"Zeroing out" triggers when using StimTracker with Acqknowledge

I have been using Acqknowledge and BioPac MP150 to measure EDA while running a task in E-Prime 3.0. During the task, we use the StimTracker to record the timing of responses. Most of the time everything seems to work fine, but occasionally when I open Acqknowledge and start recording, the channels start recording up at 5 volts (the voltage they would normally “jump” to when marking an event) rather than 0 volts. When the channels are stuck at 5 volts, I cannot see the event markers (the recording can’t “jump” to 5 volts, since it’s already at 5 volts).
I can’t figure out why this happens some times but not others, and I can’t figure out a specific fix for the issue (the “fix” that’s worked so far seems to be fiddling with the cords and turning the StimTracker and BioPac on and off until the channels reset to zero).
Has anyone had experience with this issue before or have any suggestions on a reliable fix?