208 Beam-based Correction and Optimization for Accelerators
8.1 LINAC-TO-BOOSTER TRAJECTORY STEERING
The linac-to-booster (LTB) transport line is the part of the SPEAR injector
that connects the 120-MeV linac to the Booster synchrotron. The steering of
the beam trajectory in the LTB is important for achieving a good capture
efficiency for the beam injected into the Booster. It is frequently tuned during
operations.
The Booster has a finite transverse acceptance. For the best capture efficiency, the LTB beam trajectory at the injection point should have the proper
position and angle such that beam is at the center of acceptance in both
transverse planes. There are a number of steering knobs that can affect the
trajectory at the end of the LTB, including corrector magnets in the linac,
corrector magnets and the trim coils on the bending magnets in the LTB, the
injection kicker, the injection septum, and the power of the linac klystron K3.
The K3 power knob changes the beam energy and in turn the trajectory due
to dispersion in the transport line.
A pair of steering magnets located near the end of the transport line are
chosen for the steering experiment for each plane. In the horizontal plane, the
trim coils on the B3 bending magnet (B3trim) and the injection septum are
used. The correctors COR3V and COR5V are used for the vertical plane. The
two knobs in each plane can effectively change both the position and angle
coordinates of the beam at the injection point.
The tuning of one steering magnet shifts the beam trajectory along a line
in the (x, x
) or (y, y
) phase space. Ideally, the lines traced by the two knobs
in each plane should be orthogonal; in such a case, the steering of one knob
is independent of the other. Such knobs can be formed by combining the
two steering magnets with a proper ratio of the strengths. For example, one
combined knob is for the x coordinate, and the other for the x
coordinates,
and similarly for the vertical plane. However, in this experiment we simply
used the setpoints of the individual magnets as knobs. The parameter ranges
are (−1, 1) A for the three steering magnets, B3trim, COR3V, and COR5V,
and (−0.5, 0.5) A for the injection septum, relative to the initial setpoints. The
ranges are large enough such that the deviation of any knob can significantly
reduce the capture efficiency before reaching the limit.
The Booster beam current (the monitor is referred to as the Q-meter)
measured near the end of the ramping cycle is used as the optimization objective. The intensity of the linac beam is stable in a short period of time.
The injection loss only occurs near the beginning of the cycle. Therefore, any
change to the capture efficiency will be reflected on the Q-meter reading.
After the knobs are changed, the code waits for 3 seconds for the magnets
to settle to the new setpoints. The beam is then turned on and monitored
for 2 seconds. Since the injector runs with a 10-Hz repetition rate, there are
20 valid Q-meter data points. The average value, with a minus sign added, is
used as the objective function. The minus sign is inserted to make maximizing
the beam current a minimization problem.
8.1 LINAC-TO-BOOSTER TRAJECTORY STEERING
The linac-to-booster (LTB) transport line is the part of the SPEAR injector
that connects the 120-MeV linac to the Booster synchrotron. The steering of
the beam trajectory in the LTB is important for achieving a good capture
efficiency for the beam injected into the Booster. It is frequently tuned during
operations.
The Booster has a finite transverse acceptance. For the best capture efficiency, the LTB beam trajectory at the injection point should have the proper
position and angle such that beam is at the center of acceptance in both
transverse planes. There are a number of steering knobs that can affect the
trajectory at the end of the LTB, including corrector magnets in the linac,
corrector magnets and the trim coils on the bending magnets in the LTB, the
injection kicker, the injection septum, and the power of the linac klystron K3.
The K3 power knob changes the beam energy and in turn the trajectory due
to dispersion in the transport line.
A pair of steering magnets located near the end of the transport line are
chosen for the steering experiment for each plane. In the horizontal plane, the
trim coils on the B3 bending magnet (B3trim) and the injection septum are
used. The correctors COR3V and COR5V are used for the vertical plane. The
two knobs in each plane can effectively change both the position and angle
coordinates of the beam at the injection point.
The tuning of one steering magnet shifts the beam trajectory along a line
in the (x, x
) or (y, y
) phase space. Ideally, the lines traced by the two knobs
in each plane should be orthogonal; in such a case, the steering of one knob
is independent of the other. Such knobs can be formed by combining the
two steering magnets with a proper ratio of the strengths. For example, one
combined knob is for the x coordinate, and the other for the x
coordinates,
and similarly for the vertical plane. However, in this experiment we simply
used the setpoints of the individual magnets as knobs. The parameter ranges
are (−1, 1) A for the three steering magnets, B3trim, COR3V, and COR5V,
and (−0.5, 0.5) A for the injection septum, relative to the initial setpoints. The
ranges are large enough such that the deviation of any knob can significantly
reduce the capture efficiency before reaching the limit.
The Booster beam current (the monitor is referred to as the Q-meter)
measured near the end of the ramping cycle is used as the optimization objective. The intensity of the linac beam is stable in a short period of time.
The injection loss only occurs near the beginning of the cycle. Therefore, any
change to the capture efficiency will be reflected on the Q-meter reading.
After the knobs are changed, the code waits for 3 seconds for the magnets
to settle to the new setpoints. The beam is then turned on and monitored
for 2 seconds. Since the injector runs with a 10-Hz repetition rate, there are
20 valid Q-meter data points. The average value, with a minus sign added, is
used as the objective function. The minus sign is inserted to make maximizing
the beam current a minimization problem.