|
|
|
#include <LASAlignmentTubeAlgorithm.h>
Public Member Functions | |
| LASBarrelAlignmentParameterSet | CalculateParameters (LASGlobalData< LASCoordinateSet > &, LASGlobalData< LASCoordinateSet > &) |
| double | GetTEC2TECAlignmentParameterCorrection (int, int, int, LASGlobalData< LASCoordinateSet > &, LASBarrelAlignmentParameterSet &) |
| double | GetTIBTOBAlignmentParameterCorrection (int, int, int, LASGlobalData< LASCoordinateSet > &, LASBarrelAlignmentParameterSet &) |
| LASAlignmentTubeAlgorithm () | |
| void | ReadMisalignmentFromFile (const char *, LASGlobalData< LASCoordinateSet > &, LASGlobalData< LASCoordinateSet > &) |
implementation of the alignment tube analytical algorithm
Definition at line 20 of file LASAlignmentTubeAlgorithm.h.
| LASAlignmentTubeAlgorithm::LASAlignmentTubeAlgorithm | ( | ) |
Definition at line 8 of file LASAlignmentTubeAlgorithm.cc.
{
}
| LASBarrelAlignmentParameterSet LASAlignmentTubeAlgorithm::CalculateParameters | ( | LASGlobalData< LASCoordinateSet > & | measuredCoordinates, |
| LASGlobalData< LASCoordinateSet > & | nominalCoordinates | ||
| ) |
Definition at line 18 of file LASAlignmentTubeAlgorithm.cc.
References funct::cos(), gather_cfg::cout, LASBarrelAlignmentParameterSet::GetBeamParameter(), LASBarrelAlignmentParameterSet::GetParameter(), LASCoordinateSet::GetPhi(), LASGlobalData< T >::GetTEC2TECEntry(), LASGlobalData< T >::GetTIBTOBEntry(), LASCoordinateSet::GetZ(), pos, funct::pow(), funct::sin(), LASGlobalLoop::TEC2TECLoop(), and LASGlobalLoop::TIBTOBLoop().
Referenced by LaserAlignment::endRun().
{
std::cout << " [LASAlignmentTubeAlgorithm::CalculateParameters] -- Starting." << std::endl;
// for debugging only
//######################################################################################
//ReadMisalignmentFromFile( "misalign-var.txt", measuredCoordinates, nominalCoordinates );
//######################################################################################
// loop object
LASGlobalLoop globalLoop;
int det, beam, disk, pos;
// phi positions of the AT beams in rad
const double phiPositions[8] = { 0.392699, 1.289799, 1.851794, 2.748894, 3.645995, 4.319690, 5.216791, 5.778784 };
std::vector<double> beamPhiPositions( 8, 0. );
for( beam = 0; beam < 8; ++beam ) beamPhiPositions.at( beam ) = phiPositions[beam];
// the radii of the alignment tube beams for each halfbarrel.
// the halfbarrels 1-6 are (see TkLasATModel TWiki): TEC+, TEC-, TIB+, TIB-. TOB+, TOB-
// in TIB/TOB modules these radii differ from the beam radius..
// ..due to the radial offsets (after the semitransparent mirrors)
const double radii[6] = { 564., 564., 514., 514., 600., 600. };
std::vector<double> beamRadii( 6, 0. );
for( int aHalfbarrel = 0; aHalfbarrel < 6; ++aHalfbarrel ) beamRadii.at( aHalfbarrel ) = radii[aHalfbarrel];
// the z positions of the halfbarrel_end_faces / outer_TEC_disks (in mm);
// parameters are: det, side(0=+/1=-), z(0=lowerZ/1=higherZ). TECs have no side (use side = 0)
std::vector<std::vector<std::vector<double> > > endFaceZPositions( 4, std::vector<std::vector<double> >( 2, std::vector<double>( 2, 0. ) ) );
endFaceZPositions.at( 0 ).at( 0 ).at( 0 ) = 1322.5; // TEC+, *, disk1 ///
endFaceZPositions.at( 0 ).at( 0 ).at( 1 ) = 2667.5; // TEC+, *, disk9 /// SIDE INFORMATION
endFaceZPositions.at( 1 ).at( 0 ).at( 0 ) = -2667.5; // TEC-, *, disk9 /// MEANINGLESS FOR TEC -> USE .at(0)!
endFaceZPositions.at( 1 ).at( 0 ).at( 1 ) = -1322.5; // TEC-, *, disk1 ///
endFaceZPositions.at( 2 ).at( 1 ).at( 0 ) = -700.; // TIB, -, outer
endFaceZPositions.at( 2 ).at( 1 ).at( 1 ) = -300.; // TIB, -, inner
endFaceZPositions.at( 2 ).at( 0 ).at( 0 ) = 300.; // TIB, +, inner
endFaceZPositions.at( 2 ).at( 0 ).at( 1 ) = 700.; // TIB, +, outer
endFaceZPositions.at( 3 ).at( 1 ).at( 0 ) = -1090.; // TOB, -, outer
endFaceZPositions.at( 3 ).at( 1 ).at( 1 ) = -300.; // TOB, -, inner
endFaceZPositions.at( 3 ).at( 0 ).at( 0 ) = 300.; // TOB, +, inner
endFaceZPositions.at( 3 ).at( 0 ).at( 1 ) = 1090.; // TOB, +, outer
// reduced z positions of the beam spots ( z'_{k,j}, z"_{k,j} )
double detReducedZ[2] = { 0., 0. };
// reduced beam splitter positions ( zt'_{k,j}, zt"_{k,j} )
double beamReducedZ[2] = { 0., 0. };
// the z positions of the virtual planes at which the beam parameters are measured
std::vector<double> disk9EndFaceZPositions( 2, 0. );
disk9EndFaceZPositions.at( 0 ) = -2667.5; // TEC- disk9
disk9EndFaceZPositions.at( 1 ) = 2667.5; // TEC+ disk9
// define sums over measured values to "simplify" the beam parameter formulas
// all these have 6 entries, one for each halfbarrel (TEC+,TEC-,TIB+,TIB-,TOB+,TOB-)
std::vector<double> sumOverPhiZPrime( 6, 0. );
std::vector<double> sumOverPhiZPrimePrime( 6, 0. );
std::vector<double> sumOverPhiZPrimeSinTheta( 6, 0. );
std::vector<double> sumOverPhiZPrimePrimeSinTheta( 6, 0. );
std::vector<double> sumOverPhiZPrimeCosTheta( 6, 0. );
std::vector<double> sumOverPhiZPrimePrimeCosTheta( 6, 0. );
// these have 8 entries, one for each beam
std::vector<double> sumOverPhiZTPrime( 8, 0. );
std::vector<double> sumOverPhiZTPrimePrime( 8, 0. );
// define sums over nominal values
// all these have 6 entries, one for each halfbarrel (TEC+,TEC-,TIB+,TIB-,TOB+,TOB-)
std::vector<double> sumOverZPrimeSquared( 6, 0. );
std::vector<double> sumOverZPrimePrimeSquared( 6, 0. );
std::vector<double> sumOverZPrimeZPrimePrime( 6, 0. );
std::vector<double> sumOverZPrimeZTPrime( 6, 0. );
std::vector<double> sumOverZPrimeZTPrimePrime( 6, 0. );
std::vector<double> sumOverZPrimePrimeZTPrime( 6, 0. );
std::vector<double> sumOverZPrimePrimeZTPrimePrime( 6, 0. );
// all these are scalars
double sumOverZTPrimeSquared = 0.;
double sumOverZTPrimePrimeSquared = 0.;
double sumOverZTPrimeZTPrimePrime = 0.;
// now calculate them for TIBTOB
det = 2; beam = 0; pos = 0;
do {
// define the side: 0 for TIB+/TOB+ and 1 for TIB-/TOB-
const int theSide = pos<3 ? 0 : 1;
// define the halfbarrel number from det/side
const int halfbarrel = det==2 ? det+theSide : det+1+theSide; // TIB:TOB
// this is the path the beam has to travel radially after being reflected
// by the AT mirrors (TIB:50mm, TOB:36mm) -> used for beam parameters
const double radialOffset = det==2 ? 50. : 36.;
// reduced module's z position with respect to the subdetector endfaces (zPrime, zPrimePrime)
detReducedZ[0] = measuredCoordinates.GetTIBTOBEntry( det, beam, pos ).GetZ() - endFaceZPositions.at( det ).at( theSide ).at( 0 ); // = zPrime
detReducedZ[0] /= ( endFaceZPositions.at( det ).at( theSide ).at( 1 ) - endFaceZPositions.at( det ).at( theSide ).at( 0 ) );
detReducedZ[1] = endFaceZPositions.at( det ).at( theSide ).at( 1 ) - measuredCoordinates.GetTIBTOBEntry( det, beam, pos ).GetZ(); // = zPrimePrime
detReducedZ[1] /= ( endFaceZPositions.at( det ).at( theSide ).at( 1 ) - endFaceZPositions.at( det ).at( theSide ).at( 0 ) );
// reduced module's z position with respect to the tec disks +-9 (for the beam parameters)
beamReducedZ[0] = ( measuredCoordinates.GetTIBTOBEntry( det, beam, pos ).GetZ() - radialOffset ) - disk9EndFaceZPositions.at( 0 ); // = ZTPrime
beamReducedZ[0] /= ( disk9EndFaceZPositions.at( 1 ) - disk9EndFaceZPositions.at( 0 ) );
beamReducedZ[1] = disk9EndFaceZPositions.at( 1 ) - ( measuredCoordinates.GetTIBTOBEntry( det, beam, pos ).GetZ() - radialOffset ); // ZTPrimePrime
beamReducedZ[1] /= ( disk9EndFaceZPositions.at( 1 ) - disk9EndFaceZPositions.at( 0 ) );
// residual in phi (in the formulas this corresponds to y_ik/R)
const double phiResidual = measuredCoordinates.GetTIBTOBEntry( det, beam, pos ).GetPhi() - nominalCoordinates.GetTIBTOBEntry( det, beam, pos ).GetPhi();
// sums over measured values
sumOverPhiZPrime.at( halfbarrel ) += phiResidual * detReducedZ[0];
sumOverPhiZPrimePrime.at( halfbarrel ) += phiResidual * detReducedZ[1];
sumOverPhiZPrimeSinTheta.at( halfbarrel ) += phiResidual * detReducedZ[0] * sin( beamPhiPositions.at( beam ) );
sumOverPhiZPrimePrimeSinTheta.at( halfbarrel ) += phiResidual * detReducedZ[1] * sin( beamPhiPositions.at( beam ) );
sumOverPhiZPrimeCosTheta.at( halfbarrel ) += phiResidual * detReducedZ[0] * cos( beamPhiPositions.at( beam ) );
sumOverPhiZPrimePrimeCosTheta.at( halfbarrel ) += phiResidual * detReducedZ[1] * cos( beamPhiPositions.at( beam ) );
sumOverPhiZTPrime.at( beam ) += phiResidual * beamReducedZ[0]; // note the index change here..
sumOverPhiZTPrimePrime.at( beam ) += phiResidual * beamReducedZ[1];
// sums over nominal values
sumOverZPrimeSquared.at( halfbarrel ) += pow( detReducedZ[0], 2 ) / 8.; // these are defined beam-wise, so: / 8.
sumOverZPrimePrimeSquared.at( halfbarrel ) += pow( detReducedZ[1], 2 ) / 8.;
sumOverZPrimeZPrimePrime.at( halfbarrel ) += detReducedZ[0] * detReducedZ[1] / 8.;
sumOverZPrimeZTPrime.at( halfbarrel ) += detReducedZ[0] * beamReducedZ[0] / 8.;
sumOverZPrimeZTPrimePrime.at( halfbarrel ) += detReducedZ[0] * beamReducedZ[1] / 8.;
sumOverZPrimePrimeZTPrime.at( halfbarrel ) += detReducedZ[1] * beamReducedZ[0] / 8.;
sumOverZPrimePrimeZTPrimePrime.at( halfbarrel ) += detReducedZ[1] * beamReducedZ[1] / 8.;
sumOverZTPrimeSquared += pow( beamReducedZ[0], 2 ) / 8.;
sumOverZTPrimePrimeSquared += pow( beamReducedZ[1], 2 ) / 8.;
sumOverZTPrimeZTPrimePrime += beamReducedZ[0] * beamReducedZ[1] / 8.;
} while( globalLoop.TIBTOBLoop( det, beam, pos ) );
// now for TEC2TEC
det = 0; beam = 0; disk = 0;
do {
// for the tec, the halfbarrel numbers are equal to the det numbers...
const int halfbarrel = det;
// ...so there's no side distinction for the TEC
const int theSide = 0;
// also, there's no radial offset for the TEC
const double radialOffset = 0.;
// reduced module's z position with respect to the subdetector endfaces (zPrime, zPrimePrime)
detReducedZ[0] = measuredCoordinates.GetTEC2TECEntry( det, beam, disk ).GetZ() - endFaceZPositions.at( det ).at( theSide ).at( 0 ); // = zPrime
detReducedZ[0] /= ( endFaceZPositions.at( det ).at( theSide ).at( 1 ) - endFaceZPositions.at( det ).at( theSide ).at( 0 ) );
detReducedZ[1] = endFaceZPositions.at( det ).at( theSide ).at( 1 ) - measuredCoordinates.GetTEC2TECEntry( det, beam, disk ).GetZ(); // = zPrimePrime
detReducedZ[1] /= ( endFaceZPositions.at( det ).at( theSide ).at( 1 ) - endFaceZPositions.at( det ).at( theSide ).at( 0 ) );
// reduced module's z position with respect to the tec disks +-9 (for the beam parameters)
beamReducedZ[0] = ( measuredCoordinates.GetTEC2TECEntry( det, beam, disk ).GetZ() - radialOffset ) - disk9EndFaceZPositions.at( 0 ); // = ZTPrime
beamReducedZ[0] /= ( disk9EndFaceZPositions.at( 1 ) - disk9EndFaceZPositions.at( 0 ) );
beamReducedZ[1] = disk9EndFaceZPositions.at( 1 ) - ( measuredCoordinates.GetTEC2TECEntry( det, beam, disk ).GetZ() - radialOffset ); // ZTPrimePrime
beamReducedZ[1] /= ( disk9EndFaceZPositions.at( 1 ) - disk9EndFaceZPositions.at( 0 ) );
// residual in phi (in the formulas this corresponds to y_ik/R)
const double phiResidual = measuredCoordinates.GetTEC2TECEntry( det, beam, disk ).GetPhi() - nominalCoordinates.GetTEC2TECEntry( det, beam, disk ).GetPhi();
// sums over measured values
sumOverPhiZPrime.at( halfbarrel ) += phiResidual * detReducedZ[0];
sumOverPhiZPrimePrime.at( halfbarrel ) += phiResidual * detReducedZ[1];
sumOverPhiZPrimeSinTheta.at( halfbarrel ) += phiResidual * detReducedZ[0] * sin( beamPhiPositions.at( beam ) );
sumOverPhiZPrimePrimeSinTheta.at( halfbarrel ) += phiResidual * detReducedZ[1] * sin( beamPhiPositions.at( beam ) );
sumOverPhiZPrimeCosTheta.at( halfbarrel ) += phiResidual * detReducedZ[0] * cos( beamPhiPositions.at( beam ) );
sumOverPhiZPrimePrimeCosTheta.at( halfbarrel ) += phiResidual * detReducedZ[1] * cos( beamPhiPositions.at( beam ) );
sumOverPhiZTPrime.at( beam ) += phiResidual * beamReducedZ[0]; // note the index change here..
sumOverPhiZTPrimePrime.at( beam ) += phiResidual * beamReducedZ[1];
// sums over nominal values
sumOverZPrimeSquared.at( halfbarrel ) += pow( detReducedZ[0], 2 ) / 8.; // these are defined beam-wise, so: / 8.
sumOverZPrimePrimeSquared.at( halfbarrel ) += pow( detReducedZ[1], 2 ) / 8.;
sumOverZPrimeZPrimePrime.at( halfbarrel ) += detReducedZ[0] * detReducedZ[1] / 8.;
sumOverZPrimeZTPrime.at( halfbarrel ) += detReducedZ[0] * beamReducedZ[0] / 8.;
sumOverZPrimeZTPrimePrime.at( halfbarrel ) += detReducedZ[0] * beamReducedZ[1] / 8.;
sumOverZPrimePrimeZTPrime.at( halfbarrel ) += detReducedZ[1] * beamReducedZ[0] / 8.;
sumOverZPrimePrimeZTPrimePrime.at( halfbarrel ) += detReducedZ[1] * beamReducedZ[1] / 8.;
sumOverZTPrimeSquared += pow( beamReducedZ[0], 2 ) / 8.;
sumOverZTPrimePrimeSquared += pow( beamReducedZ[1], 2 ) / 8.;
sumOverZTPrimeZTPrimePrime += beamReducedZ[0] * beamReducedZ[1] / 8.;
} while( globalLoop.TEC2TECLoop( det, beam, disk ) );
// more "simplification" terms...
// these here are functions of theta and can be calculated directly
double sumOverSinTheta = 0.;
double sumOverCosTheta = 0.;
double sumOverSinThetaSquared = 0.;
double sumOverCosThetaSquared = 0.;
double sumOverCosThetaSinTheta = 0.;
double mixedTrigonometricTerm = 0.;
for( beam = 0; beam < 8; ++beam ) {
sumOverSinTheta += sin( beamPhiPositions.at( beam ) );
sumOverCosTheta += cos( beamPhiPositions.at( beam ) );
sumOverSinThetaSquared += pow( sin( beamPhiPositions.at( beam ) ), 2 );
sumOverCosThetaSquared += pow( cos( beamPhiPositions.at( beam ) ), 2 );
sumOverCosThetaSinTheta += cos( beamPhiPositions.at( beam ) ) * sin( beamPhiPositions.at( beam ) );
}
mixedTrigonometricTerm = 8. * ( sumOverCosThetaSquared * sumOverSinThetaSquared - pow( sumOverCosThetaSinTheta, 2 ) )
- pow( sumOverCosTheta, 2 ) * sumOverSinThetaSquared
- pow( sumOverSinTheta, 2 ) * sumOverCosThetaSquared
+ 2. * sumOverCosTheta * sumOverSinTheta * sumOverCosThetaSinTheta;
// even more shortcuts before we can calculate the parameters
double beamDenominator = ( pow( sumOverZTPrimeZTPrimePrime, 2 ) - sumOverZTPrimeSquared * sumOverZTPrimePrimeSquared ) * beamRadii.at( 0 );
std::vector<double> alignmentDenominator( 6, 0. );
std::vector<double> termA( 6, 0. ), termB( 6, 0. ), termC( 6, 0. ), termD( 6, 0. );
for( unsigned int aHalfbarrel = 0; aHalfbarrel < 6; ++aHalfbarrel ) {
alignmentDenominator.at ( aHalfbarrel ) = ( pow( sumOverZPrimeZPrimePrime.at( aHalfbarrel ), 2 ) - sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) ) * mixedTrigonometricTerm;
termA.at( aHalfbarrel ) = sumOverZPrimeZTPrime.at( aHalfbarrel ) * sumOverZTPrimeZTPrimePrime - sumOverZPrimeZTPrimePrime.at( aHalfbarrel ) * sumOverZTPrimeSquared;
termB.at( aHalfbarrel ) = sumOverZPrimePrimeZTPrime.at( aHalfbarrel ) * sumOverZTPrimeZTPrimePrime - sumOverZPrimePrimeZTPrimePrime.at( aHalfbarrel ) * sumOverZTPrimeSquared;
termC.at( aHalfbarrel ) = sumOverZPrimeZTPrimePrime.at( aHalfbarrel ) * sumOverZTPrimeZTPrimePrime - sumOverZPrimeZTPrime.at( aHalfbarrel ) * sumOverZTPrimePrimeSquared;
termD.at( aHalfbarrel ) = sumOverZPrimePrimeZTPrimePrime.at( aHalfbarrel ) * sumOverZTPrimeZTPrimePrime - sumOverZPrimePrimeZTPrime.at( aHalfbarrel ) * sumOverZTPrimePrimeSquared;
}
// have eight alignment tube beams..
const int numberOfBeams = 8;
// that's all for preparation, now it gets ugly:
// calculate the alignment parameters
LASBarrelAlignmentParameterSet theResult;
// can do this in one go for all halfbarrels
for( int aHalfbarrel = 0; aHalfbarrel < 6; ++aHalfbarrel ) {
// no errors yet
// rotation angles of the lower z endfaces (in rad)
theResult.GetParameter( aHalfbarrel, 0, 0 ).first = ( sumOverPhiZPrime.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosThetaSquared * sumOverSinThetaSquared
- sumOverPhiZPrimePrime.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverCosThetaSquared * sumOverSinThetaSquared
- sumOverPhiZPrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinThetaSquared
+ sumOverPhiZPrimePrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinThetaSquared
+ sumOverPhiZPrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosThetaSinTheta * sumOverSinTheta
- sumOverPhiZPrimePrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverCosThetaSinTheta * sumOverSinTheta
- sumOverPhiZPrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosThetaSquared * sumOverSinTheta
+ sumOverPhiZPrimePrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverCosThetaSquared * sumOverSinTheta
- sumOverPhiZPrime.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * pow( sumOverCosThetaSinTheta, 2 )
+ sumOverPhiZPrimePrime.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * pow( sumOverCosThetaSinTheta, 2 )
+ sumOverPhiZPrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverCosThetaSinTheta
- sumOverPhiZPrimePrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverCosThetaSinTheta )
/ alignmentDenominator.at ( aHalfbarrel );
// rotation angles of the upper z endfaces (in rad)
theResult.GetParameter( aHalfbarrel, 1, 0 ).first = ( sumOverPhiZPrimePrime.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosThetaSquared * sumOverSinThetaSquared
- sumOverPhiZPrime.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosThetaSquared * sumOverSinThetaSquared
- sumOverPhiZPrimePrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinThetaSquared
+ sumOverPhiZPrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinThetaSquared
+ sumOverPhiZPrimePrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosThetaSinTheta * sumOverSinTheta
- sumOverPhiZPrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosThetaSinTheta * sumOverSinTheta
- sumOverPhiZPrimePrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosThetaSquared * sumOverSinTheta
+ sumOverPhiZPrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosThetaSquared * sumOverSinTheta
- sumOverPhiZPrimePrime.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosThetaSinTheta * sumOverCosThetaSinTheta
+ sumOverPhiZPrime.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosThetaSinTheta * sumOverCosThetaSinTheta
+ sumOverPhiZPrimePrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverCosThetaSinTheta
- sumOverPhiZPrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverCosThetaSinTheta )
/ alignmentDenominator.at ( aHalfbarrel );
// x deviations of the lower z endfaces (in mm)
theResult.GetParameter( aHalfbarrel, 0, 1 ).first = -1. * ( sumOverPhiZPrime.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosThetaSquared * sumOverSinTheta
- sumOverPhiZPrimePrime.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverCosThetaSquared * sumOverSinTheta
- sumOverPhiZPrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinTheta
+ sumOverPhiZPrimePrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinTheta
- sumOverPhiZPrime.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverCosThetaSinTheta
+ sumOverPhiZPrimePrime.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverCosThetaSinTheta
+ sumOverPhiZPrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * numberOfBeams * sumOverCosThetaSinTheta
- sumOverPhiZPrimePrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * numberOfBeams * sumOverCosThetaSinTheta
- sumOverPhiZPrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * numberOfBeams * sumOverCosThetaSquared
+ sumOverPhiZPrimePrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * numberOfBeams * sumOverCosThetaSquared
+ sumOverPhiZPrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverCosTheta
- sumOverPhiZPrimePrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverCosTheta )
/ alignmentDenominator.at ( aHalfbarrel ) * beamRadii.at( aHalfbarrel );
// x deviations of the upper z endfaces (in mm)
theResult.GetParameter( aHalfbarrel, 1, 1 ).first = -1. * ( sumOverPhiZPrimePrime.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosThetaSquared * sumOverSinTheta
- sumOverPhiZPrime.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosThetaSquared * sumOverSinTheta
- sumOverPhiZPrimePrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinTheta
+ sumOverPhiZPrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinTheta
- sumOverPhiZPrimePrime.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverCosThetaSinTheta
+ sumOverPhiZPrime.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverCosThetaSinTheta
+ sumOverPhiZPrimePrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * numberOfBeams * sumOverCosThetaSinTheta
- sumOverPhiZPrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * numberOfBeams * sumOverCosThetaSinTheta
- sumOverPhiZPrimePrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * numberOfBeams * sumOverCosThetaSquared
+ sumOverPhiZPrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * numberOfBeams * sumOverCosThetaSquared
+ sumOverPhiZPrimePrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverCosTheta
- sumOverPhiZPrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverCosTheta )
/ alignmentDenominator.at ( aHalfbarrel ) * beamRadii.at( aHalfbarrel );
// y deviations of the lower z endfaces (in mm)
theResult.GetParameter( aHalfbarrel, 0, 2 ).first = ( sumOverPhiZPrime.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinThetaSquared
- sumOverPhiZPrimePrime.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinThetaSquared
- sumOverPhiZPrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * numberOfBeams * sumOverSinThetaSquared
+ sumOverPhiZPrimePrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * numberOfBeams * sumOverSinThetaSquared
+ sumOverPhiZPrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverSinTheta * sumOverSinTheta
- sumOverPhiZPrimePrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverSinTheta * sumOverSinTheta
- sumOverPhiZPrime.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosThetaSinTheta * sumOverSinTheta
+ sumOverPhiZPrimePrime.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverCosThetaSinTheta * sumOverSinTheta
- sumOverPhiZPrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinTheta
+ sumOverPhiZPrimePrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinTheta
+ sumOverPhiZPrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * numberOfBeams * sumOverCosThetaSinTheta
- sumOverPhiZPrimePrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeSquared.at( aHalfbarrel ) * numberOfBeams * sumOverCosThetaSinTheta )
/ alignmentDenominator.at ( aHalfbarrel ) * beamRadii.at( aHalfbarrel );
// y deviations of the upper z endfaces (in mm)
theResult.GetParameter( aHalfbarrel, 1, 2 ).first = ( sumOverPhiZPrimePrime.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinThetaSquared
- sumOverPhiZPrime.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinThetaSquared
- sumOverPhiZPrimePrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * numberOfBeams * sumOverSinThetaSquared
+ sumOverPhiZPrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * numberOfBeams * sumOverSinThetaSquared
+ sumOverPhiZPrimePrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverSinTheta * sumOverSinTheta
- sumOverPhiZPrimeCosTheta.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverSinTheta * sumOverSinTheta
- sumOverPhiZPrimePrime.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosThetaSinTheta * sumOverSinTheta
+ sumOverPhiZPrime.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosThetaSinTheta * sumOverSinTheta
- sumOverPhiZPrimePrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinTheta
+ sumOverPhiZPrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * sumOverCosTheta * sumOverSinTheta
+ sumOverPhiZPrimePrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimeZPrimePrime.at( aHalfbarrel ) * numberOfBeams * sumOverCosThetaSinTheta
- sumOverPhiZPrimeSinTheta.at( aHalfbarrel ) * sumOverZPrimePrimeSquared.at( aHalfbarrel ) * numberOfBeams * sumOverCosThetaSinTheta )
/ alignmentDenominator.at ( aHalfbarrel ) * beamRadii.at( aHalfbarrel );
}
// another loop is needed here to calculate some terms for the beam parameters
double vsumA = 0., vsumB = 0., vsumC = 0., vsumD = 0., vsumE = 0., vsumF = 0.;
for( unsigned int aHalfbarrel = 0; aHalfbarrel < 6; ++aHalfbarrel ) {
vsumA += theResult.GetParameter( aHalfbarrel, 1, 2 ).first * termA.at( aHalfbarrel ) + theResult.GetParameter( aHalfbarrel, 0, 2 ).first * termB.at( aHalfbarrel );
vsumB += theResult.GetParameter( aHalfbarrel, 1, 1 ).first * termA.at( aHalfbarrel ) + theResult.GetParameter( aHalfbarrel, 0, 1 ).first * termB.at( aHalfbarrel );
vsumC += beamRadii.at( aHalfbarrel ) * ( theResult.GetParameter( aHalfbarrel, 1, 0 ).first * termA.at( aHalfbarrel ) + theResult.GetParameter( aHalfbarrel, 0, 0 ).first * termB.at( aHalfbarrel ) );
vsumD += theResult.GetParameter( aHalfbarrel, 1, 2 ).first * termC.at( aHalfbarrel ) + theResult.GetParameter( aHalfbarrel, 0, 2 ).first * termD.at( aHalfbarrel );
vsumE += theResult.GetParameter( aHalfbarrel, 1, 1 ).first * termC.at( aHalfbarrel ) + theResult.GetParameter( aHalfbarrel, 0, 1 ).first * termD.at( aHalfbarrel );
vsumF += beamRadii.at( aHalfbarrel ) * ( theResult.GetParameter( aHalfbarrel, 1, 0 ).first * termC.at( aHalfbarrel ) + theResult.GetParameter( aHalfbarrel, 0, 0 ).first * termD.at( aHalfbarrel ) );
}
// calculate the beam parameters
for( unsigned int beam = 0; beam < 8; ++beam ) {
// parameter A, defined at lower z
theResult.GetBeamParameter( beam, 0 ).first = ( cos( beamPhiPositions.at( beam ) ) * vsumA
- sin( beamPhiPositions.at( beam ) ) * vsumB
- vsumC
+ sumOverPhiZTPrime.at( beam ) * sumOverZTPrimeZTPrimePrime - sumOverPhiZTPrimePrime.at( beam ) * sumOverZTPrimeSquared )
/ beamDenominator;
std::cout << "BBBBBBBB: " << cos( beamPhiPositions.at( beam ) ) * vsumA << " "
<< -1. * sin( beamPhiPositions.at( beam ) ) * vsumB << " "
<< -1. * vsumC << " "
<< sumOverPhiZTPrime.at( beam ) * sumOverZTPrimeZTPrimePrime - sumOverPhiZTPrimePrime.at( beam ) * sumOverZTPrimeSquared << " "
<< beamDenominator
<< std::endl;
// parameter B, defined at upper z
theResult.GetBeamParameter( beam, 1 ).first = ( cos( beamPhiPositions.at( beam ) ) * vsumD
- sin( beamPhiPositions.at( beam ) ) * vsumE
- vsumF
+ sumOverPhiZTPrimePrime.at( beam ) * sumOverZTPrimeZTPrimePrime - sumOverPhiZTPrime.at( beam ) * sumOverZTPrimePrimeSquared )
/ beamDenominator;
}
return theResult;
}
| double LASAlignmentTubeAlgorithm::GetTEC2TECAlignmentParameterCorrection | ( | int | det, |
| int | beam, | ||
| int | disk, | ||
| LASGlobalData< LASCoordinateSet > & | nominalCoordinates, | ||
| LASBarrelAlignmentParameterSet & | alignmentParameters | ||
| ) |
get global phi correction from alignment parameters for an alignment tube module in TEC(AT)
Definition at line 537 of file LASAlignmentTubeAlgorithm.cc.
References funct::cos(), LASBarrelAlignmentParameterSet::GetBeamParameter(), LASBarrelAlignmentParameterSet::GetParameter(), LASGlobalData< T >::GetTEC2TECEntry(), LASCoordinateSet::GetZ(), and funct::sin().
Referenced by LaserAlignment::ApplyATMaskingCorrections().
{
// INITIALIZATION;
// ALL THIS IS DUPLICATED FOR THE MOMENT, SHOULD FINALLY BE CALCULATED ONLY ONCE
// AND HARD CODED NUMBERS SHOULD CENTRALLY BE IMPORTED FROM src/LASConstants.h
// the z positions of the halfbarrel_end_faces / outer_TEC_disks (in mm);
// parameters are: det, side(0=+/1=-), z(0=lowerZ/1=higherZ). TECs have no side (use side = 0)
std::vector<std::vector<std::vector<double> > > endFaceZPositions( 4, std::vector<std::vector<double> >( 2, std::vector<double>( 2, 0. ) ) );
endFaceZPositions.at( 0 ).at( 0 ).at( 0 ) = 1322.5; // TEC+, *, disk1 ///
endFaceZPositions.at( 0 ).at( 0 ).at( 1 ) = 2667.5; // TEC+, *, disk9 /// SIDE INFORMATION
endFaceZPositions.at( 1 ).at( 0 ).at( 0 ) = -2667.5; // TEC-, *, disk9 /// MEANINGLESS FOR TEC -> USE .at(0)!
endFaceZPositions.at( 1 ).at( 0 ).at( 1 ) = -1322.5; // TEC-, *, disk1 ///
endFaceZPositions.at( 2 ).at( 1 ).at( 0 ) = -700.; // TIB, -, outer
endFaceZPositions.at( 2 ).at( 1 ).at( 1 ) = -300.; // TIB, -, inner
endFaceZPositions.at( 2 ).at( 0 ).at( 0 ) = 300.; // TIB, +, inner
endFaceZPositions.at( 2 ).at( 0 ).at( 1 ) = 700.; // TIB, +, outer
endFaceZPositions.at( 3 ).at( 1 ).at( 0 ) = -1090.; // TOB, -, outer
endFaceZPositions.at( 3 ).at( 1 ).at( 1 ) = -300.; // TOB, -, inner
endFaceZPositions.at( 3 ).at( 0 ).at( 0 ) = 300.; // TOB, +, inner
endFaceZPositions.at( 3 ).at( 0 ).at( 1 ) = 1090.; // TOB, +, outer
// the z positions of the virtual planes at which the beam parameters are measured
std::vector<double> disk9EndFaceZPositions( 2, 0. );
disk9EndFaceZPositions.at( 0 ) = -2667.5; // TEC- disk9
disk9EndFaceZPositions.at( 1 ) = 2667.5; // TEC+ disk9
// for the tec, the halfbarrel numbers are equal to the det numbers...
const int halfbarrel = det;
// ...so there's no side distinction for the TEC
const int theSide = 0;
// also, there's no radial offset for the TEC
const double radialOffset = 0.;
// phi positions of the AT beams in rad
const double phiPositions[8] = { 0.392699, 1.289799, 1.851794, 2.748894, 3.645995, 4.319690, 5.216791, 5.778784 };
std::vector<double> beamPhiPositions( 8, 0. );
for( unsigned int aBeam = 0; aBeam < 8; ++aBeam ) beamPhiPositions.at( aBeam ) = phiPositions[aBeam];
// the radii of the alignment tube beams for each halfbarrel.
// the halfbarrels 1-6 are (see TkLasATModel TWiki): TEC+, TEC-, TIB+, TIB-. TOB+, TOB-
// in TIB/TOB modules these radii differ from the beam radius..
// ..due to the radial offsets (after the semitransparent mirrors)
const double radii[6] = { 564., 564., 514., 514., 600., 600. };
std::vector<double> beamRadii( 6, 0. );
for( int aHalfbarrel = 0; aHalfbarrel < 6; ++aHalfbarrel ) beamRadii.at( aHalfbarrel ) = radii[aHalfbarrel];
// reduced z positions of the beam spots ( z'_{k,j}, z"_{k,j} )
double detReducedZ[2] = { 0., 0. };
// reduced beam splitter positions ( zt'_{k,j}, zt"_{k,j} )
double beamReducedZ[2] = { 0., 0. };
// reduced module's z position with respect to the subdetector endfaces (zPrime, zPrimePrime)
detReducedZ[0] = nominalCoordinates.GetTEC2TECEntry( det, beam, disk ).GetZ() - endFaceZPositions.at( det ).at( theSide ).at( 0 ); // = zPrime
detReducedZ[0] /= ( endFaceZPositions.at( det ).at( theSide ).at( 1 ) - endFaceZPositions.at( det ).at( theSide ).at( 0 ) );
detReducedZ[1] = endFaceZPositions.at( det ).at( theSide ).at( 1 ) - nominalCoordinates.GetTEC2TECEntry( det, beam, disk ).GetZ(); // = zPrimePrime
detReducedZ[1] /= ( endFaceZPositions.at( det ).at( theSide ).at( 1 ) - endFaceZPositions.at( det ).at( theSide ).at( 0 ) );
// reduced module's z position with respect to the tec disks +-9 (for the beam parameters)
beamReducedZ[0] = ( nominalCoordinates.GetTEC2TECEntry( det, beam, disk ).GetZ() - radialOffset ) - disk9EndFaceZPositions.at( 0 ); // = ZTPrime
beamReducedZ[0] /= ( disk9EndFaceZPositions.at( 1 ) - disk9EndFaceZPositions.at( 0 ) );
beamReducedZ[1] = disk9EndFaceZPositions.at( 1 ) - ( nominalCoordinates.GetTEC2TECEntry( det, beam, disk ).GetZ() - radialOffset ); // ZTPrimePrime
beamReducedZ[1] /= ( disk9EndFaceZPositions.at( 1 ) - disk9EndFaceZPositions.at( 0 ) );
// the correction to phi from the endcap algorithm;
// it is defined such that the correction is to be subtracted ///////////////////////////////// ???
double phiCorrection = 0.;
// contribution from phi rotation of first end face
phiCorrection += detReducedZ[1] * alignmentParameters.GetParameter( halfbarrel, 0, 0 ).first;
// contribution from phi rotation of second end face
phiCorrection += detReducedZ[0] * alignmentParameters.GetParameter( halfbarrel, 1, 0 ).first;
// contribution from translation along x of first endface
phiCorrection += detReducedZ[1] * sin( beamPhiPositions.at( beam ) ) * alignmentParameters.GetParameter( halfbarrel, 0, 1 ).first / beamRadii.at( halfbarrel );
// contribution from translation along x of second endface
phiCorrection += detReducedZ[0] * sin( beamPhiPositions.at( beam ) ) * alignmentParameters.GetParameter( halfbarrel, 1, 1 ).first / beamRadii.at( halfbarrel );
// contribution from translation along y of first endface
phiCorrection -= detReducedZ[1] * cos( beamPhiPositions.at( beam ) ) * alignmentParameters.GetParameter( halfbarrel, 0, 2 ).first / beamRadii.at( halfbarrel );
// contribution from translation along y of second endface
phiCorrection -= detReducedZ[0] * cos( beamPhiPositions.at( beam ) ) * alignmentParameters.GetParameter( halfbarrel, 1, 2 ).first / beamRadii.at( halfbarrel );
// contribution from beam parameters;
// originally, the contribution in meter is proportional to the radius of the beams: beamRadii.at( 0 )
// the additional factor: beamRadii.at( halfbarrel ) converts from meter to radian on the module
phiCorrection += beamReducedZ[1] * alignmentParameters.GetBeamParameter( beam, 0 ).first * beamRadii.at( 0 ) / beamRadii.at( halfbarrel );
phiCorrection += beamReducedZ[0] * alignmentParameters.GetBeamParameter( beam, 1 ).first * beamRadii.at( 0 ) / beamRadii.at( halfbarrel );
return phiCorrection;
}
| double LASAlignmentTubeAlgorithm::GetTIBTOBAlignmentParameterCorrection | ( | int | det, |
| int | beam, | ||
| int | pos, | ||
| LASGlobalData< LASCoordinateSet > & | nominalCoordinates, | ||
| LASBarrelAlignmentParameterSet & | alignmentParameters | ||
| ) |
get global phi correction from alignment parameters for an alignment tube module in TIB/TOB
Definition at line 428 of file LASAlignmentTubeAlgorithm.cc.
References funct::cos(), LASBarrelAlignmentParameterSet::GetBeamParameter(), LASBarrelAlignmentParameterSet::GetParameter(), LASGlobalData< T >::GetTIBTOBEntry(), LASCoordinateSet::GetZ(), and funct::sin().
Referenced by LaserAlignment::ApplyATMaskingCorrections().
{
// INITIALIZATION;
// ALL THIS IS DUPLICATED FOR THE MOMENT, SHOULD FINALLY BE CALCULATED ONLY ONCE
// AND HARD CODED NUMBERS SHOULD CENTRALLY BE IMPORTED FROM src/LASConstants.h
// the z positions of the halfbarrel_end_faces / outer_TEC_disks (in mm);
// parameters are: det, side(0=+/1=-), z(0=lowerZ/1=higherZ). TECs have no side (use side = 0)
std::vector<std::vector<std::vector<double> > > endFaceZPositions( 4, std::vector<std::vector<double> >( 2, std::vector<double>( 2, 0. ) ) );
endFaceZPositions.at( 0 ).at( 0 ).at( 0 ) = 1322.5; // TEC+, *, disk1 ///
endFaceZPositions.at( 0 ).at( 0 ).at( 1 ) = 2667.5; // TEC+, *, disk9 /// SIDE INFORMATION
endFaceZPositions.at( 1 ).at( 0 ).at( 0 ) = -2667.5; // TEC-, *, disk9 /// MEANINGLESS FOR TEC -> USE .at(0)!
endFaceZPositions.at( 1 ).at( 0 ).at( 1 ) = -1322.5; // TEC-, *, disk1 ///
endFaceZPositions.at( 2 ).at( 1 ).at( 0 ) = -700.; // TIB, -, outer
endFaceZPositions.at( 2 ).at( 1 ).at( 1 ) = -300.; // TIB, -, inner
endFaceZPositions.at( 2 ).at( 0 ).at( 0 ) = 300.; // TIB, +, inner
endFaceZPositions.at( 2 ).at( 0 ).at( 1 ) = 700.; // TIB, +, outer
endFaceZPositions.at( 3 ).at( 1 ).at( 0 ) = -1090.; // TOB, -, outer
endFaceZPositions.at( 3 ).at( 1 ).at( 1 ) = -300.; // TOB, -, inner
endFaceZPositions.at( 3 ).at( 0 ).at( 0 ) = 300.; // TOB, +, inner
endFaceZPositions.at( 3 ).at( 0 ).at( 1 ) = 1090.; // TOB, +, outer
// the z positions of the virtual planes at which the beam parameters are measured
std::vector<double> disk9EndFaceZPositions( 2, 0. );
disk9EndFaceZPositions.at( 0 ) = -2667.5; // TEC- disk9
disk9EndFaceZPositions.at( 1 ) = 2667.5; // TEC+ disk9
// define the side: 0 for TIB+/TOB+ and 1 for TIB-/TOB-
const int theSide = pos<3 ? 0 : 1;
// define the halfbarrel number from det/side
const int halfbarrel = det==2 ? det+theSide : det+1+theSide; // TIB:TOB
// this is the path the beam has to travel radially after being reflected
// by the AT mirrors (TIB:50mm, TOB:36mm) -> used for beam parameters
const double radialOffset = det==2 ? 50. : 36.;
// phi positions of the AT beams in rad
const double phiPositions[8] = { 0.392699, 1.289799, 1.851794, 2.748894, 3.645995, 4.319690, 5.216791, 5.778784 };
std::vector<double> beamPhiPositions( 8, 0. );
for( unsigned int aBeam = 0; aBeam < 8; ++aBeam ) beamPhiPositions.at( aBeam ) = phiPositions[aBeam];
// the radii of the alignment tube beams for each halfbarrel.
// the halfbarrels 1-6 are (see TkLasATModel TWiki): TEC+, TEC-, TIB+, TIB-. TOB+, TOB-
// in TIB/TOB modules these radii differ from the beam radius..
// ..due to the radial offsets (after the semitransparent mirrors)
const double radii[6] = { 564., 564., 514., 514., 600., 600. };
std::vector<double> beamRadii( 6, 0. );
for( int aHalfbarrel = 0; aHalfbarrel < 6; ++aHalfbarrel ) beamRadii.at( aHalfbarrel ) = radii[aHalfbarrel];
// reduced z positions of the beam spots ( z'_{k,j}, z"_{k,j} )
double detReducedZ[2] = { 0., 0. };
// reduced beam splitter positions ( zt'_{k,j}, zt"_{k,j} )
double beamReducedZ[2] = { 0., 0. };
// reduced module's z position with respect to the subdetector endfaces (zPrime, zPrimePrime)
detReducedZ[0] = nominalCoordinates.GetTIBTOBEntry( det, beam, pos ).GetZ() - endFaceZPositions.at( det ).at( theSide ).at( 0 ); // = zPrime
detReducedZ[0] /= ( endFaceZPositions.at( det ).at( theSide ).at( 1 ) - endFaceZPositions.at( det ).at( theSide ).at( 0 ) );
detReducedZ[1] = endFaceZPositions.at( det ).at( theSide ).at( 1 ) - nominalCoordinates.GetTIBTOBEntry( det, beam, pos ).GetZ(); // = zPrimePrime
detReducedZ[1] /= ( endFaceZPositions.at( det ).at( theSide ).at( 1 ) - endFaceZPositions.at( det ).at( theSide ).at( 0 ) );
// reduced module's z position with respect to the tec disks +-9 (for the beam parameters)
beamReducedZ[0] = ( nominalCoordinates.GetTIBTOBEntry( det, beam, pos ).GetZ() - radialOffset ) - disk9EndFaceZPositions.at( 0 ); // = ZTPrime
beamReducedZ[0] /= ( disk9EndFaceZPositions.at( 1 ) - disk9EndFaceZPositions.at( 0 ) );
beamReducedZ[1] = disk9EndFaceZPositions.at( 1 ) - ( nominalCoordinates.GetTIBTOBEntry( det, beam, pos ).GetZ() - radialOffset ); // ZTPrimePrime
beamReducedZ[1] /= ( disk9EndFaceZPositions.at( 1 ) - disk9EndFaceZPositions.at( 0 ) );
// the correction to phi from the endcap algorithm;
// it is defined such that the correction is to be subtracted ///////////////////////////////// ???
double phiCorrection = 0.;
// contribution from phi rotation of first end face
phiCorrection += detReducedZ[1] * alignmentParameters.GetParameter( halfbarrel, 0, 0 ).first;
// contribution from phi rotation of second end face
phiCorrection += detReducedZ[0] * alignmentParameters.GetParameter( halfbarrel, 1, 0 ).first;
// contribution from translation along x of first endface
phiCorrection += detReducedZ[1] * sin( beamPhiPositions.at( beam ) ) * alignmentParameters.GetParameter( halfbarrel, 0, 1 ).first / beamRadii.at( halfbarrel );
// contribution from translation along x of second endface
phiCorrection += detReducedZ[0] * sin( beamPhiPositions.at( beam ) ) * alignmentParameters.GetParameter( halfbarrel, 1, 1 ).first / beamRadii.at( halfbarrel );
// contribution from translation along y of first endface
phiCorrection -= detReducedZ[1] * cos( beamPhiPositions.at( beam ) ) * alignmentParameters.GetParameter( halfbarrel, 0, 2 ).first / beamRadii.at( halfbarrel );
// contribution from translation along y of second endface
phiCorrection -= detReducedZ[0] * cos( beamPhiPositions.at( beam ) ) * alignmentParameters.GetParameter( halfbarrel, 1, 2 ).first / beamRadii.at( halfbarrel );
// contribution from beam parameters;
// originally, the contribution in meter is proportional to the radius of the beams: beamRadii.at( 0 )
// the additional factor: beamRadii.at( halfbarrel ) converts from meter to radian on the module
phiCorrection += beamReducedZ[1] * alignmentParameters.GetBeamParameter( beam, 0 ).first * beamRadii.at( 0 ) / beamRadii.at( halfbarrel );
phiCorrection += beamReducedZ[0] * alignmentParameters.GetBeamParameter( beam, 1 ).first * beamRadii.at( 0 ) / beamRadii.at( halfbarrel );
return phiCorrection;
}
| void LASAlignmentTubeAlgorithm::ReadMisalignmentFromFile | ( | const char * | filename, |
| LASGlobalData< LASCoordinateSet > & | measuredCoordinates, | ||
| LASGlobalData< LASCoordinateSet > & | nominalCoordinates | ||
| ) |
allows to push in a simple simulated misalignment for quick internal testing purposes; overwrites LASGlobalData<LASCoordinateSet>& measuredCoordinates; call at beginning of LASBarrelAlgorithm::CalculateParameters method
one line per module, format for TEC: det ring beam disk phi phiErr format for TEC(at) & TIBTOB: det beam z "-1" phi phiErr
Definition at line 651 of file LASAlignmentTubeAlgorithm.cc.
References benchmark_cfg::cerr, mergeVDriftHistosByStation::file, LASGlobalData< T >::GetTEC2TECEntry(), LASGlobalData< T >::GetTECEntry(), LASGlobalData< T >::GetTIBTOBEntry(), phi, pos, relativeConstraints::ring, LASCoordinateSet::SetPhi(), LASCoordinateSet::SetPhiError(), LASGlobalLoop::TEC2TECLoop(), LASGlobalLoop::TECLoop(), LASGlobalLoop::TIBTOBLoop(), and z.
{
std::ifstream file( filename );
if( file.bad() ) {
std::cerr << " [LASAlignmentTubeAlgorithm::ReadMisalignmentFromFile] ** ERROR: cannot open file \"" << filename << "\"." << std::endl;
return;
}
std::cerr << " @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@" << std::endl;
std::cerr << " @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@" << std::endl;
std::cerr << " [LASAlignmentTubeAlgorithm::ReadMisalignmentFromFile] ** WARNING: you are reading a fake measurement from a file!" << std::endl;
std::cerr << " @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@" << std::endl;
std::cerr << " @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@" << std::endl;
// the measured coordinates will finally be overwritten;
// first, set them to the nominal values
measuredCoordinates = nominalCoordinates;
// and put large errors on all values;
{
LASGlobalLoop moduleLoop;
int det, ring, beam, disk, pos;
det = 0; ring = 0; beam = 0; disk = 0;
do {
measuredCoordinates.GetTECEntry( det, ring, beam, disk ).SetPhiError( 1000. );
} while( moduleLoop.TECLoop( det, ring, beam, disk ) );
det = 2; beam = 0; pos = 0;
do {
measuredCoordinates.GetTIBTOBEntry( det, beam, pos ).SetPhiError( 1000. );
} while( moduleLoop.TIBTOBLoop( det, beam, pos ) );
det = 0; beam = 0; disk = 0;
do {
measuredCoordinates.GetTEC2TECEntry( det, beam, disk ).SetPhiError( 1000. );
} while( moduleLoop.TEC2TECLoop( det, beam, disk ) );
}
// buffers for read-in
int det, beam, z, ring;
double phi, phiError;
while( !file.eof() ) {
file >> det;
if( file.eof() ) break; // do not read the last line twice, do not fill trash if file empty
file >> beam; file >> z; file >> ring;
file >> phi; file >> phiError;
if( det > 1 ) { // TIB/TOB
measuredCoordinates.GetTIBTOBEntry( det, beam, z ).SetPhi( phi );
measuredCoordinates.GetTIBTOBEntry( det, beam, z ).SetPhiError( phiError );
} else { // TEC or TEC(at)
if( ring > -1 ) { // TEC
measuredCoordinates.GetTECEntry( det, ring, beam, z ).SetPhi( phi );
measuredCoordinates.GetTECEntry( det, ring, beam, z ).SetPhiError( phiError );
}
else { // TEC(at)
measuredCoordinates.GetTEC2TECEntry( det, beam, z ).SetPhi( phi );
measuredCoordinates.GetTEC2TECEntry( det, beam, z ).SetPhiError( phiError );
}
}
}
file.close();
}
1.7.3