TILT/BEAM CENTRE allows any non-orthogonality of the detector to the direct beam (tilt), and the direct beam centre on the detector to be determined [10]. This is achieved by least squares fitting of powder rings (or rings from other randomly orientated samples e.g. wax).
Below is the log file from FIT2D of the present user interaction. The documentation will be fully written when the algorithm and user inputs are stable:
Fit sub-menu: ENTER COMMAND [INPUT PARAMETERS]:tilt WARNING: This is a development routine, user prompts are likely to change SAMPLE TO DETECTOR DISTANCE (MILLIMETRES) (Range: .10000 to 1.000E+05) [200.000]: NUMBER OF ANGULAR SECTIONS (Range: 10 to 360) [90]:? The powder ring is divided into a number of equal angle sections for calculating the radial centre of each section. From the radial centre and the average angle two Cartesian coordinates are calculated. These coordinates are used to fit optimum beam centre and detector plane tilt angles. Enter required number of sections for calculations. NUMBER OF ANGULAR SECTIONS (Range: 10 to 360) [90]: WEIGHTED FITTING [NO]:? The fitting of tilt angle and beam centre to the estimated ring centres may be performed using weighted fitting or unweighted fitting. If weighted fitting is chosen then the average intensity (3 pixels) around the calculated centre of each radial profile is used to weight the fit. This means that strong rings and strong angular regions of rings will have more influence than weaker ones. This should make the fitting more robust when the data has weak rings and noisy background. WEIGHTED FITTING [NO]: REJECT OUTLYING COORDINATES [YES]:? Outlying coordinate positions which are more than an input number of standard deviations radially from the fitted ring positions may be rejected from the coordinate lists. The beam centre/tilt can then be re-fitted without these coordinates. If erroneous coordinate positions are influencing the fit, this option may allow them to be removed. have more influence than weaker ones. This should make the fitting more robust when the data has weak rings and noisy background. REJECT OUTLYING COORDINATES [YES]: REJECT LIMIT (NUMBER OF STANDARD DEVIATIONS) (Range: 1.00000 to 10.0000) [3.00000]:? Enter the limit of number of standard deviations after which coordinate positions are to be rejected from the coordinate lists. A three sigma limit should be reasonable for less than about 200 coordinates (assuming they are normally distributed.) REJECT LIMIT (NUMBER OF STANDARD DEVIATIONS) (Range: 1.00000 to 10.0000) [3.00000]: FULL ALGORITHM INFORMATION [YES]:? YES: if you want step by step diagnostics information FULL ALGORITHM INFORMATION [YES]: INPUT METHOD FOR BEAM CENTRE [KEYBOARD]:? AVERAGED GRAPHICAL: Average of average symmetry centres CIRCLE COORDINATES: Least squares fit on >=3 coordinates ELLIPSE COORDINATES: Least squares fit on >=5 coordinates GRAPHICAL COORDINATE: Single graphical input coordinate KEYBOARD: Single keyboard entered X/Y coordinate INPUT METHOD FOR BEAM CENTRE [KEYBOARD]: X-PIXEL COORDINATE OF BEAM CENTRE [256.5]:256 Y-PIXEL COORDINATE OF BEAM CENTRE [256.5]:255 INFO: The search distance either side of the powder rings (mm) = 1.511 INFO: The search distance either side of the powder rings (X-pixels) = 15.11 GRAPHICAL INPUT: 4.3077289E+02 2.6403860E+02 INFO: Radius of powder ring 1 (mm) = 6.894 INFO: Radius of powder ring 2 (mm) = 17.501 INFO: Radius of powder ring 1 (X-pixels) = 68.940 INFO: Radius of powder ring 2 (X-pixels) = 175.006 REFINE BEAM CENTRE [YES]:? The entered beam centre position may be kept fixed or may be refined along with the tilt and powder ring opening angles. If the beam centre is well known e.g. by using a semi-transparent beam-stop then it is probably better not to refine the position. Enter "YES" for the beam centre position to be refined, and "NO" if the present values are to be kept fixed. REFINE BEAM CENTRE [YES]: REFINE SAMPLE DISTANCE [NO]:? Enter "YES" to refine the sample to detector distance. Normally it is probably best initially to keep this fixed, so answer "NO". Having refined all other parameters, on a second iteration the distance can also be simultaneously refined. REFINE SAMPLE DISTANCE [NO]: INFO: Calculating centre of gravity coordinates on 1 powder ring INFO: Number of acceptable coordinates on first ring = 90 INFO: Fitting ellipse to centre of gravity coordinates INFO: Number of coordinates = 90 INFO: Best fit ellipse centre (X/Y mm) = 25.61649 25.61244 INFO: Best fit ellipse centre (X/Y pixels) = 256.1649 256.1244 INFO: Best fit radius 1, radius 2 (mm) = 7.004938 6.983946 INFO: Best fit radius 1 (X pixels) = 70.04938 INFO: Best fit radius 2 (Y pixels) = 69.83945 INFO: Best fit angle of axis 1 (degrees) = 36.94184 INFO: Estimated coordinate radial position error (mm) = .1446273E-01 INFO: Estimated coordinate radial position error (X pixels) = .1446273 INFO: Calculating centre of gravity coordinates on powder rings INFO: Fitting powder rings to centre of gravity coordinates INFO: Number of iterations = 6 INFO: Number of function calls = 64 INFO: Sum of squares = .3310081043632615E-07 INFO: Number of rejected coordinates = 0 INFO: Number of iterations = 4 INFO: Number of function calls = 125 INFO: Sum of squares = .3310080011801160E-07 INFO: Fit of powder ring to inclined detector INFO: Best fit beam centre (X/Y mm) = 25.59796 25.60419 INFO: Best fit beam centre (X/Y pixels) = 255.9796 256.0420 INFO: Cone 1 best fit 2 theta angle (degrees) = 1.999999 INFO: Cone 2 best fit 2 theta angle (degrees) = 4.999608 INFO: Best fit angle of tilt plane rotation (degrees) = 28.59482 INFO: Best fit angle of tilt (degrees) = 4.826739 INFO: Estimated coordinate radial position error (mm) = .1379255E-01 INFO: Estimated coordinate radial position error (X pixels) = .1379255 INFO: Alternative fit of powder rings to centre of gravity coordinates INFO: Number of iterations = 5 INFO: Number of function calls = 46 INFO: Sum of squares = .3310209888343639E-07 INFO: Number of rejected coordinates = 0 INFO: Number of iterations = 1 INFO: Number of function calls = 18 INFO: Sum of squares = .3310145134466964E-07 INFO: Fit of powder ring to inclined detector INFO: Best fit beam centre (X/Y mm) = 25.59794 25.60424 INFO: Best fit beam centre (X/Y pixels) = 255.9794 256.0424 INFO: Cone 1 best fit 2 theta angle (degrees) = 2.000001 INFO: Cone 2 best fit 2 theta angle (degrees) = 4.999619 INFO: Best fit angle of tilt plane rotation (degrees) = 28.54871 INFO: Best fit angle of tilt (degrees) = 4.825097 INFO: Estimated coordinate radial position error (mm) = .1379269E-01 INFO: Estimated coordinate radial position error (X pixels) = .1379269 INFO: In the absence of any other information, there are two equally valid solutions to the tilt angle and beam centre which could have formed a powder ring on an inclined detector. Theoretically these should both give the same goodness of fit, so both solutions are output. If the position of the beam-stop is known, then the "correct" solution may be selected. INFO: SOLUTION 1 INFO: Best fit beam centre (X/Y mm) = 25.59796 25.60419 INFO: Best fit beam centre (X/Y pixels) = 255.9796 256.0420 INFO: Cone 1 best fit 2 theta angle (degrees) = 1.999999 INFO: Cone 2 best fit 2 theta angle (degrees) = 4.999608 INFO: Best fit angle of tilt plane rotation (degrees) = 28.59482 INFO: Best fit angle of tilt (degrees) = 4.826739 INFO: Estimated coordinate radial position error (mm) = .1379255E-01 INFO: Estimated coordinate radial position error (X pixels) = .1379255 INFO: SOLUTION 2 INFO: Best fit beam centre (X/Y mm) = 25.59794 25.60424 INFO: Best fit beam centre (X/Y pixels) = 255.9794 256.0424 INFO: Cone 1 best fit 2 theta angle (degrees) = 2.000001 INFO: Cone 2 best fit 2 theta angle (degrees) = 4.999619 INFO: Best fit angle of tilt plane rotation (degrees) = 28.54871 INFO: Best fit angle of tilt (degrees) = 4.825097 INFO: Estimated coordinate radial position error (mm) = .1379269E-01 INFO: Estimated coordinate radial position error (X pixels) = .1379269 WHICH SOLUTION (Range: 1 to 2) [1]: