*****Why does rust give a high tone on most units? *****
In my opinion, it depends on the type and orientation of an iron target in regards to the detector response.
Picture a buried iron target, slightly elongated, that is orientated in such a way that the magnetic field lines at depth are parallel to the long axis of the iron target. There is very little eddy current generation (flux does not penetrate well) but there is the enhancement of the secondary magnetic field returned due to the ferromagnetic nature of iron. That is, under a magnetic field the mag polarity is rather easily aligned within the atomic structure of electrons in iron atoms (electrons are in unpaired orbits so there is no opposing spin to cancel out their reaction to the applied field. It is the electron's spin that creates the strong magnetic moments) . When the field is applied the domains begin to switch, the maximum is reached and then the field drops away as the primary coil polarity reverses. Between polarity shifts in the current of the primary field coil there is a space where the applied field falls to zero. In a high permeable target (iron) this is the opportunity for the many magnetic domains to relax and return to normal state of orientation. The relaxing of these secondary magnetic domains produces a secondary magnetic field sensed by the receive coil just as dissipating eddies produce a secondary magnetic field.
In another way, the iron target can be thought of as having greater permeability (permeability is the ability of an object to concentrate magnetic flux). The primary field of the coil has a limited capability to affect the target based on the density of the field lines produced - which is a set amount for a particular coil/detector. The iron target will drain a portion of the energy by using the energy to realigning millions of magnetic moments - which leaves less energy available to generate eddy currents beneath the rust (eddies borrow their energy from the primary field too). The higher permeable targets (iron) tend to protect their inner-core better from eddy formation and therefore hide their conductive metal nature better. This is one reason why dual freq detector that utilizes a low frequency field can induce eddy currents deeper into the conductive layer of the target, which improves the discrimination of the ferrous/non-ferrous nature of the target rather than relying perhaps on only weaker surface eddies that may occur with a higher operating frequency.
So, despite weak eddy current generation there is stronger overriding returned magnetic field that interacts with the receive coil balance (permeability changes), and is interpreted as a ferro-signal - or iron signal (low grunt). Not all iron alloys are necessarily magnetizable to the same degree, so the variance in iron is really a variance in permeability. The higher the relative permeability the more positive the phase angle (more ferrous looking), so an apparently rusted target may provide a weak ferrous signature (high tone) because to the receive coil it is really more non-ferrous than ferrous (despite the outer layer of rust).
Back to the orientation issue. If on the other hand, we change the orientation of the coil so that the field lines now slice into the iron target and produce eddy currents beneath the oxidized outer layer, then with perpendicular field lines cutting the iron target the secondary field produced by the eddies appears to the receive coil as non-ferromagnetic and a metal signal is produced (high tone). That is, the eddies prevail over the ferro response.
The phase angle is going to vary with coil height changes, even minor ones, and the coil's axial offset - as field lines cut the target at different angles.
Also, a large ferrous target can also appear non-ferrous (high metal tone) because the greater surface area produces a larger eddy current that can override the weaker permeability change in the receive coil that would be due to the ferro-magnetic nature of the target alone.