CMU MAC Article
This August, CMU physicists working on the CMS experiment at CERN's Large Hadron Collider (LHC) successfully built and tested prototypes for the High-Granularity Calorimeter upgrade. Building functional prototypes is, on one hand, a major milestone which has been several years in the making, and on the other, only the first step in a manufacturing project that will take place over the next three years.
The CMS experiment is one of two general purpose detectors at the LHC. CMS can be thought of as a large 3D camera that records the products of the proton collisions provided by the LHC. Images collected from the detector were used to discover the Higgs boson in 2012. Since its discovery, a major focus of the field has been in studying the properties of the Higgs boson in detail. Comparing these measurements to predictions will allow new theories to be tested and will further our understanding of this unique particle. However, to significantly improve the precision of the Higgs boson measurements more data is needed.
The LHC has a 15 year program to increase the total number of proton collisions by a factor of 20. This program requires collecting more data faster and comes at a significant cost: increased radiation. In addition to producing new exotic states of matter – like the Higgs boson – LHC proton collisions produce large amounts of ionizing radiation. This radiation is similar to that produced by a nuclear reactor and is damaging to both people and the instrumentation that makes up the detector. The current CMS detector was built almost 20 years ago and was not designed to be able to handle the amount of radiation damage foreseen in the upgraded LHC running. New, upgraded detectors are needed both to improve the quality of the recorded images and to cope with the more challenging radiation environment. This is where the High-Granularity Calorimeter upgrade comes in.
The High-Granularity Calorimeter (HGC) is a new detector that will replace the current CMS detectors in the regions which face the most radiation. The HGC is a next-generation imaging calorimeter that will significantly increase the precision with which the LHC collisions are imaged; e.g.: the number of individual measurements per picture will increase from ∼ 20k in current detectors to ∼6M in HGC, an example of big data getting bigger. The measurements of individual particles will go from the handful of numbers that the current detector provides to a high-resolution 3D movie of how the particles interact as they traverse the detector.
The HGC will be built in the next five years and CMU is playing a leading role in its construction. The HGC will be composed of 30,000 8-inch hexagonal modules. The modules – essentially radiation tolerant digital cameras – will be tiled to form wheels that are several meters in diameter; these wheels will then be stacked to form the full 3D detector. In total, the HGC will require 600 m2 of active silicon sensors.
We will build and test 5000 of these modules at CMU. The remaining modules will be produced by CMS collaborators at UC Santa Barbara and Texas Tech University in the US, and by groups in China, India and Taiwan. Each module consists of a silicon sensor attached to printed circuit board housing readout electronics and to a base plate which provides cooling and overall stability. Module construction will be performed with a series of automated robots which use pattern-recognition algorithms for assembly and to make the required ~200 electrical connections per module. After a series of testing at CMU and Fermilab – a US particle physics lab outside of Chicago – the modules will be sent to CERN for installation.
The production of the first working modules this summer was part of a qualifying exercise in which the various module assembly centers convinced ourselves and the rest of the CMS collaboration that we are ready and able to build the high-quality modules needed by HGC. To get to this point, our group has established a class 1,000 clean-room on the 8th floor of Wean Hall, expanding an existing space used by the medium-energy physics group. We have installed and commissioned an 8000 pound gantry robot, used to attach the different module layers, and an automated wire bonder, used to make the electrical connections within the modules. The recently produced prototype modules have allowed us to test our automated assembly procedures and exercise the full production chain.
"XXXXXXXXX" Quote from Manfred here
Producing one module to specification is just the beginning, during full-scale module production – starting in 2024 – CMU will make up to 24 modules per day. A major challenge in ramping throughput will be recruiting and on-boarding local talent. To meet production needs we will need to grow our group to five full-time technicians and engineers who will work on the daily production line.
Implementing reliable test procedures for quality control is another major challenge going forward. The module production will be pipe-lined with a several-day start-to-finish time per module. It will thus be critical to catch and fix any flaws in production as quickly as possible. The test procedures will be developed mainly by postdocs and graduate students and will provide an opportunity for a significant number of CMU undergraduates to get hands-on experience testing modern particle physics detectors. "QUOTE here about unique time to get involved in this research perfect time for to contribute to the upgraded LHC detectors". We believe that recent advances in image processing from Machine Learning will be crucial in assuring quality control during production. This work, a mix of computer science, machine learning and robotics, is a perfect fit for CMU and we plan to tap into resources throughout the university.
(John Alison)