- ATLAS TDAQ counts the "first trigger" after L1 reset as "0"
- TIM sends L1ID values to ROD without modification
- The ABCD chip counts the "first trigger" after L1 reset as "1" :-(
- The ROD EFB adds one to the L1ID reported by TIM and compares the modules' L1IDs with this :-)
- Hence link BC errors will be flagged when
- (module L1ID) is not equal to (TIM L1ID + 1)
- The header of each event fragment reports the L1ID exactly as recieved by TIM.
- No further mechanism to offset the L1ID should be required, as long as all modules receive ECR (soft reset) at the start of each run (or orbit) to synchronise the counters.
All you ever need to know about BCIDs, but were afraid to ask:
The operations manual has an EFB register called EFB_CMND_O, at address 00402210 the top 8 bits of which [31:24] are called BCID Offset Value - which can be used to offset the expected BCID if required. HOwever I cant see any equivalent thing for the L1ID.