LEB2000

Grounding, Shielding, Cooling And Alignment

Id: 33
Corresponding Author: Greg HALLEWELL
Experiment: ATLAS
Sub-system: Tracker
Topic: Grounding Shielding Cooling And Alignment

Development of Fluorocarbon Evaporative Cooling Recirculators and Controls for the ATLAS Pixel and Semiconductor Tracking Detectors

C. Bayer (Wuppertal), M. Bosteels (CERN), P. Bonneau (CERN), H. Burckhart (CERN), D. Cragg (RAL), R. English (RAL), G. Hallewell (RAL/CPPM), B. Hallgren (CERN), S. Kersten (Wuppertal), P. Kind (Wuppertal), K. Langedrag (Oslo), S. Lindsay (Melbourne), M. Merkel (CERN), S. Stapnes (Oslo), J. Thadome (Wuppertal), V. Vacek (CERN/Czech Technical University, Prague)

Abstract:

We report on the development of evaporative fluorocarbon cooling recirculators and their control systems for the ATLAS Pixel and Semiconductor Tracking (SCT) detectors. A prototype circulator uses a hermetic, oil-less compressor and C3F8 refrigerant. The mass flow rate to each circuit is individually tuned via feedback according to the circuit load variation, using dome-loaded pressure regulators in the liquid supply lines piloted with analog compressed air from DAC-driven voltage to pressure ("V2P") converters. Evaporated C3F8 exits each circuit through an analog air-piloted back-pressure regulator, which sets the circuit operating temperature. A hard-wired thermal interlock system automatically cuts power to individual silicon modules should their temperature exceed safe values.

All elements of the circulator and control system have been implemented in prototype form. Temperature, pressure and flow measurement in the circulation system uses standard ATLAS CanBus LMB ("Local Monitor Box") DAQ and CanBus interfaced DACs in a large (300 + channel) multi-drop Can network administered through a BridgeView user interface. Prototype 16 channel interlock modules have been tested.

The performance of the circulator under steady state, partial-load, and transient conditions is discussed and future developments are outlined.

Summary:

We report on the development of evaporative fluorocarbon cooling recirculators and their control systems for the ATLAS Pixel and Semiconductor Tracking (SCT) detectors.

The front-end electronics and silicon substrates of these detectors collectively dissipate around 50kW of heat, which must be removed from the ATLAS inner detector cavity through around 400 separate evaporative cooling circuits. For an operational lifetime of around 10 years in the high radiation field close to the LHC beams, the silicon substrates of these detectors must operate at a temperature below ~ -6 C, with only short warm-up periods each year for maintenance. Evaporative cooling is chosen since it offers minimal extra material in the tracker sensitive volume.

Following our studies of evaporatively-cooled Pixel and SCT thermo-structures (LEB 99), we have addressed the development of evaporative fluorocarbon recirculators and their control systems for use with per-fluoro-n-propane (C3F8) at an evaporation temperature (pressure) of ~-25 C (~1.7 bar abs).

A prototype circulator is centered around a hermetic, oil-less piston compressor operating at an aspiration pressure of ~ 1 bar abs and an output pressure of ~ 10 bar abs. Aspiration pressure is regulated via PID variation of the compressor motor speed from zero to 100%, based on the sensed pressure in an input buffer tank. High pressure C3F8 vapor is condensed and passed to the detectors in liquid form, with optional pre-cooling to a temperature of ~ -15 C.

Coolant liquid will be split into around 400 circuits in racks on the ATLAS service platforms. The mass flow rate to each circuit will be individually tuned via feedback according to the circuit load variation, using pressure regulators in the liquid supply lines. These regulators will operate in an inaccessible, high radiation, magnetic field environment, and will be dome-loaded, using analog compressed air delivered from DAC-driven voltage to pressure ("V2P") converters.

Evaporated C3F8 will exit each circuit through an analog air-piloted dome-loaded back-pressure regulator, which will determine the boiling pressure, and hence the operating temperature. Such individual temperature control is impossible in a parallel flow liquid cooling system.

A hard-wired thermal interlock system will automatically cut power to individual silicon modules should their temperature exceed safe values for any reason.

The performance of the circulator and the temperature distribution on powered silicon modules under steady state, partial-load, interlock-trip, start-up and shutdown conditions will be discussed. Finally, aspects of a full-scale demonstrator with ~ 25 cooling circuits and 6kW cooling capacity, currently undergoing commissioning, will be outlined.

Id: 50
Corresponding Author: Martin MANDL
Experiment: ATLAS
Sub-system: Tracker
Topic: Grounding Shielding Cooling And Alignment

Grounding and Shielding of the ATLAS TRT

Martin Mandl for the TRT collaboration

Abstract:

This paper addresses practical considerations for the engineering of the grounding and shielding system of the ATLAS-TRT.

Summary:

A ground system serves three primary functions: personnel safety, equipment and facility protection, and electrical-noise reduction. Defining the potential of each conductive material to be within certain margins achieves safety. A proper signal reference system together with shielding of sensitive as well as noisy parts provide noise reduction. Defining the potential of the conductive structures and building a signal-reference system inside the TRT, can be chosen within two philosophies: either strongly connecting everything together or trying to control the currents which flow in the system. The first one yields the lowest impedance between any two points of the system, but simultaneously allows loops and shield currents to flow inside the system. The second approach allows to break these loops and to ban shield currents from intruding the system through carefully provided low-impedance paths.

Each ATLAS sub detector has to follow "The ATLAS Policy on Grounding and Power Distribution" which gives the following guidelines:

• [...] electrical isolation of all detector systems, [...]

• [...] floating low-voltage power supplies, [...]

• [...] floating high-voltage power supplies, [...]

• [...] data transmission, clock and trigger distribution through optical links
or shielded twisted-pair cables, [...]

• [...] detector located inside a faraday cage. [...]

This negates the first philosophy at an intersystem level, but still allows it inside the sub detector. Only the final system will show all systematic effects which could not be predicted from a small prototype. Implementing provisions for both philosophies allows us to postpone the choice until more experience has been acquired. Therefore a way of realizing both approaches has been defined.

Id: 79
Corresponding Author: Jan GODLEWSKI
Experiment: ATLAS
Sub-system: General Interest
Topic: Grounding Shielding Cooling And Alignment

Mono-phase cooling system for front-end electronics on the example of the ATLAS TRT detector

Magnus Andersson - Luleå University of Technology Sweden
Pierre Bonneau - CERN
Michel Bosteels - CERN
Jan Godlewski - INP Krakow Poland, CERN

Abstract :

The work presents the results of cooling tests performed for the ATLAS TRT electronics. The test installation and control equipment are described.

A model of a standard cooling unit designed for all ATLAS detectors is also presented together with its modifications corresponding to various limitations connected with experimental zone, magnetic field, limited access and localization of various detectors.

Summary:

The efficiency of cooling system for front-end electronics of TRT end-cap detector was studied both by Finite Element Analysis and experimental tests. A good agreement between the simulations and tests results was achieved. FEA model can predict the temperature of the electronics with a sufficient accuracy in a wide range of heat dissipation (better then 5ºC in 50 to 100 mW/channel range). A mono-phase cooling system was tested experimentally using a unit in which fluorocarbon was used as a coolant. The test results, which will be presented, made it possible to design a final installation.
In the next step various experiments will be included into the final configuration taking into account limitations of the experimental zone.The main goals are as follows: to install a minimum amount of active components in the cavern, to ensure safe and reliable functioning by using systems as simple as possible, enable a distant control and necessary action in the case of problems. The use of a very expensive liquid results also in the necessity of finding a reliable recuperation method.
A cooling unit is designed in such a way that it can function using any tape of liquid and at any temperature. In the ATLAS experiment one can define three distinct temperature zones cooled by mono-phase cooling liquid. On the one hand temperature screens isolating silicon detectors at about -10ºC and TRT at 14ºC, access to both of which is very limited which results in small pipes and as a consequence in high pressure drops. On the other hand, there are the remaining detectors operating at temperatures higher than the dew point in experimental cavern. The access here is less limited enabling the work at the low pressure. Depending on the localization of cooling units it is recommended to equip the pumps with hydraulic motors. Elements and the logic of control are described, while the complete design of a control system will be worked out basing on standards, which will be accepted by the whole ATLAS.

Id: 97
Corresponding Author: John EVANS
Experiment: General Interest
Sub-system: General Interest
Topic: Grounding Shielding Cooling And Alignment

Electronic Design Automation tools for high-speed electronic systems

B.J. Evans
E. Calvo Giraldo
T. Motos Lopez

CERN, IT/CE

Abstract:

The LHC detectors will produce a large amount of data that will need to be moved very quickly. The signal-speeds and interconnect-density involved lead to difficult electrical design problems, particularly regarding signal-integrity issues.

Various commercial Electronic Design Automation programs are now available to address these problems. These include 3-D full-wave electromagnetic-field solvers, SPICE-based circuit-simulation programs and printed circuit board signal-integrity point products. We will show how these seemingly disparate tools can be used in a complementary fashion to provide detailed studies of detector-electronic design. Two case studies will be presented from LHC experiments.

Summary:

This report shows how various EDA tools can be used for high-speed digital design. These will be classified into three main groups: electromagnetic field calculation, circuit simulation and PCB analysis. We will highlight how each is best suited for a particular class of problem.

Field calculation programs are used when a very detailed behaviour of the system is needed. These can be applied to several critical aspects of high-speed electronics design - connectors, cables and packaging - and will provide the most comprehensive model information. The tools directly solve Maxwells equations for a given 3D (or uniform 2D) structure and a set of boundary conditions. Two distinct methods are used to solve these problems. Pseudo-static codes are used to solve structures whose dimensions are much larger than the wavelength considered. When the structure dimensions are comparable to, or less than the considered wavelength, a full-wave code has to be used with a corresponding increased simulation time. CERN has available the set of Ansoft tools (Maxwell 2D/3D Field Simulator and HFSS) and LC from Cray Research.

The output from the field-solver tools can be used as models for SPICE-based circuit-simulation programs which allows much faster analyses in the time and frequency domains. We have made extensive use of the PSpice simulator during our investigations.

Signal-integrity analysis for a PCB presents a different kind of problem. Here, possibly thousands of signals have to be examined and a full 3D-analysis would lead to impracticably long simulation times. However, simplified models still provide extremely useful what-if analysis in the pre-layout phase as well as the possibility of highlighting possible signal integrity violations at the post-layout stage. This approach has the advantage that these programs can be very well integrated with traditional design tools. All calculations are made from the board layout itself and automatically include effects due to track widths, dielectrics and board stackup. The PCB layout itself can also be driven by a set of design constraint rules. CERN has available the SpecctraQuest programs which are fully integrated with our Cadence PCB tools.

Two case studies will be presented in this paper. The first examines the ALICE Pixel Backplane where it has been proposed (ref) to use a meshed power and ground plane for the detector PCB. This has been analysed while considering two opposing constraints - the PCB has to be as transparent as possible to the beam while still retaining sufficient signal and power-supply integrity.

The second example considers a cable design for ALICE's Time Projection Chamber (ref ). Here, crosstalk calculations were made while respecting the required cable mechanical properties.

Id: 98
Corresponding Author: John EVANS
Experiment: General Interest
Sub-system: General Interest
Topic: Grounding Shielding Cooling And Alignment

Minimizing crosstalk in a high-speed cable-connector assembly

B.J. Evans
E. Calvo Giraldo
T. Motos Lopez

CERN, IT/CE

Abstract

This paper presents the detailed signal-integrity analysis results of a connector-cable assembly linking the ALICE Time Projection Chamber (TPC) to its Front-End Electronics.

The goal was to minimize the crosstalk (electromagnetic coupling) between signal lines for a given line to ground capacitance. Both mechanical (cable flexibility and strength) and electrical (fast signal rise-times) design constraints were considered.

The design was analysed using Finite Element Method software tools to extract equivalent circuit models for the connector and cable. We will show how these programs helped us to quickly investigate different cable configurations. The resulting PSpice simulations will be presented.