Minutes of meeting held on 09 Apr 2020
==========================

slides are at:
https://beamdocs.fnal.gov/AD-public/DocDB/ShowDocument?docid=8233

(1) Dave presented what needs to be done before we can do the 20 Hz
tests during the shutdown
      (i) The reason why we want to go to 20 Hz is to increase beam
      power to the experiments. At 20 Hz, g-2 will have 16% more beam
      and mu2e will get 16.5% more beam.
      (ii) We need to create a list of what we are going to test while
      running at 20 Hz and who will be looking at the systems.
      (iii) Dave showed some of the identified issues.
      (iv) Greg Vogel went through the controls identified issues
         (a) TLG --  Right now the test TLG can run with a few events. A
      	   full working system will require 6 racks of hardware. The
      	   UCDT needs to be upgraded to a PowerPC
      	   frontend. Furthermore, we need to have a list of where
      	   these need to be piped to for people to use. The test TLG
      	   system will ony have a subset of the clock events.
	      (1) The 20 Hz cavity test was done using $BA event in
      	           the time line. This is not applicable for frontend
      	           tests because they go off on $0F and will require
      	           more events than just 1.
               (2) A new TLG system will be developed for PIPII for 20
	       Hz.
	   (b) Networks -- traffic will increase on the network for 20
	   Hz. Jim Patrick doesn't think that going to EPICs will
	   reduce the traffic.
	   (c) Datalogging -- Jim Patrick says that datalogging will
	   have to be upgraded.
	   (d) Master clock supports 15 Hz and 20 Hz. Not a problem.
	   (e) ACNET and MOOC frontends. Will need to find out how 20
	   Hz affects them.
	   (f) TCLK/beamsync triggers -- will need to find out how
	   the hardware that depends on these triggers behave.
	   (g) NML and PIPIIIT are still using the 15 Hz clock.
	   (h) Controls apps -- there will need to be a huge effort.
	 (v) Greg Vogel says that in order to do the 20 Hz
	 infrastruture test, we have to have people at the lab. For
	 example TLG programmer, Linden. Recovery from a crashed the
	 controls system is diffcult. There willl be a new test version
	 of D69 so that it knows that it is running at 20 Hz. Rich
	 Neswold will need to upgrade the code on the TLG frontend and
	 has to be at the lab to do the upgrade.
	     (a) However, Greg thinks that they should be ready with
	     the TLG frontend that is 20 Hz ready by September.
	 (vi)  Jim Patrick says that we need to prioritize what to
	 look at. He says IRMs, MOOC frontends will be upgraded so
	 don't need to worry about them.
	 (vii) Craig Drennan says that BPM and BLM systems will need
	 to be replaced. He says it will cost $14M
	    (a) MI and Recycle EchoTek's will have to be replaced.
	    (b) Booster BPM ethernet will have to upgraded.
	    (c) Booster BLMs have to be upgraded (maybe upgraded
	    IRMs?) Code is done by Bob Goodwin.
	 (viii) Tan says that we have to generate 2 lists:
	 	(a) Distribution of test TCLK signal to the correct
		places.
		(b) A big spreadsheet that has better granularity of
		what systems that will be deprecated, upgraded
		etc. and costs for 20 Hz.
		(c) Dave, Tan and Paul Derwent will get together to
		generate the lists.