Hi,
I have some problems with the stability of TPS7A49.
In the attached scheme you can see (from top right, in board_2) a LDO that generates 12 V and injects it into the AC channels that connects two boards through long cables (the two wires with differential signal are twisted).
From that DC, in the board_1, the TPS7A49 generates the 5 V Vcc which supply all the components on that board (also two buffers are present, to have two separate Vcc/2 signals, needed into Circuit_1). Board_1 uses few mA, so with those 680 Ohm resistors, there is about 8 V at the TPS7A49 input.
Then Circuit_1 generates the AC differential signal over the output wires and, in the other board, Circuit_2 receive and elaborate that signal.
We already produced many PCBs of Board_1 type, and in most cases all is functioning correctly.
But in few cases the Board_1 AC differential output was degraded and has onother oscillation frequency, of about 112 KHz.
I did some tests and the problem is caused by the TPS7A49: removing it and connecting directly a 5 V power supply at its output, all works fine.
Initially I thought the component exposed pad was not correctly welded, so I replaced it (twice), but I saw the same behaviour.
Thus I thought it was a sability issue due to the input and output capacitance. C5 and C6 are two 10 uF capacitors (±10%, 0805, X5R, 35 V). I added onother 10 uF capacitor over each of C5 and C6, and now all works fine.
Is it possible that TPS7A49 needs more than the 10 uF capacitance declared as recommended in the datasheet?
Thanks,
Clemente
