Development of readout electronics

Electronics and programs adapted to the application and to the detector chips are required for the control and readout of our detectors. For this purpose, we develop our own hardware, software and firmware and make them available to other research groups and collaboration partners. Currently we develop and support systems for detectors with Timepix3, VMM3a and Timepix chips. Furthermore, we are working on possibilities for real-time data analysis and data reduction in the readout system. In addition, we are developing other control systems that are needed to operate our detectors. These include a gas system that ensures constant gas flow and pressure in the detectors and a universal monitoring system for voltages, temperatures and other gas parameters.

Eine Wissenschaftlerin und ein Wissenschaftler arbeiten hinter einer Glasfassade und mischen Chemikalien mit Großgeräten.
© Gruber / Universität Bonn


The Timepix3 is a readout ASIC with 256x256 pixels with a pixel pitch of 55 µm. The chip features simultaneous time and charge measurements. The readout of the pixel data is data driven, which means that the data is processed as soon as a pixel has detected charge. The data is then sent to the readout electronics. As the data is zero-suppressed, there is no information from inactive pixels in the data and very high readout rates can be reached. The Timepix3 serves as a basis for future GridPix detectors.

We are developing an open readout and control system for the Timepix3 consisting of our own hardware, firmware and software. The goal is to build a system that can support the broad range of our detector application and that is also easy to use for other research groups. Furthermore, different user interfaces in form of graphical user interface, command line interface and scripts shall allow an easy start and a modular integration into different applications.


The Scalable Readout System (SRS) was developed by the RD51 collaboration as a general purpose readout system for particle detectors. The SRS can be used for smaller test setups (also in our group) with only a few dozen channels as well as for larger detectors (as for example at the European Spallation Source) with thousands of channels. Due to the modular principle, different front-end chips can be used, thus the SRS can be used for different types of detectors.

We are particularly concerned with the implementation of the VMM frontend chip in the SRS. The VMM was developed for the New Small Whell Upgrade of the ATLAS detector as a new readout chip for the Micromegas and sTGC detectors, but is universally usable for other detectors and sets new standards with a readout rate in the megahertz range and low electronic noise.

The worldwide exploitation of the SRS with the VMM for a wide variety of experiments and detector development projects has been advanced in an EU project. Our research extends from the development of the field-programmable gate array (FPGA) firmware for the SRS to SRS- and VMM-based detectors.

© M. Lupberger
© M. Ummenhofer


The Timepix is the predecessor of the Timepix3 and also offers 256x256 pixels at a distance of 55 µm. Unlike the Timepix3, either time or charge can be measured, but not both simultaneously. The Timepix data is read from all pixels simultaneously, regardless of whether a pixel was active or not. The Timepix serves as the basis for GridPix detectors.

Our group has developed two readout systems for Timepix optimized for different applications. One was primarily used on the CAST experiment and provides additional interfaces for cosmic ray vetoes and high voltage control. The other is designed for scalability and supports up to 160 Timepix chips. Based on this, a prototype for an ILC TPC with GridPixes was read out. We continue to use the readout systems for detectors that still use a Timepix ASIC. Furthermore we support other research groups that use Timepix or GridPix with our readout systems.

Hardware-based online analysis

The amount of data generated in modern particle detectors exceeds the capacities of the readout electronics, bandwidth and data storage. For some time now, the solution has been to use so-called triggers. Special chips within the readout chain decide whether the data of a particular particle collision should be stored or discarded.

Classically, so-called field-programmable gate arrays (FPGAs) are used for this purpose, on which highly parallelized logic designs can be implemented. Possibilities for triggers include discarding data based on fixed parameters or, as a result of developments in recent years, machine learning algorithms in the future. We are also one of the first research groups in the world to look into the use of Adaptive Compute Acceleration Platform (ACAP) chips. In addition to an FPGA, these chips also contain hundreds of highly specialized AI cores. We are researching the possibility of using ACAPs to implement more complex neural networks as triggers.

Hardware-basierte online Analyse
© P. Schwäbig
Eine Wissenschaftlerin und ein Wissenschaftler arbeiten hinter einer Glasfassade und mischen Chemikalien mit Großgeräten.
© Gruber / Universität Bonn

Monitoring System

In addition to the data from detectors, other data such as temperature, pressure and voltages are relevant for the interpretation of results and thus for stable operation of the detectors. Therefore, we are developing a common monitoring system for our applications. The goal is a modular system in which different measured variables from different applications can be easily integrated. The system will then take care of data storage and visualization.

One possible implementation of the system is to add a microcontroller to the readout electronics to record data independently of the event data. We are currently pursuing this approach for the control and readout system for Timepix3. Furthermore interfaces for further data sources e.g. from laboratory devices for voltage generation are offered.

Gas System

To ensure constant gas conditions in our detectors, we develop a gas system with pressure and flow regulators to which the detectors can be connected. The system is compact and self-sufficient for easy use in different applications at different locations.

In a first version we develop a system that supports premixed gases. However, the system is designed from the beginning to be able to mix multiple gases in the future. The system also has an interface to the monitoring system, so that the settings and measured values can be stored.

© J. Glowacz
Wird geladen