Higher-order-mode absorbers for energy recovery linac cryomodules at Brookhaven National Laboratory

Several future accelerator projects at Brookhaven for the Relativistic Heavy Ion Collider (RHIC) are based on energy recovery linacs (ERLs) with high-charge high-current electron beams. Their stable operation mandates effective higher-order-mode (HOM) damping. The development of HOM dampers for these projects is pursued actively at this laboratory. Strong HOM damping was experimentally demonstrated both at room and at superconducting (SC) temperatures in a prototype research and development (R&D) five-cell niobium superconducting rf (SRF) cavity with ferrite dampers. Two room-temperature mock-up five-cell copper cavities were used to study various damper configurations with emphasis on capacitive antenna dampers. An innovative type of ferrite damper over a ceramic break for an R&D SRF electron gun also was developed. For future SRF linacs longer cryomodules comprised of multiple superconducting cavities with reasonably short intercavity transitions are planned. In such a configuration, the dampers, located closer to the cavities, will be at cryogenic temperatures; this will impose additional constraints and complications. This paper presents the results of simulations and measurements of several damper configurations.

Figure 1

View of ECX five-cell cavity vacuum string assembly with ferrite dampers, but without its cryostat. HOM signals are obtained from a pickup (PU) probe, beam position monitor (BPM) buttons, and the fundamental power coupler (FPC).

Figure 2

The cross section of the ECX five-cell cavity vacuum string assembly (dimensions are in mm).

Figure 3

HOM spectra of the ECX cavity at room temperature. The red curve shows the transmission through the cavity (BPM 6-6), while the black curve shows transmission between two probes in the upstream tube (BPM 2-10).

Figure 4

Plots of the longitudinal electric field component at 1 cm off axis for a dipole beam tube mode (red), and for a quasitrapped mode (blue). The fields are plotted against the location of the cell centers.

Figure 5

Resonances in the upstream (blue) and downstream (red) beam tube. The latter houses the FPC. Its additional resonant structure is caused by the FPC’s waveguide connection to the rf source.

Figure 6

Spectra of HOMs of the ERL five-cell cavity in the normal conducting (NC) and in the superconducting state (at 4 K).

Figure 7

Comparison of the resonance curves of the quasitrapped mode (left) with the neighboring damped mode (right). The direct $Q$ measurements at left are 9 500, 44 200, and 47 800 for NC, 4, and 2 K. Also seen is the frequency shift in the left resonance due to the cooldown.

Figure 8

Model of the ECX five-cell cavity used for MWS simulation.

Figure 9

Simulated $R/Q$ values, transverse shunt impedance values, and the estimated worst-case BBU limit.

Figure 10

An energy recovery linac cryomodule with six cavities.

Figure 11

The models for the BNL1 (top) and BNL3 (bottom) cavities.

Figure 12

Brillouin diagram for the BNL1 and BNL3 cavity models.

Figure 13

On-axis transverse electric field of the $0\pi /5$ modes in BNL1 and BNL3. The fields are plotted against the location of the cell centers.

Figure 14

MWS models for the BNL3 cavity with 22 cm diameter beam tubes and end tapers as presently being fabricated.

Figure 15

Simulated accelerating field ($E_X$ @ 703 MHz) and transverse electric field ($E_Z$ @ 1023 MHz) of a beam tube HOM resonance in the BNL3 cavity. Also shown are the cell centers (red) and the location of the cavity with the beam tubes (blue).

Figure 16

Quasitrapped modes in the BNL1 and BNL3 cavities. The fields are plotted against the location of the cell centers and the damper.

Figure 17

The HOM damper ring with capacitive probes between two copper cavities.

Figure 18

Frequency splitting of HOMs in the combined ERL Cu cavities.

Figure 19

Connecting tube with damper ring between Cu cavities.

Figure 20

Damper ring with four mushroom capacitive probes each with a 51 mm diameter disk on a 25 mm-long stem. A 1 ft ruler is added for scale.

Figure 21

Frequency spectrum of a capacitive damper ring with staggered dampers between Cu cavities.

Figure 22

$Q$ external of fundamental versus damper position and probe size ($\mathrm{\text{disk diameter}}\times \mathrm{\text{stem length}}\times \mathrm{\text{stem diameter}}$ in mm), probe position is defined in Fig. 19.

Figure 23

Sketch of a capacitive HOM damper with a high-pass filter attached (dimensions in mm).

Figure 24

Prototype HOM ferrite damper.

Figure 25

Ferrite HOM damper installed between two five-cell Cu cavities.

Figure 26

Frequency spectrum of ferrite damper between Cu cavities.

Figure 27

Surface impedance of the ferrite damper, treated as a cylinder with 6 cm inner diameter radius and unbound outer diameter (dashed), or with 3 mm wall thickness (solid). Ferrite parameters are $\epsilon =13$ and $\mu =29/(1+j2,5f_{\mathrm{GHz}})$.

Figure 28

Ferrite HOM damper with 10 cm inner diameter ceramic break. The plates with ferrite tiles are held in place by the clamshell holder (dimensions are in cm).

Figure 29

The ceramic-ferrite damper with some ferrite tiles removed (top) and the SRF gun with the ceramic-ferrite damper assembled (bottom).

Figure 30

Damping of dipole gun resonances (black) by ferrite alone (red) and by the ceramic/ferrite combination (blue).