Energy Distribution in Hostile Environment Power Converters and

Energy Distribution in Hostile Environment: Power Converters and Devices Mauro Citterio on behalf of the INFN-APOLLO project Mauro Citterio ICATPP Como – 10/4/2011

INDEX • The ATLAS LAr Calorimeter System …. a test case • The Proposed Power Distribution for an Upgraded LAr System • Characteristics of Power MOSFETs under irradiation • - exposed to ionizing radiation (gamma 60 Co) • - exposed to heavy ions (75 Br at 155 Me. V) • - exposed to protons (216 Me. V) • Conclusions Mauro Citterio ICATPP Como – 10/4/2011 2

The ATLAS experiment LAr barrel calorimeter The power distribution and conversion scheme in the detector area Mauro Citterio ICATPP Como – 10/4/2011 3

ATLAS Experiment: Lar Barrel Calorimeter Details of the Front End Electronics and Main Power Converter The required qualification doses for this application are: 4. 5 x 104 rad and 2 x 1012 particles/cm 2 (> 20 Me. V) Ten times higher for Hi-LHC scenario (70 safety factor)!!! Mauro Citterio ICATPP Como – 10/4/2011 4

ATLAS Experiment: Present Status LAr Calorimeter Front-End Board (FEB) Power Distribution 19 LDO regulators/FEB Mauro Citterio ICATPP Como – 10/4/2011 5

Proposed Power Supply Distribution Scheme for a LAr Upgrade MORE INFO TAKE A LOOK AT THE DEDICATED POSTER !!! CRATE Card #3 Card #2 POL LDO Converter Card #1 POL LDO Converter LDO 280 Vdc ni. POL Converter. LDO Main DC/DC Converter POL POL Converter LDO Converter POL ni. POL Converter LDO Converter Regulated DC bus ni. POL Converter POL POL Converter with high step-down ratio Characteristics: • • • Main isolated converter with N+1 redundancy High DC bus voltage (12 V or other) Distributed Non-Isolated Point of Load Converters (ni. POL) with high step-down ratio Mauro Citterio ICATPP Como – 10/4/2011 6

Critical Elements for a LAr Upgrades The Main Converter The Point of Load C+ 4 C+ V 3 in C+ 2 C+ 1 L T i. L 1 S S L 4 1 2 1 Q T C 3 3 o + V - uto Uin Q T i. T 2 Q T 2 2 Q 1 C L R Uo - o 2 S 4 3 Vout = 12 V - Required Mosfet Voltage Breakdown: ~ 200 Volt or higher - Mosfets, diodes and controller must be qualified against radiation D<50% Uo = Uin. D/2 POL Specifications: Switched In Line Converter SILC Mauro Citterio + S C UC 4 -1 1 + Input voltage: Output current: Operating frequency: Ug = 12 V Uo = 2. 5 V Io = 3 A fs = 1 MHz 350 n. H air core inductors ICATPP Como – 10/4/2011 7

Power Mosfets exposed to gamma rays Devices under test: 30 V STP 80 NF 03 L-04 30 V LR 7843 200 V IRF 630 Used doses: For each type of device 20 samples were tested, 5 for each dose value (tested at the ENEA Calliope Test Facility) Measurements : Breakdown Voltage @ VGS=-10 V I 1600 Gray Threshold Voltage @ VDS=5 V II 3200 Gray ON Characteristic @ VGS=10 V III 5890 Gray Gate Leakage @ VDS=10 V IV 9600 Gray Mauro Citterio ICATPP Como – 10/4/2011 8

30 V Mosfet: STP 80 NF 03 L-04 Mauro Citterio ICATPP Como – 10/4/2011

30 V Mosfet: LR 7843 Mauro Citterio ICATPP Como – 10/4/2011 10

200 V Mosfet: IRF 630 Mauro Citterio ICATPP Como – 10/4/2011 11

Mosfet Exposed to Heavy Ions. The SEE framework Destructive Single Event Effects in Power MOSFETS (tested at INFN Catania) Source Gate Source Body Gate Body + N P P _ + N + _ _ + P _ N N P N + Drain Single Event Burnout Single Event Gate Rupture Mauro Citterio ICATPP Como – 10/4/2011 12

The SEE experimental set-up The IGSS evolution during irradiation Source Vgs Gate Vds Parameter Analyzer 1 MW Body 1 MW Cd + N P _ 50 W P Cg 50 W The current DUT pulses Impacting Ion + _ N N Mauro Citterio + Drain ICATPP Como – 10/4/2011 Fast Sampling Oscilloscope 13

The SEE analysis TIME DOMAIN WAVEFORMS SCATTER PLOT NUMERICAL INTEGRATION Γ-LIKE DISTRIBUTION FUNCTION PARAMETERS EXTRACTION MEAN CHARGE vs BIAS VOLTAGE Mauro Citterio Γ-LIKE DISTRIBUTION FUNCTION ICATPP Como – 10/4/2011 14

The SEE experimental results 200 V Mosfet: IRF 630 Devise TID D 21 D 22 D 06 D 10 D 14 D 16 D 17 0 Gy 1600 Gy 3200 Gy 5600 Gy 9600 Gy Bias Conditions during Irradiation Vds=20 V-110 V vgs=-2 V Vds=20 V-120 V vgs=-6 V Vds=20 V-70 V vgs=-2 V Vds=20 V-50 V vgs=-6 V Vds=20 V-55 V vgs=-6 V Vds=20 V-50 V vgs=-6 V Vds=20 V-45 V vgs=-6 V Drain Damage Gate Damage Vds=100 V-110 V Vds=110 V-120 V Vds=60 V-70 V Vds=40 V-50 V Vds=50 V-55 V Vds=45 V-50 V Vds=40 V-45 V Vds=100 V-110 V Vds=60 V-70 V Vds=40 V-50 V Vds=40 V-45 V The increase of the ϒ-dose causes a reduction of the critical bias condition at which drain and gate damages appear Mauro Citterio ICATPP Como – 10/4/2011 15

The SEE experimental results D 21 0 Gy Vds=110 V Vgs=-2 V Two different sensitive areas D 21 0 Gy Vds=110 V Vgs=-2 V Mean charge vs Vds Mauro Citterio ICATPP Como – 10/4/2011 The SEB current pulse 16

The SEE experimental results Scatter-plot Vds=50 V The increase of the ϒ-dose causes a widening of the current pulses Mauro Citterio ICATPP Como – 10/4/2011 17

Mosfet Exposed to Protons SEB characterization Characterization requires that an SEB circumvention method be utilized SEB characterization produces a cross-sectional area curve as a function of LET for a fixed VDS and VGS. Generally SEB is not sensitive to changes in the gate bias, VGS. However, the VGS bias shall be sufficient to bias the DUT in an “off” state (a few volts below VTH), allowing for total dose effects that may reduce the VTH. The only difference in the test set -up was that the current probe was on the Mosfet Source Mauro Citterio ICATPP Como – 10/4/2011 18

Mosfet Exposed to Protons The results are still preliminary. Only the 200 V Mosfets (IRF 630, samples from two different manufacturers) were exposed Proton energy: 216 Me. V Ionizing Dose: < 30 Krads (facility at Massachusetts General Hospital, Boston) An “absolute” cross section will require the knowldege of the area of the Mosfet die which is unknown. Mauro Citterio ICATPP Como – 10/4/2011 19

Mosfet Exposed to Protons Work still in progress ……………. . § The number of SEB events recorded at each VDS was small less then 30 events for the ST less than 150 events for the IR devices Large statistical errors affect the measurements § The cross section at VDS = 150 V (“de-rated” operating voltage) can not be properly estimated § Dependence from manufacturer § “Knee” not well defined • To effectively qualify the devices for 10 years of operation at Hi-LHC, the cross section has to be of the order of 10 -17/ cm 2, which puts the failure rate at <1 for 10 years of operation • Proton irradiation campaigns with increased fluences and more samples are planned. Mauro Citterio ICATPP Como – 10/4/2011 20

Conclusions § Distributed Power Architecture has been proposed § Main converter (SILC topology) § Point of load converter (IBDV topology) § Critical selcction of components to proper withstand radiation § Controller, Driver and Isolator § FPGA for overall monitoring § MOSFETS, both for main converter and POL have been selected and tested § Gamma ray § Heavy ions § Protons § Some results are encouraging, however more systematic validation is on-going § Novel devices based on Si. C and Ga. N, are also under investigation Mauro Citterio ICATPP Como – 10/4/2011 21