its relationship with wall current. Figure 3A The location of the current source formation on the strip BPM is shown. Figure 3B The direction of the wall current formed by the current source on the strip BPM is shown.

[0052] Figure 4 This is a typical spectrum of the output signal of a striped BPM.

[0053] Figure 5 This is a schematic diagram showing the connection between the coaxial lead-out terminal and the extremely low input impedance high-impedance amplifier in the beam position detector based on wall current of the present invention.

[0054] Figure 6 This is a circuit diagram of the signal acquisition system of the beam position detector based on wall current of the present invention.

[0055] Figure 7 These are cross-sectional views of the isolation electrode model and the DC blocking isolation electrode model used for comparison with the present invention.

[0056] Figure 8A This is the time-domain waveform of the output signal of the isolation electrode model.

[0057] Figure 8B This is the frequency domain waveform of the output signal of the isolation electrode model.

[0058] Figure 8C This is a diagram of the position sensitivity coefficients of the isolation electrode model.

[0059] Figure 9A This is the time-domain waveform of the output signal of the DC isolation electrode model.

[0060] Figure 9B This is the frequency domain waveform of the output signal of the DC isolation electrode model.

[0061] Figure 9C This is a diagram of the position sensitivity coefficients of the DC isolation electrode model.

[0062] Figure 10A , Figure 10B These are time-domain and frequency-domain waveforms of a set of relative electrode signals based on the SBPM model of wall current according to the present invention.

[0063] Figure 10C This is a graph showing the electrode sensitivity coefficients of the SBPM model based on wall current in this invention when the load resistance is 6.3 ohms.

[0064] Figure 10D The above is the signal-to-beam spectrum response diagram of the SBPM model based on wall current of this invention when the load resistance is 6.3 ohms.

[0065] Figure 10E This is the differential-mode signal spectrum of the SBPM model based on wall current of this invention when the load resistance is 6.3 ohms.

[0066] Figure 11A The characteristic impedance is 3 ohms, mu = 400, the electrode angle is 20 degrees, and the spectrum of the electrode sensitivity coefficient of Model 3 is shown.

[0067] Figure 11B This is the frequency domain distribution diagram of the differential mode signal when the characteristic impedance is 3 ohms, mu = 400, and the electrode angle is 20°.

[0068] Figure 12 This is a simulation result of the differential mode signal after reducing the electrode angle.

[0069] Figure 13 This is the spectrum of the differential signal when the permeability increases to 6400.

[0070] Figure 14 This is the frequency domain distribution of position sensitivity when the permeability increases to 6400.

[0071] Figure 15 This is the time-domain waveform of the electrode output signal when the magnetic core permeability is 6400.

[0072] Figure 16A This is the spectrum of the output signals of the two opposite electrodes when the load is 6.3 ohms and the simulation duration is 0.35μs.

[0073] Figure 16B This is the spectrum of the detector position sensitivity coefficient obtained from simulation with a load of 6.3 ohms and a simulation duration of 0.35 μs.

[0074] Figure 17A This is the time-domain waveform of the output signal of the two opposite electrodes when the load is 3.1 ohms.

[0075] Figure 17B This is the spectrum of the output signal from the relative electrode and the signal spectrum diagram when the load is 3.1 ohms.

[0076] Figure 17C This is the spectrum of the detector position sensitivity coefficient obtained from the simulation with a load of 3.1 ohms and a simulation duration of 0.35 μs. Detailed Implementation

[0077] The preferred embodiments of the present invention are given below with reference to the accompanying drawings and described in detail.

[0078] like Figure 1 , Figures 2A-2C The diagram shows a beam position detector based on wall current according to an embodiment of the present invention, which includes a vacuum conduit 10, the vacuum conduit 10 including an electrode mounting section 11 and conventional sections other than the electrode mounting section. The conduit wall of the vacuum conduit 10 has an inner surface and an outer surface, the inner surface being the surface facing the central axis of the vacuum conduit 10.

[0079] The outer diameter of the vacuum conduit is the same at the electrode mounting section 11 and the conventional section. The inner diameter of the vacuum conduit at the electrode mounting section 11 is larger than that at the conventional section, thus forming an annular electrode groove. Specifically, the outer diameter of the vacuum conduit is D0, the inner diameter at the conventional section is D1, and the inner diameter at the electrode mounting section 11 is D2. In this embodiment, there are two conventional sections: a first conventional section 13 located upstream of the electrode mounting section 11 and a second conventional section 14 located downstream of the electrode mounting section 11.

[0080] Four strip electrodes 20 are installed in the electrode recess, extending along the length of the vacuum pipe 10 and spaced apart from the inner surface of the vacuum pipe at the electrode mounting section 11. Each strip electrode 20 has an electrode start end 21 and an electrode end end 22 at both ends along the length of the vacuum pipe 10. The electrode start end 21 is spaced apart from both the electrode mounting section 11 and the conventional section, and is connected only to the inner conductor 31 of a coaxial lead-out terminal 30 on one electrode mounting section 11, allowing the electrode start end 21 to connect to a signal acquisition system outside the vacuum pipe 10 via this coaxial lead-out terminal 30. The electrode end end 22 of each strip electrode 20 is fixed to the conventional section, preferably to the end face of the conventional section facing the electrode mounting section 11, thereby fixing the strip electrode 20.

[0081] Thus, the electrode mounting section 11 and its strip electrode 20 form a quasi-coaxial structure, with the strip electrode 20 serving as the inner conductor and the electrode mounting section 11 as the outer conductor. In this embodiment, the resistance of the quasi-coaxial structure is 50 ohms.

[0082] The outer surface of the vacuum pipe is grounded and is continuous, with two circular holes provided only at the coaxial lead-out terminals 30 on the electrode mounting section 11 for mounting the coaxial lead-out terminals 30.

[0083] Each strip electrode 20 is 2 mm thick, and the gap between each strip electrode 20 and the inner surface of the vacuum pipe at the electrode mounting section 11 is 5 mm. The length of the electrode is 200 mm. The total length of the vacuum pipe 10 is 250 mm, and the angle between the strip electrode 20 and the center of the detector is 15°.

[0084] Among them, the electrode subtend angle is Figure 2B The angle subtended by the electrode relative to the center of the vacuum tube within any specific cross-section. The electrode angle of the strip electrode 20 is independent of the electrode length. A larger electrode angle results in a stronger signal, but also stronger coupling. The former is beneficial for improving measurement resolution, while the latter will increase low-frequency coupling. Therefore, the choice of electrode angle always involves a balance between these two factors.

[0085] The distance Gap between the electrode start end 21 and the first conventional segment 13 is 3 mm.

[0086] A soft magnetic core 40 is provided between the electrode start end 21 and the inner surface of the vacuum pipe at the electrode mounting section 11. The soft magnetic core 40 is made of manganese-zinc ferrite with a permeability better than 10000 and an applicable frequency range of at least DC-5MHz. In this embodiment, the soft magnetic core includes nine magnetic ring structures arranged sequentially along the length of the vacuum pipe 10. Each magnetic ring structure has a length of 21.5mm, and its inner diameter and outer diameter are 33.55mm and 38.45mm, respectively. The inner diameter has a positive tolerance, and the outer diameter has a negative tolerance.

[0087] The inner surface of the vacuum tube is continuous in the conventional section. The inner surface of the vacuum tube 10 is gold-plated to make the resistance of the tube wall of the vacuum tube 10 sufficiently low, thereby reducing the low-frequency cutoff frequency of the output signal by reducing the circulating current.

[0088] In addition, the surface of the strip electrode 20 is also gold-plated, which reduces the wall current transmission impedance.

[0089] like Figure 5 and Figure 6 As shown, the signal acquisition system includes four ultra-low input impedance high-impedance amplifiers 41, each corresponding to one of the four coaxial leads 30; a radio frequency front-end 42 connected to all the ultra-low input impedance high-impedance amplifiers 41; and an ADC module 43 connected to the radio frequency front-end 42. Furthermore, Figure 6 The feedback resistor Rt of the ultra-low input impedance high-impedance amplifier 41 is a characteristic parameter of the current-voltage conversion of the ultra-low input impedance high-impedance amplifier 41, which should be determined according to the magnitude of the beam current to be measured and the input amplitude of the subsequent ADC module. The negative terminal of the input of each ultra-low input impedance high-impedance amplifier 41 is connected to the output terminal of the coaxial lead 30, and the positive terminal of the input of the ultra-low input impedance high-impedance amplifier is grounded. Each ultra-low input impedance high-impedance amplifier 41, as a device connected across the coaxial lead 30, has an input resistance R1 that is the resistance value Ring connected across the coaxial lead 30. Therefore, significantly reducing the resistance value Ring connected across the coaxial lead 30 can improve the low-frequency cutoff frequency of the electrode signal. The feedback resistor Rt and the input resistance R1 of the ultra-low input impedance high-impedance amplifier 41 are characteristic values ​​displayed to the outside world by the input terminal of the ultra-low input impedance high-impedance amplifier 41, and are its overall properties.

[0090] The ADC module 43 is preferably a digital signal processing module with FPGA functionality. The coaxial lead 30 is preferably an SMA type coaxial feed-through. The inner core and inner wall of the coaxial lead 30 are gold-plated, and the inner core of the conductors of the ultra-low input impedance high-impedance amplifier 41 are all gold-plated. The input impedance of the ultra-low input impedance high-impedance amplifier is less than 10μΩ, thereby minimizing the resistance value across the coaxial lead.

[0091] The principle behind the beam position detector based on wall current of the present invention being able to reduce the low-frequency cutoff frequency is explained in detail below.

[0092] According to the research of this invention, the limiting factors of the low-frequency response of strip BPM include the following:

[0093] For a beam operating within a well-conducting vacuum tube, due to Gauss's law, there is no electromagnetic field distribution within the tube wall. That is, the transverse electric field emitted by the beam terminates entirely on the inner wall of the vacuum tube, forming a mirror charge. When the beam is relatively long and moves stably at a certain speed, the electromagnetic field distribution is equivalent to the electromagnetic field distribution formed by a pulsed current along the same axis. Correspondingly, an induced charge flow, i.e., wall current, forms on the inner wall of the vacuum tube, with the opposite sign and the same direction of motion as the source charge. The angular distribution of the wall current on the inner wall is determined by the position of the source charge's center of mass; when the beam's center of mass deviates slightly from the tube center, the angular distribution of the wall current is linearly dependent on the beam's center of mass. When the beam charge moves to a discontinuous structure on the inner wall of the vacuum tube, due to the departure or entry of the beam charge, a current source is formed at that structure due to the change in the induced charge of the beam. Since the wall current is driven by the beam, it can be considered a current source in most cases. For a continuous metallic vacuum inner wall, the wall current and the beam current are equal in magnitude but opposite in direction, except for the DC component. For the DC component of the beam, since it has a steady distribution, only an induced charge distribution forms on the vacuum inner wall, not an induced current. In the case of a single beam passage, since the DC component of the beam tends to 0, the induced wall current is equal in magnitude but opposite in direction to the beam current.

[0094] Based on the principle of strip BPM, through theoretical derivation and simulation verification, it was found that the main mechanism limiting the low-frequency response of strip BPM (SBPM) lies in the existence of a loop current path (wall current path) between the strip electrode of the strip BPM and the inner wall of the vacuum tube. If this current path is blocked, the frequency response of the SBPM will approach the beam spectrum itself infinitely, avoiding the defect that the efficiency of inductive detectors is proportional to the operating frequency. Theoretically, by increasing the impedance of the loop current path and improving the conductivity of the vacuum chamber wall, the extracted component of the loop current path can be reduced.

[0095] Typically, strip BPMs output beam position information by sensing the time-domain changes in their wall current. The following section combines... Figure 3A and Figure 3B Specific theoretical derivations are performed to explain the basic working principle of strip BPM and its relationship with wall current, and further to explain how the beam position detector based on wall current of this invention reduces the low-frequency cutoff frequency.

[0096] like Figure 3A and Figure 3B As shown, due to Gauss's law, when the bundle reaches the end of the first conventional section 13 near the electrode mounting section 11 (i.e., point SA), due to the structural abrupt change, the wall current charge originally induced at that point loses the driving force of the bundle charge and will inevitably dissipate at point SA. Considering that there is no current inside the metal for high-frequency components, the accumulated induced charge at point SA is equivalent to forming a first current source with a charge outflow value of Ib. Half of this current source charge, 0.5Ib, is reflected along the inner surface of the vacuum pipe (i.e., propagating in the reverse direction along the inner surface of the first conventional section 13), while the other half, 0.5Ib, continues to propagate along the inner surface of the vacuum pipe (i.e., propagating along the inner surface of the vacuum pipe in the electrode mounting section 11), that is, flowing along the inner surface of the outer conductor of the coaxial structure formed by the electrode mounting section 11 and its strip electrode 20; after a delay of 2Lg / c1 (Lg is the electrode length), this part of the current is output from the coaxial lead-out terminal 30 through the outer surface of the strip electrode 20.

[0097] Similarly, such as Figure 3A and Figure 3B As shown, at time t1, at the electrode end (i.e., point SB) of the strip electrode 20, due to the driving force of the beam charge, an equivalent second current source with an induced current flowing into it is formed at that point, and the current value of the second current source is Ib. Half of this induced current comes from the coaxial lead-out terminal 30; the other half comes from the outer surface of the strip electrode 20 (i.e., the side of the strip electrode 20 facing the beam), that is, the outer surface of the inner conductor of the coaxial structure formed by the electrode mounting section 11 and the strip electrode 20.

[0098] In other words, the current of both the first and second current sources is Ib, which is represented as Ib(t) in the time domain and Ib(w) in the frequency domain.

[0099] The output signal of the coaxial lead-out terminal 30 of SBPM is partly derived from the induced current at point SB at time t1. Part of the induced current at point SB provides a signal to the coaxial lead-out terminal 30. The other part is derived from the induced current at point SA at time t0, which forms a ring current on the inner surface of the coaxial structure. After a delay of 2Lg / c1, it is output through the coaxial lead-out terminal 30.

[0100] Therefore, the output signal of coaxial lead-out terminal 30 is:

[0101] I(t)=0.5(Ib(t1)-Ib(t0-2Lg / c1)) (5)

[0102] Where Ib(t1) is the induced current of the current source at time t1, and Ib(t0-2Lg / c1) is the current from the induced current of the current source at time t0 that reaches the coaxial lead-out terminal through the inner surface of the coaxial structure.

[0103] To reduce high-frequency impedance, the gap between the electrode start end 21 and the inner surface of the first conventional section 13 of the vacuum pipe is very small (i.e., the gap between point SB and point SA is very small), so t0 and t1 can be considered to be the same.

[0104] The frequency domain expression I(w) of the output signal of the coaxial lead-out terminal 30 is:

[0105] I(w)=0.5(1-exp(jw×2Lg / c1))×Ib(w)(6)

[0106] Where w is the angular frequency, Lg is the electrode length, c1 is the group velocity of the current, and Ib(w) is the current of the current source with angular frequency w.

[0107] The strip electrode 20 is a long strip of metal. The two opposing surfaces of the strip electrode 20 can be divided into an inner wall surface facing the center of the pipe, where the current is directly driven by the beam current, and an outer wall surface facing away from the center of the pipe, where the current is only indirectly driven by the beam current.

[0108] On strip electrode 20, there exists simultaneously a current flowing through the outer wall surface of strip electrode 20 and a wall current driven by cluster charges flowing through the inner wall surface of strip electrode 20, and the two currents are in opposite directions. Therefore, the total current magnitude Ie(t) on strip electrode 20 is:

[0109] Ie(t)=Ib(t1)-0.5Ib(t1)=0.5Ib(t1)(7)

[0110] Where Ie(t) is the total current on the strip electrode 20, Ib(t1) is the induced current of the current source at time t1, which represents the wall current driven by the bundle charge flowing through the inner wall of the strip electrode 20, and 0.5Ib(t1) is half of the induced current of the current source at time t1, which represents the current flowing through the outer wall of the strip electrode 20.

[0111] As can be seen from expression (6), the DC response of the coaxial lead-out terminal 30 is 0, and the low-frequency response is proportional to the operating frequency.

[0112] The spectral response of the output signal at the coaxial terminals of a typical SBPM (strip BPM) is as follows: Figure 4 As shown, where, Figure 4 These results are based on classic SBPM parameters, but the analysis theory is not based on classic directional coupling theory. The main peak frequency on the spectral response curve is equal to c1 / 4Lg; from DC to the first main peak, the signal amplitude is approximately proportional to the frequency. Since the amplitude of the first main peak is only related to the beam current parameters and the electrode angle, increasing the electrode length or decreasing the group velocity c1 of the current at that point can increase the low-frequency response amplitude of the SBPM. However, due to spatial limitations in electrode length, the ω0 of the SBPM is generally above 500MHz. Therefore, the spectral intensity of the SBPM output signal at 10kHz is only about -94dB of its peak value, which is insufficient for long-beam clusters.

[0113] For SBPM applications involving longer proton beams, a natural approach is to extend the electrical length of the strip electrode 20 until the first peak of the output signal falls within the effective spectral bandwidth of the beam. Since the spatial length of the electrode is limited by the accelerator, extending the electrical length of the strip electrode 20 can be achieved by increasing the permeability mu of the medium between the strip electrode 20 and the wall of the vacuum tube. Because the group velocity is proportional to the 0.5th power of mu, increasing the permeability of the medium can at most reduce the first peak of the SBPM to the MHz level. However, this is clearly insufficient for low-energy proton beam applications.

[0114] When in Figure 1 After the gap in the coaxial structure formed by the strip electrode 20 and the wall of the vacuum pipe is filled with magnetic medium, since the characteristic impedance of the coaxial structure depends on the permeability of the medium, at the current source SA point, the magnitudes of the current Ir(t) reflected along the wall of the vacuum pipe and the current If(t) flowing through the loop current path become:

[0115] Ir(t)=Ib(t0)*(R0+jωL) / (2R0+jωL) (8)

[0116] If(t)=Ib(t0)*R0 / (2R0+jωL) (9)

[0117] Where Ib(t0) is the induced current of the current source at time t0, L is the inductance of the loop current path, and R0 is the resistance of the vacuum tube wall. When the frequency approaches 0, the above two equations degenerate into the strip BPM case mentioned above.

[0118] Correspondingly, at point SB, the current Iin(t1) flowing into the coaxial lead-out terminal 30 and the current Icir(t) flowing into the outer surface of the strip electrode 20 are respectively:

[0119] Iin(t1)=Ib(t1)*(R2+jwL) / (R1+R2+jwL) (10)

[0120] Icir(t)=Ib(t1)*R1 / (R1+R2+jwL) (11)

[0121] Where Ib(t1) is the induced current of the current source at time t1, R1 is the input resistance of the ultra-low input impedance high-impedance amplifier, i.e., the resistance value across the coaxial lead terminal; R2 is the equivalent characteristic impedance after introducing soft magnetic material into the quasi-coaxial structure; L is the inductance of the ring current path; ω is the angular frequency.

[0122] Similarly, as the frequency approaches 0, the above two equations degenerate into the case of strip BPM. When the frequency increases to a point where the inductive reactance is much greater than the resistance R0 of the vacuum tube wall and the equivalent characteristic impedance R2 after the introduction of soft magnetic material into the coaxial structure, the current in the loop current path decreases sharply.

[0123] For a typical frequency band of interest, R0 and R2 are both much smaller than the inductive reactance, so the output signal is the difference between equations (9) and (10) (current reversed), that is, the output signal is:

[0124] Iou(t)=Ib(t)*jwL / (R1+jwL)-Ib(t-2Lg / c)*R0 / jwL (12)

[0125] As can be seen from formula (12), the smaller the resistance value Rin (i.e., the input resistance R1 of the ultra-low input impedance high-impedance amplifier 41) connected across the coaxial lead terminal, the larger the current at the output terminal, especially in the low-frequency range. The smaller R0 is, the smaller the loop current component is, and consequently the smaller the amplitude of its cancellation of the output current; correspondingly, the larger the low-frequency component of the output signal. The larger the inductance L is, the larger the output current is, and at the same time, the smaller the loop current component is; thus, the larger the low-frequency component of the signal is. In addition, the introduction of high permeability material leads to a decrease in the group velocity of the current signal in the coaxial line segment, which is equivalent to indirectly lengthening Lg, and can also improve the low-frequency component of the signal.

[0126] In fact, as long as the equivalent characteristic impedance R2 of the coaxial structure after introducing soft magnetic material is much greater than the resistance value across the coaxial lead 30 (i.e., the input resistance R1 of the extremely low input impedance high impedance amplifier 41), then regardless of the frequency band, formula (10) will degenerate into the form we most desire, namely:

[0127] Iin(t1)=Ib(t1)*(R2+jwL) / (R1+R2+jwL)=Ib(t1) (13)

[0128] That is, all wall currents are input and output through the coaxial terminal 30. Due to the introduction of soft magnetic material, the preconditions are easily met. Equation (9) is actually very close to 0 in most frequency bands. In particular, when the conductivity of the inner wall surface of the electrode mounting section 11 with soft magnetic material is much smaller than that of other sections, equation (9) can be considered 0. Therefore, at this time, equation (12), which is the output signal of the coaxial terminal, is actually Ib(t).

[0129] For typical wall current detectors used for bundle longitudinal waveform measurement and charge measurement, a large high-frequency bandwidth is required to achieve distortion-free waveform measurement. Simultaneously, minimal signal tailing is also desired to reduce measurement errors; therefore, a low-frequency cutoff frequency is also necessary. When the input resistance R1 of the extremely low input impedance high-impedance amplifier 41 approaches R0, the detector's low-frequency cutoff frequency approaches a minimum. At this point, the high-frequency cutoff frequency is primarily controlled by the gap distributed capacitance, which is not considered in the model of this invention.

[0130] Simulation results:

[0131] Because the low-energy proton beam is exceptionally long, its electromagnetic field bandwidth may only be on the order of kHz, which does not meet the differential-mode signal bandwidth operating range mentioned above. To further optimize the WSBPM design, in-depth modeling and analysis of the relevant models were conducted. The two main focuses of optimization are: first, reducing the low-frequency cutoff frequency of the electrode output signal, at which point the maximum amplitude of the output signal is proportional to the total charge, rather than the current intensity; and second, significantly reducing the low-frequency coupling between the detector electrodes to lower the low-frequency cutoff frequency of the differential-mode signal, thereby increasing the application range of position measurement in low frequencies. In this simulation example, the following three basic physical models were established.

[0132] In the simulation, the beam length selected was a Gaussian pulse, and the output result of the detector's low-frequency response was obtained through frequency domain analysis.

[0133] (I) Isolation Electrode Model

[0134] The cross-section of the isolation electrode model is as follows Figure 7 As shown, it only includes two position electrodes (i.e., strip electrodes 20) connected to the coaxial lead-out terminal 30. On both sides of the position electrodes, isolation electrodes 20' with a larger opening angle are added. The two ends of the isolation electrodes 20' are directly connected to the vacuum pipe.

[0135] At this time, the time-domain and frequency-domain waveforms of the output signal are as follows: Figure 8A and Figure 8B As shown, where Figure 8A The horizontal axis represents time, and the vertical axis represents voltage; Figure 8B Then it is Figure 8AThe FFT result is such that the horizontal axis represents frequency and the vertical axis represents V / Hz.

[0136] Clearly, the low-frequency bandwidth of the output signal at this point is reduced to only about 50MHz due to the presence of a quasi-reflection wave, which is insufficient for space BPM applications. Further analysis reveals that the quasi-reflection wave is not generated by the ring current at the magnetic core, as changing the core permeability does not alter the bandwidth. Simulation verification shows that the quasi-reflection wave is caused by the wall current introduced by the isolation electrode, which counteracts the echo signal generated by the beam. Its main peak frequency is only related to the vacuum chamber aperture. At first glance, this model seems to offer no advantage, but in-depth analysis reveals its greatest advantage lies in the almost complete isolation between the two position electrodes, meaning the differential-mode and common-mode signal spectral characteristics are completely consistent, with no differential-mode attenuation due to low-frequency coupling. Furthermore, the position sensitivity coefficient of this model for position measurement perfectly matches the theoretical model (R / 2). Figure 8C The frequency domain distribution of the position sensitivity coefficient of the isolation electrode model is shown (position sensitivity coefficient = common-mode signal × beam position / differential-mode signal). Figure 8C The horizontal axis represents frequency, and the vertical axis represents sensitivity coefficient, with the unit being mm. The larger the value, the greater the measurement error of the system under the same signal-to-noise ratio. When the detector is actually working, its value will increase due to various reasons, that is, the low-frequency cutoff frequency will rise. The physical meaning is shown in formula (14) below.

[0137] After in-depth thinking and multiple modeling, it was found that the realization of perfect isolation lies in the fact that the wall current on the isolation electrode is 0 potential. At this time, the boundary conditions on the position electrode are mainly determined by the isolation electrode and the adjacent 0 potential vacuum chamber, and 0 potential completely conforms to the theoretical model of wall current distribution.

[0138] (II) DC Isolation Electrode Model

[0139] To avoid the influence of reverse current in the isolation electrode, a DC-blocking isolation electrode model is established. At this time, with... Figure 7 Similarly, it only includes two position electrodes (i.e., strip electrodes 20) connected to the coaxial lead-out terminal 30, with isolation electrodes 20' of a larger opening angle added on both sides of the position electrodes. The difference is that the isolation electrodes 20' are not directly connected to the starting end, but are separated by a DC interruption, and high-frequency electrical connection is achieved through the capacitance formed by this interruption.

[0140] At this time, the time-domain and frequency-domain waveforms of the output signal are as follows: Figure 9A and Figure 9B As shown. At this point, because DC does not pass through the isolation electrode, the reflected wave is significantly reduced, but it still exists. This is reflected in the frequency spectrum as a decrease in the low-frequency components. Figure 9BThe electrodes closer to the cluster show a significant decrease in low-frequency sensitivity due to inter-electrode coupling. This indicates strong inter-electrode coupling, which is reflected in the spectral distribution of the electrode sensitivity coefficient as a sharp increase in the low-frequency sensitivity coefficient.

[0141] Since resolution is proportional to the electrode sensitivity coefficient and inversely proportional to the system signal-to-noise ratio, that is:

[0142] Δx=k / SNR(14)

[0143] Δx is the resolution, k is the electrode sensitivity coefficient, and SNR is the system signal-to-noise ratio.

[0144] Assuming the signal-to-noise ratio of the electrode output signal is constant. Figure 9C The sharp increase in the sensitivity coefficient at low and high frequencies means that, according to formula (14), the position measurement resolution of the system in the low-frequency band will deteriorate drastically, implying difficulties in applying the model to long-beam clusters. Furthermore, the sensitivity coefficient has increased from the theoretical value of 12.5 to around 20 over a larger bandwidth. Finally, the appearance of a reflection-like wave also suggests that the DC-isolated isolation electrode model is unsuccessful; however, there is theoretically a possibility for continuous improvement. However, analyzing a large isolation capacitor inevitably leads to a situation with an extremely small isolation distance, and the simulation program's calculation time increases quadratically to distinguish this extremely small isolation distance. Therefore, it is difficult to continuously optimize the DC-isolated isolation electrode model through simulation.

[0145] (III) SBPM model based on wall current without isolation electrodes (i.e., this invention)

[0146] The wall current-based SBPM model of this invention is shown in the figure below, completely eliminating the isolation electrode. The signal pickup method is a wall current pickup scheme, but the electrode structure optimization adopts a conventional SBPM scheme to reduce inter-electrode coupling. The cross-section of the wall current-based SBPM model is shown below. Figure 2B As shown.

[0147] Since optimizing the isolation electrode is difficult to continue, the beam detector based on wall current in this invention adopts a wall current-based SBPM model, in which the isolation electrode is eliminated. Considering the importance of the signal extraction load, the system performance under different load conditions is compared in detail. Figure 10A , Figure 10B This is a set of time-domain waveforms and spectra of relative electrode signals based on the SBPM model of wall current according to the present invention, wherein the magnetic core permeability is 400.

[0148] Depend on Figure 10BIt can be seen that, although the appropriately optimized SBPM model of the beam detector based on wall current of this invention still exhibits low-frequency coupling and some echo signal under low permeability conditions, the coupling is actually less severe compared to the DC-isolated isolation electrode model. This is reflected in the sensitivity coefficient, where the low-frequency sensitivity coefficient increases, but not significantly. Figure 10C As shown; simultaneously, the bandwidths of the system's differential-mode and common-mode signals are as follows: Figure 10D , Figure 10E As shown.

[0149] Depend on Figure 10D As can be seen, the signal experiences no attenuation relative to the beam across the entire bandwidth. In other words, this model exhibits no attenuation across frequency bands, except for the DC cutoff limitation. This represents a significant improvement over results reported in the literature.

[0150] from Figure 10C Distribution of k-values ​​of the system position measurement sensitivity coefficient and Figure 10E From the differential-mode signal spectrum, the SBPM model based on wall current of this invention still exhibits low-frequency coupling between electrodes. However, compared to the DC-isolated isolation electrode model, the K value has decreased from over 300 to 58, indicating a significant improvement in coupling performance. Figure 10E The results show that the differential-mode signal at low frequencies decreased by about 10 dB compared to the uncoupled frequency band. However, due to low-frequency coupling, the decrease in differential-mode signal at low frequencies in existing technologies is all above 40 dB.

[0151] Due to the electrode sensitivity coefficient, even very close to DC, it only increases from around 20 to around 58, which translates to a 2.9-fold degradation in system resolution. However, since this model is in a fully pass-through state except for the DC component, with appropriate parameter matching, reflection-like waves almost completely disappear, meaning the cutoff frequency will be very low. Preliminary assessment suggests that, with reasonable optimization of system parameters, the resolution of the wall current-based SBPM model in this invention can be achieved to be close to the performance of a standard SBPM.

[0152] Current simulation results show that Figure 10C The inflection point frequency and amplitude of the sensitivity coefficient increase should be positively correlated with the load resistance. However, simulating smaller electrode loads requires a smaller simulation grid, resulting in excessively long computation times.

[0153] The parameter optimization process of the beam position detector based on wall current in this invention:

[0154] Based on the preceding analysis, the low-frequency coupling between electrodes in the SBPM model based on wall current of this invention should continue to improve as the load resistance decreases. Since the simulation program requires a smaller mesh to simulate a smaller load resistance, and the simulation calculation time is inversely proportional to the fourth power of the minimum mesh size, after considerable effort, the minimum load impedance currently calculated in the simulation is 3.1 ohms. Under the same conditions for other parameters, the degree of low-frequency coupling does indeed continue to decrease. Figure 11A , Figure 11B This refers to the sensitivity coefficient distribution and differential signal spectrum at this time.

[0155] from Figure 11A , Figure 11B It can be seen that after reducing the load resistance, the position sensitivity coefficient of the WSBPM near the low-frequency cutoff frequency only increases to about 28 compared to about 20 in the higher frequency band; the relative amplitude of the differential mode signal also decreases by only about 1.7 dB. With the system signal-to-noise ratio remaining unchanged, the position measurement resolution is only twice that of a conventional SBPM measuring the same charge in the higher frequency band. Considering that the beam charge in low-energy proton beam applications is much larger than that in high-energy electron applications, the system resolution can be expected to be controlled within 10 μm.

[0156] The table below shows the simulation results of the SBPM model based on wall current of this invention under different load resistance conditions:

[0157] Table 1. Simulation results of the SBPM model based on wall current of the present invention under different load conditions with an electrode angle of 20 degrees and mu = 40°.

[0158]

[0159]

[0160] In addition, preliminary calculations were performed for different electrode angles and core permeability to verify the theoretical analysis results above.

[0161] Figure 12 The simulation results are for a differential-mode signal with an electrode angle of 12 degrees, a characteristic impedance of 3.1 ohms, and a permeability of 400 Ω. From... Figure 12 The results show that, apart from the reduction in fluctuations at high frequencies when the electrode angle is decreased, there is no difference near DC.

[0162] Figure 13 , Figure 14 These are the simulation results of the differential mode signal and position sensitivity when the magnetic core permeability is 6400.

[0163] from Figure 13 , Figure 14It can be seen that increasing the permeability slightly improves both the low-frequency coupling and position sensitivity of the differential-mode signal. For example, if the K value decreases from 28 to 27, it seems unnecessary to pursue excessively high magnetic permeability. However, if... Figure 15 As shown, from the perspective of the time-domain signal, the electrode output time-domain signal at this time is compared to... Figure 10A The result can be considered perfect, and it seems there is no risk of any additional low-frequency cutoff. Further experimental verification is needed to determine whether there is any additional low-frequency cutoff under low permeability conditions.

[0164] In-depth analysis of the low-frequency cutoff risk of the beam position detector based on wall current of the present invention:

[0165] Since the aforementioned simulation results were obtained under relatively short beam lengths, involving only short time-domain simulations, and then transformed to the frequency domain, the accuracy of the results is limited by the total computation time or beam length. Low-frequency results with characteristic times longer than the bundle length used in the simulation may yield erroneous analysis results due to the computation time cutoff during simulation. In other words, the risk of low-frequency cutoff cannot be eliminated through simulation analysis with limited time.

[0166] Considering the impact of computation time and beam length on the reliability of the computation results, simulations were performed with a load of 6.3 ohms and 3.1 ohms under beam parameters of beta = 0.1, beam length of 1m, and total simulation time of 0.35us. The simulation time was increased by about 10 times compared to the aforementioned model.

[0167] At this point, the simulation results for a load of 6.3 ohms are as follows: Figure 16A , 16B As shown. From Figure 16A The results show that the output signal spectrum is still quite perfect, with no additional low-frequency cutoff; however, it is obvious that the amplitude difference between the two signals decreases near the DC end. Correspondingly, in Figure 16B As can be seen above, the electrode sensitivity coefficient shows a significant increase around ~DC-8MHz. Compared to... Figure 10C The results show that, due to differences only in beam length and computation time, there are significant differences in the inflection point frequency of low-frequency coupling and the deterioration of position sensitivity. In other words, the longer the beam length and simulation time are set during simulation, the closer the analysis of the detector's low-frequency behavior will be to reality. It can be expected that as the simulation time and beam length increase, the inflection point of the detector's position sensitivity coefficient will continue to decrease until it approaches reality. This results in a lower, more realistic low-frequency cutoff frequency for the differential mode signal, and a longer beam length application range.

[0168] Simulation results with a load of 3.1 ohms are as follows: Figures 17A-17C As shown. Figure 17C Compared to Figure 11A Similarly, the only differences are in simulation duration and beam length. The differences in the calculation results are similar to those when the load is 6.3 ohms, that is, the inflection frequency of the sensitivity coefficient drops significantly, and the differential mode signal becomes smaller at low frequencies due to coupling, meaning that the increase in the detector sensitivity coefficient is more severe. Similarly, the common mode signal does not attenuate at low frequencies.

[0169] at the same time, Figure 17C and Figure 16B The only difference is the load impedance. As expected, the lower load impedance reduced the inflection point frequency of the sensitivity coefficient from approximately 8 MHz to 4 MHz. Similarly, the differential-mode signal attenuation due to coupling at low frequencies was also reduced, with the sensitivity coefficient decreasing from approximately 2000 to 200.

[0170] Simulation models verified the technical solution provided by this invention, significantly reducing the low-frequency cutoff frequency of the SBPM. Further simulation analysis revealed that while the low-frequency cutoff frequency of the common-mode signal continuously decreased, the low-frequency cutoff frequency of the differential-mode signal exhibited an abnormal increase at low frequencies. Further theoretical and simulation analysis indicated that this abnormal increase stemmed from the fact that the potential on the strip electrodes of the SBPM was not zero during operation, violating the boundary condition that the potential of each electrode must be zero under uncoupled operation. The introduction of non-zero potential caused differential-mode signal coupling between the electrodes. Since the strength of signal coupling is positively correlated with wavelength, this led to the abnormal increase in the low-frequency cutoff frequency of the differential-mode signal.

[0171] Simulation results show that, when applied to macropulses of low-energy proton beams at the 100μs level, the common-mode signal output by the electrodes can perfectly respond to macropulse beams at the 100μs or even ms level because the system does not have a significant low-frequency cutoff. Electrode coupling will result in a low-frequency cutoff frequency in the differential-mode signal output by the detector, thus degrading the resolution of low-frequency position measurements.

[0172] With control over the system load impedance, the low-frequency cutoff and corresponding resolution degradation can be continuously improved. By introducing a transimpedance amplifier with extremely low input impedance into the lead-out path to perform current-to-voltage conversion, differential-mode signal coupling can be significantly reduced, thereby improving the low-frequency cutoff frequency. Reducing the angle between the electrodes and the detector center can also reduce differential-mode signal coupling between the electrodes.

[0173] In summary, through theoretical analysis and simulation studies, this invention proposes the following three improvements to reduce the low-frequency cutoff frequency:

[0174] 1) This invention improves the characteristic impedance of the strip electrode by introducing soft magnetic material, and at the same time, the inner surface of the vacuum tube is plated with gold to improve the conductivity of the inner surface of the vacuum tube outside the detector electrode, so that the resistance of the tube wall of the vacuum tube is small enough, thereby reducing the low frequency cutoff frequency of the output signal (including common mode and differential mode signals) by reducing the loop current.

[0175] 2) The output terminal of the coaxial lead of the present invention is connected to an amplifier with extremely low input impedance, which greatly reduces the resistance value across the coaxial lead, reduces the electrode voltage of the beam position detector of the present invention during operation, avoids damage to the deviation of its 0 potential boundary condition, reduces the coupling between electrodes, and thus reduces the low frequency cutoff frequency of the differential mode signal.

[0176] 3) The present invention reduces the center angle of the electrode pair to the detector, thereby reducing the low-frequency coupling between the electrodes, reducing the main peak frequency on the spectral response curve, and thus increasing the low-frequency response amplitude of the beam position detector based on wall current of the present invention.

[0177] This invention is expected to reduce the low-frequency cutoff frequency of the differential mode signal of the SBPM detector to 1kHz, which can meet the requirements for high-precision, non-destructive beam position measurement of ultra-long beams of 100µs. It solves the current difficulty of lacking real-time, non-destructive, high-precision beam position measurement on high-energy transport lines after slow extraction in related devices.

[0178] Theoretical demonstration and simulation practice have proven that the input resistance R1 of the ultra-low input impedance high-impedance amplifier 41 can also significantly reduce the low-frequency coupling between the four electrodes of WSBPM, thereby improving the shortcomings of the existing technology where the low-frequency cutoff frequency of the differential mode signal used for detector position measurement is much greater than the low-frequency cutoff frequency of the common mode signal used for current intensity measurement.

[0179] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of the invention. Various variations can be made to the above embodiments of the present invention. All simple and equivalent changes and modifications made based on the claims and description of this application fall within the protection scope of the claims of this patent. All aspects not described in detail in this invention are conventional technical content.

Claims

1. A beam position detector based on wall current, comprising a vacuum tube, the vacuum tube including an electrode mounting section and a conventional section other than the electrode mounting section; the inner diameter of the vacuum tube at the electrode mounting section is larger than the inner diameter at the conventional section, thereby forming an electrode groove; four strip electrodes extending along the length direction of the vacuum tube and spaced apart from the inner surface of the vacuum tube at the electrode mounting section are mounted at the electrode groove, each strip electrode having an electrode start end and an electrode end at both ends along the length direction of the vacuum tube, the electrode start end being connected to the inner conductor of a coaxial lead-out terminal on an electrode mounting section, such that the electrode start end is connected to a signal acquisition system outside the vacuum tube through the coaxial lead-out terminal, characterized in that... The signal acquisition system includes four ultra-low input impedance high-impedance amplifiers connected one-to-one with four coaxial leads, a radio frequency front-end connected to all the ultra-low input impedance high-impedance amplifiers, and an ADC module connected to the radio frequency front-end. A soft magnetic core is provided between the electrode start end and the inner surface of the vacuum pipe at the electrode mounting section, and the inner surface of the vacuum pipe is gold-plated.

2. The beam position detector based on wall current according to claim 1, characterized in that, The electrode is 200 mm long and the total length of the vacuum tube is 250 mm, such that the angle between the strip electrode and the center of the beam position detector based on wall current is 15°.

3. The beam position detector based on wall current according to claim 1, characterized in that, Each strip electrode has an electrode start and an electrode end at both ends along the length of the vacuum pipe. The electrode start is spaced apart from the electrode mounting section and the conventional section, and the electrode end is fixed on the end face facing the electrode mounting section.

4. The beam position detector based on wall current according to claim 1, characterized in that, Each strip electrode has a thickness of 2 mm, and the gap between each strip electrode and the inner surface of the vacuum pipe at the electrode mounting section is 5 mm. The conventional section includes a first conventional section upstream of the electrode mounting section and a second conventional section downstream of the electrode mounting section. The distance between the electrode start end and the first conventional section is 3 mm.

5. The beam position detector based on wall current according to claim 1, characterized in that, The surface of the strip electrode is plated with gold.

6. The beam position detector based on wall current according to claim 1, characterized in that, The negative terminal of the input of each ultra-low input impedance high-impedance amplifier is connected to the output terminal of the coaxial lead, and the positive terminal of the input of the ultra-low input impedance high-impedance amplifier is grounded.

7. The beam position detector based on wall current according to claim 1, characterized in that, The inner core and inner wall of the coaxial lead-out terminal are gold-plated, and the inner core of the wires of the ultra-low input impedance high-impedance amplifier are all gold-plated. The input impedance of the ultra-low input impedance high-impedance amplifier is less than 10μΩ.

8. The beam position detector based on wall current according to claim 1, characterized in that, The ADC module is an FPGA, and the coaxial output terminal is an SMA type coaxial feed-through.

Images