James & Steven noticed that the PA5 boards have the LCD 'Rain' problem.  Verified with Ricahrd's board, it has it too.


The root cause is too much ripple on the 3.3V power supply.  I have not been able to determine the reason for the differences in ripple from board to board, but there is a significant difference from one board to another.  

Presumably, this is due to capacitor or inductor variation.  The capacitors look like not too much variance. The bad board's capacitors were the specified 10uF.   I have not measured the inductors yet.

This issue was reported by SZ on PB1, but not on PB2.  We got 1 PCB in MV from SZ that was supposed to have the rain problem, but we didn't see the issue on that board.  Since we never saw the issue here, we never knew to look at the power supply for the problem.  

By doubling the capacitance from 2x10uF to 4x10uF on the 3.3V rail, it virtually eliminates the rain on the worst offending board.  There is still a very slight amount left, but it's nearly undetectable.

By going another 50% to 6x10uF, it's undetectable.

We can get 22uF and even 47uF capacitors in the same 0805 size, so we will be able to resolve this issue with a component change only (no PCB change).  

The plan is to change the 2x10uF to 2x22uF caps (111650-A226).  

Some of these caps are on order from digikey, and the solution will be verified july 15.

MPQ will proceed as planned.