Subaru Outback Forums banner

P0420 Diagnosis

999K views 3K replies 324 participants last post by  bluehawk91 
#1 · (Edited by Moderator)
There are several reasons a P0420 will come up as an error and illuminate the MIL. A lot of people are scared of it. They get all excited and search all over the internet to try and guess at what they should do. Reading good advice and bad trying to save money and hoping it's not a Catalytic converter gone bad.

Well, I have been trying to inform as many people as I can that the 420 code is not a scary animal that is going to eat your wallet. This code simply means that somewhere in the chain of events that controls the overall performance of the engine there is a problem and you have to find it.

List of possibilities:
  • Fuel quality (low-quality fuel a/k/a Murphy Gas)
  • Fuel quantity (filter or pump issue)
  • Ignition issues (plugs, wires, coils injector flow)
  • Temperature control (too high or too low)
  • Vacuum leaks
  • Exhaust leaks (not only an annoying noise, but it also affects proper value reporting to the ECM by the AF and O2 sensors)
  • Poor amperage conductivity (i.e. battery or cabling)
Sensor issues can be:
  • MAF (Mass Air Flow Meter)
  • MAP (Manifold Absolute Pressure)
  • ECT (Engine Coolant Temperature)
  • AF (Air/Fuel ratio a.k.a. Front O2 which can be a single bank 1 for 4 cyl or bank 1 or bank 2 H6)
  • O2 (This is the actual O2 behind the CAT that lets the computer know how the CAT is functioning)

So, in an attempt to show some data and explain simply what I found, I decided to video some diag segments and show the live data from the vehicle. I will do this to show differences as they come up. Today I had 2 Subarus with emission issues. The first was a 96 Legacy wagon AWD 2.2 that was low on coolant due to a crack in the radiator and the ECT value was causing issues with starting because it controls the IAC position. It reported high temperature to the ECM which closed the IAC. No air - No start. The second is the following.

I am uploading video now. It is from a 2001 LL Bean with just shy of 199,000 miles. The MIL was on for a week and went off this morning as the owner was driving it to me. I checked all the systems on the car for error codes and the only one was a P0420 still stored as active. So I began watching data from the car.

I first let it sit for a couple of hours to cool down. Checked the oil, dirty but full. The coolant level was good. Battery amperage checked out excellent at 705 A output and conductivity measured 650 A at the block so grounding was good. Air filter was mediocre so I left it, for now, to keep the car in the same condition it was in when the MIL set. The owner hadn't driven it much since the light, so the same gas was in the tank.

There is a point in the video where I thought one thing and said another. I referred to the oxygen sensor as storing oxygen when in fact I meant the sensor was reflecting the CAT storing oxygen then it let go. There still may be an issue with the CAT, but at this time I cannot condemn a part that is working through another issue with the engine.

In this case, the engine is overheating due to a stuck thermostat and the HC's are actually being burned due to overheating before reaching the CAT. Since this is occurring and the ECM is programmed to expect other values from the O2 sensor, it reports that the CAT is not functioning as it was designed and illuminates the MIL. So a thermostat is in order as well as a fuel filter which you will see why toward the end in the fuel trim value for bank 2. Since the engine runs almost perfectly up to the point of overtemp, there is no need to replace the ignition plugs at this time. It was recommended based on the mileage as preventive maintenance. The owner is considering selling the vehicle since she doesn't drive it much anymore and has a smaller car she uses for the higher MPG's, so the car is parked until she decides to repair it or its sold.

2001 Subaru Outback LLBean P0420 Diag - YouTube

You also need to know that if any of the above issues go on too long, it will damage the catalytic converter to the point it will necessitate replacement.
 
See less See more
#1,700 ·
Might get you over the hump of this event, but there's no certainty of longevity. I'm a frugal buyer, but learned the hard way not to skimp on Subaru OEM for certain parts...axles, sensors and such. Even if that means used. A knock sensor however, may be a fair bet on aftermarket.
 
#1,703 ·
All, posted this on the Forester Fourm but this one seems to be a bit more "hands on".

2005 Forester bought last year for my son who is in school about 3 hours away.
Chasing a re-occurring P0420 code, I have replaced or checked the following:
PCV valve – OEM – original valve was not clogged
Plugs – NKG – removed plugs looked good, no evidence of oil or unusual wear.
Wires- NKG
Front and Rear 02 sensors – Denso
Cat – Eastern – original front cat disintegrated
Air Filter – Fram
Checked the coil pack resistance
Cleaned the MAP sensor
Checked for exhaust leaks, plugged the tail pipe with car on ramps and listened/ felt for leaks.
Checked for vacuum leaks, sprayed MAP cleaner around gaskets, injectors, etc.
The car runs strong, although the A/C compressor drags the idle down a little.
Replaced the clutch and had the A/C serviced at a shop.
The only other thing the car will do is occasionally after sitting overnight, it’ll crank an unusual amount before starting. Maybe coolant temp sensor?
I’ve asked my son to record some live data but being old enough that my first four cars had points I’m not really sure what to look for in the data.
I’m leaning towards replacing the MAP sensor but would like to at least test the one in the car first.
Any help, or suggestions would be greatly appreciated.
Thanks
John
 
#1,714 ·
I just saw this post so I apologize.

Rom Raider is good for logging the data and here's what you need to log:
AF 1
AF Correction 1
AF Learning 1
AF Learning 3
CPC Duty Cycle
Fuel Tank Pressure (it may or may not be an option on your data list)
O2
Engine Temp
RPM
Throttle Position in % and V
MAP (relative or absolute - doesn't matter which)
MAF Measured
Engine Load

Record a cold start up for 2 minutes, then drive it at 45 mph or so for a couple minutes to record. Get on the highway and record another 2-3 minutes at highway speeds over 55 and up to 70.

Put all 3 data logs in to a zip folder and upload the zip folder as an attachment to your post. Then we download it and look it over.
 
#1,704 ·
We have seen issues come up with failed EVAP purge valves or vent valves. These would cause errors in air metering measurements which effects fuel trims but not create a knock where the fuel trims are not really high. EVAP issues are akin to a vacuum leak. None of my Subarus expel pressure when I release the cap and the EVAP systems are working as designed.
 
#1,707 ·
Hi Cardoc, I'm still trying to figure out the P0420 on my 2002 OB (4 cyl). I now have a 2003 OB to compare it with (4 cyl, 4EAT, which has no CEL). Since my last post, I did manage to source a good used Cat from a 2003 Legacy that I'm parting and this car was running with no CEL. When installed on the 2002, it threw up the P0420 in around 50 miles. Studying the RR plots, I can see the knock correction and AF correction are behaving badly still. But another thing I noticed was the drastic difference in intake air temp. Although these logs were not taken the same day, the ambient conditions were close (if anything, a little cooler out for the 2002 logs). What's odd is that I noticed this in a log from a few days ago and swapped in a used OEM spare temp/pressure sensor before doing the attached logs but I'm still getting high intake air temps (so I'm not sure if it's a bad sensor or what). I have stock intakes on both. I will try another sensor but I'm wondering if this may be the root cause of the CEL? The Cat does seem to be working on the 2002.
 

Attachments

#1,709 ·
I'll look at the logs a bit later, it's a busy morning.

IAT feedback is a determinate factor in engine load and air mass calculations that the ECM uses to adjust the engine's fueling. If the IAT is high, that indicates a lower air mass and the ECM will reduce fuel. Reduced fuel based on IAT could mean that the cylinder is getting a lean shot of fuel due to a higher mass of air entering combustion than is recorded and this could cause detonation which would be a knock. The ECM would correct for the knock with increased fuel and would also increase the fuel based on AF sensor feedback.

The fuel map determination would be altered and you would see a positive percentage increase in STFT and over time, the LTFT. Constant adjustment of the fuel trim affects the catalytic converter's ability to maintain proper temperature to reburn the gases in the exhaust stream.

I'll post up after I've had opportunity to look at the logs and see if this is the case.
 
#1,710 ·
You left out the rear O2 sensor data. You can remove the roughness pids.

Did you ever try running the engine without the catalytic converter/manifold?

I currently have a 11 Forester I'm working with that is throwing a P0171. The STFT is good a lot of the time, the LTFT is decent, but it's the Learning Sensor 3 that is high. I see yours is high as well, over 45%.

I don't see this too often and most times it's due to an air leak or exhaust leak. In the case of the Forester I believe the cat is restricted because there is not a waveform that runs low/high, but rather a high voltage feedback in the .85-.95V range which the ECM determines as a rich condition exiting the catalytic converter and therefore alters it's trim to try to maintain the stoichiometric volume and get the cat to function better. The lean code is not referring to the AF data feedback but is reference to the O2 feedback based on the cat's inability to burn. I have a cat on the way for the Forester and am 95% confident that the cat is the P0171 issue. We'll see when it arrives later in the week.
 
#1,721 ·
Just a early observation regarding the cat:

At around rows 94-95 in the log, the throttle is wide open (100%), the engine speed is around 5000 rpm (and rising), and the MAF is around 100 g/s and rising. We also know from the previous log that the MAF can reach as high as 120 g/s. The data suggests flow-through is good, and doesn't point to a restricted converter.
 
#1,729 ·
Hi Cardoc, relative to post #1715, I was going to try swapping the IAT/Pressure sensors but discovered that there was a change from 2002 to 2003; the '03 sensor is only for pressure (and has a smaller hole in intake manifold). They moved the IAT (alone) sensor to the air filter box. As such, it would seem the '03 would tend to read lower temperatures at the airbox location. So I guess my question is, for ambient conditions of 65 F, fully warmed up engine, what range should I expect to read on the intake manifold location IAT?
 
#1,730 ·
Probably close to 100° or higher if the engine has been running for a length of time with the hood closed and the thermostat has cycled. Higher temperatures if the engine sits idling with the hood closed since there's no air pushing in from the outside in to the air ram or circulating around the engine bay from driving.

If you have a laser thermostat, or access to one, you could get a round about number by measuring the temperature at the throttle body and compare that to the sensor reading.

The heated element is in the tip of the MAP. Have you taken it out and looked for carbon build up? Carbon from the PCV system circulation can build up and alter the temp measurement.
 
#1,731 ·
Thanks Cardoc. I do have a laser thermometer somewhere and will have to try that. The tip of the MAP/Temp sensors I've tried looked clean. I swapped in another used sensor I had, cleared the codes but engine light popped on again in around 60 miles. I'm thinking to try putting a Subaru knock sensor back in (I have a couple of used ones). Maybe the cheapie ebay sensor I have in there is not doing the job properly.
 
#1,735 · (Edited)
Not sure if I asked before: Does Romraider Logger display the ECU ID at the bottom of the Logger screen, and perhaps even the [strike]ECU ID[/strike] CAL ID?

There might be a specific ECU definition file available that will provide more information, including the short term "Feedback Knock Correction", Fine Learning Knock Correction (by engine rpm/load) and Ignition Advance Multiplier (IAM). These might be able to provide more clarity regarding the Knock Correction Advance, which is made up of several distinct knock-related parameters.

Also, is the car a FED-spec or CA-spec version?
 
#1,749 ·
I'm working my way through this thread but I jumped to the end to post a question.

2003 WRX (turbo). Bought it from a guy after he blew his 2nd engine. Bought a short block and went from there. It was a father/son project. And it was like drinking from a fire hose for me after a 40 year break from serious engine work. I am learning. The car has 130,00 miles total and 17,000 on the engine.

Around 4,500 miles ago I replaced the rear O2 with a Denso after a P0420. I thought the problem was solved.

Then after another 1,500 we got P0420 again. A smoke test revealed a couple of vacuum leaks and a new MAF sensor raised the fuel mileage considerably.

Sure enough, after another 2,500 miles P0420 is back. I bought a scanner back when we were doing the engine but never learned to use it. I thought I was getting what I needed from the wireless OBDII/Android w Torque Pro. Today I buckled down and tried the scanner. I have a lot to learn. In the driveway the downstream O2 (if I'm doing this right) was flat. After a 30 mile drive it's another story. Does seeing this waveform tell me that I need a CAT? I won't stop reading up on the subject but I can go ahead and order a CAT if I've proven that it's bad.

And just so it gets said, thanks for the education cardoc and everyone!

 
#1,751 ·
Yep, that's a problem cat.

How much oil are you going through?
Have you checked the turbine side of the turbo for oil?
How much is in the compressor side and the intercooler piping?
Did you put in a new cat with the new engine? If the other engine blew oil or coolant down the pipes, that could be what trashed the cat.
 
#1,753 ·
Ok, I think it's the CAT on the down pipe (comes out of the turbo and goes down to the normal exhaust) that I need. There is yet another CAT before the turbo but I'm reading that it's just to help when the engine isn't up to operating temp and won't affect the P0420.

It's tempting to live with this but I'm just not bent that way. I'm used to logging with a cheap bluetooth unit and an Android phone but I ordered the Vag Com cable to use with RomRaider. If I'm gonna ask for more help from you guys I want to make it as easy as possible...

I realize that the name of this forum has Outback in it. Is it acceptable to ask for help with my WRX here? There is a lot of WRX information online but I'm not interested in more performance, just a stock setup and clean air.

When the new cable comes, assuming I can get one of you folks to help me with the logs, do I log with the old CAT or the NEW one? Once I got the fuel trims down and the mileage up I really thought I was golden until the old P0420 came back.

I'm up to page 108 in this thread. I've learned a lot. Still, I appreciate any help with my 'homework' that anyone wants to give. I have 4 Subarus and this probably won't be my last rodeo. The more I know, the better.
 
#1,754 ·
@tmcquinn

I don't mind if you post up questions about a WRX, Forester, Impreza, what ever. It's all Subaru.

This thread also applies to other makes and models when it comes to diagnosing a P0420 and the root cause for the lack of efficiency on the cat's part so the info here can be used with other vehicles, even German cars.

There should only be piping between the cylinder heads and the turbo. If you have a CAT prior to the turbo, something isn't right. You don't want any restrictions prior to the turbo and a catalytic converter is a restriction. The exhaust manifolds feed to the turbo, then you have a CAT in the downpipe coming out of the turbo followed by another pipe with either a second cat and/or a resonator.

All you need is the downpipe cat. After that, you can put in whatever you like for sound, or no sound.

As far as what killed it, I'd say it was damaged by the engine failure or it started before the engine failed and it's digressed to it's current state. Rich fueling doesn't help any either.

When you get the CAT installed, clear/reset the ECM and drive it for a couple days so it will relearn the fuel trim, then you can log data from the car and post it up here in a zip folder.

Rom Raider is relatively easy to use. Once you choose the parameters to log, you can save the profile and every time you connect to the car it will automatically load that profile.

To Log:
AF Sensor (either voltage or Lambda/Stoich)
O2 Sensor (voltage - don't need the heater circuit or resistance)
ECT
TPS (Voltage and %)
MAF
MAP
Wastegate
STFT
LTFT
Fine Knock Correction
Ignition timing
Injector pulse
RPM
Battery Voltage (at least for the first few logs)

You may also want to think about getting it tuned proper, either for performance or economy. Plenty of tuners work via email. A few are on this forum.
 
#1,762 ·
So I have the new cat and the new A/F sensor installed and we have Romraider going. I'm ready to log and try to convince myself that the cat killing issue is no longer with me. The battery was disconnected while I worked on the car.
Do I start logging now or wait until the ECU gets out of 'learning mode'? (Not too familiar with learning mode...)
 
#1,755 ·
Thank you, sir! It will take me a while to get everything and get the parts installed but I WILL be back.

Hah, I think Subaru agrees with you on the CAT between the heads and the turbo. Not only did it do very little, there is always the chance it will be damaged and spew debris into the $1,400 turbo, ruining it. After a couple of years, I think it was sometime around 2005, Subaru removed this CAT from the WRX. And just about all of the performance guys either replace it with an aftermarket unit or butcher all the innards out of it, making it just another pipe. I'm willing to take my chances with it. My son does not push this car very hard, though I guess that's no guarantee it won't fail catastrophically some day.

I'm still slogging through this thread and looking for a reasonable aftermarket replacement. I'm leaning towards Rock Auto, though they offer more than one brand. Any recommendations would be appreciated. At the rate companies are being acquired and having their facilities shipped off to China I'm half afraid to rely on any reviews more than a few months old. A lot of companies that I used to think of as good quality vendors seem to be deteriorating these days.
 
#1,756 ·
@tmcquinn
Is it 50 state, FED or California emissions? You would need to look at the emission sticker on the hood or run your VIN with the dealer parts dept and ask them.

Davidco seems to work most times. The Bosal is the one for California (CARB) emission ratings and complient with the CARB ECM programming. Of course, if you aren't in an area that does tailpipe analyzing, you could get by with the FED/50 state cat. I don't think your ECM will disagree with it, especially if you get the car tuned right.

Oh Canada: https://www.ebay.com/itm/fits-Subar...m1a335b9256:g:19MAAOSwrRlZl1t3&frcectupt=true

Me, I'd get a header to replace that manifold cat. The turbo will work more efficiently and the car will breathe better. https://www.ebay.com/itm/For-02-07-...ap0Jc:sc:UPSGround!78745!US!-1&frcectupt=true
 
#1,757 ·
I was going to say it isn't California compliant but now I'm confused. Here is the sticker. It clearly mentions California...

The Canada CAT tempts me. But I must be on a list or something because every time I order something from Canada it takes 2 weeks from the time it's shipped until it arrives here. Plus, I asked them to verify that it has the bung for the A/F sensor and they didn't answer. I will probably go for the Davico from RockAuto. I know I can count on them and it will come with gaskets.

I respect what you're saying but let me get this issue behind me before I propose to eliminate the other CAT. The Family Finance Committee has been testy lately.

I'm going to take one more run at the car today. It's in the garage now. I'm still making my way through this thread and I want to test for an exhaust leak before I order a CAT. Another day of delay isn't going to hurt. I have a spare car that my son can drive. Anyway, I'm starting to feel like I'm learning. I put the scanner on my 2012 n/a Impreza and made another highway run. The downstream O2 sensor would certainly show movement when the throttle was moved but at steady highway speeds it was flat and very close to .5 volts (if memory serves). And the cable came today. I am going to fix this one way or another.
 

Attachments

#1,758 ·
Wow, it doesn't take long for a new exhaust to start looking nasty, does it?

I did the smoke test. The photo might be misleading but I was under there. It's leaking. It's leaking under the heat shield on the downpipe and then shooting out from under the heat shield. The CAT is in the downpipe. It's time to spend money. It would have been nice to have found a loose bolt or a bad gasket but ya can't be lucky all the time.

Everything downstream from there is less than 2 years old and showed no sign of a leak.

Downpipe & front of car to the left, exhaust and rear of car to the right.
 

Attachments

Top