CSC Digitization Rick Wilkinson 1 CSC design and

CSC Digitization Rick Wilkinson 1

CSC design and readout • Large chambers n Wires groups: 5 cm wide n Cathode strips: 8 -16 mm wide 3. 3 m wire-group hits every 25 ns same information for trigger/offline Level-1 Trigger: half-strip hits every 25 ns HLT*/offline: 12 -bit digitization every 50 ns *HLT – High Level Trigger 2

Code Structure Ionize, Delta rays, drift, avalanche Sim. Hit Wire. Hit Analog. Signal Wire. Digi Shape & sum (shares code) Strip. Hit Analog. Signal Strip. Digi Comparator. Digi 3

Effects From Calibrations • Used: – – – Gains Pedestal Width Correlated Noise Crosstalk Bad Chambers • Not Used: – Bad Channels 4

Correlated Noise in Strips • 1 st, 2 nd, and 8 th samples use pedestal width / 2 – Should hardcode a 25% correlation for nearest neighbors • Other samples use correlated noise from database 5

Crosstalk in Strips • Simulation models crosstalk as: – Constant times neighboring signal (resistive) – Constant times the slope of the neighboring signal (capacitive) • Crosstalk calibrations are measured as a linear fit: – Crosstalk fraction = m * (t-tpeak) + b • Reconstruction surprisingly insensitive to crosstalk 6

Raw Data Packing • Complex – Over 40 data structures – Significant changes between versions • Needs to be perfect – The Examiner module will drop an entire FED if any problems are seen with headers, wordcounts, CRCs 7

Plans • Queued for 3_5_X – Tweaked timing so that downstream products (Rec. Hits and L 1 trigger segments) are centered in BX. – Updated to newest packing format • Needed for data mixing – Data. Mixer code to merge strip signals • To do: – Validate crosstalk 8