Voltage should not be fluctuating wildly for 02 sensor after the cat. The fact that your 02 sensor is varying the voltage leads me to think you have a bad 02 sensor but at bank 1 not bank 2.

Also your MAF 18K/hour equates to 18,000/3600 = 5g/s. 5g/s at idle is kind of high so. If you look at the video I posted, my MAF g/s was 3.6g/s at 700 rpm.
I think your MAF is over-reporting the airflow through it and the computer is trying to compensate for this data pid. You can try cleaning the MAF with MAF cleaner, don't use carb cleaner and make sure it is dried before starting the engine. Don't buy a cheap MAF from eBay or Amazon. Better to go to a junkyard and buy a genuine Siemens MAF.

Thanks, I meant to edit this post. The voltage is actually pretty static for the after cat sensors. My mistake.

Bank 1 after cat: voltage is 0.75 to 0.76 V
Bank 2 after cat: voltage is 0.58 to 0.59 V

Interesting theory, as far as I know the MAF is original, now 188,000 miles.

The MAF has been cleaned recently as well, no change. Yes, used MAF cleaner!

The car's exhaust definitely smells rich. The coil packs are also original, except one that was replaced.

The M54 engine utilizes a "Hot Wire" in the MAF. Air passing through the MAF cools the "Hot Wire" and from the voltage that is used to keep the "Hot Wire" hot the computer can determine the amount of air rushing through the MAF. Like a light bulb, the "hot Wire" eventually starts to deteriorate until it fails. Since the "Hot wire" is not getting as hot as when it was new, the computer is just assuming that a lot of air is rushing through the MAF, so to accommodate this rush of air, the computer commands the injectors to provide additional fuel to the mixture.

I've finally replaced the MAF with a new VDO sensor from FCP here:

https://www.fcpeuro.com/products/bmw...em-13627567451

I'm not sure I really notice much of a drivability difference, maybe slightly better throttle response. But maybe that is all in my head lol.

Here are the updated readings with the new MAF installed:

15 kg/h at idle
35 kg/h at 2000 rpm
44 kg/h at 2500 rpm
57 kg/h at 3000 rpm

With the old, high mileage original MAF, the readings were:

18 kg/h at idle
44 kg/h at 2,000 rpm
55 kg/h at 2,600 rpm

I'll wait for the gurus to chime in as to how much of a difference that is... next step is to replace those post-cat oxygen sensors.

Post grams/second at idle with all accessories off. (lights, radio, AC/fan, etc). Should be hovering around 3.2 - 3.8 g/s

Remember that the MAF is measuring the density of the air, so temp, humidity and barometric pressure all play a role in trying to compare readings.

If the car sits outside, check for wires chewed by animals (squirrels etc.)...

Took the readings again tonight, with daytime running lights and radio off. Everything off.

14 kg/h at idle
32 kg/h at 2000 rpm
41 kg/h at 2500 rpm
55 kg/h at 3000 rpm

Seems to be driving fine, the one after-cat oxygen sensor says no signal so definitely going to replace both.

I'm going to say that based on the info above, the new MAF has not improved the cocktail of codes condition haha.

edited correction to the percentage
3.88 g/s. vs 5.00 g/s. difference 28%
8.88 g/s. vs. 12.22 g/s. difference 37%
11.38 g/s. vs. 15.27 g/s. difference. 34%
15.27 g/s

Yep must be something else. Data doesn't lie

You are comparing the wrong data sets actually. Keeping all other factors constant, I think we would want to compare the first numbers reported. Those were the before/after readings where the headlights were on both before and after the replacement.

The data shared yesterday with the headlights off cannot be accurately compared to the numbers with the old maf, where the headlights were on. It did end up making a difference in the readings with the daytime lights left on.