From Philippe.Farthouat@cern.ch Sun Apr 5 03:00:53 2020 Date: Sun, 5 Apr 2020 07:00:42 +0000 From: Philippe Farthouat To: Marco Verzocchi Cc: Steve Herbert Kettell , Philippe Farthouat Subject: Meetingg on Tuesday Dear Marco, Thanks for your answers which address mainly all the points. Please find below a few questions we would like to be clarified during our meeting on Tuesday.  - We understand that the intermediate WIB can work with the new PTC. This requires a change of firmware to have the WIB driving the I2C lines. Has this modification already been implemented ? - The interface PTC/SC has still a lot of options opened. When and how the decisions will be made ? Will it be during the Fall with the new requirement document ? - We understand there is an engineer in charge of coordinating the firmware/software work. Could we have details about the process to be put in place ? Will there be a mechanism to remotely update PL firmware ( “Gateware” ) and PS firmware ? Will there be a mechanism to remotely recover the WIB if the firmware becomes corrupted and renders the board inoperable ? - The DUNE-SP timing system periodically interrogates the status of every timing endpoint in the system. The endpoint must be able to turn on and off its return signal to the timing master when commanded by the timing master. Is it envisaged that in normal operation all five WIBs in a WIEC will function as timing endpoints? Or will the PTC be the single endpoint in the WIEC ? - Lines in the WIEC will be configured as an I2C  “backplane” . If the WIBs are timing endpoints will they be able to be I2C masters and control the single optical transmitter if they are interrogated by the timing master? How long will it take for these I2C transactions to complete and the transmitter to be activated ? - A minor question: Every timing endpoint in the system must have  unique identifier. Is it anticipated that this will be the unique hardware identifier associated with the PCB ( if I understand correctly from response (14) each board will have ROM with a unique ID ) or will it be associated with geography ( slot, crate number )? - It looks like p17of https://indico.fnal.gov/event/22807/contribution/6/material/slides/0.pdf has the wrong text. What text was intended to be there? - Concerning the BER, in your answer you mention that a BER of  10^-12 leads to 0.44% down time. How do you get this number when at most you loose 10^6 samples per day I.e. 0.5 second per day ? - The question of the interlock is still a bit unclear. For instance, disabling the power of a FEMB could be locally decided without requiring communication with DDSS etc. Would it be possible to have a diagram showing the interlock generation ? - Could you clarify the schedule for getting all the interface and requirements documents finalised ? As well as a revised schedule for the next steps of review and start of production ? - Finally a remark concerning Grounding: it may be more valuable to use a signal generator that can produce radiated and conducted noise at the maximum possible level permitted by 150 PCs  for  the tests with ProtoDUNE. Cheers,  Philippe _______________________________________________________ Philippe Farthouat   EP Department Tel:  +41 (0)22 767 6221 CERN Mobile: +41 (0)75 411 5318 CH-1211 Geneva 23 Email. Philippe.Farthouat@cern.ch  Switzerland _______________________________________________________