Conversions in the Tracker... 7/8/06

photon gun with Geant4 simulation

The well understood FED ... 7/8/06

Timing Gain Scan Noise

Understanding the FED ... 17/7/06

ADC counts vs FED raw word for FE4, fiber5, for a timing run (20179). I'd like to reconstruct a tick from this ...

List of Tracker questions

• Calibration Questions

  .
  • The synchronization procedures described in CMS IN 2003-021 & IN 2004-019 state that the optimal FED delays for sampling analog APV pulses are determined by varying the opto-hybrid PLL delay 1 ns steps.
    
    • The reconstructed APV ticks at the FED appear at ~250 ns. Is the FED sampling triggered externally for this study (ie: it doesn't lock to The APV frames as it would in real running)? Is the cable delay between the FEC and CCU insignificant here? Otherwise I'd expect an additional 2-300 ns latency.
    • The APV ticks are ~50ns wide and are sampled at 40 MHz at the FED. This leads to an "image" of the tick that consists of 3 samples? By changing the PLL delay one samples 3 other positions of the tick pulse shape? Splice all of these snapshots together to get the picture above from that below?
    • Assuming we've imaged the tick and locate the optimal sampling positions, how does this translate to a FED delay. Haven't we been playing with the opto-hybrid PLL delay in imaging the ticks? Do we turn this into a timing advance for sampling on the FED? Why don't we "image" using the FED fine-delay instead?
      
    • Ticks are full current range output. Is the amplifier's baseline resoration timing independent of pulse height? If not, how can we use ticks for calibration?
      
  
  
  

• Other Questions.

  • This link suggests that the MTCC will use a CMSSW based DAQ with a "global" trigger to achiveve O(kHz) readout over slink. How is this different from the planned high-rate tests?
    

---