O2 sensors work like a mini fuel cell, outputting a voltage based on a oxygen content difference between the exhaust and a reference (air).
I agree it acts like a power cell, producing electricity due to oxygen level difference across the element, but I'd not say
fuel cell. It really is closer to a simple chemical "battery".
Your sensors' complete systems not only put out signals based on oxygen content differential, but pressure differential as well. The document referenced earlier may go into that, but the sensor's own closed-loop controllers cause somewhat of a pressure differential as needed (either direction) to get the cell itself to stay at the equivalent of "stoich" engine fuel burn, and the controllers put out their signal based on what it took to create that condition. Your sensor systems don't measure the pressures, so it's possible the output is just a WAG at times if other than a "hard-coded" pressure differential actually exists.
Does D&D know how to "do" lambda probe bungs? I don't know. If you're using one of their complete systems, gasket to muffler tail, unmodified, then by your photos I'd say perhaps they don't. If all they did was emulate a stock headpipe's bungs but didn't emulate the amount of backpressure in the system then that may be the problem. To get yours to work correctly the mounting may have to be a bit more unsightly.
An additional possibility may be that you're using -'09 bung locations with an ECM programmed for '10- locations, with the time in the cycle for the sensor poll being too late and the main mass of gas already beyond the sensors. But that's just a WAG on
my part. Some of the ECM's sensor poll timings
are pretty critical, like the MAP. I don't know about the O2 sensors but imagine that'd be pretty critical too, especially at the lower engine speeds.
Is that front sensor mounted in a location that makes for an eddie? Exhaust gas will follow the outside of a curve as much as possible, just like water does.