Integration of Band 5 receivers at ALMA/JAO. Giorgio Siringo Senior RF Engineer & FrontEnd Technical Lead - JAO on behalf of the B5 integration team

Transcription

1 Integration of Band 5 receivers at ALMA/JAO Giorgio Siringo Senior RF Engineer & FrontEnd Technical Lead - JAO on behalf of the B5 integration team

2 Who is Who B5 Management ESO Project Manager: Pavel Yagoubov ESO Integration Manager at JAO: Gianni Marconi NA Development Program Manager: William Randolph

3 Who is Who Who designed the CCA: Pre-production: GARD, Sweden Production: GARD + NOVA Who delivers the CCA: GARD+NOVA for ESO Who delivers the WCA (and cold multipliers): NRAO Technology Center (CV)

4 Who is Who Several people involved at GARD/Chalmers, at NOVA, at ESO, at NRAO......and at JAO: 2 electronics engineers (50%) 3 technician (100%) workforce distributed in the two shifts supervised by Gianni Marconi and myself

5 B5 Receiver B5 project: just a bit of history Pre-production design started in 2005 at Chalmers (GARD) 6 cartridges were delivered to JAO Tested by SIST in 2012 Production project started in 2013 Some important changes ASTE, APEX-SEPIA, LLAMA

6 B5 Receiver 2012: Astro-holography at 209 GHz DV23xDA51 (at the OSF) AIV DV23

7 B5 Receiver 2012: Astro-holography at 209 GHz DV23xDA51 (at the OSF) AIV DA51

8 B5 Receiver pre-production design

9 B5 Receiver production design YTO range: GHz LO driver: GHz WCA multiplier: Modified to 6x CCA multipliers: 2x at 110 K Extended Sky frequency range: (158) GHz

10 Integration The integration process starts before shipment: the cartridges are carefully tested by the manufacturer before shipment (the CCAs cold in a test cryostat at NOVA, the WCAs at NTC) in a process called Preliminary Acceptance In-house (PAI) The results go in a report uploaded to the EDM, globally available within the ALMA consortium. We have done 70 WCAs and 44 CCAs PAI so far.

11 Integration When non-compliances are detected, these are evaluated case by case. Minor non-compliances are usually waived, typical CCA waivers for: image rejection ratio (14), cross-talk (11), gain compression (2), IF power variations (27) Major non-compliances would result in blocking the shipment.

12 Integration If all PAI results are acceptable and eventual waivers approved, the cartridge is shipped to JAO/OSF. When we receive it, we do a new series of tests at room temperature, the so-called Provisional Acceptance on-site (PAS), mainly to certify integrity after shipment by cross-check with PAI results.

13 Integration The cartridges (a CCA + WCA pair) are then integrated in a suitable Front-End (FE) (when available...) The integration process is not limited to the installation of the two cartridges only, but also of the window, IR filter, band-specific power supply module (CPDS), M&C configuration parameters (CCA thermometers calibration, optimal bias voltages and currents of the SIS mixers and LNA amplifiers,...)

14 Integration The cartridges (a CCA + WCA pair) are then integrated in a suitable Front-End (FE) Before cooling down, another warm health check is done to verify that the hardware survived the integration process. After cool down, a set of verification tests are done on the antenna simulator (FETMS) and optimization based on parameters provided by NOVA, for 3.5 K (FE operation) and 4.0 K (specifications) reference temperatures. (FETMS=FE Test & Measurement System)

15 Integration The cartridges (a CCA + WCA pair) are then integrated in a suitable Front-End (FE) after-integration verification on the FETMS (at OSF): vacuum leak check at CCA and window; LO waveguide mapping; warm health check; WCA locking tests at warm; cool-down and cryogenics performance verification; cold health check (did it survive cool down?); workmanship verification (elevation dip, lose cables?); noise noise temperature check/optimization; (Cycle5!) IF output power level and variations; beam pattern verification (including polarization efficiency)

16 Integration: FETMS The cartridges (a CCA + WCA pair) are then integrated in a suitable Front-End (FE)

17 Integration: FETMS The cartridges (a CCA + WCA pair) are then integrated in a suitable Front-End (FE)

18 In antenna verification The FE + integrated B5 cartridges is then installed on a suitable antenna element In antenna engineering verification (at AOS): signals-path connectivity receiver noise temperature verification/optimization amplitude stability (Allan variance) ambient load elevation dip azimuth slew

19 In antenna verification The FE + integrated B5 cartridges is then installed on a suitable antenna element In antenna engineering verification (at AOS)

20 In antenna verification The FE + integrated B5 cartridges is then installed on a suitable antenna element In antenna engineering verification (at AOS)

21 Integration Progress Requirements for Cycle5 (October 2017): 43x12m + 10x7m + 3xTP total: 56 AE/FE blocker: 41x12m +9x7m + 3xTP = 53 (not yet official - ALMA Technical Handbook) Already integrated: 32 / FEs to go: 38 We keep track of the progress in several places: FETMS database, JIRA, CMMS, EDM and two twiki pages: one for the pre-acceptance (FETMS/OSF) the other for the on-sky verification (AE/AOS)

22 Integration Progress We keep track in FETMS database, JIRA, CMMS, EDM. We also have two twiki pages to track activities. If pre-accepted move to antenna at the AOS

23 Integration Progress 32 FEs integrated (as of January 2017) 28 already fully tested on sky

24 On sky verification DV01 / FE61 / CCA5-08 spectral check: Trx Tsys

25 On sky verification

26 On sky verification Birdie at GHz in both polarizations associated to the 12th harmonic of WVR LO drive signal at

27 On sky verification Water line detection in Orion OMC 1

28 On sky verification Science verification projects: VY Cma: H2O and SiO, polarization capabilities demonstrated Arp 220: H2O, CS and HNC observations Sgr B2: full band spectral scan from 163 to 211 GHz ( Science verification program already completed between June and November 2016 topic of next talk by Liz Humphreys

29 Ancillary story #1 Some problems with cryogenics are slowing down the integration rate We are facing unexpected difficulties with the cryogenics, we switched from planned preventive maintenance (FE maintenance every hours ~ 1 year of operation) to condition-based maintenance = we only take a FE out of an AE if the cryogenics performance is bad impossibility to integrate new cartridges! (with consequences on the short and long term)

30 Ancillary story #1

31 Ancillary story #2 We found that the ALMA software was not able to optimize the LO power for B5 receivers We then found that actually the problem was not limited to B5 but other bands (in particular B10) were also affected: problem unnoticed until the introduction of B5. We are now testing a new algorithm based on probabilistic bisectioning (Baesian approach) going to be used for all bands, using mostly B5 for testing and fine-tuning of some parameters

32 Integration Enjoy ALMA Band 5 receiver!

33 Integration We keep track in FETMS database, JIRA, CMMS, EDM. We have some other tools...

34 Integration We keep track in FETMS database, JIRA, CMMS, EDM. We have some other tools...