Ciao Antonio,

Thanks for your mail. I would say we don't get have everything you mention 100% covered, but I'll go through and link what we have.

1) The overall installation plan, in terms of what goes where, is maintained by Markus in a spreadsheet on sharepoint:

https://espace.cern.ch/ATLAS-FELIX/_layouts/15/WopiFrame.aspx?sourcedoc=/ATLAS-FELIX/Shared%20Documents/Procurement%20and%20installation/USA15%20space%20planning/ReadOut%20racks%20lay-out.xlsx&action=default

(Some elements, such as hosting of TTC hardware and associated cable routing, are in the hands of the subdetectors rather than FELIX)

2) At the assembly house (PRIME) a series of tests are carried out before qualifying the boards for shipment to CERN. There are specified in the following documents:

a) Appendix B of the BNL-712 manual, specifying low level tests:

https://espace.cern.ch/ATLAS-FELIX/Shared%20Documents/Procurement%20and%20installation/FLX-711%20cards/FLX-711%20IT/Technical%20files/FLX-712%20Test%20Procedure/Documents/BNL-712_manual.pdf

b) Higher level test manual:

https://espace.cern.ch/ATLAS-FELIX/_layouts/15/WopiFrame.aspx?sourcedoc=/ATLAS-FELIX/Shared%20Documents/Procurement%20and%20installation/FLX-711%20cards/FLX-711%20IT/Technical%20files/FLX-712%20Test%20Procedure/Documents/Test_procedure_FLX712_new.odt&action=default


3) Once the boards arrive at CERN they are subjected to another series of functional + stress tests, defined here:

https://espace.cern.ch/ATLAS-FELIX/_layouts/15/WopiFrame.aspx?sourcedoc=/ATLAS-FELIX/Shared%20Documents/Procurement%20and%20installation/FLX-711%20cards/FLX-712-bulk-production-testing.docx&action=default

-> these details need to be refined somewhat, which a reference configuration for the test server added. This is currently in progress as we ramp up and train the testing team. At this point we will also be able to create a more detailed test schedule.

4) Finally, for installation at P1, we need to request slots in agreement with each subdetector team to do the work. We will start on this once we have fully qualified cards and servers.

Please don't hesitate to ask any questions or make suggestions. For the later stages things are still fluid, so if you think there are extra tests we should go please let us know. On the specific case of the dirty fibre test, I don't know if any such test was done at the design stage, but if this has proven useful in the past for LHCb we could certainly consider it.

Many thanks

Will



-----Original Message-----
From: Antonio Pellegrino <antonio@nikhef.nl>
Sent: 20 January 2020 14:31
To: William Panduro Vazquez <j.panduro.vazquez@cern.ch>
Cc: Joseph Vermeulen <i73@nikhef.nl>
Subject: Installation and Commissioning procedures


Ciao Wil (Cc: Jos),

I would like to get more insight on the organization of the installation and commissioning for Phase I, things whether tests are done at intermediate places/phases (e.g. any electrical test prior to installation in servers, tests FLX+servers prior to installation at Point 1, etc.), organization of tests (people responsible for instructions and schedule etc.), interaction with infrastructure (e.g. inserting a dirty fibers for tests can induce damage to optical transceivers), etc.

Sorry, lots of questions (I have some experience from LHCb of how difficult this is, but little knowledge of how we are gonna tackle this in Felix), if you have no time to point me to the documentation, I will try to make a first iteration with Carlo,

thanks,
sincerely
Antonio

********************************************************************************
Antonio Pellegrino (Nikhef and Rijksuniversiteit Groningen) c/o NIKHEF, P.O.Box 41882
1009 DB Amsterdam - The Netherlands
email: antonio@nikhef.nl
Cell. NL +31 6 46 81 20 02
Cell. CH +41 75 411 2572
Car NL +31 6 10 45 01 43
Tel. Nikhef +31 20 592 5002
FAX +31 20 592 5155
********************************************************************************