P0420 diagnosis

[Mark R]

I've had this code a couple of times on a 2006 Lexus IS250. It's got 150k, so it's well used, but is running fine.

I've done some crude diagnostics with scan software and the O2 sensors all appear to be switching. I can't see or hear any kind of exhaust leak.

However, I did notice one thing a bit odd with the scan software trace when I manually ran the ECU in open loop and tweaked the fuel trims manually.

In this case, it appears that B1S2 stays lean a lot of the time, and is very slow to switch to rich, but very fast to switch back to lean.

I'm thinking maybe this is not a cat problem but instead a B1S2 problem, or a very minor exhaust leak.

I don't want to go changing cats unless essential, because lexus' quoted price is preposterous, and an aftermarket cat would likely require a full custom exhaust system (still cheaper than 1 lexus maniverter). Unfortunately, the two shops I took it too, both recommended 4 new O2 sensors first, and if that doesn't fix it a new cat.

What do the experts here think?

 

[bamx2]

My non-expert suggestions-

I would try cleaning the intake and o2 sensors , running something like Techron in the fuel first. - Worked for one vehicle


If you find that the B1S2 is not the problem. I would check your cat in and out temps and exhaust gas levels too. If those check out and it is well within emission specs but still get the PO420 code , consider this "fix" that gets the post cat sensor directly out of the exhaust stream- http://honda-tech.com/acura-integra-type-r-8/spark-plug-non-fouler-1841240/ - Worked on another vehicle when the step above did not.

 

[SearchMaster]

If the O2 sensors are original, I'd definitely replace them first. At 150K they're due...

I've had two high mileage cars throw P0420 and replacing the O2 sensors fixed it both times.

 

[thedarkwolf]

My old 200k pontiac vibe would throw that code sometimes, I just reset it and ignored it. Sometimes it would stay off for a couple weeks sometimes for a couple months. If I remember right it just means the cat dropped below 90% efficient for a long enough time to trigger the code. Of coarse it could also mean your after cat o2 is starting to get flaky. One redneck trick you can do is to pull the o2 sensor and mount it in a spark plug non-fouler like
http://www.355nation.net/forum/how-...fouler-eliminate-check-engine-light-pics.html

That is what I was going to do if the check engine light starting coming on more often on my vibe. I wasn't going to pay to replace a cat on that car just because the cat was a little below 90% efficient but I sold it first.

 

[rumpleforeskin]

Any chance you can get "fuel trim" data from the time the code was logged?
More likely that the Cat message is a result of the problem and not the cause, fouled or damaged 02 sensors can definitely cause this.

 

[Mark R]

Any chance you can get "fuel trim" data from the time the code was logged?
More likely that the Cat message is a result of the problem and not the cause, fouled or damaged 02 sensors can definitely cause this.

Yup. Got the freeze frame data:

The middle B1 fuel trims are a bit strange, but I wonder if that is simply the ECU running a catalyst test by leaning out the mixture on one bank only, because the before and after trims look OK.

 

[sontakke]

You need larger sample. What you want to see is lots and lots of transition on the primary sensor and few transitions on secondary. The computer looks at the ratio of the corss count aka transitions and turns on the P04x0 if secondayr sensor starts following the primary sensor.

Either put the new converter or google O2 sensor extension. There is no third option.

Anybody who tells you to replace secondary O2 sensor to fix P0420 does not understand how P0420 is triggered.

 

[tortillasoup]

I've had this code a couple of times on a 2006 Lexus IS250. It's got 150k, so it's well used, but is running fine.

I've done some crude diagnostics with scan software and the O2 sensors all appear to be switching. I can't see or hear any kind of exhaust leak.

However, I did notice one thing a bit odd with the scan software trace when I manually ran the ECU in open loop and tweaked the fuel trims manually.

In this case, it appears that B1S2 stays lean a lot of the time, and is very slow to switch to rich, but very fast to switch back to lean.

Bad oxygen sensors fail to lean.

Buy a Denso or NGK oxygen sensor depending on what the car uses from the factory. Absolutely unnecessary to buy an O2 sensor from Lexus though NEVER put a Bosch into a Japanese car and vice versa.

 

[drnickriviera]

Had that code on a 01' ES300. Turned out to be a bad MAF. I even tested the MAF with the factory manual test

 

[thomsbrain]

It's either your Bank 1 in-cat O2 sensor or a dying Bank 1 cat.

I actually have the same code coming and going on my Accord at the moment. I have a replacement cat and sensor in my garage and will install them one of these days. They were cheap enough I figured I'd just get both and be sure to fix it in the first try.

 

[Mark R]

Bad oxygen sensors fail to lean.

Buy a Denso or NGK oxygen sensor depending on what the car uses from the factory. Absolutely unnecessary to buy an O2 sensor from Lexus though NEVER put a Bosch into a Japanese car and vice versa.

So, if bad O2 sensors tend to read rich, then the B1S2 sensor may not be the problem.

I was planning to get the pre-cat sensors done next service (they are the original ones and have 150k miles on them), so I may well order these any way. (They aren't available in Europe except from Lexus, so I'll have to import some denso ones from the US or Japan)

So conceivably, if the pre-cat sensor is reading a bit rich, it could drive the mixture a bit lean, which might explain why the cat/B1S2 isn't working quite right? Am I on the right track?

 

[tortillasoup]

So, if bad O2 sensors tend to read rich, then the B1S2 sensor may not be the problem.

I don't understand how you could so easily misinterpret what I just said. Oxygen SENSORS FAIL LEAN. When there is a rich condition, voltage is produced. Where there is a "lean" condition, no voltage is produced. An oxygen sensor makes a tiny amount of electricity based upon the differential in oxygen content of the exhaust stream and the atmosphere. If the exhaust is similar in composition to the atmosphere (lean) then no electricity is produced.

 

[Mark R]

I don't understand how you could so easily misinterpret what I just said. Oxygen SENSORS FAIL LEAN. When there is a rich condition, voltage is produced. Where there is a "lean" condition, no voltage is produced. An oxygen sensor makes a tiny amount of electricity based upon the differential in oxygen content of the exhaust stream and the atmosphere. If the exhaust is similar in composition to the atmosphere (lean) then no electricity is produced.

Oh. OK. I see. I had understood it as fail to go lean. Got it.

That makes a lot more sense. Thanks.

 

[sontakke]

You guys don't get it do you? If O2 sensor fails either being lean all the time or being rich all the time, ECM is smart enough to turn on the O2 code and will NOT run the cat efficiency test. The enabling criteria is NOT satisfied when there is an O2 code.

When you get cat efficiency code, ECM is right 99% of the time that it is the cat that is low on efficiency.

The only fly in the ointment is that ECM has tendency to apply extremely stringent test aka > 90% efficiency which is not really mandated by Feds or EPA. So if you take the sensor extension route, you don't have to feel guilty.

Another corollary of the above fact is that if you put an aftermarket cat, you will be back to square one within less than year. Aftermarket cats do not have the same amount of precious material and their efficiency falls off at much higher rate.

Just revisit this topic in next twelve months and tell us the outcome.

 

[tortillasoup]

You guys don't get it do you? If O2 sensor fails either being lean all the time or being rich all the time, ECM is smart enough to turn on the O2 code and will NOT run the cat efficiency test. The enabling criteria is NOT satisfied when there is an O2 code.

When you get cat efficiency code, ECM is right 99% of the time that it is the cat that is low on efficiency.

You have no idea what you're talking about. On any OBD-II car, there are a minimum of two oxygen sensors. If one of the sensors fails, any catalyst efficiency codes are invalid. A catalyst efficiency code comes up with a failed o2 sensor because it's not switching like it's suppose to. The computer is expecting a slightly varying oxygen mixture after the catalytic converter. Typically, on non wideband oxygen sensor equipped cars, what the computer is looking for is for the oxygen mixture to vary greatly pre-catalytic converter. But post catalytic converter, the oxygen mixture is suppose to be significantly less varied, but definitely still varied. It's basically trying to see that the catalytic converter is doing its job.


For a car of this vintage and mileage, it would make absolute sense that the oxygen sensor would be done for and in need of replacement. A catalytic converter failing however, that's pretty unlikely unless the car has a major oil burning/contamination or physical shock damage has been done to the catalytic converter.

Yes, OP could take the catalytic converter off the car and have it tested to see if it's bad, but it's cheaper to just buy a $50 oxygen sensor and give it a whirl. No fucking way would I just throw a new catalytic converter onto the car without trying anything else first.

 

[razel]

If you are interested in a solution, take it to a shop. If you are interested in learning find a shop willing to show you how to diagnose it. Either way, a shop will cost you money, but you simply do not have enough tools to further correctly investigate the data. Or the right tools will cost more than the cost for the shop to teach you.

If I were you though and I do encounter oxygen sensor engine codes often in my car that also has 150k miles. They are all emissions related. Only one time did an oxygen sensor cause my car to start rough. Regardless, I don't bother to fix it until I need a smog check.

 

[sontakke]

Only if you are talking about changing the primary O2 sensor; he should go ahead and throw a new one in. It can't hurt. If he is thinking of the secondary one to fix P04x0 code, he is making a big mistake.

 

[tortillasoup]

Well the o2 sensor output on the wave forms looks identical. It's interesting to see how/why the car decided that one bank is not performing sufficiently. Get an O2 sensor spacer and put it on that O2 sensor that is causing trouble. If your code goes away, then you probably have a failing but not necessarily failed catalytic converter. Sounds like the car has dual exhaust and dual catalytic converters.


The way that B1S2 behaves just looks like a failing Oxygen sensor rather than a bad catalytic converter. The reason the computer may think it's the cat rather than the o2 sensor is because of the duration of it showing voltage is short. A clogged/failing oxygen sensor may require a lot of inertia in order to produce a voltage so the amount of time it produces a voltage will be brief. This can look like the o2 sensor is following the pre-cat o2 sensor by having a similar profile as untreated exhaust, but really it's just a bad sensor.

 

[Mark R]

Update:

O2 sensors changed. Had to order them internationally as they weren't available locally, except from the stealership, at astronomical prices. As the pre-cats arrived first, they went in first.

No change after replacing all O2 sensors. Code came back the next day.

An aftermarket catalyst was installed and seems to have fixed the problem. It's only been a few days, so I'm not absolutely sure, but there has been a dramatic change in the O2 voltage traces (see below), suggesting that it probably was the issue.

The old catalyst showed an intact substrate, so the reason for failure is not clear. The car has no other running issues.

A replacement OEM cat was approximately $1250, so was out of the question. A road legal aftermarket manicat was obtained for $450, which came with a lifetime warranty, and a certificate of OBD compliance.

For people who are interested (in particular, note the 5 minute period of steady cruising, indicated by a relative lack of switching).

1. Post-cat O2 traces, after O2 sensor replacement (bank 1 on top, bank 2 on bottom)

2. Post-cat O2 traces, following bank 1 catalyst replacement

 

[tortillasoup]

"The old catalyst showed an intact substrate, so the reason for failure is not clear. The car has no other running issues. "

Do you have the old cat? Try cleaning it like they've mentioned on youtube and see if you can get it to work properly.

 

[sontakke]

I had told you that changing rear O2 sensor will not fix the code! Please revisit this in 6-12 months when you get that code back again. Mark my words; the aftermarket converter will not make the full year.

 

[tortillasoup]

I had told you that changing rear O2 sensor will not fix the code! Please revisit this in 6-12 months when you get that code back again. Mark my words; the aftermarket converter will not make the full year.

You did call it and I agree that aftermarket converters definitely can be bad but not always.

 

[sontakke]

His came with lifetime warranty. I will be curious as to how that will be handled when the time comes to collect on it.

The sad thing is OP could have spent $5 and researched in to O2 sensor extension instead.

 

[tortillasoup]

His came with lifetime warranty. I will be curious as to how that will be handled when the time comes to collect on it.

The sad thing is OP could have spent $5 and researched in to O2 sensor extension instead.

You mean using an o2 sensor spacer instead? Some places still do tail pipe with the smog check.

 

[Mark R]

You mean using an o2 sensor spacer instead? Some places still do tail pipe with the smog check.

Round here, it's a tail pipe emissions check. However, the testers are told in the official emissions testing manual to inspect the exhaust system and sensors for evidence of tampering, and that any sign of tampering should result in an emissions fail.

My brother decatted a car once, and used to get emissions done by a mechanic who could be bribed with a case of beer. This worked fine until he sold the car. At which point the buyer realised it was decatted and couldn't get through emissions, and that caused him serious headaches, with lawyers letters going backwards and forwards. That's not something that I particularly want to repeat.