October 2, 2013 11:28 WSPC - Proceedings Trim Size: 9in x 6in Final˙Paper

1

POWER SUPPLY SYSTEM FOR THE ATLAS

EXPERIMENT: DESIGN SPECIFICATIONS,

IMPLEMENTATION, TEST AND FIRST RESULTS∗

M. LAZZARONI∗∗

Dipartimento di Fisica, Universita degli Studi di Milano

Via Celoria, 16 - 20133 Milan, Italy

andI.N.F.N. Milano,

Via Celoria, 16 - 20133 Milan, Italy∗∗E-mail: massimo.lazzaroni@unimi.it

www.unimi.it

M. CITTERIO and S. LATORRE

I.N.F.N. Milano,

Via Celoria, 16 - 20133 Milan, Italy

A. LANZA

I.N.F.N. Pavia,

Via Bassi, 6 - 27100 Pavia, Italy

P. COVA and N. DELMONTE

Dipartimento di Ingegneria Informazione, University of Parma

Viale G.P. Usberti 181/A - 43124 Parma, Italyand

I.N.F.N. Pavia,Via Bassi, 6 - 27100 Pavia, Italy

F. GIULIANI

Dipartimento di Ingegneria Informazione, University of ParmaViale G.P. Usberti 181/A - 43124 Parma, Italy

The planned upgrade of instrumentation sensitivity in the ATLAS experiment

of the Large Hadron Collider (LHC), at CERN, calls for a new type of power

distribution architecture. Moreover, power supplies require DC-DC power con-

∗This work is supported by I.N.F.N. Italy Apollo collaboration.

October 2, 2013 11:28 WSPC - Proceedings Trim Size: 9in x 6in Final˙Paper

2

verters able to work in very hostile environment and maintaining high level ofReliability, Availability, Maintainability and Safety (denoted as RAMS require-

ments) during the experimental activity. Two main issues need to be discussed:

first, electronic devices and equipment must operate in very high backgroundof both charged and neutral particles and high static magnetic field and, sec-

ond, the increase of the radiation background and the requirements of new

front-end electronics are indeed incompatible with the current capability ofthe actual distribution system. The APOLLO R&D collaboration, funded by

the Italian Istituto Nazionale di Fisica Nucleare (INFN), aims to study dedi-

cated topologies of both distribution system and DC-DC power converters andto design, build and test demonstrators, developing the needed technology for

the industrialization phase. The collaboration has designed a 3kW, 280V-12V

converter (MC) based on the Switch in Line architecture (SIL), a DC to DCphase-shifted converter characterized by a disposition in line of the MOSFETs

with good soft switching performances, and in the last year many steps have

been taken to enhance the power dissipation and the reliability and to im-prove the general features of the designed converter. In particular a new water

heat sink was designed on the basis of TFD simulation accounting for the lay-out of the specific converter. Experimental activities in order to characterize

both thermal and electrical features of the MC confirm the correctness of the

adopted design criteria.

Keywords: Power Supply; Power Converter; Reliability; LHC.

1. Introduction

The Large Hadron Collider (LHC) is, nowadays, the world’s largest and

highest-energy particle operative accelerator, built by the European Or-

ganization for Nuclear Research (CERN) able to extends the frontiers of

particle physics thanks to its high energy and luminosity. The Phase-2 up-

grade of the LHC is planned to take place during the long shutdown of the

year 2022 when the detectors will be upgraded or substituted. An increase

of the radiation background will cause the accumulation of a higher Total

Ionizing Dose (TID). Furthermore a significant increase of power demand of

the detectors implies a complete re-design of the power distribution system,

water cooling and very stringent thermal constraints1 in order to maximize

both performances and RAMS requirements.2,3

2. The Main Converter

The Main Converter is based on a Switch in Line Converter (SIL) topol-

ogy where the circuit consists in a series connected half bridge converter as

described in previous papers.4,5 The first, of three in 2oo3 redundancy con-

figuration, implemented prototype able to deliver 1.5 kW i.e. 125 A at 12 V

is depicted in Fig. 1 (parallel connection is possible because SILC topology

acts like a controlled current generator).

October 2, 2013 11:28 WSPC - Proceedings Trim Size: 9in x 6in Final˙Paper

3

Fig. 1. MainConverter Fig. 2. Cold Plate

In order to improve the performance and the reliability of the con-

verter in the last year the prototype has been deeply modify: the planar

transformer PCBs has been replaced with copper windings separated with

Kapton sheets, a specific cold plate has been designed, simulated, tested

and manufactured (Fig. 2), the thermal coupling between the transformer

and the cold plate has been re-designed and improved, the connections to

the external power cables have been improved and, finally, the auxiliary

power supply and the controller have been replaced. In particular, the cold

plate has been desinged taking into account the necessity of dissipation in

proximity of single devices in such a way to have a power dissipation that

Table 1. Module devices: description and dissi-

pated power (output power: 1.2 kW).

N Device Pd

(W)

1 Planar transformer - core 130

2 Planar transformer - windings 70

3 Diodes (ISOTOP) 104 Primary - MOSFETs (TO247) 18

5 Primary - inductor core 4

6 Primary - inductor windings 37 Auxiliary - MOSFET (TO247) 0.2

8 Auxiliary - MOSFET (D2PAK) 0.2

9 Auxiliary - transformer core 0.510 Auxiliary - transformer windings 0.5

11 Capacitors and other devices 4.512 Copper traces at secondary 5.5

Total power dissipation 246.4

October 2, 2013 11:28 WSPC - Proceedings Trim Size: 9in x 6in Final˙Paper

4

Fig. 3. The utilized evaluation board. Fig. 4. The tested demo-board.

fit locally with the installed component (Table 1). The test results have led

to the adoption of a sink to a 2C shaped cold plate where the maximum

experimentally measured temperatures at 1.5 kW of delivered power was

108 C.

3. Magnetic field test of a commercial DC-DC converter

An integrated DC-DC step-down converter by Linear Technologies

(LTM4619) has been tested in order to check its magnetic field tolerance.

Test have been carried out at the LASA, a center for research and devel-

opment in the fields of accelerators and of superconductivity of University

of Milan and INFN. The tested evaluation board is shown in the follow-

ing Fig. 3. The tests have been carried out changing the position of the

evaluation board in order to apply the magnetic field in different directions

(Fig. 3). The experimental setup is depicted in Fig. 4.

The two outputs of the evaluation board are both configured to provide

an output voltage of 1.8 V on a nominal load of 0.6 Ω. The current supplied

is therefore equal to 3 A per output, 5.4 W. The board power input was

supplied at 20 V.

It was also added a heat sink which enabled to maintain the module at

approximately 35 C. This allows the device to work as much as possible at

the conditions of nominal efficiency. The tests have been carried out with

the applied magnetic field in three directions.

The best performance was obtained with the magnetic field orthogonal

to the device plane, up to about 0.4 T. The worst performance was observed

with the magnetic field parallel to the device and with direction right to

left, as depicted in Fig. 5.

October 2, 2013 11:28 WSPC - Proceedings Trim Size: 9in x 6in Final˙Paper

5

Fig. 5. Experimental results.

4. Conclusions

The designed prototype has proved capable of providing the performance

required by the design specifications. Further tests will be conducted with

the aim to verify both the radiation and magnetic field effects. Some results

on a commercial POL which showed a good behavior until 0.2− 0.4 T have

been also presented.

References

1. P. Tenti, G. Spiazzi, S. Buso, M. Riva, P. Maranesi, F. Belloni, P. Cova,R. Menozzi, N. Delmonte, M. Bernardoni, F. Iannuzzo, G. Busatto, A. Porzio,F. Velardi, A. Lanza, M. Citterio and C. Meroni, Journ. of Instrumentation6 (2011).

2. M. Lazzaroni, L. Cristaldi, L. Peretto, P. Rinaldi and M. Catelani, Reliabilityengineering: basic concepts and applications in ICT (ISBN 978-3-642-20982-6,e-ISBN 978-3-642-20983-3, DOI: 10.1007/978-3-642-20983-3, Springer, BerlinHeidelberg, 2011).

3. P. Cova, N. Delmonte, F. Giuliani, M. Citterio, S. Latorre, A. Lanza andM. Lazzaroni, Thermal modeling and characterization of power converters forlhc power supplies, in Proc. of 12th IMEKO TC10, Workshop on TechnicalDiagnostics, 202–207 , (Florence, Italy, june, 2013).

4. M. Alderighi, M. Citterio, S. Latorre, M. Riva, P. Cova, N. Delmonte,A. Lanza, M. Bernardoni, R. Menozzi, A. Costabeber, A. Paccagnella, F. Sichi-rollo, G. Spiazzi, M. Stellini, P. Tenti, S. Baccaro, F. Iannuzzo, A. Sanseverino,G. Busatto and V. D. Luca, Power distribution architecture for high energyphysic hostile environment, in Proc. of 13th ICATPP Conference, 890–894 ,(Villa Olmo, Como, Italy, october, 2011).

5. M. Alderighi, M. Citterio, M. Riva, S. Latorre, A. Costabeber, A. Paccagnella,F. Sichirollo, G. Spiazzi, M. Stellini, P. Tenti, P. Cova, N. Delmonte,A. Lanza, M. Bernardoni, R. Menozzi, S. Baccaro, F. Iannuzzo, A. Sansev-erino, G. Busatto, V. D. Luca and F. Velardi, Journ. of Instrumentation 7(2012).