CSRCSBEND

Like CSBEND, but incorporates a simulation of Coherent Synchrotron radiation.

Parameter Name	Units	Type	Default	Description
L	$M$	double	0.0	arc length
ANGLE	$RAD$	double	0.0	bend angle
K1	$1/M^{2}$	double	0.0	geometric quadrupole strength
K2	$1/M^{3}$	double	0.0	geometric sextupole strength
K3	$1/M^{4}$	double	0.0	geometric octupole strength
K4	$1/M^{5}$	double	0.0	geometric decapole strength
E1	$RAD$	double	0.0	entrance edge angle
E2	$RAD$	double	0.0	exit edge angle
TILT	$RAD$	double	0.0	rotation about incoming longitudinal axis
H1	$1/M$	double	0.0	entrance pole-face curvature
H2	$1/M$	double	0.0	exit pole-face curvature
HGAP	$M$	double	0.0	half-gap between poles
FINT		double	0.5	edge-field integral
DX	$M$	double	0.0	misalignment
DY	$M$	double	0.0	misalignment
DZ	$M$	double	0.0	misalignment
FSE		double	0.0	fractional strength error
ETILT		double	0.0	error rotation about incoming longitudinal axis
N_KICKS		long	4	number of kicks
NONLINEAR		long	1	include nonlinear field components?
LINEARIZE		long	0	use linear matrix instead of symplectic integrator?
SYNCH_RAD		long	0	include classical synchrotron radiation?
EDGE1_EFFECTS		long	1	include entrace edge effects?
EDGE2_EFFECTS		long	1	include exit edge effects?
EDGE_ORDER		long	1	order to which to include edge effects
INTEGRATION_ORDER		long	2	integration order (2 or 4)

Like CSBEND, but incorporates a simulation of Coherent Synchrotron radiation.

Parameter Name	Units	Type	Default	Description
BINS		long	0	number of bins for CSR wake
BIN_ONCE		long	0	bin only at the start of the dipole?
BIN_RANGE_FACTOR		double	1.2	Factor by which to increase the range of histogram compared to total bunch length. Large value eliminates binning problems in CSRDRIFTs.
SG_HALFWIDTH		long	0	Savitzky-Golay filter half-width for smoothing current histogram
SG_ORDER		long	1	Savitzky-Golay filter order for smoothing current histogram
SGDERIV_HALFWIDTH		long	0	Savitzky-Golay filter half-width for taking derivative of current histogram
SGDERIV_ORDER		long	1	Savitzky-Golay filter order for taking derivative of current histogram
TRAPAZOID_INTEGRATION		long	1	Select whether to use trapazoid-rule integration (default) or a simple sum.
OUTPUT_FILE		STRING	NULL	output file for CSR wakes
OUTPUT_INTERVAL		long	1	interval (in kicks) of output to OUTPUT_FILE
OUTPUT_LAST_WAKE_ONLY		long	0	output final wake only?
STEADY_STATE		long	0	use steady-state wake equations?
USE_BN		long	0	use b$<$n$>$ instead of K$<$n$>$?
B1	$1/M$	double	0.0	K1 = b1*rho, where rho is bend radius
B2	$1/M^{2}$	double	0.0	b2 = B2*rho
B3	$1/M^{3}$	double	0.0	b3 = B3*rho
B4	$1/M^{4}$	double	0.0	b4 = B4*rho
ISR		long	0	include incoherent synchrotron radiation (scattering)?

Like CSBEND, but incorporates a simulation of Coherent Synchrotron radiation.

Parameter Name	Units	Type	Default	Description
CSR		long	1	enable CSR computations?
BLOCK_CSR		long	0	block CSR from entering CSRDRIFT?
DERBENEV_CRITERION_MODE		STRING	disable	disable, evaluate, or enforce
PARTICLE_OUTPUT_FILE		STRING	NULL	name of file for phase-space output
PARTICLE_OUTPUT_INTERVAL		long	0	interval (in kicks) of output to PARTICLE_OUTPUT_FILE
SLICE_ANALYSIS_INTERVAL		long	0	interval (in kicks) of output to slice analysis file (from slice_analysis command)
HIGH_FREQUENCY_CUTOFF0		double	-1	Spatial frequency at which smoothing filter begins. If not positive, no frequency filter smoothing is done. Frequency is in units of Nyquist (0.5/binsize).
HIGH_FREQUENCY_CUTOFF1		double	-1	Spatial frequency at which smoothing filter is 0. If not given, defaults to HIGH_FREQUENCY_CUTOFF0.
WAKE_FILTER_FILE		STRING	NULL	Name of file supplying wakefield filtering data.
WFF_FREQ_COLUMN		STRING	NULL	Name of column supplying frequency values for wakefield filtering data.
WFF_REAL_COLUMN		STRING	NULL	Name of column supplying real values for wakefield filtering data.
WFF_IMAG_COLUMN		STRING	NULL	Name of column supplying imaginary values for wakefield filtering data.

For a discussion of the method behind this element, see M. Borland, ``Simple method for particle tracking with coherent synchrotron radiation,'' Phys. Rev. ST Accel. Beams 4, 070701 (2001).

Recommendations for using this element. The default values for this element are not the best ones to use. They are retained only for consistency through upgrades. In using this element, it is recommended to have 50 to 100 k particle in the simulation. Setting BINS=600 and SG_HALFWIDTH=1 is also recommended to allow resolution of fine structure in the beam and to avoid excessive smoothing. It is strongly suggested that the user vary these parameters and view the histogram output to verify that the longitudinal distribution is well represented by the histograms (use OUTPUT_FILE to obtain the histograms). For LCLS simulations, we find that the above parameters give essentially the same results as obtained with 500 k particles and up to 3000 bins.

In order to verify that the 1D approximation is valid, the user should also set DERBENEV_CRITERION_MODE = ``evaluate'' and view the data in OUTPUT_FILE. Generally, the criterion should be much less than 1.

In order respects, this element is just like the CSBEND element, which provides a symplectic bending magnet that is accurate to all orders in momentum offset. The field expansion is available to fourth order.

One pitfall of symplectic integration is the possibility of orbit and path-length errors for the reference orbit if too few kicks are used. This may be an issue for rings. Hence, one must verify that a sufficient number of kicks are being used by looking at the trajectory closure and length of an on-axis particle by tracking. Using INTEGRATION_ORDER=4 is recommended to reduce the number of required kicks.

Normally, one specifies the higher-order components of the field with the K1, K2, K3, and K4 parameters. The field expansion in the midplane is $B_y(x) = B_o * (1 + \sum_{n=1}^4\frac{K_n\rho_o}{n!}x^n)$. By setting the USE_bN flag to a nonzero value, one may instead specify the b1 through b4 parameters, which are defined by the expansion $B_y(x) = B_o * (1 + \sum_{n=1}^4\frac{b_n}{n!}x^n)$. This is convenient if one is varying the dipole radius but wants to work in terms of constant field quality.

Setting NONLINEAR=0 turns off all the terms above K_1 (or b_1) and also turns off effects due to curvature that would normally result in a gradient producing terms of higher order.

Edge effects are included using a first- or second-order matrix. The order is controlled using the EDGE_ORDER parameter, which has a default value of 1. N.B.: if you choose the second-order matrix, it is not symplectic.