SPIDER is the negative ion source prototype for the ITER Heating Neutral Beam Injector. Its upcoming experimental campaign introduces new solid-state RF amplifiers which require a strict Voltage Standing Wave Ratio (VSWR ≤ 1.15) to deliver nominal power. Since the plasma load varies significantly with operating conditions, this thesis aims to characterize the equivalent RF driver impedance and optimize the Matching Network (MN) to minimize the VSWR. As the total load impedance is composed of no-load and on-load components, vacuum impedance measurements were initially acquired from SPIDER, with circuit parameters extracted using Particle Swarm Optimization and Differential Evolution algorithms. Based on these initial evaluations and factors from literature, a preliminary MN was designed and installed on SPIDER for the upcoming campaign, prioritizing a wide-bandwidth operation within the VSWR limit. Subsequently, the on-load (plasma) parameters were obtained from the MINION testbed, with maximum power of 75 kW, by comparing multiple extraction methods from different measurements; this allowed for the characterization of the variation of resistance and inductance with respect to pressure and power. Furthermore, the plasma behavior was extrapolated up to the 100 kW SPIDER target; by combining the SPIDER vacuum parameters with the MINION plasma contributions, a refined MN configuration was derived. The primary limitation of this study lies in this necessary extrapolation beyond the directly measured 75 kW limit, alongside a slight operating frequency gap between the testbed and SPIDER. Consequently, while this predictive analysis provides a crucial engineering basis for the MN setup, the extrapolated trends will require direct validation during future full-power operations.

SPIDER is the negative ion source prototype for the ITER Heating Neutral Beam Injector. Its upcoming experimental campaign introduces new solid-state RF amplifiers which require a strict Voltage Standing Wave Ratio (VSWR ≤ 1.15) to deliver nominal power. Since the plasma load varies significantly with operating conditions, this thesis aims to characterize the equivalent RF driver impedance and optimize the Matching Network (MN) to minimize the VSWR. As the total load impedance is composed of no-load and on-load components, vacuum impedance measurements were initially acquired from SPIDER, with circuit parameters extracted using Particle Swarm Optimization and Differential Evolution algorithms. Based on these initial evaluations and factors from literature, a preliminary MN was designed and installed on SPIDER for the upcoming campaign, prioritizing a wide-bandwidth operation within the VSWR limit. Subsequently, the on-load (plasma) parameters were obtained from the MINION testbed, with maximum power of 75 kW, by comparing multiple extraction methods from different measurements; this allowed for the characterization of the variation of resistance and inductance with respect to pressure and power. Furthermore, the plasma behavior was extrapolated up to the 100 kW SPIDER target; by combining the SPIDER vacuum parameters with the MINION plasma contributions, a refined MN configuration was derived. The primary limitation of this study lies in this necessary extrapolation beyond the directly measured 75 kW limit, alongside a slight operating frequency gap between the testbed and SPIDER. Consequently, while this predictive analysis provides a crucial engineering basis for the MN setup, the extrapolated trends will require direct validation during future full-power operations.

Load characterization and matching network optimization of high-power RF drivers in SPIDER and MINION experiments

FRASSON, EMANUELE
2025/2026

Abstract

SPIDER is the negative ion source prototype for the ITER Heating Neutral Beam Injector. Its upcoming experimental campaign introduces new solid-state RF amplifiers which require a strict Voltage Standing Wave Ratio (VSWR ≤ 1.15) to deliver nominal power. Since the plasma load varies significantly with operating conditions, this thesis aims to characterize the equivalent RF driver impedance and optimize the Matching Network (MN) to minimize the VSWR. As the total load impedance is composed of no-load and on-load components, vacuum impedance measurements were initially acquired from SPIDER, with circuit parameters extracted using Particle Swarm Optimization and Differential Evolution algorithms. Based on these initial evaluations and factors from literature, a preliminary MN was designed and installed on SPIDER for the upcoming campaign, prioritizing a wide-bandwidth operation within the VSWR limit. Subsequently, the on-load (plasma) parameters were obtained from the MINION testbed, with maximum power of 75 kW, by comparing multiple extraction methods from different measurements; this allowed for the characterization of the variation of resistance and inductance with respect to pressure and power. Furthermore, the plasma behavior was extrapolated up to the 100 kW SPIDER target; by combining the SPIDER vacuum parameters with the MINION plasma contributions, a refined MN configuration was derived. The primary limitation of this study lies in this necessary extrapolation beyond the directly measured 75 kW limit, alongside a slight operating frequency gap between the testbed and SPIDER. Consequently, while this predictive analysis provides a crucial engineering basis for the MN setup, the extrapolated trends will require direct validation during future full-power operations.
2025
Load characterization and matching network optimization of high-power RF drivers in SPIDER and MINION experiments
SPIDER is the negative ion source prototype for the ITER Heating Neutral Beam Injector. Its upcoming experimental campaign introduces new solid-state RF amplifiers which require a strict Voltage Standing Wave Ratio (VSWR ≤ 1.15) to deliver nominal power. Since the plasma load varies significantly with operating conditions, this thesis aims to characterize the equivalent RF driver impedance and optimize the Matching Network (MN) to minimize the VSWR. As the total load impedance is composed of no-load and on-load components, vacuum impedance measurements were initially acquired from SPIDER, with circuit parameters extracted using Particle Swarm Optimization and Differential Evolution algorithms. Based on these initial evaluations and factors from literature, a preliminary MN was designed and installed on SPIDER for the upcoming campaign, prioritizing a wide-bandwidth operation within the VSWR limit. Subsequently, the on-load (plasma) parameters were obtained from the MINION testbed, with maximum power of 75 kW, by comparing multiple extraction methods from different measurements; this allowed for the characterization of the variation of resistance and inductance with respect to pressure and power. Furthermore, the plasma behavior was extrapolated up to the 100 kW SPIDER target; by combining the SPIDER vacuum parameters with the MINION plasma contributions, a refined MN configuration was derived. The primary limitation of this study lies in this necessary extrapolation beyond the directly measured 75 kW limit, alongside a slight operating frequency gap between the testbed and SPIDER. Consequently, while this predictive analysis provides a crucial engineering basis for the MN setup, the extrapolated trends will require direct validation during future full-power operations.
NBI
matching network
fitting algorithms
RF power
plasma load
File in questo prodotto:
File Dimensione Formato  
Frasson_Emanuele.pdf

embargo fino al 11/09/2027

Dimensione 8.45 MB
Formato Adobe PDF
8.45 MB Adobe PDF

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/113084