Roumyana Hadjiiska
Th C CS (R inf m dat det K
The CMS trigger system explores two levels – the hardware L1 (Level One) trigger level and software algorithm level – so called HLT (High Level Trigger). During Run-1 the L1 trigger logic was based on the particular local triggers decisions from any of the sub-detectors. In order to cope with the drastically increased rate, in the beginning of the Run-2, L1 trigger has been re-based on a regional principle, collecting an input from all the detectors in a given pseudorapidity region in order to make its decision. Thus, the RPC system contributes to three different pseudorapidity regions relevant to BMTF, OMTF and EMTF (Barrel, Overlap and Endcap Muon Track Finders). The CMS RPCs explore double gap design with a common readout strips for both the gaps [2]. The space resolution is ~ 1 cm and the time resolution is ~2 ns [3]. The RPCs are used in CMS mainly as a trigger detector. They contribute to L1 mainly with hits as main trigger primitives. However, the existence of two RPC layers on the first and second barrel stations allows to build RPC segments and improve the trigger efficiency in the given region. The time window of the readout electronics is 25 ns. The fast time response ensure correct association of the measured particles with the collision time and relevant bunch crossing (BX). BX 1. en col in int tha sta tra ins a s det thr the (D m Ch m en an association of the muon, coming from the proton-proton collisions measured in Run-2 are shown in (Fig.2). As might be seen from the plots (note the logarithmic Y scale) more than 99% of the measured muons are associated in the central BX window. Small fraction in the non-central BX windows are caused by background particles and post-collision effects. 2. 2. ca me an bo L the the ele RP pr as ba for the the (~ the 3. RPC detector performance 3.1 Main objects and monitored quantities 3.1.1 Cluster Size
There are many quantities which are carefully monitored on-line (during the time of the data taking) and offline (in the end of every particular data taking set). The main focus of the offline monitoring is on the cluster size and hit efficiency. The so called RPC rechits (reconstructed hits) are the output of the RPC local reconstruction and are built by the clustered adjacent strips, fired in the same BX window with response to the passing particles. The average cluster sizes range from 2 to 3 strips. The strip pitch is smaller for the RPCs installed at lower radii and larger for chambers on the outermost stations. Because of this the cluster size varies for the different RPC stations and it is larger for the innermost station RB1. The effect is well visible on the cluster size plot in (Fig.6). 3. (in on ma cal the H lu ins me stu Tr tra to eff de sta av 3. 3. eff po tak ap mi co are an ob sta (Fi 3. the en hardware problems - threshold control, chambers switched off because of gas leak problems. The numbers given on the plots show the average efficiency for the well performing and the fraction of the problematic RPC rolls. The RPC efficiencies measured in 2015-2018 are comparable and in agreement with the design requirements of efficiency above 95%. 4. eff ins est lu eff un the 5. ke ac ad R [1]