wav trigger - power and serial comms problems

wav trigger updated to 1.32 firmware, 5v jumper bridged. arduino micro rx / tx pins connected directly to the wav trigger's corresponding rx/tx on the ftdi header area. currently the wav trigger setup is quite "bare bones", no sd card installed, no trigger inputs set up or anything. i'm simply trying to establish minimal communication with the serial pins.

i'm having trouble with signal "chatter" on potentiometers with the arduino micro if i have the wav trigger 5vin and ground pins sharing the power/ground rails with the arduino and pots (detailed in another thread). despite the signal chatter, i am able to have the arduino request and successfully receive/display the wav trigger firmware version during the setup function. so, i know that basic comms should be functional.

to work around the signal chatter issue, i tried powering the wav trigger separately via a sparkfun 5v ftdi that is plugged into a 5v 1A usb wall adapter, using just the power and ground pins from the ftdi . this seems to power the wav trigger successfully (i have the blinking led on the wav trigger going). however, the arduino (using the same rx/tx pin connections noted above) isn't communicating with the wav trigger (doesn't receive the firmware version), like it does if the wav trigger is sharing the 5v and ground rails with the arduino and pots.

what's going on here?

further investigation and confusion...

i put the ftdi into a breadboard, and i have the wav trigger 5vin connected to the ftdi 5v and wav trigger ground connected to the ftdi ground.

i also put another connection from the ftdi ground to the arduino/rail ground.

if i leave the ftdi-powered wav trigger on, and unplug/replug the arduino usb to the laptop, the arduino doesn't power on UNTIL i pull and reconnect the ftdi -> ground rail connection. doing this (quickly), the arduino powers up AND reads the firmware from the wav trigger, and the pot has no noise.

i'm suprised to have to share the ground between the arduino and the wav trigger (especially in this odd sequence) in order to get the serial communications working -- maybe this is further insight into what is happening?