ATCA Crate Infrastructure in place Rack in D

  • Slides: 11
Download presentation
ATCA Crate / Infrastructure in place - Rack in D 3: dimension - Electrical

ATCA Crate / Infrastructure in place - Rack in D 3: dimension - Electrical power availability - Tell 1 racks running in D 3 (air ventilation, safety aspect) - Temperatures, power of crates in D 3 - Conclusion 13 Octobre 2011 Laurent Roy 1

Rack in D 3 dimensions: 2 m 54 x 60 cm x 90 cm

Rack in D 3 dimensions: 2 m 54 x 60 cm x 90 cm - Useful space : height 56 U (2 m 49) ~10 U need to be free at the bottom for cabling and water valve access. width 19’’ (48, 3 cm) depth ~73 cm - Level 1 crates : 11 U (9 U+2 U Fan) x 19” x ~72 cm - ATCA crate (ex: ELMA catalogue) dimensions OK, few mechanical adaptation (26 U instead of 22 U used for 2 crates). 13 Octobre 2011 Laurent Roy 2

Electrical Power for Frond End racks in D 3 Maximum available: ~ 410 k.

Electrical Power for Frond End racks in D 3 Maximum available: ~ 410 k. W Consumption total: ~ 70 k. W (in ‘Physics’ mode) No problem with ‘general power’ (17% of the capacity used) 13 Octobre 2011 Laurent Roy 3

- Electrical power available per rack distribution designed for 11 k. W/rack max. 16

- Electrical power available per rack distribution designed for 11 k. W/rack max. 16 A circuit breakers (Merlin Gerin C 60 H – D 16 A). 7 single phase power outlets available (burndy connector). - ATCA crate consumption: ? ? ex. ELMA catalogue: Output PSU 2799 watt maxi -> input (80%) ~ 3. 5 k. W maxi = ~ 16 A circuit breaker should be Ok (to be check with the manufacturers) 13 Octobre 2011 Laurent Roy 4

- Cooling: vertical air flow mixed water exchanger Air temperature inside the rack with

- Cooling: vertical air flow mixed water exchanger Air temperature inside the rack with 2 exchangers: ~ 18 -19 ˚C - Safety aspect (DSS): Thermo switch: 40 ˚C (increase to 60 ˚C possible), Smoke detection, CO 2 automatic triggering. ATCA Crate need to be adapted (by the company? ) to be compatible with vertical air flow. water: 16 ˚C 13 October 2011 Laurent Roy 5

24 Tell 1+L 1 crates location / 8 racks spare in D 2 ‘Detector

24 Tell 1+L 1 crates location / 8 racks spare in D 2 ‘Detector Extension’ row enough space Air Temperature in Rack (‘Physic mode’): 13 Octobre 2011 Laurent Roy 6

Cooling / power dissipated - Tell 1 crate Velo: 17 Tell 1+2 Hugin power

Cooling / power dissipated - Tell 1 crate Velo: 17 Tell 1+2 Hugin power ~820 watts / 865 watts in ‘Physics’ -> total 2 crates = ~ 1. 7 k. W/rack ( Temp inside rack 19˚C ) Other example: TT 16 Tell 1 + 1 Hugin ~645 watts / 705 watts in ‘Physics’ - ATCA crate: maxi 14 boards x 250 W = ~3. 5 k. W ->total 2 crates = ~7. 0 k. W maxi /rack (4 times more !) Tests done in 2002 by PH electronic group (3 exchangers, 3 crates): http: //cdsweb. cern. ch/record/38031/files/epess-2003 -014. pdf Conclusion: “ The air-water rack cooling system developed for LEP and used for the measurements in this report can cope with a total power dissipation of the order of 10 k. W with less than 2% power loss to the surroundings. This corresponds to a useful power dissipation per subrack of 2. 3 k. W. In these conditions most of the electronics is below 40 ˚C ” = total 6. 9 k. W for 3 crates Cooling in place should be Ok (but 40˚C inside? ) , maybe a third exchanger need to be installed, need to be check more in detail, tests ? … 13 Octobre 2011 Laurent Roy 7

General power summary Total Electrical power: (Level 1+LV+HV+…) 70 k. W Heat removed by

General power summary Total Electrical power: (Level 1+LV+HV+…) 70 k. W Heat removed by water: Temp input: 16 ˚C Temp output: ~ +0. 5 ˚C Flow ~ 60 m 3/h ~ 35 k. W Rack Temp inside: 18 -19 ˚C Heat ‘lost’ in room removed by air conditioning units Air temperature in the rooms: ~ 20. 5 ˚C ~ few k. W (D 3 A unit capacity: 26 k. W maximum) (D 3 B unit capacity: 32 k. W maximum) Dissipation in detector side ~ 30 k. W ? 13 Octobre 2011 Laurent Roy 8

General power summary Total Electrical power: (Level 1+LV+HV+…) 70 k. W Heat removed by

General power summary Total Electrical power: (Level 1+LV+HV+…) 70 k. W Heat removed by water: Temp input: 16 ˚C Temp output: ~ +0. 5 ˚C Flow ~ 60 m 3/h ~ 35 k. W (+ 30 -70 k. W with ACTA) Rack Temp inside: 18 -19 ˚C Heat ‘lost’ in room removed by air conditioning units Air temperature in the rooms: ~ 20. 5 ˚C ~ few k. W (few additional) Remove L 1 crates ~ -12 k. W Add ATCA crates ~ +40 -80 k. W ? (depend on boards quantity) Total ~ 30 -70 k. W ? additional (D 3 A unit capacity: 26 k. W maximum) (D 3 B unit capacity: 32 k. W maximum) Dissipation in detector side ~ 30 k. W ? Maxi ~ 105 k. W need to be removed with water = output + 1. 5 ˚C (instead of 0. 5 ˚C) Ok with general cooling (ex. + 4˚C for PC farm in ‘Physics mode’) 13 Octobre 2011 Laurent Roy 9

CONCLUSION: - ATCA crate 19 ’’ dimensions compatible with Rack used few mechanical adaptation

CONCLUSION: - ATCA crate 19 ’’ dimensions compatible with Rack used few mechanical adaptation needed 13 U instead of 11 U - General electrical distribution already in place enough - Maximum breaker in place 16 A (but could be changed if needed) - Crate should be adapted for vertical air cooling - Heating power to be evacuated per rack ~ 4 times more / current L 1 rack need to be check with more detail (boards quantity/crate) installation of additional 2 U exchanger ? - General cooling power enough - Rk: cabling arrangement should be also taken into account early 13 Octobre 2011 Laurent Roy 10

spare slide 13 Octobre 2011 Laurent Roy 11

spare slide 13 Octobre 2011 Laurent Roy 11