What tools do I need? (P0420 Bank 1 catalytic inefficiency on a 15 year old dual-cat vehicle)

You need a scanner.

FIRST you need to look at the value coming out of the upstream oxygen sensor. Is it within range? If so, the system is running locked and the mixture is good.

If not, then deal with the engine running issue that is the real problem.

If the value coming out of the downstream sensor is totally out of range or the pointer moves very slowly as the car warms up, then the downstream sensor is likely bad.

If the value coming out of the downstream senor pretty much tracks the value coming out of the upstream sensor, then the catalytic converter is bad. You can gun the engine and watch the two graphs move and they are almost exactly the same (because the converter is doing nothing and so the composition of the input and output ARE almost exactly the same).

Do actual diagnosis, find the problem and fix it.

If it is the converter, sometimes cleaning it will solve the problem, and sometimes it won't. If you have more time than money and the vehicle has the original converter it's worth trying.

Usually cleaning helps when the converter is gunked up with tar from a previous running problem, but it won't help if the converter is just worn out.

This may have gunked up the converter. If that is the case, a lot of highway driving may clean it out, or cleaning it by hand may help. But the scanner will tell you for sure when you look at the value coming out of the upstream sensor and see if it's in range.

That's not a bad plan, but before doing that you should check the upstream sensors. This is a good way to isolate whether it's a sensor issue or a converter issue, but if there is an underlying problem you'll know.

This is a lot more work than just using the scanner and looking at the sensor values, and again it doesn't tell you if your exhaust is clean or not.

These mostly look silly. Cleaning the converter may be a good plan, if it's a problem that was caused by earlier running too rich. But first make sure it's not still running too rich and that the upstream sensors are good.

The cheap aftermarket converters will not make you happy. If it's bad, get an OEM one from Bavarian Motorsport. You will need kroil, an acetylene torch, an asbestos mat to put behind it while you heat up the bolts, break-free, a breaker bar and a large vocabulary of curse words.

You MAY find it's easier to drop the whole exhaust system and take it apart outside of the car, if you're working in a driveway.

--scott

Here are the values coming out of the scanner just now, with the engine warm, and with my foot on the throttle with steady pressure.

Which numbers look abnormal to you? ========================= P0132=O2 Sensor Circuit High Voltage BAnk 1 Sensor 1 P0430=Catalyst System Efficiency Below Threshhold Bank 2 ========================= MIS=OK FUE=OK CCM=OK CAT=OK EVAP=NOT OK AIR=OK O2S=OK HRT=OK ========================= RPM held steady at approximately 2000 O2B1S1(V)=1.275 O2B1S2(V)=0.760 O2B2S1(V)=0.665? O2B2S2(V)=0.635G ========================= RPM held steady at approximately 2000 O2B1S1(V)=1.275 O2B1S2(V)=0.750 O2B2S1(V)=0.720? O2B2S2(V)=0.635G ========================= RPM held steady at approximately 2000 O2B1S1(V)=1.275 O2B1S2(V)=0.740 O2B2S1(V)=0.150? O2B2S2(V)=0.635G ========================= RPM held steady at approximately 2000 O2B1S1(V)=1.275 O2B1S2(V)=0.740 O2B2S1(V)=0.720? O2B2S2(V)=0.685G ========================= RPM held steady at approximately 2000 O2B1S1(V)=1.275 O2B1S2(V)=0.740 O2B2S1(V)=0.110? O2B2S2(V)=0.660G ========================= I realized I should probably record actual RPM so I added it to the mix ========================= RPM=3227 O2SLOC B1S12--B2S12-- O2B1S1(V)=1.275? O2B1S2(V)=0.670G O2B2S1(V)=0.630? O2B2S2(V)=0.710 ========================= RPM=3244 O2SLOC B1S12--B2S12-- O2B1S1(V)=1.275? O2B1S2(V)=0.645G O2B2S1(V)=0.680 O2B2S2(V)=0.570 ========================= RPM=2226 O2SLOC B1S12--B2S12-- O2B1S1(V)=1.275? O2B1S2(V)=0.780G O2B2S1(V)=0.795 O2B2S2(V)=0.750 ========================= RPM=2178 O2SLOC B1S12--B2S12-- O2B1S1(V)=1.275? O2B1S2(V)=0.685G O2B2S1(V)=0.285 O2B2S2(V)=0.640 ========================= RPM=778 O2SLOC B1S12--B2S12-- O2B1S1(V)=1.275? O2B1S2(V)=0.850G O2B2S1(V)=0.695 O2B2S2(V)=0.765 ========================= RPM=763 O2SLOC B1S12--B2S12-- O2B1S1(V)=1.275? O2B1S2(V)=0.850G O2B2S1(V)=0.120 O2B2S2(V)=0.725 ========================= RPM=751 O2SLOC B1S12--B2S12-- O2B1S1(V)=1.275? O2B1S2(V)=0.845G O2B2S1(V)=0.650 O2B2S2(V)=0.745 ========================= Does anyone here know how to help me interpret those numbers?

This is the data you want to see, but if possible you want to see it as graphs, because a lot of what you care about are trends and not just the stock voltages. Your software also should be able to put lines on the graph showing the nominal values.

However, since it's a BMW, you can guess that the nominal value is somewhere around 1.0. High voltages indicate something running too rich, low voltages indicate something running too lean.

So.. we look at bank one sensor one... it's running kind of rich, which is bad... and then we look at bank two sensor one.... it is running super lean if that voltage is to believed. So we have a problem.

Downstream on sensor two the value on bank 1 is kind of high, so you may be getting the error being set due to that, but that's not your problem. Your PROBLEM is that half the engine is running too rich and the other half is running too lean (probably because the ECU is trying to compensate).

The FIRST thing that I would do would be to swap the upstream sensors from side to side and see if the readings move from side to side. If they do, you have a bad oxygen sensor. If they don't, you need to start looking on the engine side for vacuum leaks and bad injectors.

You can put a new converter onto this, but it won't stay good for very long until you sort out the mixture problems.

--scott

Oh, and I am assuming that your software is correct in enumerating them, that O2B1S1 is upstream bank one and O2B1S2 is downstream bank one. Make sure you trust your software either by validating it with a mirror or pulling a wire from one sensor and making sure it shows up as pulled on the software.

What IS that questionmark? What does it indicate? If you had a plot you could guess.

--scott

I agree. It seems that the Cen-Tech item 60693 can do graphs so I'll look at them tomorrow when I drive the vehicle warm again.

Apparently, from my googling of how the oxygen sensors work, as you said, the more "reducing agent" there is (e.g., a rich mixture), the more oxygen diffuses from ambient to the exhaust gas, hence, as you said, that generates a higher voltage.

If there aren't many hydrocarbons in the mixture (e.g., a lean mixture), then there aren't any reducing agents for the oxygen to diffuse from the outside air into the inside of the oxygen sensor, so it barely generates a voltage.

I agree with you that bank 1 sensor 1 is pegged at a high non-fluctuating voltage of 1.275 volts, while bank 2 sensor 1 is fluctuating from about

The two upstream sensors are brand new, the P0132 code indicates Bank 1 Sensor 1 has a "O2 Sensor Circuit High Voltage", which is true, since it's steady at 1.275 volts.

If we assume the two upstream sensors are working properly, and if their input voltages and heater circuits are correct, then that would indicate a rich mixture - but - there are plenty of codes for rich mixtures - none of which are indicated.

So I'm not sure what to make of the steady 1.275 volts on the bank 1 upstream lambda sensor - but your suggestion of switching the two upstream sensors should help isolate whether it's a bad sensor from the factory or something in the fuel mixture or input electrical circuitry.

I agree that the downstream sensor output isn't all that useful until I figure out why the upstream lambda sensors are indicating such wacky dichotomy of very rich (bank 1) and not very rich (bank 2).

Yup. I'll do that tomorrow in the daylight.

Yup. Either it's really running rich, or the voltage to the sensor is bad, or the sensor itself, even though it's brand new, is bad.

Switching the two upstream sensors should help eliminate some of those choices.

Agreed. Here are some readings from today which I generated on the highway before I read your post above (which repeats for today, basically, the readings that you already commented on from the prior day). ========================= RPM 749 O2B1S1(V) 1.275 O2B2S1(V) 0.680 ? O2B1S2(V) 0.830 G O2B2S2(V) 0.715 ========================= RPM 1468 O2B1S1(V) 1.275 O2B2S1(V) 0.095 ? O2B1S2(V) 0.780 G O2B2S2(V) 0.740 ========================= RPM 2104 O2B1S1(V) 1.275 O2B2S1(V) 0.670 ? O2B1S2(V) 0.755 G O2B2S2(V) 0.565 ========================= RPM 2182 O2B1S1(V) 1.275 O2B2S1(V) 0.675 ? O2B1S2(V) 0.700 G O2B2S2(V) 0.615 =========================

Hi Scott,

Thanks for that advice, which is good advice, and which I only received just now, so I haven't been able to take you up on that test to confirm what I would have assumed, which is the assumption that...

• O2B1S1 (I would assume is the upstream sensor for bank 1, cylinders 123)

• O2B1S2 (I would assume is the downstream sensor for cylinders 123)

• O2B2S1 (I would assume is the upstream sensor for bank 2, cylinders 456)

• O2B2S2 (I would assume is the downstream sensor for cylinders 456)

I'm not at all sure what you mean by using a "mirror" but it's easy enough to disconnect a sensor, especially for the two upstream sensors which are tucked away but accessible from the passenger side of the engine bay.

I have the owners manual for the Cen-Tech Harbor Freight item 60693 OBDII scan tool, which I had borrowed from a friend. The "G" indicates a "graph" is available, while the "?" indicates "help" is available.