Subaru P0420

2003 Subaru WRX

I am an ex parts changer who occasionally relapses...

The consensus in the Subaru community seems to be that P0420 is something you reset and live with. That just rubs me the wrong way. The exhaust has about 15,000 miles on it and that includes the rear CAT. As near as I can tell this little 4 cylinder only has one 'bank' so I can't do comparisons that way (watching the oxygen storage test on the premium channel). I really, really thought this was behind me after I replaced the MAF sensor, fixed some vacuum leaks, and got a huge jump in fuel mileage and really low fuel trims. But it's back. This is my son's car and he did admit to a tank of cheap gas just before the code came up. I'm ready to bite the bullet and try to get to the bottom of this. Of course, it's nothing to go 1,000 miles between P0420 CELs but I am really tired of this.

So today is P0420 day. I keep seeing that an exhaust leak can cause this. I'll get it up on jack stands today and see if I can find a leak. Is there any value in trying to use my smoke machine to test for a leak or should I be able to hear any leak that's bad enough to cause the code?

And I'm not easily insulted on this one. I'll read any advice folks have to offer. I've never really learned to use my Genisys EVO scanner but I'm ready to get off of my behind and give it a shot.

Well, I was overly optimistic about how much data would be in the 'freeze frame'!

tmcquinn wrote: this little 4 cylinder only has one 'bank' so I can't do comparisons that way (watching the oxygen storage test on the premium channel)

The oxygen storage test is not a comparison of bank-to-bank (left and right side of a "V" engine) oxygen sensors. It is a comparison of BEFORE and AFTER the Catalytic Converter oxygen sensors. You are right that the 4 cylinder engine only has one bank. But, it has two oxygen sensors on that bank. They would be identified as Bank 1 Sensor 1 (upstream, or BEFORE the Catalytic Converter), and Bank 1 Sensor 2 (downstream, or AFTER the Catalytic Converter). You can, and should, do an oxygen storage test.

I've never really learned to use my Genisys EVO scanner but I'm ready to get off of my behind and give it a shot.

Find and graph your O2 sensors. Try the Oxygen Storage test.

Is there any value in trying to use my smoke machine to test for a leak

Absolutely. A smoke machine can, often, make light work of finding exhaust leaks.

Thanks. The clock ran out on me today. My scanner has a user interface that confuses the living hell out of me. I want to try it again in the morning when I'm fresh and it's cool in the garage. There's a Subaru guy on youtube who uses the same scanner. I know it will do exactly what I want. It's just a matter of sticking with it. Or maybe even seeing if the CD has a manual on it!

I am on my way. Sometimes a good night's sleep can cure a bad case of the dumbass. One of the buttons was wonky, it's getting better with use, and reading the manual didn't hurt. Just don't tell my wife about the manual thing...

I'm going to take a drive to capture some data and sit down and do a LOT more reading. But getting this tool going is a big step for me. I've gotten some pretty good use out of the $10 OBDII bluetooth adapter with an Android phone but this should be a big step up.

This is still my favorite forum, by far, but I've found a Subaru specific forum where this issue has been discussed in great detail. Still, I figure it isn't polite to just disappear without an update.

I'm getting better with my scanner. It's been discontinued and there's no help from the manufacturer but it's paid for and I think I can get a lot of use out of it. Still, I'm pretty jealous of the hardware some of you have.

The photo I posted above was while the car was sitting in the driveway. I took it for a drive and got quite a different result. The downstream O2 is anything but flat. Unless I'm totally not understanding Paul's videos, the CAT (the bloody car has 3 of them) is not working. The CAT on the pipe coming down from the turbo is one of the few original parts. It probably had a good dose of oil, coolant, and metal shavings when the prior owner grenaded the engine.

(I hope the photo shows. I never seem to get it on the first try.)

What is the time base on that screen shot? Is that going rich and lean with throttle input?

Let me see what I can find. Right now I'm struggling with translating the terms Subaru/OTC/Torque Pro use and how to translate back and forth. It's confusing.
That photo is my OTC Genisys, looking at the defaults it brought up for my car. There are many more parameters in the menu that can be selected but most of them just say 'no data'. And the amount of time it takes to configure is substantial. I take it that this is because there are a LOT of different cars with many different configurations. I will get better at this.

That probably represented a steady 30 mph in 3rd gear since I was almost home. But I had a 30 minute drive at a steady 65 mph in 5th gear and the waveform was just as ragged.

I want to correct something I said earlier. The manufacturer (I think that it's just Bosch who bought the product line) couldn't give me support but one of the techs did point me to a youtube channel where I'm learning about the scanner. So it's not like they just gave me a 'Good luck, Gomer' and sent me on my way.

I also want to put the scanner on my 2012 Impreza and take a drive. I'd like to see a flat wave on that rear O2 sensor to help convince me I'm not doing something stupid.

Sorry, I didn't mean for that to sound so technical. Just wondering if we're looking at a few seconds or few minutes on the screen

My goal is to become technical enough to run with the big dogs!

The spikes in red indicate a throttle snap of some kind, and are consistent between the MAF and the AFR wideband sensor.

The yellow is a relatively stable sample of steady light throttle, based on the MAF and AFR sensor. This same yellow sample in the post-cat sensor is "no bueno"... the purple line is roughly your .5v territory, and there shouldn't be all those cross counts there. A P0420 sets because of because of too many downstream cross counts, which you have. Based on the limited data here, I say you either have an exhaust leak or a junk Cat.

Also, since you are new to the Genisys, (you may have already figured this out) think of the light blue box as a magnifying glass. Dotted trace is displayed in the bottom. Numbers on the right are Min, Max, and Avg.

These things were hot stuff when they came out back in the day. Man, 10-15 years ago it seems like every shop had one. It was a great tool for its time for sure, and still a great one to learn on if you are newer to using scanners. I still run across some old techs that use them occasionally.
Unfortunately I think they haven't made their updates since 2013, maybe even before that.

Thanks, that's what I thought, but wanted to make sure I was seeing what I thought I was seeing.

Somehow, my brain completely omitted the MAF signal. Maybe because my Autel display looks the same but only shows 2 graphs at a time.

TheTechWhisperer wrote:
The spikes in red indicate a throttle snap of some kind, and are consistent between the MAF and the AFR wideband sensor.

The yellow is a relatively stable sample of steady light throttle, based on the MAF and AFR sensor. This same yellow sample in the post-cat sensor is "no bueno"... the purple line is roughly your .5v territory, and there shouldn't be all those cross counts there. A P0420 sets because of because of too many downstream cross counts, which you have. Based on the limited data here, I say you either have an exhaust leak or a junk Cat.

Also, since you are new to the Genisys, (you may have already figured this out) think of the light blue box as a magnifying glass. Dotted trace is displayed in the bottom. Numbers on the right are Min, Max, and Avg.

These things were hot stuff when they came out back in the day. Man, 10-15 years ago it seems like every shop had one. It was a great tool for its time for sure, and still a great one to learn on if you are newer to using scanners. I still run across some old techs that use them occasionally.
Unfortunately I think they haven't made their updates since 2013, maybe even before that.

Very cool. Thanks for taking the time to write all of that.

Cross counts means deviations above and below .5 volts?

My compressor is out of town on a house I'm remodeling. So my smoke tester is useless. How much of an exhaust leak would it take to hose things up that much? I'm thinking that if I get the car on jack stands and go over the exhaust with a stethoscope maybe I can dismiss the exhaust leak scenario. I'd rather do the smoke test but that would mean more of a delay. I hereby release anyone and everyone who comments from any liability, moral, financial or otherwise, if I buy a new CAT and then find out it was something else!

I wonder, I have one of those 5 gallon portable air tanks that holds 120 psi. Maybe that would be enough for the smoke tester? It runs the air pressure through a regulator that reduces the pressure down to 'light breeze' levels...

Man, I haven't seen a Genisys in years! TTW is right - these things were everywhere at one point. They even had five gas and DSO attachments.

I'd definitely at least listen for exhaust leaks as a matter of good practice. Otherwise, the cats are done. The downstream switching that TTW highlighted (during steady state cruise) is classic inefficiency.

If your state doesn't do tailpipe testing, then I'd suggest only replacing the cats ahead of the downstream sensor. If you mean to keep this car, then I'd suggest buying the best cats you can afford. Aftermarket is a roll of the dice.

Well, I know the Genisys is out of fashion but unless they start giving away SnapOns mine is here to stay! It won't work on my newer cars but neither will I, at least not while they're still under warranty.

My five gallon portable air tank was sufficient to drive my cheapo smoke tester. It looks like it's leaking out of the down pipe that contains the monitored cat. That cat and the a/f sensor are two of the last original parts on this engine. Replacements have been ordered.

My state doesn't do tailpipe testing. I'm rolling the dice on parts but the dealer price on the cat. would make a nun curse.

It took me a while to come up to speed on this car. It has a pre turbo cat. that just about everyone removes, a cat between the a/f and downstream O2 sensors, and a 3rd cat. that is downstream and isn't monitored. Maybe someone gave Subaru a deal on cats back on 03. Anyway, I hope that I haven't ruined cat 3 with my ignorance (it's relatively new) but the best I think I can do is take before/after temperatures on it when I'm done.

The amount of reading I had do on this was beyond belief. The devil on one shoulder wanted to just reset the code every few months while the angel on the other shoulder reminded me what the air in Los Angeles looked like in the 70s.

I'm a little behind. Things are breaking faster than I can fix them. At least it all waited until it's warm...

I have the new cat installed, along with a new A/F sensor. I'm behind on logging some real data but I had to drive all over town this morning to get parts for my other son's car. I will get to it. I had my spare phone with Torque Pro and my bluetooth OBDII unit running for the drive. Anyway, the one thing I noticed was that the A/F ratio changes a lot more with the new sensor. It's not like it's holding steadily on 14.7 with hardly any change now. I'm not worried yet, the fuel trims are still pretty low. But, this is what I should expect to see if the old sensor was tired, right?

(The old sensor was a wreck. My guy who specializes in bolt extraction struggled with it and then had to carefully remove all the threads that it left in the exhaust.)

The upstream A/F behavior doesn't sound unreasonable, especially during throttle changes. Does the downstream sensor hold steady while cruising at a steady load?

If you've driven it for awhile, you can also check the catalyst readiness monitor status. If it completed with no codes set, you're good to go.

Thanks!

I will absolutely be checking the downstream O2. I'm hoping that the prior owner ruined the cat but I will be checking. I don't want to buy another one and the labor, in a tiny garage on jack stands, was no day at the beach.

I will look for 'catalyst readiness monitor status', thanks. What's killing me right now is that Subaru, Torque Pro, Genisys, and RomRaider, the tools I have available, all seem to use different terms for the same thing. I might have a future question about that one.

No worries, ask away! I find that Torque Pro is not good for emissions monitors or Mode $06, so you might try the Genisys instead.

Wilco.

I was just on a parts run for my 2nd broken WRX and figured a little information was better than none.