Learning Objective: This module introduces basic concepts and features of superconducting accelerating cavities.
Skills:
- Gain a basic understanding of the motivation for use of superconducting cavities
- Understand the terms and features unique to superconducting cavities
- Generally know the kinds of diagnostics and controllable parameters available for SRF cavity operation
Introduction:
The LCLS SC-Linac will employ superconducting (SC) cavities to accelerate electron beam at up to 1 MHz repetition rate.
Most slide images are from Dan's, Sebastian's, and Janice's talks on 31 Aug 2021. Some are from Andy Benwell's talk 22 May 2020.
Also highly recommended is the Cryomodule Internal Photo Tourwhich details the components of a cryomodule and their functions.
Motivation
To make operation at such a high repetition rate economically viable, it is critical to minimize the amount of input RF power that is wasted as heat loss. Cavity performance is characterized by the quality factor, Q0, a ratio of energy stored inside the cavity to power lost through cavity walls. Because the nitrogen-doped niobium cavities of the SC-Linac are designed to dissipate far less heat through the walls than the copper cavities of the original LCLS linac, they have a higher Q0 so can transfer more of the power generated by the RF to the electron beam. Due to the superfluid nature of low temperature liquid helium, it conducts the heat to the surface of the liquid, away from the cavities. From Dan Gonnella's talk 31 Aug 2021, if the cavities were made of copper and you fed in 20 MV of CW RF, the dissipated power in the cavity walls would be 15 MW. With niobium at 2K, the dissipated power is 15W. The cryogenic costs increase the wall power to 15 kW, three orders of magnitude lower than power lost through copper walls!
Overview
Cavities and Cryomodules
The SRF cavities used for LCLS-SC are 1.038m long and made of nitrogen-doped niobium. Each cavity contains nine cells. Strings of eight cavities are bolted together with a button design BPM and a multifunction magnet inside a cryomodule.
RF Source
The 1.3 GHz CW RF is produced by the low level RF (LLRF) system then amplified by a solid state amplifier (SSA). Each cavity has its own SSA. The RF generated by the SSA is directed into the tunnel through rectangular waveguides and fed into the downstream end of the cavity through a fundamental power coupler (FPC) .
HOMs
Each cavity also has two higher-order-mode couplers (HOM couplers), small cylinderical cans (a few cm in diameter and ~10cm long) attached to each end of the cavity that absorb frequencies above 1.3 GHz. They also contain antenna for signal monitoring, but these aren't connected in the housing.
Frequency Tuning
When the RF drive frequency matches the resonant frequency of an accelerating cavity, standing waves can form inside the cavity storing the electromagnetic energy used to accelerate a particle beam. Due to thermal expansion, a cavity's volume will change with temperature, shifting the resonant frequency of the cavity. Cavities also experience Lorentz force detuning which is a distortion of the cavity walls due to the pressure from the electromagnetic fields of the RF.
Frequency tuners are used to change the shape of the cavity to get it back on resonance. There are two types of tuners: stepper motors that can make slow, coarse changes to the cavity frequency and piezo tuners that make fast, incremental adjustments to maintain the resonant frequency during normal running.
Cryo
Inside each cryomodule, the cavities are welded into titanium "helium vessels" which contain the volume of liquid helium used to cool cavities to their 2K operating temperature. There are three stages of cooling down from ambient temperatures to 40K, 5k, and finally 2K. (See Introduction to LCLS-SC Cryo Systems article for more details.)
Magnets
Due to spatial constraints, there is one multifunction conductively-cooled magnet in each cryomodule. The magnet has three concentric coils at each pole wired as quadrupole, x corrector, and y corrector.
Important Concepts
Amplitude vs Gradient
RF Amplitude is given in units of volts. An electron will have 1 MeV of energy (E=qV) after passing through a 1MV potential field. In our system, the LLRF is given a desired RF amplitude in MV (ADES) which is the total energy imparted to the beam if the cavity is phased correctly. Gradient (in MV/m) is amplitude divided by the cavity length. The LLRF also takes as input the desired phase (PDES). After initially phasing the cavities with the beam and calibrating the amplitude, a PDES of 0 should be the phase that imparts the most energy to the beam (fully forward phased). If PDES=0, the energy imparted to the beam by the cavity should be ADES.
Quench
If there is a problem that causes part of the cavity to increase in temperature such that it is no longer superconducting, the RF will no longer resonate and all the power will reflect back from the cavity. This is called a quench. This is typically caused by impurities in the cavity or on the surface and each cavity is measured to ascertain where it quenches. When the CMs were tested at the partner labs before shipping, the cavities were processed up to 21 MV/m or other limiting factor. For LCLS-SC the plan is to run the cavities at a maximum at 16 MV/m (16.6 MV). If there is a known problem, the cavity max PV (ADES_MAX) will be set to the operation limit.
From Andy's 22May2020 talk:
