Topics

Electronics experts connect in Aachen

25 January 2011

TWEPP-2010 scrutinizes topics from chip design to power provision.

CCtwe1_01_11

Each year, the Topical Workshop on Electronics for Particle Physics (TWEPP) provides the opportunity for experts to come together to discuss electronics for particle-physics experiments and accelerator instrumentation. Established in 2007, it succeeds the workshops initiated in 1994 to focus on electronics for LHC experiments, but with a much broader scope. As the LHC experiments have now reached stable operating conditions, the emphasis is shifting further towards R&D for future projects, such as the LHC upgrades, the studies for the Compact Linear Collider and the International Linear Collider, as well as neutrino facilities and other experiments in particle- and astroparticle physics.

The latest workshop in the series, TWEPP-2010, took place on 20–24 September at RWTH Aachen University and attracted 190 participants, mainly from Europe but also from the US and Japan. It covered a wide variety of topics, including electronics developments for particle detection, triggering and data acquisition; custom analogue and digital circuits; optoelectronics; programmable digital logic applications; electronics for accelerator and beam instrumentation; and packaging and interconnect technology. The programme of plenary and parallel sessions featured 16 invited talks together with 63 oral presentations and 66 poster presentations selected from a total of 150 submissions – an indication of the attractiveness of the workshop concept. The legacy of the meeting as a platform for the discussion of common LHC electronics developments is reflected in several electronics working groups for the super-LHC (sLHC) project holding their bi-annual meeting during the workshop, namely the Working Groups for Power Developments and for Optoelectronics, as well as the Microelectronics User Group. In addition, two new working groups on Single Event Upsets and on development of electronics in the emerging xTCA standard had “kick-off” meetings during TWEPP-2010.

After a welcome and introduction to particle physics in the host country and the host institute (see box), the opening session continued with “Physics for pedestrians”, a talk by Patrick Michel Puzo of the Laboratoire de l’Accélérateur Linéaire, Orsay, in which he explained the Standard Model of particle physics, as well as experimental measurement techniques, to the audience of hardware physicists and engineers. DESY’s Peter Göttlicher went on to present the European X-ray Free Electron Laser project (XFEL) currently under construction at DESY. This fourth-generation light source will provide ultra-short flashes of intense and coherent X-ray light for the exploration of the structure and dynamics of complex systems, such as biological molecules. Dedicated two-dimensional camera systems, such as the Adaptive Gain Integrating Pixel Detector (AGIPD), are being developed to record up to 5000 images a second with a resolution of 1 megapixel. The session closed with a summary of the status of the LHC by CERN’s Ralph Assmann, who also discussed the expected and observed limitations and prospects for further increases in intensity, luminosity and beam energy at the LHC, as well as short- and long-term planning.

From ASICs to optical links

For the next three days, morning and afternoon sessions began with plenary talks, after which the audience separated into two parallel sessions. With 20 presentations, the session on application-specific integrated circuits (ASICs) was again by far the most popular, demonstrating the demand of chip designers for a forum to present and discuss their work. One increasingly important aspect in the next generation of experiments with high radiation levels is the mitigation of single-event effects (SEE), such as single event upsets (SEU), which are caused by the interaction of particles with the semiconductor material. Deep-submicron integrated circuit technologies with low power consumption are becoming increasingly sensitive to SEEs and this must be carefully taken into account at both the system level and the ASIC design level. Invited speaker Roland Weigand of the European Space Agency gave insight into the various approaches of SEE mitigation that are employed in space applications, where integrated circuits are exposed to solar and cosmic radiation.

A relatively new development is the 3D integration of circuits, where several circuit layers are stacked on each other and interconnected, for example by through-silicon-vias. The advantages include the reduction of the chip area, reduced power consumption, a high interconnection density and the possibility to combine different processes in one device. Within particle physics, a possible future application is in the upgrades of the large silicon trackers of the LHC experiments. Kholdoun Torki from Circuits Multi-Projets, Grenoble, presented the plans for a 3D multiproject wafer run for high-energy physics, which allows several developers to share the cost of low-volume production by dividing up the reticle area.

The parallel session on “Power, grounding and shielding” focused mainly on novel power-provision schemes for upgrades of the LHC experiments, namely serial powering and DC–DC conversion. An increase in the number of readout channels and the possible implementation of additional functionality, such as a track trigger, in the tracker upgrades of ATLAS and CMS will lead to higher front-end power consumption and consequently larger power losses in the supply cables (already installed) and to an excessive increase in the material budget of power services. New ways to deliver the power therefore need to be devised. Both of the new schemes discussed solve this problem by lowering the current to be delivered. In serial powering, this is done by daisy-chaining many detector modules, while DC–DC conversion schemes provide the power at a higher voltage and lower current, with on-detector voltage conversion. These topics were further expanded in the session of the Working Group for Power Developments.

CCtwe3_01_11

Another parallel session was devoted to the topic of optoelectronics and optical links. Data transmission via optical links is already standard in the LHC experiments because such links do not suffer from noise pick-up and contribute less material than the classic copper wires. In the session and in the following working-group meeting, presentations focused on experience with installed systems as well as on new developments, in particular for the Versatile Link project, which will develop high-speed optical-link architectures and components suitable for deployment at the sLHC. In an inspiring talk, invited speaker Mark Ritter of IBM expanded on optical technologies for data communication in large parallel systems. He explained that scaling in chip performance is now constrained by limitations on electrical communication bandwidth and power dissipation and he described how optical technologies can help overcome these constraints. The combination of silicon nanophotonic transceivers and 3D integration technologies might be the way forwards, with a photonic layer integrated directly into the chip such that on-board data transmission between the individual circuit layers is performed optically.

First LHC experience

A highlight of this year’s workshop was the topical session devoted to the performance of LHC detectors and electronics under the first beam conditions. Gianluca Aglieri Rinella of CERN presented the experience with ALICE, a detector designed specifically for the reconstruction of heavy-ion collisions, where high particle-multiplicities and large event sizes are expected. He showed that more than 90% of the channels are alive for most of the ALICE detector subsystems, with the data-taking efficiency being around 80%. The ALICE collaboration’s goal for proton–proton collisions is to collect a high-quality, minimum-bias sample with low pile-up in the time projection chamber, corresponding to an interaction rate of 10 kHz. For this reason, the peak luminosity at ALICE is deliberately reduced during proton–proton running.

CCtwe4_01_11

Thilo Pauly of CERN presented the ATLAS report. He showed that more than 97% of the channels are operational for all detector systems and that 94% of the delivered data are good for physics analysis. The ATLAS momentum scale for tracks at low transverse momentum is measured with a precision of a few per mille, while the energy scale for electromagnetic showers is known from the reconstruction of neutral pions to better than 2%. The experience of CMS, presented by Anders Ryd of Cornell University, is similarly positive, with all subsystems being 98% functional with a data-taking efficiency of 90%. He explained that the collaboration struggled for a while with the readout of high-occupancy beam-induced events in the pixel detector – the main reason for detector downtime – but managed to solve the problem.

Last but not least, Karol Hennessy of the University of Liverpool reported on LHCb, which is optimized to detect decays of beauty and charm hadrons for the study of CP violation and rare decays. This experiment has had a detector uptime of 91% and a fraction of working channels above 99% in most subdetectors. One specialty is the Vertex Locator – a silicon-strip detector consisting of retractable half-discs whose innermost region is only 8 mm away from the beam. This detector reaches an impressive peak spatial resolution of 4 μm.

Posters and more

The well attended poster session took place in the main lecture hall and featured 66 posters. Discussions were so lively that the participants had to be reminded to stop because they would otherwise miss the guided city tour. The workshop dinner took place in the coronation hall of the town hall, where participants were welcomed by the mayor of Aachen. The dinner saw the last speech by CERN’s François Vasey as Chair of the Scientific Organizing committee. He became Workshop Chair in 2007, shaping the transition to TWEPP and after four successful workshops he now passes the baton to Philippe Farthouat, also of CERN. The next workshop in the series will take place on 26–30 September 2011 in Vienna.

TWEPP-10 was organized by the Physikalisches Institut 1B, RWTH Aachen University, with support from Aachen University, CERN and ACEOLE, a Marie Curie Action at CERN funded by the European Commission under the 7th Framework Programme.

bright-rec iop pub iop-science physcis connect