In considering possible external factors for the gearbox failures I was making the assumption that the AH engineers had factored in all of the normal stresses and strains to be expected in flight.
There is a possible alternative which may not have been picked up by the monitoring programs.
Continuous vibration monitoring is great at picking up sudden changes in vibration levels. A sudden jump indicates that something in the rotating system has changed. Visual inspection is usually sufficient to identify the fault at that stage.
However constant vibration at a particular frequency that does not change over time is considered to be fine and a good sign that everything is functioning normally. This view may be totally wrong. Vibration at infrasound frequencies of 10-20Hz has been proved to cause embrittlement of metal and other substances.
Helicopters being a collection of vibrating rotary parts, a constant level of vibration at around 20Hz might be accepted as perfectly normal if it doesn't change over time. The vibration is however slowly degrading certain components. Rotors could delaminate, gearboxes fail, fastening bolts suddenly snap under load.
A rotor blade starting to delaminate should be detected very quickly. Failed fastenings should also be picked up during standard walkaround inspections. Internal gearbox components are hidden away and difficult to assess without a full strip down.
The point being that if infrasound is a factor there may well be no indication of impending failure until a gear tooth or something else actually snaps off.
Interestingly these sound frequencies frequently cause feelings of foreboding and discomfort in humans exposed to them.