But this may not be too relevant, since this seems only to make problems when the channel is open.
Theres huge uncharted territory between open channel and shorted channel. This need scientific approach. Suggest 3D chart with following axes: X - frequency of stimulation (can be done using ultra powerful piezo etc, rigidly attached to scope chassis), Y - channel impedance ("load"), Z - scope response. Should take no more than a week of testing
Of course there should be 3D acceleration sensors in various places to analyze stimulation wave propagation etc. Ideally it would be not constant wave but single soliton, otherwise standing waves could build up. Altering resonant frequency of chassis, PCB by adding sensors should be also taken into account.
So if done properly bit more than week, maybe year or so...