I tend to agree with Old Beefer and the other poster who correctly stated that a high frequency pitch oscillation can be misinterpreted as a vertical bounce. The Bell 212 with the basic SCAS sometimes suffered from a 'chattering' pitch gyro (loose wires) which would induce a severe vibration, as have other types (AS332 sometimes prone.) But, if the AP was 'out' in the first place then there should be no interaction at all with the controls from the AFCS. Reference to the collective channel is a red hering I think, because even where it to be functional, it would only be a trim actuator, and not a a SAS actuator, so would be unable to generate the frequency you describe.
You should not be surprised at how quickly the AP killed the problem. The disc is typically moved by the SAS far more rapidly than any pilot inputs and the fact it only has 10%, or so, authority is misleading too - try moving the cyclic back and to within 10% of it's range and see what the aircraft does!
I strongly suspect that there is something unrelated to the specific AFCS which has led to a PIO, and because it happens to be a unique AFCS fit within the fleet that is disracting your attenton.
I'm sure we will all be very interested to know the final outcome
PS. As FBH machines, do they have CVFDRs fited, or HUMS? I'm sure there will be some clues in their recordings.