sddsemitproc

• description:

  sddsemitproc analyzes quadrupole scan emittance measurement data. It accepts a file containing the transport matrix for each data point and measured beam sizes. Because sddsemitproc uses the matrix rather than a thin-lens model, it can analyze data from arbitrarily complex scans, involving, for example, multiple thick-lens quadrupoles.

  The matrix data can be prepared using elegant. For example, the vary_element command can be used to vary one or more quadrupoles. In addition, the beam size data may be prepared using elegant, to allow simulation of emittance measurements.

  sddsemitproc will perform error analysis using a Monte Carlo technique. A user-specified number of random error sets are generated and added to all measurements. Analysis is performed for each error set. Statistics over all the error sets provide most likely values and error bars.

  The beam parameters computed by sddsemitproc pertain to the beginning of whatever system is simulated in elegant.

• examples:
  elegant quadScan.ele sddscollapse quadScan.fin -pipe=out
  | sddsxref -pipe=in quadScan.data -take=SigmaX,SigmaY
  | sddsemitproc -pipe=in emitResults.sdds

• synopsis:
  sddsemitproc [inputfile] [outputfile] [-pipe=[input][,output]] [-sigmaData=xName,yName] [-errorData=xName,yName | -errorLevel=valueInm,[{gaussian,nSigmas | uniform}]] [-nErrorSets=number] [-seed=integer] [-limitMode=resolution | zero[,reject] [-deviationLimit=xLevelm,yLevelm] [-resolution=xResolutionm,yResolutionm] [-verbosity=level]

• files:
  • inputfile -- An SDDS file containing one or more pages with columns named Rij, where ij is 11, 12, 33, and 34. These give elements of the horizontal and vertical transport matrices from the beginning of a system to the observation point. The sigma matrix inferred will be that for the beginning of the system. Typically, one starts with the final file from the run_setup command in elegant, and collapses it using sddscollapse. Each page of inputfile corresponds to a different emittance measurement.

    In addition to this data, inputfile must also contain columns giving the rms beam sizes in x and y. The user supplies the names of the columns using the -sigmaData option; otherwise, they default to Sx and Sy. These columns may be from elegant (e.g., Sx and Sy), if one wants to simulate an emittance measurement. Note that the theory behind the emittance measurement is strictly correct only for true RMS beamsize measurements. Use of FWHM or some other measure will give unreliable results.

  • outputfile -- A file containing one page for each page of inputfile. The parameters of outputfile give the measured geometric rms emittance, sigma matrix, and Twiss parameters of the beam in the horizontal and vertical planes. If error sets were requested (using -nErrorSets), then there are also parameters giving the error bars (``sigma's'') of the measured values.

• switches:
  • -sigmaData=xName,yName -- Supplies the names of the columns in inputfile from which the x and y rms beam sizes are to be taken. Default values are Sx and Sy, which are the data provided by elegant.

  • -errorLevel=valueInm,[gaussian,nSigmas | uniform] -- Supplies the standard deviation of random errors to be added to the measured beam sizes for Monte Carlo error analysis.

  • -errorData=xName,yName -- May be used to supply the names of columns in the input file that contain the error level for each measurement. This is an option instead of using -errorLevel, which allows varying the measurement error for each point.

  • -nErrorSets=number -- The number of sets of random errors to generate and add to the measurements. Each error set is used to perturb the original measurement data. The results are analyzed separately for each error set, then combined to give means and error bars.

  • -seed=integer -- Seed for the random number generator. Recommend a large, positive, odd integer less than $2^31$. If no seed is given or if the given seed is negative, then a seed is generated from the system clock.

  • -resolution=xResolutionm,yResolutionm -- The resolution of the beam size measurements, in meters. These values are subtracted in quadrature from the measured beam sizes to obtain the true beam sizes.

  • -limitMode=resolution | zero[,reject] -- If measured or perturbed beam sizes are less than the resolution or less than zero, then errors will result. One can use this option to limit minimum beam size values or reject points. In general, if one has to do this the measurement is probably bad.

  • -deviationLimit=xLevelm,yLevelm -- Specifies the maximum deviation, in meters, from the fit that data points may have and still be included. An initial fit is performed for each randomized set or the raw data, as appropriate. Outliers are then removed and the fit is repeated.

  • -verbosity=level -- Higher values of level result in more informational printouts as the program runs.

• author: M. Borland, ANL/APS.