Hub SN30 Trailer Addendum

17 JUN 2020
- When SN30 went into the 14SS, it ran for a period of time in Slot #2,
  but somewhere over that period, the SW_1V2 supply "died". 
- After-the-fact measurements showed SW_1V2 was only making roughly
  600mV and around 300mA current.
- Dan recommended that I verify and/or just change out R1202 and C1201.
- R1202 gave what seemed a reasonable measurement in-circuit - about 15K,
  and read 20K out-of-circuit.
- Replaced both parts as recommended. 
- Powered the unit up and saw the exact same output - roundabout 
  600mV and 300mA.
- Subsequent investigation showed R1202 was again reading low.
- Again, replaced R1202.
- At this point, my suspicions turned to the DC converter itself.
- Used hot air to R&R DCDC5. (Put it back on with leaded solder.)
- Double- and triple-checked all associated components vs mentor
  comps file for DCDC5 and final assy. instructions.
- Put unit into bench test stand and powered up.
- All voltages measured their given outputs.
- Used DPI_CLI.exe to configure the new DCDC5.
- Unit is ready to go back through the 14SS. Maybe just in Slot #1
  so that Spencer can run GbE, which wasn't possible before.

25 JUN 2020
- Unit ran in Slot 2 and passed IBERT.
- Pawel was going to do some tests, and he said he saw a problem with
  the MGTs, but that it disappeared.
- When moving to Slot #1, the Hub's PLL wouldn't lock to TTC.
- I asked Pawel to investigate, and he said he saw the same problem as
  before - MGT initialization.
- I happened to catch the scope display as I was walking by, and saw
  that the PLL was locked again, but it quickly came unlocked.
- I watched it for a while to see if there was any kind of pattern,
  i.e., locked for so long, unlocked for so long. But it was totally
  random: 10 seconds locked, 3 minutes not; 1m5s locked, 30s not.

- I know the MGT AVTT supply is used for these links and for making
  Vref. I'm going to investigate that for now.
  
29 JUN 2020
- MGT_AVCC ended up being the problem.
- Powered on the bench, the MGT_AVCC supply was only making about
  600mV out. 
- In dpi_cli.exe, the UVOUT flag was on.
- No warnings/flags were evident in dpi_gui, but it was showing the
  low output.
- Took to measuring all components in this supply, using the relevant
  drawings, and the NETS & COMPS files.
- The resistors all gave reasonable in-circuit reading, and read 
  their given values out of circuit. 
- C1051 read about 100 Ohms across it, though, and that was with 
  everything else out of circuit, so that didn't seem right, given
  its location in the circuit.
- Removed C1051 and measured. It looked fine.
- Re-installed C1051 and re-measured. Now it was reading several
  hundred thousand Ohms (300-something K, or so), which seemed
  far more reasonable.
- Reassembled the MGT_AVCC supply, applied external cooling, and
  powered SN30 back up.
- Let unit run for one hour with my meter, my scope, and dpi_gui
  monitoring the supplies' outputs.
- No further problems/oddities seen.
- This unit is ready to go back in the 14SS for tests.

30 JUN 2020
- I powered on SN30 on the bench and let it run for an hour while
  monitoring DCDC2 output. It never strayed much more than
  +/- 10mV from it's designated output (1.0V).
- I will be removing the IPMC bypass and installing it in the 
  14SS for IBERT test.