I think some folks are still confusing iron bias not just with iron discrimination, but also with unmasking. To put it simply, Iron Bias (and I assume, Iron Filter on the Legend) allows the user a way to influence how the machine responds to mixed-signal targets that might cause a “falsing” ferrous item to indicate in the non-ferrous range, without changing how the machine responds to solid, non-ferrous targets. Discrimination and/or tone break adjusts how the machine responds to different conductivity characteristics across the board…”fine tuning” a tone break or discrimination means any item assigned a particular conductivity value looks and sounds the same on the machine. It's an important difference.
To use the numbers from Detector’s requested example, iron bias won’t give an iron grunt on a +4 ferrous, and a high tone on a +4 non-ferrous – that’s just not how it works. Instead, when set properly, iron bias will turn a target into a -5 if it’s ferrous, when it might otherwise read +4. Meanwhile, a +4 non-ferrous target will remain +4. That’s what iron bias accomplishes, and it needs more than one simultaneous data point from multifrequency to make it happen… it’s impossible on a single frequency machine or mode because a +4 hit is a +4 hit - there’s nothing to compare it to. The video that I made below shows this effect directly. The best I have laying around the house that has matching VDI numbers at low iron bias is a rusty iron horse tack ring and a small gold pin. For those who don’t watch videos, it shows both targets individually reading at 12-13 VDI at iron bias setting F2 0. At a slightly increased iron bias of F2 3, the horse tack ring reads -2, while the gold pin continues to read 12-13…no other settings changed beyond iron bias. Don’t get too caught up in the idea that the horse tack ring is a large target, the same effect can be demonstrated with a smaller iron target, as long as it is dense enough that it’s conductivity “fools” the detector into assigning a higher VDI reading. Honestly, I don’t think you can demonstrate iron bias on “a small iron wire” with a non-ferrous target in an air test as Detector requested, because the detector won’t get fooled by the small wire on the surface…a small wire on the surface doesn’t generate enough conductivity to alter the detector’s field and trigger “falsing”, or a high VDI (see the Iffy Signal video posted earlier in the thread - the nail stays as iron, the detector isn't fooled). Bury that same nail in the dirt several inches and let it marinate in the ground for awhile, and yes, a detector will start to false on the tips, and iron bias will become useful in certain situations again.
I consider true unmasking to be a situation where a detector can be set up to see two targets when it would otherwise see only one. Iron bias is not intended to be an unmasker, although it can sometimes look that way (again, Iffy Signal’s video posted by Digalicious earlier in the thread is a good example). Unmasking is what recovery speed (or reactivity for Deus), slower sweep speeds, and/or smaller coils, etc, are for. If a ferrous and non-ferrous target are too close together and recovery speed is too low, a target is more likely to get missed. Iron bias can exacerbate this effect, of course - if iron bias is set too high, and the machine sees the two targets as one co-mingled target signal, a high iron bias set by the user will mark the signal as ferrous, and you’ll never hear the high tone. This is exactly what the earlier Iffy Signal’s video is showing, and I can demonstrate a very similar test with the Equinox 800 (frankly, the Legend may perform slightly better based on my own tests compared to his video). While watching that video, you can easily convince yourself that the change in Iron Filter “unmasked” the coin from the nail – that the Legend was seeing two individual targets and that as the iron bias was increased, the detector “erased” the nail like a magic wand and left the coin, but that’s not really what is happening. I make the case that the Legend was seeing the nail and dime as a co-mingled single target, presenting both ferrous and non-ferrous levels of conductivity as the target was sampled in multifrequency mode. With the Iron Filter set at 8, the detector was doing as the operator told it to do – “call that kind of signal iron”. As he changed the Iron Filter down toward the lower numbers, again, the detector was doing as it was told, and assigning the mixed multifrequency signals non-ferrous VDI numbers and tones instead.
But again, Iron Bias (or Iron Filter) is not a magic wand, it’s just a tool. If you combine what I show in my video with what you see in Iffy Signal’s video, you can see the obvious downside to iron bias. In both of our videos, we have the benefit of seeing what the targets are…a nail next to a coin, a chunky piece of iron, and a piece of gold. But in the real world, you still have no real idea what you might be digging when the machine alerts you with a solid non-ferrous signal…it might be a masked coin, it might be gold, it might still be balled up aluminum. But hopefully, if the iron bias is set up properly, the target is much less likely to be a falsing iron nail.