Triggers

Page maintainer: Lara

This page is considered done. It been reviewed by an expert. There may be missing elements, but they are all flagged and the text has no errors.

Intro

The first level trigger system is designed to produce a fast readout signal for the entire spectrometer whenever it finds an event of possible interest for physics analysis. The main physics trigger (Trigger 21) was therefore designed to detect deep-inelasting scattering events by looking for scattered leptons in the fast detector components.

Various triggers were implemented for different physics studies and some technical triggers were implemented to measure trigger efficiencies or background. The efficiencies of the triggers depend on the calorimeter threshold, as for a calo threshold of 1.4 GeV the trigger is not very efficient, while it is for the higher calo threshold of 3.5 GeV, as can be seen in the trigger efficiency section. The main triggers are combinations of H0, H1, H2 hodoscope signals, and calorimeter.

Essentials

Main Triggers

It should be noted that many of the triggers underwent various changes over the years. It is therefore crucial to check the exact trigger requirements for each of the years (and even subsets during a year) when analyzing data (see section Trigger Lists per year). Only trigger 21 stayed the same for 1996-2007 up to changes of the calorimeter threshold. In the following we will give some examples of triggers taken from 2005.

Physics Triggers

These are the main triggers used in analyses.

Trigger 21:(H0T*H1T*H2T*CALOT)+(H0B*H1B*H2B*CALOB)

This is the main physics trigger of our experiment. It looks for a single track corresponding to a scattered beam particle. The calorimeter requirement increases the likelihood that the particle is in fact a lepton.

Trigger 24: (H0T*H1T*H2T*CALOT) 
Trigger 25: (H0B*H1B*H2B*CALOB) 

Same as trigger 21, just only for top or bottom.

Trigger 17:H0>8*(2H0T*2H1T*2H2T*2BCT)*(H0B*H1B*H2B*BCB) and 
Trigger 27:H0>8*(2H0B*2H1B*2H2B*2BCB)*(H0T*H1T*H2T*BCT)

Trigger 17 looks for three tracks events:2 tracks in top and one in bottom, and similarly, Trigger 27 looks for two tracks in bottom and one in top.

Trigger 28: (H0mult<8)*[(H0T*H1T*BCT)*(H0B*H1B*BCB)]

These triggers look for two tracks, one in the top and one in the bottom. Since the calorimeter is not involved, these tracks can be and often are hadron tracks, hence the name "photo-production trigger". The H0 multiplicity cut and the BCs (MCs for trigger 22) are used to clean up the sample. Our high-pT hadron paper was obtained from an analysis of data coming from this trigger. Since the calorimeter is missing, this trigger can also be used to obtain the calorimeter trigger efficiency.

Trigger 22: (H0mult<8)*[(H0T*H1T*MCT)+(H0B*H1B*MCB)]

This is the same as Trigger 28, only with the MCs instead of BCs.

Trigger 9: (LUMI>25GeV)*(not GMS)

This quasi-physics trigger looks for a big energy deposit in either the left of right lumi calorimeter. This is looking for elastic scattering of the beam particle from the target.

Calibration Triggers

These triggers are used for diagnostics. For example, Trigger 18 is used to obtain the efficiency of H0, important because it enters Trigger 21, our main physics trigger. By looking at the ratio of events that fire trigger 18 (which is the same as trigger 24, but without H0top) and trigger 24 (which is the top trigger 21, and includes H0top), one gets the amount of times when H0top should have fired but did not, thus obtaining its efficiency. In turn, the efficiency of trigger 21 is the product of the efficiencies of all detectors entering its definition.

Trigger 18: (H1T*H2T*CALOT) 
Trigger 26: (H1B*H2B*CALOB)

These are the same as Trigger 21 (or rather 24 and 25), but without H0. They are used to obtain the trigger efficiency of H0.

Trigger 19: (H0T*H2T*CALOT)+(H0B*H2B*CALOB) 

These are the same as Trigger 21, but without H1. They are used to obtain the trigger efficiency of H1.

Trigger 20: (H0T*H1T*CALOT)+(H0B*H1B*CALOB) 

These are the same as Trigger 21, but without H2. They are used to obtain the trigger efficiency of H2.

Trigger 29: Hera Clock

It gives the lepton beam bunch signal, indicating the time at which the lepton bunches pass through the target and thereby defines a time window for all detector triggers to arrive.

Trigger Lists per year

1996

1997

1998

1999

2000 before July 16,2000

2000

2002

2003 before October, 19

2003 before October 29

2003

2004

2006

2007

Prescale Factors

In conditions of high luminosity the event rate could increase up to a point that a high deadtime would be achieved. To avoid this situation some triggers were prescaled. A prescale factor n means that only 1 event out of every n events that were generated is actually going to the "trigger OR" If the DAQ is not busy at this point of time, the event then gets accepted.

Note that in the trigger livetime calculation (triggers accepted vs. triggers generated), the number of triggers after prescaling goes in, i.e., the lifetime can be calculated like (number of triggers-after-prescaling accepted)/(number of triggers-after-prescaling generated).

In analyses that involve asymmetries it is generally not needed to know whether a trigger was prescaled or not; it is crucial instead in the case of absolute cross section measurements.

The prescale factors can be accessed from the udst as g1Trigger_iPreScalFactor. Care has nevertheless to be taken, since in some years these cases (the full 1996 and 1997 productions, or the first bursts of a run) can be zero, or a negative value, or be still set to the value it had in the previous run. A better estimate of prescale factors then comes from the ratio

g1Trigger_rGenTrigBefPS/g1Trigger_rGenTrigAftPS 

The following table provides a list of triggers per year and target, telling whether the main triggers were prescaled or not. It should be remembered that each year the trigger definitions changed.

Percentage of times that the prescale factor (obtained as g1Trigger.rGenTrigBefPS/g1Trigger.rGenTrigAftPS) was set to 1 for each trigger. The calculation was done by requiring that the bursts were analysable (g1Quality_iExpment=1) and that the ratio of trigger generated before prescaling to trigger generated after prescaling was less than 1.5.

		Tr17	Tr18	Tr19	Tr20	Tr21	Tr22	Tr24	Tr25	Tr26	Tr27	Tr28
1996	Pol.	2	1	58	58	100	1	100	100	5	4	98
	Unpol.H	2	2	100	100	100	3	84	100	6	4	12
	Unpol. D	3	1	98	98	100	1	85	100	5	3	6
	Unpol. He	4	2	100	100	100	4	87	100	8	5	9
1997	Pol. H	1	2	81	42	100	1	45	82	8	1	99
	Unpol. H	1	2	98	1	99	1	8	5	1	1	77
	Unpol. D	1	10	98	9	100	10	4	3	1	2	45
	Unpol. N	1	5	89	37	100	1	48	54	1	8	94
1998	Pol.D	23	44	100	100	100	100	100	100	73	100	100
	Unpol.H	46	2	100	99	100	100	100	100	0	100	100
	Unpol.D	45	1	100	100	100	96	100	100	0	96	96
	Unpol.Kr	0	2	100	100	100	100	100	100	0	100	100
1999	Pol. D	23	15	100	100	100	100	100	100	96	78	100
	Unpol.H	0	0	100	100	100	100	100	100	0	100	100
	Unpol.D	1	4	99	99	99	88	99	99	57	78	81
	Unpol.N	1	0	100	100	100	100	100	100	49	97	100
	Unpol.Kr	0	0	100	100	100	79	100	100	2	100	56
2000	Pol. D	5	64	100	100	100	95	100	100	78	55	86
	Unpol. H	9	36	69	69	100	8	100	100	0	2	8
	Unpol. D	2	14	93	93	100	6	99	100	0	1	6
	Unpol. He	3	0	19	19	100	20	100	100	20	0	3
	Unpol. N	0	0	100	100	100	5	100	100	10	0	0
	Unpol. Kr	0	100	100	100	100	1	100	100	1	1	1
	Unpol. Ne	11	50	100	100	100	8	100	100	0	4.2	6
	Unpol. H 12GeV	0	0	100	100	100	0	100	100	100	0	0
	Unpol. D 12GeV	0	2	35	35	100	0	100	100	100	0	0
	Unpol. N 12GeV	0	0	0	0	100	0	100	100	100	0	0
	Unpol. Kr 12GeV	0	3	64	64	100	0	100	100	100	0	0
2002	Pol. H	78	99	100	100	100	100	100	100	95	12	100
	Unpol. H	11	87	100	100	100	100	100	100	100	29	100
	Unpol. D	8	94	100	100	100	100	100	100	100	46	100
	Unpol. Kr	0	100	100	100	100	100	100	100	100	99	100
2003	Pol. H	89	100	100	100	100	100	100	100	100	89	100
	Unpol. H	100	100	100	100	100	30	100	100	100	100	87
	Unpol. Kr	100	100	100	100	100	9	100	100	100	100	8
2004	Pol. H	96	80	100	100	100	98	100	100	80	96	98
	Unpol. H	100	100	100	100	100	100	100	100	100	100	100
	Unpol. D	92	68	88	88	100	75	90	90	68	93	75
	Unpol. Kr	99	72	84	84	100	83	90	90	72	99	84
	Unpol. Xe	100	50	69	69	100	66	89	89	50	100	69
2005	Pol. H	99	91	100	100	100	96	100	100	91	99	94
	Unpol.H	99	99	100	100	100	91	100	100	99	99	91
	Unpol.D	99	94	38	100	100	18	100	100	95	99	18
	Unpol.Kr	100	89	97	100	100	98	100	100	89	100	98
	Unpol.Xe	100	95	96	98	100	85	100	100	95	100	84
2006	Unpol. H	100	98	100	100	100	97	100	100	98	100	98
	HD Unpol.H	100	100	3	100	100	3	100	100	100	100	3
	HD Unpol.D	100	100	0	100	100	0	100	100	100	100	0
2007	Unpol. H	100	100	100	100	100	100	100	100	100	100	100
	HD Unpol.H	100	100	31	100	100	30	100	100	100	100	30
	HD Unpol.D	100	100	97	100	100	97	100	100	100	100	97
		Tr17	Tr18	Tr19	Tr20	Tr21	Tr22	Tr24	Tr25	Tr26	Tr27	Tr28

More Info

Lists of prescale factors for different data-taking years.

uDST Documentation

See also Trigger efficiencies