The goal of the DT upgrade is to maintain the present system performance, trigger and reconstruction, at the HL-LHC background rates (instant and integrated) and under the HL-LHC CMS Trigger/DAQ conditions (750 kHz L1 trigger rate and 12.5 μs L1 trigger latency).
The present DT detectors will stay for HL-LHC operation. However, preliminary studies show that they may suffer detector ageing over the lifetime of HL-LHC. An extensive R&D upgrade program is underway to mitigate the potential ageing-associated problems. The results of ageing studies and the discussion of possible mitigation measures are presented in the Phase 2 Muon TDR.
The rate of failures of DT on-detector electronics, the so-called Minicrates (MiC), is projected to be unsustainably high at HL-LHC and the new HL-LHC CMS Trigger/DAQ requirements exceed the present MiC capabilities. Therefore, all MiC will be replaced together with the associated back-end electronics.
A new on detector electronic board is being designed, so called OBDT (On Board electronics for Drift Tubes), and CIEMAT will be responsible of the design and fabrication of the 180 boards that cover the theta view. An image of the first prototype is shown here:
Image of the OBDT prototype board.
In addition, a new backend system is being designed, using the most performant ATCA boards with ten´s Gbps optical links and highest performant FPGAs. At CIEMAT we have developed a new Trigger algorithm to operatate under HL-LHC taking profit of the full resolution of the DT signals which will be available from the OBDT. This algorithm has been implemented in the Phase 1 uTCA boards for validating its performance.
Image of the AB7 backend prototype at the CIEMAT´s lab.
A first version of this algorithm is being operated satisfactorily at the CMS Slice Test at CERN.
Once the Sector Collector relocation phase was achieved, we proceeded with the replacement of the second level of trigger and readout electronics by a higher performant system.
The trigger electronics, TSC, had to be upgraded as part of the Level 1 Trigger upgrade project in order to allow improved performance that satisfies the optimal physics in a higher luminosity regime. On the other hand, the ROS-25 board needed to be replaced in order to increase the processing speed to cope with the increased occupancy due to the higher luminosity (up to a factor 2 the LHC nominal design).
A single design was foreseen for both types of boards since the main difference is the input data format and speed (480 Mbps for the trigger and 240 Mbps for the readout), which could be easily accomodated. The board is built around a Virtex 7 FPGA and contains up to 72 input optical links and 10 Gbps outputs. It follows the uTCA telecommunications standard. It is called TwinMux board for the trigger case and uROS for the readout case, with different firmware implemented in the FPGA.
Image of the TwinMux/uROS board.
Image of the uROS slice test (2017) at CMS operating under collisions.
Image of the full uROS system installed and running in CMS during 2018 proton-proton collisions.
Image of the uROS online software system during operation.
The use of DT technology in the CMS barrel has been possible due to the low hit rate and the relatively small strength of the local magnetic field. By the time of the HL-LHC start-up, the DT system will be more than 20 years old and will need to operate for other 10 years integrating ∼10 times luminosity more than it was designed for.
Accordingly, an extensive program to study the performance of the Drift Tube chambers under the expected accumulated charge during HL-LHC was started.
Many test beams and long term irradiation campaigns have taken place at the CERN GIF++ (Gamma Irradiation Facility). Present results show that even if a mild degradation is to be expected in the most exposed chambers, the barrel system will still provide very high efficiency and performance.
Image of a DT chamber at the GIF++ (Gamma Irradiation Facility).
The Sector Collector relocation was the first stage of the upgrade program for the Drift Tubes subdetector of the CMS experiment. It was accomplished during Long Shutdown 2013-2014, and consisted in the relocation of the second-level of trigger and readout electronics from the experimental (UXC) to the service cavern (USC), relieving the environmental constraints and improving accessibility for maintenance and future upgrades.
Extending the 30 meters copper electrical links for additional 100 m would degrade reliability, so the information was converted to optical with a custom system capable of dealing with the DC-unbalanced data. Initially, present electronics was used (during 2015 and up to 2017 for the ROS-25), so optical-to-copper conversion was also been installed.
Image of the different electronics pieces designed by the DT group for the Sector Collector Relocation upgrade during LS1 (2013-2014).