I had a similar problem on my '89 240. The check engine light came on and the engine started to run progressively worse. After some troubleshooting, I discovered the O2 sensor was not receiving 12 volts to the heater coil. The cause of this problem was the "in tank" fuel pump had shorted out and blew the fuse. The O2 sensor heater coil circuit shares the same fuse. With no 12 volt signal going to the O2 heater coil, the sensor wasn't getting hot enough to operate correctly. I disconnected the fuel pump harness in the trunk and installed a new fuse. The problem went away and I replaced the fuel pump about a week later (car ran fine with just the secondary pump).

I thought back in October I gave you a heads-up on the O2 sensor. I remember reading your Post. Maybe I thought about the O2 sensor but since the cat was untouched, I figured neither was the O2 sensor.

Yes, a bad O2 sensor can cause billows of black smoke worth of raw gas.
it happend to me. I wonder if the broke a wire on the Sensor by twisting banging or something
--
'75 Jeep CJ5 345Hp ChevyPwrd, two motorcycles, '85 Pickup: The '89 Volvo is the newest vehicle I own. it wasn't Volvos safety , it was Longevity that sold me http://home.no.net/ebrox/Tony's%20cars.htm

Check the fuel pressure regulator, it's round, at the front of the fuel rail, and has a vacuum line attached.

Pull off the vacuum line, and smell the hose for the scent of raw gas. If you smell raw gas, the FPR is bad.
--
Bruce S. near D.C.

O2 sensor is used as the feedback mechanism for the fuel injection computer. The computer makes a good faith guess at how much gas it should be injecting, based on the way things worked in tip-top factory fresh condition. Then it checks the O2 sensor reading to see how that went. As the miles pile on, things fall out of whack. The fuel pressure might go up or down a little. The injectors might get a little clogged. The air mass meter might read off a little. So the fuel injection computer pays attention to the end result. And if the O2 sensor says the mixture is too lean or too rich, then the computer will fudge the fuel amounts in an attempt to correct the situation. If it can fudge it a modest amount and get good readings from the sensor, it's happy. However, if the sensor is just plain busted, and keeps sending out a certain reading (either rich or lean) regardless of what the motor is doing, then the ECU will keep on fudging in that false direction. Eventually, it will reach the end of of it's programmed authority to fudge, and assume that something is seriously wrong with the motor, and light the CEL.

So in your case, it sounds like the O2 sensor is stuck on 'lean', the ECU keeps richening the mixture in a vain attempt to correct it, and eventually both leaves the motor running really rich and with a lit CEL.
--
'63 PV544 rat rod, '93 Classic #1141 245 +t