SuperKEKB vacuum system operation in the last 6 years operation

The operation of the SuperKEKB has been ongoing since 2016. The vacuum systems of the main ring (MR) which consists of a 7-GeV electron ring (HER) and a 4-GeV positron ring (LER), the damping ring (DR) for 1.1 GeV positrons in the middle of the injector linac have been working well as a whole. As of June 2022, the maximum stored beam currents of MR are 1.46 and 1.14 A for the LER and the HER, respectively, and approximately 30 mA for the DR. The pressure increase per unit beam current is steadily decreasing and the new vacuum components developed for the SuperKEKB have been working as expected. No significant electron cloud effect has been observed in the LER after installing solenoids in drift spaces in 2017 which apply magnetic fields in the beam direction. The recent pressure behavior of the LER with increased beam current is explained by considering thermal gas desorption induced by the beam as well as photon-stimulated gas desorption. Currently, the beam lifetime is primarily limited by the Touschek effect rather than the vacuum pressure, and their degrees of contribution are evaluated. The challenges associated with high beam currents, such as damage to beam-collimator heads and excess heating of beam pipes at wiggler sections have become more apparent as beam currents are increased. The status of the SuperKEKB vacuum system and the experiences during the past 6 years of operation are presented here.

Figure 1

SuperKEKB at KEK Tsukuba campus.

Figure 2

Layout of the SuperKEKB Main Ring (MR). One ring consists of four arc sections, four straight sections, and one collision point at Tsukuba (Belle II).

Figure 3

Layout of the beam pipes, bellows chambers, ion pumps, and CCGs for LER and HER in the MR tunnel.

Figure 4

Trends of pressures and beam currents since Phase-1 commissioning in 2016 for (a) LER, (b) HER, and (c) DR.

Figure 5

Total operation time and the number of faults by year. The number of faults related to the vacuum system is indicated in red.

Figure 6

Typical temperatures of (a) bellows chambers and (b) gate valves at wiggler and arc sections during a recent operation.

Figure 7

Pressure increase per unit beam current ($\mathrm{\Delta }P/\mathrm{\Delta }I$) of the whole rings and arc sections as a function of the beam dose, and $\eta$ of arc sections as a function of photon dose for (a) LER, (b) HER, and (c) DR from 2016.

Figure 8

Dependences of $P$ on $I$ (a) measured, (b) calculated using the regression curve derived assuming $\mathrm{\Delta }{P}_t\propto {(\mathrm{\Delta }T)}^2$, and (c) those derived assuming $\mathrm{\Delta }{P}_t\propto {(\mathrm{\Delta }T)}^3$ for several $N_b$ of 393–1662 for LER, where the $P$ values higher than $6\times {10}^{-8}\mathrm{Pa}$ from the 16th of March to 25th of April 2022 were used for the MRA. Gray points represent all data.

Figure 9

(a) Experimental setup to measure the dependence of $\mathrm{\Delta }{P}_t$ on $\mathrm{\Delta }T$, and (b) the measured results and calculated $\mathrm{\Delta }{P}_t$ using the regression curves derived assuming Eqs. (4) and (5).

Figure 10

Dependences of $P$ on $I$ for $N_b$ of 393–2249 for the HER from the 19th of May to the 22nd of June 2022.

Figure 11

Behaviors of pressure increase per unit beam current ($\mathrm{\Delta }P/\mathrm{\Delta }I$) and $I\tau$ as a function of beam dose from 2016 for (a) LER and (b) HER.

Figure 12

Trends of $I$, $\tau$, and $N_b$ used for the MRA to estimate the influence of the pressure and Touschek effect on $\tau$ from the 15th of May to the 21st of June 2022 for (a) LER and (b) HER (single-beam mode).

Figure 13

Dependence of $\tau$ on $I$ (a) measured and (b) calculated using the regression curve for several $N_b$ of 978–2249 for LER, where the $P$ values higher than $6\times {10}^{-8}\mathrm{Pa}$, from the 15th of May to the 21st of June 2022, were used for the MRA. Gray points correspond to all measured data.

Figure 14

Dependence of $\tau$ on $I$ (a) measured and (b) calculated using the regression curve for several pressure ranges of $6\times {10}^{-8}–1.6\times {10}^{-7}\mathrm{Pa}$ for the LER, where the data from the 15th of May to the 21st of June 2022, were used for the MRA. Gray points represent all measured data.

Figure 15

Dependence of $1/\tau$ on (a) $I/(N_b{\sigma }_z\sqrt{{ϵ}_x{ϵ}_y})$ and (b) $I/(N_b{\sigma }_z)$ at $P=6–8\times {10}^{-8}\mathrm{Pa}$ for the LER.

Figure 16

Ratio of the first term ($1/{\tau }_p$) to the second term ($1/{\tau }_t$) of the right-hand side of Eq. (14) for the several pressure ranges of $6\times {10}^{-8}–1.6\times {10}^{-7}\mathrm{Pa}$ for the LER.

Figure 17

Dependence of $\tau$ on $I$ (a) measured and (b) calculated using the regression curve for several $N_b$ of 783–2346 from the 15th of March to the 22nd of June 2022 for the HER. Gray points correspond to all measured data.

Figure 18

Ratio of the first term ($1/{\tau }_p$) to the second term ($1/{\tau }_t$) of the right-hand side of Eq. (18) for the HER.

Figure 19

Typical views of countermeasures adopted to the SuperKEKB LER: (a) beam pipes with antechambers, (b) TiN-film coating, (c) clearing electrode, (d) groove structure, magnetic fields in the beam direction by (e) permanent magnets and (f) solenoids.

Figure 20

Vertical beam sizes as a function of the current line density ($I_d$) for several bunch filling patterns measured (a) before and (b) after attaching PM units to Al-alloy bellows chambers in Phase-1 commissioning, (c) Phase-2 commissioning, and (d) Phase-3 commissioning.

Figure 21

PM units attached to an Al-alloy bellows chamber, and Type-1 and Type-2 PM units for beam pipes at drift spaces.

Figure 22

Bunch-by-bunch luminosity for a bunch filling pattern of 2/1173/2.04rf on the 13th of June 2022. The vertical axis shows the number of hits at each ZDLM channel.

Figure 23

Typical pressure bursts accompanied by beam loss (beam abort) observed in Phase-1 commissioning.

Figure 24

(a) Number of bursts occurring per 50 h of operation time (red bars), the beam currents at which pressure bursts occurred (blue and green squares), $I_{\max }$ (pink dots), and (b) locations of the pressure bursts along the ring versus the operation time from 2016 for the LER (left) and HER (right). Green solid squares and light-blue solid squares in (b) are the locations at beam collimators and at just downstream of IR, respectively.

Figure 25

Schematic of beam collimator structures and those installed in the tunnel for (a) LER and (b) HER.

Figure 26

Location of collimators in the MR, where “H” and “V” in the collimator names refer to the horizontal and vertical collimators, respectively.

Figure 27

Typical damaged head in a vertical collimator, where the traveling direction of the beam is from left to right.

Figure 28

(a) Wiggler section in the LER (Oho) and (b) bellows chamber with cooling blowers.

Figure 29

(a) Adjustment of the vertical orbit at a wiggler section, where blue dots are the horizontal and vertical positions of the beam measured by beam position monitors placed almost every 10 m. The vertical orbit was adjusted from the dashed blue line to the green line by correction magnets. (b) Change in temperatures of several beam pipes in the wiggler section at the timing of the vertical orbit adjustment shown in (a).

Figure 30

Behaviors of pressures and beam currents when air leak occurred at flanges in a wiggler section. The pressure increased at the time of beam aborts. The leaks were finally repaired by a liquid sealant (i.e., VACSEAL).

Figure 31

Inside of a bellows chamber with SR masks at antechambers developed for wiggler sections.

Figure 32

Trends of beam current and pressures around a beam pipe at the downstream of Oho wiggler section from 2022.

Figure 33

(a) Beam pipe at the downstream of Oho wiggler section and (b) spray of liquid sealant into the cooling channel.

Figure 34

Behaviors of the beam current, pressure, and temperatures of a gate valve (GV) and a bellows chamber (BL) on the 18th and 19th of June, 2022.