I've got major complaints about this graphic that lead me to believe it's about useless, or at best unreliable for any practical reference.
First off is its representation of a HEGO. According to this, a HEGO's common switching range is only between what visually looks like .49v and .51v and that the range from the mid .4v down to 0v is somehow excessively lean. I completely disagree with that...at least the way my HEGOs behave. Normal operation for my HEGOs is that the 0-.6v range corresponds to 14.9-14.4 AFRg (1.02-0.98 lambda).
Next, its showing a Wideband as going from 0-5v with 0v being lean and 5v being rich. WB sensors are NOT voltage-based devices. They are current-based. And to get any kind of meaningful reading from them, you need a WB controller to manage and control their supply current flow while monitoring their sensor-side current direction and amount. But with a properly operating WB controller managing & monitoring the sensor, a WB sensor's current to lambda transfer is linear. The controller converts this linear current characteristic to a linear voltage output...which is much easier for most devices to use. Thus a voltage transfer
of a WB is a characteristic of its controller, not of the sensor itself. That said, notice which direction the voltage goes as the mix goes from lean to rich. I don't know of a WB controller out there with a transfer that represents 0v as leanest possible value and 5v as richest possible value. Most WB controllers I'm familiar with have the 0v range representing the richest possible detection value and the 5v range representing the leanest possible detection value. Granted, a programmable controller like the Innovate controllers can be configured this way, but not by default.
Finally, they are showing the WB's transfer as an S curve. As already mentioned, WB controllers transfer a linear translation (straight line) from 0-5v, not an S-curve. The only place I would expect a bend in the transfer to occur is as you approach the max possible voltage value but the vast majority of the transfer should be a straight line. There is NO reason a WB would ever be expressed as an S-curve since they are inherently linear feedback devices.
There are enough technical SNAFUs on this graph that I say it shouldn't be used as reference to make any kind of argument.
As for the accuracy of WBs vs HEGOs, they usually parallel each other fairly well, but there can be some discrepancies and I think I know where some of this comes from. HEGOs are purely an O2 sensor. Presence of significant O2 will trigger them lean. However a WB sensor is a combination of an O2 sensor and a Hydrocarbon detection chamber (the thing that needs all the heat). In the presence of O2, the O2 sensor produces no voltage and is actually a resistance to current flow. The WB controller injects a current and the amount of resistance the O2 sensor puts up to the current push is an indication as to just how much oxygen there is present (how lean the burn was). In the absence of O2, the O2 sensor conducts current and in fact produces voltage as it does in a normal HEGO. So the controller isn't having to push current. In this case, the current flow is actually reversed. When this happens, the hydrocarbon sensor gives feedback as to how much hydrocarbon is present to give an accurate rich lambda.
I've seen my HEGOs and WB not fully agree when running Closed Loop on cold starts when I'm sure there's more unburned hydrocarbon and unburned oxygen near the condition my HEGOs are calling stoic due to incomplete (cold-engine) combustion. The incomplete burn makes the HEGOs report lean which causes the EEC to add more fuel to get what appears as a more complete burn. But under these conditions, the WB can report slightly rich due to the hydrocarbon sensor overriding some of the O2 sensor's influence due to the un-normally high amount of HC present in the exhaust. But under normal
hot-cruise, the two sensors parallel each other rather nicely because generally you don't have both an abundance of HC and O2 in the exhaust. Whichever one is in shorter supply will more completely combust. The only time this doesn't happen is under abnormal running conditions which include cold engine running, badly retarded timing, or excessive EGR flow impeding complete combustion similar to cold engine combustion.
89 Ranger Supercab, 331 w/GT40p heads, ported Explorer lower, Crane Powermax 2020 cam, 1.6RRs, FMS Explorer (GT40p) headers, Slot Style MAF, aftermarket T5 'Z-Spec', 8.8" rear w/3.27s, Powertrax Locker, Innovate LC-1, GUFB, Moates QuarterHorse tuned using BE&EA