Quote:
Actually I’m not so sure there would be current returning to a different plate if both generators were floating. The reason comes down to current circuits, how can the return current get back to it’s origin via another generator. If current leaves a generator, the same current has to return else the circuit is broken. If no other route can be found, the only option is via the correct return plate.
The voltages from each active, assuming identical electrosurgery units, become referenced to each other by virtue of the fact that the neutral plates and actives are connected a common circuit. Electrodes on either unit are floating but commoned via the patient. If each active floats relative to the other and the patient completes the common circuit between neutral plates then current will flow between plates when each plate is at a different voltage relative to each other and when the non-synchronised actives are operated simultaneously as they might be under dual operation.

Simplifying things; I think Kirchoff's law still holds for HF current within a bulk conducting volume (patient) - this law still has to be satified whatever the circuit - most of the current required to satisfy this must return to the respective neutral, it's just the route that is unexpected. The implication being that there could, under certain circumstances, be current between actives at different potentials. I'm not certain, as I'm definitely not an expert, but it wouldn't surprise me if some electrosurgery units can detect this and alarm under these circumstances (as I've posted before).

If there is a difference in voltages between the floating neutral electrodes, they are referenced via the 'commoned' actives, that are producing output, and the likelihood of this is dependent upon the impedances in the circuit (between each actives and each respective neutral electrode and between the neutral electrodes). That's why it's important that the plate contact area is maximised and a good contact is made with the patient to minimise current density and what I posted earlier is important:

Quote:
Yes, as far as I'm aware, recommendations are usually for the return path to be kept as short as possible i.e. the site of electrosurgery to neutral plate distance, hence return pathway impedance, is minimised; as is the RF current pathway within the patient. The benefit is that a lower power settings are usually required for the desired effect.

Also there's less opportunity or likelihood for RF leakage currents or even return currents to preferentially pass through lower impedance tissue (that may also be more sensitive to RF heating effects) than flow through some of the tissues between the surgical site and return circuit. A longer active to return distance may increase the opportunity for this.
Your comments:

Quote:
However if there was an interaction between the high frequencies causing a beat frequency the LF blocking circuitry would be useless. This is because only high frequency would be leaving the generators; the low frequency would be generated at the site due to the mixing.
Actually I think the predominant contribution is from the arc (or arcing of the active) rectifying HF current. Yes the generator is outputting HF but what if the filtering blocks the return of dc? My point is that if a rectifier (the effect due to the arc rectifying HF current) is placed at the floating output of a HF generator then the HF current output will be produced though not necessarily referenced to the return because of filtering or circuitry dedicated to blocking it - The dc or low frequency current components will have no return path through the generator if blocked.

Could an electrosurgery expert or someone who has some literature or a text post something perhaps?