CMS 3D CMS Logo

List of all members | Public Types | Public Member Functions | Private Attributes
HGCFEElectronics< DFr > Class Template Reference

models the behavior of the front-end electronics More...

#include <HGCFEElectronics.h>

Public Types

enum  HGCFEElectronicsFirmwareVersion { TRIVIAL, SIMPLE, WITHTOT }
 
enum  HGCFEElectronicsTOTMode { WEIGHTEDBYE, SIMPLETHRESHOLD }
 

Public Member Functions

float getADClsb ()
 returns the LSB currently configured More...
 
float getADCThreshold ()
 
hgc_digi::FEADCPulseShapegetDefaultADCPulse ()
 getter for the default ADC pulse configured by python More...
 
float getMaxADC ()
 
float getMaxTDC ()
 
int getTargetMipValue ()
 
std::array< float, 3 > getTDCForToAOnset ()
 
float getTDClsb ()
 
float getTDCOnset ()
 
float getTimeJitter (float totalCharge, int thickness)
 
 HGCFEElectronics (const edm::ParameterSet &ps)
 CTOR. More...
 
void runShaper (DFr &dataFrame, hgc::HGCSimHitData &chargeColl, hgc::HGCSimHitData &toa, const hgc_digi::FEADCPulseShape &adcPulse, CLHEP::HepRandomEngine *engine, uint32_t thrADC=0, float lsbADC=-1, uint32_t gainIdx=0, float maxADC=-1, int thickness=1, float tdcOnsetAuto=-1)
 switches according to the firmware version More...
 
void runShaper (DFr &dataFrame, hgc::HGCSimHitData &chargeColl, hgc::HGCSimHitData &toa, CLHEP::HepRandomEngine *engine, uint32_t thrADC=0, float lsbADC=-1, uint32_t gainIdx=0, float maxADC=-1, int thickness=1)
 
void runShaperWithToT (DFr &dataFrame, hgc::HGCSimHitData &chargeColl, hgc::HGCSimHitData &toa, CLHEP::HepRandomEngine *engine, uint32_t thrADC, float lsbADC, uint32_t gainIdx, float maxADC, int thickness, float tdcOnsetAuto, const hgc_digi::FEADCPulseShape &adcPulse)
 implements pulse shape and switch to time over threshold including deadtime More...
 
void runShaperWithToT (DFr &dataFrame, hgc::HGCSimHitData &chargeColl, hgc::HGCSimHitData &toa, CLHEP::HepRandomEngine *engine, uint32_t thrADC, float lsbADC, uint32_t gainIdx, float maxADC, int thickness, float tdcOnsetAuto)
 
void runSimpleShaper (DFr &dataFrame, hgc::HGCSimHitData &chargeColl, uint32_t thrADC, float lsbADC, uint32_t gainIdx, float maxADC, const hgc_digi::FEADCPulseShape &adcPulse)
 applies a shape to each time sample and propagates the tails to the subsequent time samples More...
 
void runSimpleShaper (DFr &dataFrame, hgc::HGCSimHitData &chargeColl, uint32_t thrADC, float lsbADC, uint32_t gainIdx, float maxADC)
 
void runTrivialShaper (DFr &dataFrame, hgc::HGCSimHitData &chargeColl, uint32_t thrADC, float lsbADC, uint32_t gainIdx, float maxADC)
 converts charge to digis without pulse shape More...
 
void setADClsb (float newLSB)
 
void SetNoiseValues (const std::vector< float > &noise_fC)
 
void setTDCfsc (float newTDCfsc)
 
uint32_t toaMode () const
 returns how ToT will be computed More...
 
 ~HGCFEElectronics ()
 DTOR. More...
 

Private Attributes

float adcLSB_fC_
 
hgc_digi::FEADCPulseShape adcPulse_
 
float adcSaturation_fC_
 
float adcThreshold_fC_
 
std::array< bool, hgc::nSamples > busyFlags
 
uint32_t fwVersion_
 
std::array< float, 3 > jitterConstant2_ns_
 
std::array< float, 3 > jitterNoise2_ns_
 
hgc::HGCSimHitData newCharge
 
std::vector< float > noise_fC_
 
hgc_digi::FEADCPulseShape pulseAvgT_
 
uint32_t targetMIPvalue_ADC_
 
std::vector< float > tdcChargeDrainParameterisation_
 
std::array< float, 3 > tdcForToAOnset_fC_
 
float tdcLSB_fC_
 
uint32_t tdcNbits_
 
float tdcOnset_fC_
 
float tdcResolutionInNs_
 
float tdcSaturation_fC_
 
bool thresholdFollowsMIP_
 
std::array< bool, hgc::nSamples > toaFlags
 
hgc::HGCSimHitData toaFromToT
 
float toaLSB_ns_
 
uint32_t toaMode_
 
std::array< bool, hgc::nSamples > totFlags
 

Detailed Description

template<class DFr>
class HGCFEElectronics< DFr >

models the behavior of the front-end electronics

Definition at line 24 of file HGCFEElectronics.h.

Member Enumeration Documentation

◆ HGCFEElectronicsFirmwareVersion

Enumerator
TRIVIAL 
SIMPLE 
WITHTOT 

Definition at line 26 of file HGCFEElectronics.h.

◆ HGCFEElectronicsTOTMode

template<class DFr >
enum HGCFEElectronics::HGCFEElectronicsTOTMode
Enumerator
WEIGHTEDBYE 
SIMPLETHRESHOLD 

Definition at line 27 of file HGCFEElectronics.h.

Constructor & Destructor Documentation

◆ HGCFEElectronics()

template<class DFr >
HGCFEElectronics< DFr >::HGCFEElectronics ( const edm::ParameterSet ps)

CTOR.

Definition at line 12 of file HGCFEElectronics.cc.

References edm::ParameterSet::getParameter().

13  : fwVersion_{ps.getParameter<uint32_t>("fwVersion")},
14  adcPulse_{},
15  pulseAvgT_{},
17  adcSaturation_fC_{-1.0},
18  adcLSB_fC_{},
19  tdcLSB_fC_{},
20  tdcSaturation_fC_{-1.0},
22  tdcOnset_fC_{},
23  toaLSB_ns_{},
24  tdcResolutionInNs_{1e-9}, // set time resolution very small by default
28  noise_fC_{},
30  edm::LogVerbatim("HGCFE") << "[HGCFEElectronics] running with version " << fwVersion_ << std::endl;
31  if (ps.exists("adcPulse")) {
32  auto temp = ps.getParameter<std::vector<double> >("adcPulse");
33  for (unsigned i = 0; i < temp.size(); ++i) {
34  adcPulse_[i] = (float)temp[i];
35  }
36  // normalize adc pulse
37  for (unsigned i = 0; i < adcPulse_.size(); ++i) {
38  adcPulse_[i] = adcPulse_[i] / adcPulse_[2];
39  }
40  temp = ps.getParameter<std::vector<double> >("pulseAvgT");
41  for (unsigned i = 0; i < temp.size(); ++i) {
42  pulseAvgT_[i] = (float)temp[i];
43  }
44  }
45  if (ps.exists("adcNbits")) {
46  uint32_t adcNbits = ps.getParameter<uint32_t>("adcNbits");
47  adcSaturation_fC_ = ps.getParameter<double>("adcSaturation_fC");
49  edm::LogVerbatim("HGCFE") << "[HGCFEElectronics] " << adcNbits << " bit ADC defined"
50  << " with LSB=" << adcLSB_fC_ << " saturation to occur @ " << adcSaturation_fC_
51  << std::endl;
52  }
53 
54  if (ps.exists("tdcNbits")) {
55  tdcNbits_ = ps.getParameter<uint32_t>("tdcNbits");
56  setTDCfsc(ps.getParameter<double>("tdcSaturation_fC"));
57  edm::LogVerbatim("HGCFE") << "[HGCFEElectronics] " << tdcNbits_ << " bit TDC defined with LSB=" << tdcLSB_fC_
58  << " saturation to occur @ " << tdcSaturation_fC_
59  << " (NB lowered by 1 part in a million)" << std::endl;
60  }
61  if (ps.exists("targetMIPvalue_ADC"))
62  targetMIPvalue_ADC_ = ps.getParameter<uint32_t>("targetMIPvalue_ADC");
63  if (ps.exists("adcThreshold_fC"))
64  adcThreshold_fC_ = ps.getParameter<double>("adcThreshold_fC");
65  if (ps.exists("tdcOnset_fC"))
66  tdcOnset_fC_ = ps.getParameter<double>("tdcOnset_fC");
67  if (ps.exists("tdcForToAOnset_fC")) {
68  auto temp = ps.getParameter<std::vector<double> >("tdcForToAOnset_fC");
69  if (temp.size() == tdcForToAOnset_fC_.size()) {
70  std::copy_n(temp.begin(), temp.size(), tdcForToAOnset_fC_.begin());
71  } else {
72  throw cms::Exception("BadConfiguration") << " HGCFEElectronics wrong size for ToA thresholds ";
73  }
74  }
75  if (ps.exists("toaLSB_ns"))
76  toaLSB_ns_ = ps.getParameter<double>("toaLSB_ns");
77  if (ps.exists("tdcChargeDrainParameterisation")) {
78  for (auto val : ps.getParameter<std::vector<double> >("tdcChargeDrainParameterisation")) {
79  tdcChargeDrainParameterisation_.push_back((float)val);
80  }
81  }
82  if (ps.exists("tdcResolutionInPs"))
83  tdcResolutionInNs_ = ps.getParameter<double>("tdcResolutionInPs") * 1e-3; // convert to ns
84  if (ps.exists("toaMode"))
85  toaMode_ = ps.getParameter<uint32_t>("toaMode");
86 
87  if (ps.exists("jitterNoise_ns")) {
88  auto temp = ps.getParameter<std::vector<double> >("jitterNoise_ns");
89  if (temp.size() == jitterNoise2_ns_.size()) {
90  std::copy_n(temp.begin(), temp.size(), jitterNoise2_ns_.begin());
91  } else {
92  throw cms::Exception("BadConfiguration") << " HGCFEElectronics wrong size for ToA jitterNoise ";
93  }
94  }
95  if (ps.exists("jitterConstant_ns")) {
96  auto temp = ps.getParameter<std::vector<double> >("jitterConstant_ns");
97  if (temp.size() == jitterConstant2_ns_.size()) {
98  std::copy_n(temp.begin(), temp.size(), jitterConstant2_ns_.begin());
99  } else {
100  throw cms::Exception("BadConfiguration") << " HGCFEElectronics wrong size for ToA jitterConstant ";
101  }
102  }
103 }
Log< level::Info, true > LogVerbatim
std::array< float, 3 > jitterConstant2_ns_
T getParameter(std::string const &) const
Definition: ParameterSet.h:307
bool exists(std::string const &parameterName) const
checks if a parameter exists
std::array< float, 3 > jitterNoise2_ns_
uint32_t targetMIPvalue_ADC_
hgc_digi::FEADCPulseShape adcPulse_
std::vector< float > noise_fC_
hgc_digi::FEADCPulseShape pulseAvgT_
void setTDCfsc(float newTDCfsc)
std::vector< float > tdcChargeDrainParameterisation_
Power< A, B >::type pow(const A &a, const B &b)
Definition: Power.h:29
std::array< float, 3 > tdcForToAOnset_fC_

◆ ~HGCFEElectronics()

template<class DFr >
HGCFEElectronics< DFr >::~HGCFEElectronics ( )
inline

DTOR.

Definition at line 170 of file HGCFEElectronics.h.

170 {}

Member Function Documentation

◆ getADClsb()

template<class DFr >
float HGCFEElectronics< DFr >::getADClsb ( )
inline

returns the LSB currently configured

Definition at line 91 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::adcLSB_fC_.

91 { return adcLSB_fC_; }

◆ getADCThreshold()

template<class DFr >
float HGCFEElectronics< DFr >::getADCThreshold ( )
inline

Definition at line 94 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::adcThreshold_fC_.

94 { return adcThreshold_fC_; }

◆ getDefaultADCPulse()

template<class DFr >
hgc_digi::FEADCPulseShape& HGCFEElectronics< DFr >::getDefaultADCPulse ( )
inline

getter for the default ADC pulse configured by python

Definition at line 165 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::adcPulse_.

165 { return adcPulse_; }
hgc_digi::FEADCPulseShape adcPulse_

◆ getMaxADC()

template<class DFr >
float HGCFEElectronics< DFr >::getMaxADC ( )
inline

Definition at line 95 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::adcSaturation_fC_.

95 { return adcSaturation_fC_; }

◆ getMaxTDC()

template<class DFr >
float HGCFEElectronics< DFr >::getMaxTDC ( )
inline

Definition at line 96 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::tdcSaturation_fC_.

96 { return tdcSaturation_fC_; }

◆ getTargetMipValue()

template<class DFr >
int HGCFEElectronics< DFr >::getTargetMipValue ( )
inline

Definition at line 93 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::targetMIPvalue_ADC_.

93 { return targetMIPvalue_ADC_; }
uint32_t targetMIPvalue_ADC_

◆ getTDCForToAOnset()

template<class DFr >
std::array<float, 3> HGCFEElectronics< DFr >::getTDCForToAOnset ( )
inline

Definition at line 98 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::tdcForToAOnset_fC_.

98 { return tdcForToAOnset_fC_; }
std::array< float, 3 > tdcForToAOnset_fC_

◆ getTDClsb()

template<class DFr >
float HGCFEElectronics< DFr >::getTDClsb ( )
inline

Definition at line 92 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::tdcLSB_fC_.

92 { return tdcLSB_fC_; }

◆ getTDCOnset()

template<class DFr >
float HGCFEElectronics< DFr >::getTDCOnset ( )
inline

Definition at line 97 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::tdcOnset_fC_.

97 { return tdcOnset_fC_; }

◆ getTimeJitter()

template<class DFr >
float HGCFEElectronics< DFr >::getTimeJitter ( float  totalCharge,
int  thickness 
)
inline

Definition at line 80 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::jitterConstant2_ns_, HGCFEElectronics< DFr >::jitterNoise2_ns_, HGCFEElectronics< DFr >::noise_fC_, funct::pow(), mathSSE::sqrt(), and Calorimetry_cff::thickness.

80  {
81  float A2 = jitterNoise2_ns_.at(thickness - 1);
82  float C2 = jitterConstant2_ns_.at(thickness - 1);
83  float X2 = pow((totalCharge / noise_fC_.at(thickness - 1)), 2.);
84  float jitter2 = A2 / X2 + C2;
85  return sqrt(jitter2);
86  };
std::array< float, 3 > jitterConstant2_ns_
std::array< float, 3 > jitterNoise2_ns_
T sqrt(T t)
Definition: SSEVec.h:19
std::vector< float > noise_fC_
Power< A, B >::type pow(const A &a, const B &b)
Definition: Power.h:29

◆ runShaper() [1/2]

template<class DFr >
void HGCFEElectronics< DFr >::runShaper ( DFr &  dataFrame,
hgc::HGCSimHitData chargeColl,
hgc::HGCSimHitData toa,
const hgc_digi::FEADCPulseShape adcPulse,
CLHEP::HepRandomEngine *  engine,
uint32_t  thrADC = 0,
float  lsbADC = -1,
uint32_t  gainIdx = 0,
float  maxADC = -1,
int  thickness = 1,
float  tdcOnsetAuto = -1 
)
inline

switches according to the firmware version

Definition at line 37 of file HGCFEElectronics.h.

References hgceeDigitizer_cfi::adcPulse, HGCFEElectronics< DFr >::fwVersion_, HGCFEElectronics< DFr >::runShaperWithToT(), HGCFEElectronics< DFr >::runSimpleShaper(), HGCFEElectronics< DFr >::runTrivialShaper(), HGCFEElectronics< DFr >::SIMPLE, Calorimetry_cff::thickness, and HGCFEElectronics< DFr >::WITHTOT.

Referenced by HGCFEElectronics< DFr >::runShaper().

47  {
48  switch (fwVersion_) {
49  case SIMPLE: {
50  runSimpleShaper(dataFrame, chargeColl, thrADC, lsbADC, gainIdx, maxADC, adcPulse);
51  break;
52  }
53  case WITHTOT: {
55  dataFrame, chargeColl, toa, engine, thrADC, lsbADC, gainIdx, maxADC, thickness, tdcOnsetAuto, adcPulse);
56  break;
57  }
58  default: {
59  runTrivialShaper(dataFrame, chargeColl, thrADC, lsbADC, gainIdx, maxADC);
60  break;
61  }
62  }
63  }
void runShaperWithToT(DFr &dataFrame, hgc::HGCSimHitData &chargeColl, hgc::HGCSimHitData &toa, CLHEP::HepRandomEngine *engine, uint32_t thrADC, float lsbADC, uint32_t gainIdx, float maxADC, int thickness, float tdcOnsetAuto, const hgc_digi::FEADCPulseShape &adcPulse)
implements pulse shape and switch to time over threshold including deadtime
void runSimpleShaper(DFr &dataFrame, hgc::HGCSimHitData &chargeColl, uint32_t thrADC, float lsbADC, uint32_t gainIdx, float maxADC, const hgc_digi::FEADCPulseShape &adcPulse)
applies a shape to each time sample and propagates the tails to the subsequent time samples ...
void runTrivialShaper(DFr &dataFrame, hgc::HGCSimHitData &chargeColl, uint32_t thrADC, float lsbADC, uint32_t gainIdx, float maxADC)
converts charge to digis without pulse shape

◆ runShaper() [2/2]

template<class DFr >
void HGCFEElectronics< DFr >::runShaper ( DFr &  dataFrame,
hgc::HGCSimHitData chargeColl,
hgc::HGCSimHitData toa,
CLHEP::HepRandomEngine *  engine,
uint32_t  thrADC = 0,
float  lsbADC = -1,
uint32_t  gainIdx = 0,
float  maxADC = -1,
int  thickness = 1 
)
inline

Definition at line 64 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::adcPulse_, HGCFEElectronics< DFr >::runShaper(), and Calorimetry_cff::thickness.

72  {
73  runShaper(dataFrame, chargeColl, toa, adcPulse_, engine, thrADC, lsbADC, gainIdx, maxADC, thickness);
74  }
void runShaper(DFr &dataFrame, hgc::HGCSimHitData &chargeColl, hgc::HGCSimHitData &toa, const hgc_digi::FEADCPulseShape &adcPulse, CLHEP::HepRandomEngine *engine, uint32_t thrADC=0, float lsbADC=-1, uint32_t gainIdx=0, float maxADC=-1, int thickness=1, float tdcOnsetAuto=-1)
switches according to the firmware version
hgc_digi::FEADCPulseShape adcPulse_

◆ runShaperWithToT() [1/2]

template<class DFr >
void HGCFEElectronics< DFr >::runShaperWithToT ( DFr &  dataFrame,
hgc::HGCSimHitData chargeColl,
hgc::HGCSimHitData toa,
CLHEP::HepRandomEngine *  engine,
uint32_t  thrADC,
float  lsbADC,
uint32_t  gainIdx,
float  maxADC,
int  thickness,
float  tdcOnsetAuto,
const hgc_digi::FEADCPulseShape adcPulse 
)

implements pulse shape and switch to time over threshold including deadtime

Definition at line 204 of file HGCFEElectronics.cc.

References gpuClustering::adc, hgceeDigitizer_cfi::adcPulse, ALCARECOTkAlJpsiMuMu_cff::charge, debug, MillePedeFileConverter_cfg::e, f, myMath::fast_expf(), createfilelist::int, SiStripPI::max, SiStripPI::min, mps_check::msg, HGCSample::set(), HGCSample::setToAValid(), HGCalUncalibRecHit_cfi::tdcOnset, and Calorimetry_cff::thickness.

Referenced by HGCFEElectronics< DFr >::runShaper(), and HGCFEElectronics< DFr >::runShaperWithToT().

214  {
215  busyFlags.fill(false);
216  totFlags.fill(false);
217  toaFlags.fill(false);
218  newCharge.fill(0.f);
219  toaFromToT.fill(0.f);
220 
221 #ifdef EDM_ML_DEBUG
222  constexpr bool debug_state(true);
223 #else
224  constexpr bool debug_state(false);
225 #endif
226 
227  bool debug = debug_state;
228 
229  float timeToA = 0.f;
230 
231  //configure the ADC <-> TDC transition depending on the value passed as argument
232  double tdcOnset(maxADC);
233  if (maxADC < 0) {
234  maxADC = adcSaturation_fC_;
236  }
237 
238  //configure the ADC LSB depending on the value passed as argument
239  if (lsbADC < 0)
240  lsbADC = adcLSB_fC_;
241 
242  //first look at time
243  //for pileup look only at intime signals
244  //ToA is in central BX if fired -- std::floor(BX/25.)+9;
245  int fireBX = 9;
246  //noise fluctuation on charge is added after ToA computation
247  //do not recheck the ToA firing threshold tdcForToAOnset_fC_[thickness-1] not to bias the efficiency
248  //to be done properly with realistic ToA shaper and jitter for the moment accounted in the smearing
249  if (toaColl[fireBX] != 0.f) {
250  timeToA = toaColl[fireBX];
251  float jitter = getTimeJitter(chargeColl[fireBX], thickness);
252  if (jitter != 0)
253  timeToA = CLHEP::RandGaussQ::shoot(engine, timeToA, jitter);
254  else if (tdcResolutionInNs_ != 0)
255  timeToA = CLHEP::RandGaussQ::shoot(engine, timeToA, tdcResolutionInNs_);
256  if (timeToA >= 0.f && timeToA <= 25.f)
257  toaFlags[fireBX] = true;
258  }
259 
260  //now look at charge
261  //first identify bunches which will trigger ToT
262  //if(debug_state) edm::LogVerbatim("HGCFE") << "[runShaperWithToT]" << std::endl;
263  for (int it = 0; it < (int)(chargeColl.size()); ++it) {
264  debug = debug_state;
265  //if already flagged as busy it can't be re-used to trigger the ToT
266  if (busyFlags[it])
267  continue;
268 
269  if (tdcOnsetAuto < 0) {
270  tdcOnsetAuto = tdcOnset_fC_;
271  }
272  //if below TDC onset will be handled by SARS ADC later
273  float charge = chargeColl[it];
274  if (charge < tdcOnset) {
275  debug = false;
276  continue;
277  }
278 
279  //raise TDC mode for charge computation
280  //ToA anyway fired independently will be sorted out with realistic ToA dedicated shaper
281  float toa = timeToA;
282  totFlags[it] = true;
283 
284  if (debug)
285  edm::LogVerbatim("HGCFE") << "\t q=" << charge << " fC with <toa>=" << toa << " ns, triggers ToT @ " << it
286  << std::endl;
287 
288  //compute total charge to be integrated and integration time
289  //needs a loop as ToT will last as long as there is charge to dissipate
290  int busyBxs(0);
291  float totalCharge(charge), finalToA(toa), integTime(0);
292  while (true) {
293  //compute integration time in ns and # bunches
294  //float newIntegTime(0);
295  int poffset = 0;
296  float charge_offset = 0.f;
297  const float charge_kfC(totalCharge * 1e-3);
298  if (charge_kfC < tdcChargeDrainParameterisation_[3]) {
299  //newIntegTime=tdcChargeDrainParameterisation_[0]*pow(charge_kfC,2)+tdcChargeDrainParameterisation_[1]*charge_kfC+tdcChargeDrainParameterisation_[2];
300  } else if (charge_kfC < tdcChargeDrainParameterisation_[7]) {
301  poffset = 4;
302  charge_offset = tdcChargeDrainParameterisation_[3];
303  //newIntegTime=tdcChargeDrainParameterisation_[4]*pow(charge_kfC-tdcChargeDrainParameterisation_[3],2)+tdcChargeDrainParameterisation_[5]*(charge_kfC-tdcChargeDrainParameterisation_[3])+tdcChargeDrainParameterisation_[6];
304  } else {
305  poffset = 8;
306  charge_offset = tdcChargeDrainParameterisation_[7];
307  //newIntegTime=tdcChargeDrainParameterisation_[8]*pow(charge_kfC-tdcChargeDrainParameterisation_[7],2)+tdcChargeDrainParameterisation_[9]*(charge_kfC-tdcChargeDrainParameterisation_[7])+tdcChargeDrainParameterisation_[10];
308  }
309  const float charge_mod = charge_kfC - charge_offset;
310  const float newIntegTime =
311  ((tdcChargeDrainParameterisation_[poffset] * charge_mod + tdcChargeDrainParameterisation_[poffset + 1]) *
312  charge_mod +
313  tdcChargeDrainParameterisation_[poffset + 2]);
314 
315  const int newBusyBxs = std::floor(newIntegTime / 25.f) + 1;
316 
317  //if no update is needed regarding the number of bunches,
318  //then the ToT integration time has converged
319  integTime = newIntegTime;
320  if (newBusyBxs == busyBxs)
321  break;
322 
323  //update charge integrated during ToT
324  if (debug) {
325  if (busyBxs == 0)
326  edm::LogVerbatim("HGCFE") << "\t Intial busy estimate=" << integTime << " ns = " << newBusyBxs << " bxs"
327  << std::endl;
328  else
329  edm::LogVerbatim("HGCFE") << "\t ...integrated charge overflows initial busy estimate, interating again"
330  << std::endl;
331  }
332 
333  //update number of busy bunches
334  busyBxs = newBusyBxs;
335 
336  //reset charge to be integrated
337  totalCharge = charge;
338  if (toaMode_ == WEIGHTEDBYE)
339  finalToA = toa * charge;
340 
341  //add leakage from previous bunches in SARS ADC mode
342  for (int jt = 0; jt < it; ++jt) {
343  const unsigned int deltaT = (it - jt);
344  if ((deltaT + 2) >= adcPulse.size() || chargeColl[jt] == 0.f || totFlags[jt] || busyFlags[jt])
345  continue;
346 
347  const float leakCharge = chargeColl[jt] * adcPulse[deltaT + 2];
348  totalCharge += leakCharge;
349  if (toaMode_ == WEIGHTEDBYE)
350  finalToA += leakCharge * pulseAvgT_[deltaT + 2];
351 
352  if (debug)
353  edm::LogVerbatim("HGCFE") << "\t\t leaking " << chargeColl[jt] << " fC @ deltaT=-" << deltaT << " -> +"
354  << leakCharge << " with avgT=" << pulseAvgT_[deltaT + 2] << std::endl;
355  }
356 
357  //add contamination from posterior bunches
358  for (int jt = it + 1; jt < it + busyBxs && jt < dataFrame.size(); ++jt) {
359  //this charge will be integrated in TDC mode
360  //disable for SARS ADC
361  busyFlags[jt] = true;
362 
363  const float extraCharge = chargeColl[jt];
364  if (extraCharge == 0.f)
365  continue;
366  if (debug)
367  edm::LogVerbatim("HGCFE") << "\t\t adding " << extraCharge << " fC @ deltaT=+" << (jt - it) << std::endl;
368 
369  totalCharge += extraCharge;
370  if (toaMode_ == WEIGHTEDBYE)
371  finalToA += extraCharge * toaColl[jt];
372  }
373 
374  //finalize ToA contamination
375  if (toaMode_ == WEIGHTEDBYE)
376  finalToA /= totalCharge;
377  }
378 
379  newCharge[it] = (totalCharge - tdcOnset);
380 
381  if (debug)
382  edm::LogVerbatim("HGCFE") << "\t Final busy estimate=" << integTime << " ns = " << busyBxs << " bxs" << std::endl
383  << "\t Total integrated=" << totalCharge << " fC <toa>=" << toaFromToT[it]
384  << " (raw=" << finalToA << ") ns " << std::endl;
385 
386  //last fC (tdcOnset) are dissipated trough pulse
387  if (it + busyBxs < (int)(newCharge.size())) {
388  const float deltaT2nextBx((busyBxs * 25 - integTime));
389  const float tdcOnsetLeakage(tdcOnset * vdt::fast_expf(-deltaT2nextBx / tdcChargeDrainParameterisation_[11]));
390  if (debug)
391  edm::LogVerbatim("HGCFE") << "\t Leaking remainder of TDC onset " << tdcOnset << " fC, to be dissipated in "
392  << deltaT2nextBx << " DeltaT/tau=" << deltaT2nextBx << " / "
393  << tdcChargeDrainParameterisation_[11] << " ns, adds " << tdcOnsetLeakage << " fC @ "
394  << it + busyBxs << " bx (first free bx)" << std::endl;
395  newCharge[it + busyBxs] += tdcOnsetLeakage;
396  }
397  }
398 
399  //including the leakage from bunches in SARS ADC when not declared busy or in ToT
400  auto runChargeSharing = [&]() {
401  int ipulse = 0;
402  for (int it = 0; it < (int)(chargeColl.size()); ++it) {
403  //if busy, charge has been already integrated
404  //if(debug) edm::LogVerbatim("HGCFE") << "\t SARS ADC pulse activated @ " << it << " : ";
405  if (!totFlags[it] & !busyFlags[it]) {
406  const int start = std::max(0, 2 - it);
407  const int stop = std::min((int)adcPulse.size(), (int)newCharge.size() - it + 2);
408  for (ipulse = start; ipulse < stop; ++ipulse) {
409  const int itoffset = it + ipulse - 2;
410  //notice that if the channel is already busy,
411  //it has already been affected by the leakage of the SARS ADC
412  //if(totFlags[itoffset] || busyFlags[itoffset]) continue;
413  if (!totFlags[itoffset] & !busyFlags[itoffset]) {
414  newCharge[itoffset] += chargeColl[it] * adcPulse[ipulse];
415  }
416  //if(debug) edm::LogVerbatim("HGCFE") << " | " << itoffset << " " << chargeColl[it]*adcPulse[ipulse] << "( " << chargeColl[it] << "->";
417  //if(debug) edm::LogVerbatim("HGCFE") << newCharge[itoffset] << ") ";
418  }
419  }
420 
421  if (debug)
422  edm::LogVerbatim("HGCFE") << std::endl;
423  }
424  };
425  runChargeSharing();
426 
427  //For the future need to understand how to deal with toa for out of time signals
428  //and for that should keep track of the BX firing the ToA somewhere (also to restore the use of finalToA)
429  /*
430  float finalToA(0.);
431  for(int it=0; it<(int)(newCharge.size()); it++){
432  if(toaFlags[it]){
433  finalToA = toaFromToT[it];
434  //to avoid +=25 for small negative time taken as 0
435  while(finalToA < -1.e-5) finalToA+=25.f;
436  while(finalToA > 25.f) finalToA-=25.f;
437  toaFromToT[it] = finalToA;
438  }
439  }
440  */
441  //timeToA is already in 0-25ns range by construction
442 
443  //set new ADCs and ToA
444  if (debug)
445  edm::LogVerbatim("HGCFE") << "\t final result : ";
446  for (int it = 0; it < (int)(newCharge.size()); it++) {
447  if (debug)
448  edm::LogVerbatim("HGCFE") << chargeColl[it] << " -> " << newCharge[it] << " ";
449 
450  HGCSample newSample;
451  if (totFlags[it] || busyFlags[it]) {
452  if (totFlags[it]) {
453  //brute force saturation, maybe could to better with an exponential like saturation
454  const float saturatedCharge(std::min(newCharge[it], tdcSaturation_fC_));
455  //working version for in-time PU and signal
456  newSample.set(
457  true, true, gainIdx, (uint16_t)(timeToA / toaLSB_ns_), (uint16_t)(std::floor(saturatedCharge / tdcLSB_fC_)));
458  if (toaFlags[it])
459  newSample.setToAValid(true);
460  } else {
461  newSample.set(false, true, gainIdx, 0, 0);
462  }
463  } else {
464  //brute force saturation, maybe could to better with an exponential like saturation
465  const uint16_t adc = std::floor(std::min(newCharge[it], maxADC) / lsbADC);
466  //working version for in-time PU and signal
467  newSample.set(adc > thrADC, false, gainIdx, (uint16_t)(timeToA / toaLSB_ns_), adc);
468  if (toaFlags[it])
469  newSample.setToAValid(true);
470  }
471  dataFrame.setSample(it, newSample);
472  }
473 
474  if (debug) {
475  std::ostringstream msg;
476  dataFrame.print(msg);
477  edm::LogVerbatim("HGCFE") << msg.str() << std::endl;
478  }
479 }
Definition: start.py:1
std::array< bool, hgc::nSamples > busyFlags
Log< level::Info, true > LogVerbatim
hgc::HGCSimHitData toaFromToT
wrapper for a data word
Definition: HGCSample.h:13
float getTimeJitter(float totalCharge, int thickness)
double f[11][100]
void setToAValid(bool toaFired)
Definition: HGCSample.h:47
#define debug
Definition: HDRShower.cc:19
std::array< bool, hgc::nSamples > totFlags
tuple msg
Definition: mps_check.py:285
hgc_digi::FEADCPulseShape pulseAvgT_
float fast_expf(float x)
std::vector< float > tdcChargeDrainParameterisation_
void set(bool thr, bool mode, uint16_t gain, uint16_t toa, uint16_t data)
Definition: HGCSample.h:49
hgc::HGCSimHitData newCharge
uint16_t *__restrict__ uint16_t const *__restrict__ adc
std::array< bool, hgc::nSamples > toaFlags

◆ runShaperWithToT() [2/2]

template<class DFr >
void HGCFEElectronics< DFr >::runShaperWithToT ( DFr &  dataFrame,
hgc::HGCSimHitData chargeColl,
hgc::HGCSimHitData toa,
CLHEP::HepRandomEngine *  engine,
uint32_t  thrADC,
float  lsbADC,
uint32_t  gainIdx,
float  maxADC,
int  thickness,
float  tdcOnsetAuto 
)
inline

Definition at line 143 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::adcPulse_, HGCFEElectronics< DFr >::runShaperWithToT(), and Calorimetry_cff::thickness.

152  {
154  dataFrame, chargeColl, toa, engine, thrADC, lsbADC, gainIdx, maxADC, thickness, tdcOnsetAuto, adcPulse_);
155  }
void runShaperWithToT(DFr &dataFrame, hgc::HGCSimHitData &chargeColl, hgc::HGCSimHitData &toa, CLHEP::HepRandomEngine *engine, uint32_t thrADC, float lsbADC, uint32_t gainIdx, float maxADC, int thickness, float tdcOnsetAuto, const hgc_digi::FEADCPulseShape &adcPulse)
implements pulse shape and switch to time over threshold including deadtime
hgc_digi::FEADCPulseShape adcPulse_

◆ runSimpleShaper() [1/2]

template<class DFr >
void HGCFEElectronics< DFr >::runSimpleShaper ( DFr &  dataFrame,
hgc::HGCSimHitData chargeColl,
uint32_t  thrADC,
float  lsbADC,
uint32_t  gainIdx,
float  maxADC,
const hgc_digi::FEADCPulseShape adcPulse 
)

applies a shape to each time sample and propagates the tails to the subsequent time samples

Definition at line 146 of file HGCFEElectronics.cc.

References gpuClustering::adc, hgceeDigitizer_cfi::adcPulse, ALCARECOTkAlJpsiMuMu_cff::charge, debug, f, createfilelist::int, SiStripPI::min, mps_check::msg, and HGCSample::set().

Referenced by HGCFEElectronics< DFr >::runShaper(), and HGCFEElectronics< DFr >::runSimpleShaper().

152  {
153  //convolute with pulse shape to compute new ADCs
154  newCharge.fill(0.f);
155  bool debug(false);
156  for (int it = 0; it < (int)(chargeColl.size()); it++) {
157  const float charge(chargeColl[it]);
158  if (charge == 0.f)
159  continue;
160 
161 #ifdef EDM_ML_DEBUG
163 #endif
164 
165  if (debug)
166  edm::LogVerbatim("HGCFE") << "\t Redistributing SARS ADC" << charge << " @ " << it;
167 
168  for (int ipulse = -2; ipulse < (int)(adcPulse.size()) - 2; ipulse++) {
169  if (it + ipulse < 0)
170  continue;
171  if (it + ipulse >= (int)(dataFrame.size()))
172  continue;
173  const float chargeLeak = charge * adcPulse[(ipulse + 2)];
174  newCharge[it + ipulse] += chargeLeak;
175 
176  if (debug)
177  edm::LogVerbatim("HGCFE") << " | " << it + ipulse << " " << chargeLeak;
178  }
179 
180  if (debug)
181  edm::LogVerbatim("HGCFE") << std::endl;
182  }
183 
184  for (int it = 0; it < (int)(newCharge.size()); it++) {
185  //brute force saturation, maybe could to better with an exponential like saturation
186  const uint32_t adc = std::floor(std::min(newCharge[it], maxADC) / lsbADC);
187  HGCSample newSample;
188  newSample.set(adc > thrADC, false, gainIdx, 0, adc);
189  dataFrame.setSample(it, newSample);
190 
191  if (debug)
192  edm::LogVerbatim("HGCFE") << adc << " (" << std::min(newCharge[it], maxADC) << "/" << lsbADC << " ) ";
193  }
194 
195  if (debug) {
196  std::ostringstream msg;
197  dataFrame.print(msg);
198  edm::LogVerbatim("HGCFE") << msg.str() << std::endl;
199  }
200 }
Log< level::Info, true > LogVerbatim
wrapper for a data word
Definition: HGCSample.h:13
double f[11][100]
#define debug
Definition: HDRShower.cc:19
tuple msg
Definition: mps_check.py:285
void set(bool thr, bool mode, uint16_t gain, uint16_t toa, uint16_t data)
Definition: HGCSample.h:49
hgc::HGCSimHitData newCharge
uint16_t *__restrict__ uint16_t const *__restrict__ adc

◆ runSimpleShaper() [2/2]

template<class DFr >
void HGCFEElectronics< DFr >::runSimpleShaper ( DFr &  dataFrame,
hgc::HGCSimHitData chargeColl,
uint32_t  thrADC,
float  lsbADC,
uint32_t  gainIdx,
float  maxADC 
)
inline

Definition at line 124 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::adcPulse_, and HGCFEElectronics< DFr >::runSimpleShaper().

125  {
126  runSimpleShaper(dataFrame, chargeColl, thrADC, lsbADC, gainIdx, maxADC, adcPulse_);
127  }
hgc_digi::FEADCPulseShape adcPulse_
void runSimpleShaper(DFr &dataFrame, hgc::HGCSimHitData &chargeColl, uint32_t thrADC, float lsbADC, uint32_t gainIdx, float maxADC, const hgc_digi::FEADCPulseShape &adcPulse)
applies a shape to each time sample and propagates the tails to the subsequent time samples ...

◆ runTrivialShaper()

template<class DFr >
void HGCFEElectronics< DFr >::runTrivialShaper ( DFr &  dataFrame,
hgc::HGCSimHitData chargeColl,
uint32_t  thrADC,
float  lsbADC,
uint32_t  gainIdx,
float  maxADC 
)

converts charge to digis without pulse shape

Definition at line 107 of file HGCFEElectronics.cc.

References gpuClustering::adc, debug, MillePedeFileConverter_cfg::e, createfilelist::int, SiStripPI::min, mps_check::msg, and HGCSample::set().

Referenced by HGCFEElectronics< DFr >::runShaper().

108  {
109  bool debug(false);
110 
111 #ifdef EDM_ML_DEBUG
112  for (int it = 0; it < (int)(chargeColl.size()); it++)
113  debug |= (chargeColl[it] > adcThreshold_fC_);
114 #endif
115 
116  if (debug)
117  edm::LogVerbatim("HGCFE") << "[runTrivialShaper]" << std::endl;
118 
119  if (lsbADC < 0)
120  lsbADC = adcLSB_fC_;
121  if (maxADC < 0)
122  // lower adcSaturation_fC_ by one part in a million
123  // to ensure largest charge converted in bits is 0xfff==4095, not 0x1000
124  // no effect on charges loewer than; no impact on cpu time, only done once
125  maxADC = adcSaturation_fC_ * (1 - 1e-6);
126  for (int it = 0; it < (int)(chargeColl.size()); it++) {
127  //brute force saturation, maybe could to better with an exponential like saturation
128  const uint32_t adc = std::floor(std::min(chargeColl[it], maxADC) / lsbADC);
129  HGCSample newSample;
130  newSample.set(adc > thrADC, false, gainIdx, 0, adc);
131  dataFrame.setSample(it, newSample);
132 
133  if (debug)
134  edm::LogVerbatim("HGCFE") << adc << " (" << chargeColl[it] << "/" << adcLSB_fC_ << ") ";
135  }
136 
137  if (debug) {
138  std::ostringstream msg;
139  dataFrame.print(msg);
140  edm::LogVerbatim("HGCFE") << msg.str() << std::endl;
141  }
142 }
Log< level::Info, true > LogVerbatim
wrapper for a data word
Definition: HGCSample.h:13
#define debug
Definition: HDRShower.cc:19
tuple msg
Definition: mps_check.py:285
void set(bool thr, bool mode, uint16_t gain, uint16_t toa, uint16_t data)
Definition: HGCSample.h:49
uint16_t *__restrict__ uint16_t const *__restrict__ adc

◆ setADClsb()

template<class DFr >
void HGCFEElectronics< DFr >::setADClsb ( float  newLSB)
inline

Definition at line 99 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::adcLSB_fC_.

99 { adcLSB_fC_ = newLSB; }

◆ SetNoiseValues()

template<class DFr >
void HGCFEElectronics< DFr >::SetNoiseValues ( const std::vector< float > &  noise_fC)
inline

Definition at line 76 of file HGCFEElectronics.h.

References hgceeDigitizer_cfi::noise_fC, and HGCFEElectronics< DFr >::noise_fC_.

76  {
77  noise_fC_.insert(noise_fC_.end(), noise_fC.begin(), noise_fC.end());
78  };
std::vector< float > noise_fC_

◆ setTDCfsc()

template<class DFr >
void HGCFEElectronics< DFr >::setTDCfsc ( float  newTDCfsc)
inline

Definition at line 100 of file HGCFEElectronics.h.

References MillePedeFileConverter_cfg::e, funct::pow(), HGCFEElectronics< DFr >::tdcLSB_fC_, HGCFEElectronics< DFr >::tdcNbits_, and HGCFEElectronics< DFr >::tdcSaturation_fC_.

100  {
101  tdcSaturation_fC_ = newTDCfsc;
103  // lower tdcSaturation_fC_ by one part in a million
104  // to ensure largest charge converted in bits is 0xfff and not 0x000
105  tdcSaturation_fC_ *= (1. - 1e-6);
106  }
Power< A, B >::type pow(const A &a, const B &b)
Definition: Power.h:29

◆ toaMode()

template<class DFr >
uint32_t HGCFEElectronics< DFr >::toaMode ( ) const
inline

returns how ToT will be computed

Definition at line 160 of file HGCFEElectronics.h.

References HGCFEElectronics< DFr >::toaMode_.

160 { return toaMode_; }

Member Data Documentation

◆ adcLSB_fC_

template<class DFr >
float HGCFEElectronics< DFr >::adcLSB_fC_
private

◆ adcPulse_

template<class DFr >
hgc_digi::FEADCPulseShape HGCFEElectronics< DFr >::adcPulse_
private

◆ adcSaturation_fC_

template<class DFr >
float HGCFEElectronics< DFr >::adcSaturation_fC_
private

Definition at line 178 of file HGCFEElectronics.h.

Referenced by HGCFEElectronics< DFr >::getMaxADC().

◆ adcThreshold_fC_

template<class DFr >
float HGCFEElectronics< DFr >::adcThreshold_fC_
private

Definition at line 178 of file HGCFEElectronics.h.

Referenced by HGCFEElectronics< DFr >::getADCThreshold().

◆ busyFlags

template<class DFr >
std::array<bool, hgc::nSamples> HGCFEElectronics< DFr >::busyFlags
private

Definition at line 187 of file HGCFEElectronics.h.

◆ fwVersion_

template<class DFr >
uint32_t HGCFEElectronics< DFr >::fwVersion_
private

Definition at line 174 of file HGCFEElectronics.h.

Referenced by HGCFEElectronics< DFr >::runShaper().

◆ jitterConstant2_ns_

template<class DFr >
std::array<float, 3> HGCFEElectronics< DFr >::jitterConstant2_ns_
private

Definition at line 181 of file HGCFEElectronics.h.

Referenced by HGCFEElectronics< DFr >::getTimeJitter().

◆ jitterNoise2_ns_

template<class DFr >
std::array<float, 3> HGCFEElectronics< DFr >::jitterNoise2_ns_
private

Definition at line 181 of file HGCFEElectronics.h.

Referenced by HGCFEElectronics< DFr >::getTimeJitter().

◆ newCharge

template<class DFr >
hgc::HGCSimHitData HGCFEElectronics< DFr >::newCharge
private

Definition at line 188 of file HGCFEElectronics.h.

◆ noise_fC_

template<class DFr >
std::vector<float> HGCFEElectronics< DFr >::noise_fC_
private

◆ pulseAvgT_

template<class DFr >
hgc_digi::FEADCPulseShape HGCFEElectronics< DFr >::pulseAvgT_
private

Definition at line 175 of file HGCFEElectronics.h.

◆ targetMIPvalue_ADC_

template<class DFr >
uint32_t HGCFEElectronics< DFr >::targetMIPvalue_ADC_
private

Definition at line 180 of file HGCFEElectronics.h.

Referenced by HGCFEElectronics< DFr >::getTargetMipValue().

◆ tdcChargeDrainParameterisation_

template<class DFr >
std::vector<float> HGCFEElectronics< DFr >::tdcChargeDrainParameterisation_
private

Definition at line 177 of file HGCFEElectronics.h.

◆ tdcForToAOnset_fC_

template<class DFr >
std::array<float, 3> HGCFEElectronics< DFr >::tdcForToAOnset_fC_
private

Definition at line 176 of file HGCFEElectronics.h.

Referenced by HGCFEElectronics< DFr >::getTDCForToAOnset().

◆ tdcLSB_fC_

template<class DFr >
float HGCFEElectronics< DFr >::tdcLSB_fC_
private

◆ tdcNbits_

template<class DFr >
uint32_t HGCFEElectronics< DFr >::tdcNbits_
private

Definition at line 184 of file HGCFEElectronics.h.

Referenced by HGCFEElectronics< DFr >::setTDCfsc().

◆ tdcOnset_fC_

template<class DFr >
float HGCFEElectronics< DFr >::tdcOnset_fC_
private

Definition at line 178 of file HGCFEElectronics.h.

Referenced by HGCFEElectronics< DFr >::getTDCOnset().

◆ tdcResolutionInNs_

template<class DFr >
float HGCFEElectronics< DFr >::tdcResolutionInNs_
private

Definition at line 178 of file HGCFEElectronics.h.

◆ tdcSaturation_fC_

template<class DFr >
float HGCFEElectronics< DFr >::tdcSaturation_fC_
private

◆ thresholdFollowsMIP_

template<class DFr >
bool HGCFEElectronics< DFr >::thresholdFollowsMIP_
private

Definition at line 185 of file HGCFEElectronics.h.

◆ toaFlags

template<class DFr >
std::array<bool, hgc::nSamples> HGCFEElectronics< DFr >::toaFlags
private

Definition at line 187 of file HGCFEElectronics.h.

◆ toaFromToT

template<class DFr >
hgc::HGCSimHitData HGCFEElectronics< DFr >::toaFromToT
private

Definition at line 188 of file HGCFEElectronics.h.

◆ toaLSB_ns_

template<class DFr >
float HGCFEElectronics< DFr >::toaLSB_ns_
private

Definition at line 178 of file HGCFEElectronics.h.

◆ toaMode_

template<class DFr >
uint32_t HGCFEElectronics< DFr >::toaMode_
private

Definition at line 183 of file HGCFEElectronics.h.

Referenced by HGCFEElectronics< DFr >::toaMode().

◆ totFlags

template<class DFr >
std::array<bool, hgc::nSamples> HGCFEElectronics< DFr >::totFlags
private

Definition at line 187 of file HGCFEElectronics.h.