D ZERO NOTE 827 CHANGES to the FIRST LEVEL TRIGGER TIMING ------------------------------------------- D. Edmunds Michigan State University 10-APR-1989 This notes describes modifications to the definition of the control signals that are sent from the Trigger Framework to the Front-End Data Acquisition Crates. The functional definitions of the signals have not changed, rather there have been three changes in the details of the timing of these signals. These three timing changes are the following: 1. The time after the beam crossing at which the Level 1 Trigger decision is made has been moved up from 3.0 microseconds to 2.5 microseconds. This is to give the analog front end electronics more time to acquire a new baseline before the next beam crossing. This implies that any information that will be used in making the Level 1 Trigger decision (e.g. Busy Signals, and data for the AND-OR Network) must now reach the Trigger Framework by 2.4 microseconds after beam crossing 2. The timing of the falling edges of the Start Digitization Signal and the Hold Transfer Signal have been changed so that they always straddle the point in time of the beam crossing. This arrangement insures that the analog front-end baseline will not shift in any of the trigger scenarios. Because the answer from a Level 1.5 Trigger can arrive at the First Level Trigger Framework asynchronously with respect to the 3.5 microsec beam crossing clock, in general the Framework will need to wait for the end of a 3.5 microsec cycle before "advertizing", via the sequence in the dropping of the Start Digitization and Hold Transfer Signals, the result of the Level 1.5 decision. The separation in the falling edges of the signals will be about 400 nanoseconds. The three First Level Trigger scenarios are shown in the figures at the end of this note. 1) First Level Trigger with no L1.5 decision required, 2) First Level Trigger requiring a L1.5 decision and passing the L1.5 decision, 3) First Level Trigger requiring a L1.5 decision and failing the L1.5 decision. 3. The third change is to add a skew between the rising edge of the Start Digitization Signal and the rising edge of the Hold Transfer Signal. This skew will delay the rising edge of the Hold Transfer Signal by about 200 nanosec. Note: There is no added delay to the Start Digitization Signal. Notes: During a normal readout sequence the various control signal ----- will have about the following durations: Start Digitization; up for about 1.2 microsec. Hold Transfer; up for about 0.6 microsec. Specific Triggers Fired Strobe; up for about 0.5 microsec. Because of the sensitivity of the front-end systems to the edges of these control signals, it would improve the overall stability of the data acquisition system if the various front-end systems filtered these signals (e.g. a 20 nsec RC and schmitt circuit) before using them. This can and should be done for the following reasons: Because these are all relatively long duration signals. Because it is not necessary for any of the front-end systems to respond to these signals immediately (i.e. in less than 20 nsec). Because the front-end systems are going to respond to the edges of these signals noise could be a problem. Because these signals will travel over long cable runs before being used. FIGURE 1 -------- READOUT SEQUENCE NOT REQUIRING A LEVEL 1.5 DECISION ===------------------------------- _________________________________ START | | DIGITIZE /|\ | _______________| |________ ______________________ HOLD | | TRANSFER | \|/ __________________| |________________ __________ _____________________________________________ SPECIFIC \ / TRIGGERS X FIRED __________/ \_____________________________________________ _________ SPECIFIC | | TRIGGERS /|\ | FIRED _________________________________________| |______ STROBE THE "X" SHOWS THE ---------------------------------------------X------------ BEAM CROSS FIGURE 2 -------- READOUT SEQUENCE REQUIRING A LEVEL 1.5 DECISION AND PASSING THE LEVEL 1.5 TRIGGER ----=======---------------------- ____________ ________________ START | | DIGITIZE /|\ | _______________| |________ _________ ________ HOLD | | TRANSFER | \|/ __________________| |________________ __________ _______________ _________________________ SPECIFIC \ / TRIGGERS X FIRED __________/ \_______________ _________________________ _________ SPECIFIC | | TRIGGERS /|\ | FIRED ____________________________ ________| |______ STROBE THE "X" SHOWS THE ---------------------------- ------------X------------ BEAM CROSS | /|\ CONTROL SIGNALS REMAIN | STATIC WHILE ONE OR MORE | 3.5 uSEC. BEAM CROSSING ____| CYCLES ARE SPENT MAKING THE LEVEL 1.5 DECISION FIGURE 3 -------- READOUT SEQUENCE REQUIRING A LEVEL 1.5 DECISION AND NOT PASSING THE LEVEL 1.5 TRIGGER ----===------------------------------ ____________ ________ START | | DIGITIZE /|\ | _______________| |________________ _________ ________________ HOLD | | TRANSFER | \|/ __________________| |________ __________ _______________ _________________________ SPECIFIC \ / TRIGGERS X FIRED __________/ \_______________ _________________________ SPECIFIC TRIGGERS FIRED ____________________________ _________________________ STROBE THE "X" SHOWS THE ---------------------------- ------------X------------ BEAM CROSS | /|\ CONTROL SIGNALS REMAIN | STATIC WHILE ONE OR MORE | 3.5 uSEC. BEAM CROSSING ____| CYCLES ARE SPENT MAKING THE LEVEL 1.5 DECISION NOTES: Anytime a Start Digitize signal is sent to the Front END Crates it will be accompanied by a Hold Transfer signal. This is true even for those Level 1 Triggers that do not require a subsequent Level 1.5 validation. The Specific Triggers Fired Strobe Signal is sent out only for those Level 1 Triggers not requiring a Level 1.5 Trigger confirmation, and for those Level 1 Triggers requiring and passing the Level 1.5 Trigger. ----------- REFERENCES ---------- 1. "D0 Trigger Framework" Maris Abolins, Daniel Edmunds and James Linnemann, D0 Note 328, February 24, 1986. 2. "The Level One Framework: D0 Note 328 Revised" Maris Abolins, Daniel Edmunds, Philippe Laurens, James Linnemann, and Bo Pi, D0 Note 705, May 26,1988.