Subaru Outback Forums banner
  • Hey everyone! Enter your ride HERE to be a part of October's Outback of the Month Challenge!
1 - 5 of 5 Posts

·
Registered
2006 2.5i Limited Manual
Joined
·
74 Posts
Discussion Starter · #1 ·
Hey all. A few weeks ago I got P0420 for the first time, read cardoc's seminal thread on the issue, and suspected it was the sketchy rural Maine gas I had recently filled up with. Since then it's gone and come back a few times, so I figured it warranted a little more looking into. I logged some data on a recent trip and was hoping to get some help looking through it. This trip included a cold start, some brief slow speed driving, and then some highway speeds. Note that it just so happened that on this trip the CEL was not lit. Data was grabbed with a cheap bluetooth ODB-II reader & Torque pro

Coolant temp looks good, climbs pretty linearly then evens out:
Rectangle Slope Font Parallel Plot


Engine load looks pretty normal I guess, I don't think there's too much to learn here
Azure Rectangle Font Parallel Plot


Fuel flow:
Rectangle Azure Font Parallel Slope


Now we're getting into the ones I don't understand as well. Here are the three fuel trim values (Long term, bank 1 sensor 2, and short term). In this post on the P0420 thread @cardoc mentions that he knows it's running well because long term fuel trim is 0. Mine fluctuates, but is almost never 0...
Rectangle Slope Parallel Font Diagram


Finally, there's the O2 sensor readings. In @cardoc 's thread, he compares the front and rear O2 sensor readings, which I wasn't able to find in my data. The single O2 sensor I do have (02 bank 1 sensor 2) I believe is the rear one, based on @cardoc 's video. Are the equivalence ratios or wide range o2 sensor voltage useful to diagnose potential cat/engine issues? Any help or resources on diagnosing this would be greatly appreciated.

Rectangle Slope Font Plot Parallel


I can also post the full data, or grab some more under specific conditions if that's needed. Thanks!
 

·
Registered
2012 OB , 2017 Impreza
Joined
·
3,792 Posts
Any help or resources on diagnosing this would be greatly appreciated.
I would suggest reviewing the hours of excellent troubleshooting videos on uTube. Specifically from "south main auto" and "pine hollow auto diagnostics"

Eric and Ivan are masters at interpting this kind of data. People drive for hours just to have these guys FIX problems where others failed.

Do not get hung up on Subaru specific videos. The operation, troubleshooting and diagnosis is similar on all obd2 systems.
 

·
Premium Member
01 VDC, 05 R Sedan, 06 BAJA EJ257
Joined
·
16,253 Posts
LTFT within 7% +/- is okay. Closer to zero is better. With your LTFT hanging around +5% and higher, I need to ask how old are the spark plugs, when was the timing done and how many miles on the car? Is maintenance up to date? How old is the engine oil at the time you logged this and is it full?

The AF sensor Voltage will be up/down with changes in fuel trims, on throttle, off throttle, and will closely match the STFT when overlaying a graph in terms of the spikes and drops. It just gives you the voltage measurement sent back to the ECM regarding exhaust oxygen levels before the cat. Rich exhaust is a low voltage value, lean exhaust is a high voltage. The front AF sensor is a monitor that the ECM uses to adjust fuel delivery and timing to get the performance from the engine it was designed for, manage the catalytic converter temperature and keep emissions cleaner. The only time the AF sensor data is ignored is WOT; the ECM just enriches fuel and makes timing adjustments based on other factors and feedback.

Other issues that can cause LTFT to increase is a worn AF sensor. It can send improper feedback and the ECM is erroneaously correcting the fuel. In this case, if the sensor is worn, the higher voltage feedback is telling the computer the exhaust is lean and it will add fuel. This is also what is relayed with a high or low LTFT calculation. When it adds fuel the cat doesn't, or can't, adjust fast enough and you get a short waveform from the rear O2. The front and rear sensors are maintenance items on any car. The constant heat and exhaust flow burns off the coating on the sensor and it doesn't reacte as it should. Or the seal on the sensor gets worn, contamination from oil/debris is on the wiring, or weak heater circuit, and the variation between atmosphere oxygen and the exhaust oxygen is compromised. Usually the sensors last about 85-100k miles.

The rear O2 data on your car is indicative of an engine performance, or feedback issue. Being a manual trans, when you let off the throttle between shifts the rear sensor will fluctuate slightly with fuel changes between gear shifts. What you look for in a manual is a the feedback while at a steady throttle drive and compare the rear O2 voltage with either the STFT value or the AF voltage feedback. A Rear O2 waveform that fluctuates up/down matching or closely matching the STFT means the cat is not able to burn all the fuel; it's not as efficient as it should be.

Other things are battery power supply and the grounding. If the battery is weak, the cables corroded, or there's poor connections to the engine and body, it will affect the way the sensors and ECM function. Inputs and outputs have to have a good supply of electricity to work. Since your car is manual, a good output for the battery would be 550 CCA, unless you have added accessories like amplifiers, then you want a higher output battery. Conductance through the grounds should be no more than 0.01 ohm from the negative post to the bracket by the starter. You want the same from the post to the alternator case. If it's higher, clean grounds or replace/add cables. The same measurement is included for the positive cable between the positive post and the fuse box lug and alternator to main fuse.

Fuel octane also makes a difference. Higher octane fuels resist detonation. The knock correction factor indicates how much the ECM is correcting for detonation. Years of low octane use allows carbon to build up in the combustion chamber and on the valves and this in conjunction with oil deposits from the PCV system creates hot spots on the piston head and cylinder head which assist in early detonation of the fuel and instead of a steady burn, you get an explosion; the knock. The ECM most times adds fuel to correct for the situation and this exacerbates the issue with engine performance and the function of the cat suffers. Correcting for carbon and oil deposit build up in combustion is easily done with an upper induction/combustion cleaner that you can do in your driveway. You can also add a cleaner to the fuel tank at a fill up and this will help dissolve gum and varnish build up in the fuel system and clean the injectors so they can spray more efficiently.

There's alot involved in engine performance and oftentimes with older cars and alot of miles, there can be several things that together create what seems to be a major issue. The caveat is finding the one thing, or several, that is contributing to the engine and catalytic converter performance. The code is not "Catalytic Converter Failure", it is "Catalytic Converter Efficiency is Below Threshhold". The efficiency of the cat is determined by the efficiency of the engine and transmission. In your case, the transmission efficiency is determined by your use of the clutch and the clutch wear.

If you can put a log up on Drive or similar and post a link to it here, I will look it over to narrow it down.
The sensors are labeled as bank 1 for both sensor. Bank 1 sensor 1 is the front AF, and bank 1 sensor 2 is the rear O2.

Log the following to start with:
AF Correction (STFT)
AF Learning (LTFT)
AF Lambda or you can use the AF voltage or AF ratio which will be in af 14:1 type format
AF Resistance in ohms
Rear O2 voltage
RPM
TPS
Knock Correction
Engine Load
MAF
IAT

Don't really need to see ECT unless you have seen a large fluctuation in temps after the engine warms up. What you posted above indicates the engine temperature management is good.
 

·
Registered
2006 2.5i Limited Manual
Joined
·
74 Posts
Discussion Starter · #4 ·
@cardoc, your response went well above and beyond my expectations and wildest dreams, and I genuinely appreciate you taking the time to walk me through the minutia of diagnosing my vehicle. You clearly spent time and effort in crafting your response so I'll attempt to do the same below.

how old are the spark plugs
Plugs & wires done at 149k, the old ones looked ok but not great. I made a post about their condition and used a borescope to peek at the piston heads

when was the timing done
Both my front and rear crank seals decided to piss themselves in Wyoming during a cross country road trip, and I had a dealer in Colorado replace them and the timing belt at 152373. Haven't even thought about the timing since.

how many miles on the car?
173k

Is maintenance up to date?
Generally speaking, I belive so. I do my own oil changes of full synthetic mobile 1 at an interval of ~3-5k and take a peek at everything else while I'm down there. I'm not aware of any maintenance/wear items that are behind schedule (unrelated to the P0420, I need new front brakes & rear end links).

How old is the engine oil at the time you logged this and is it full?
I logged this data shortly after changing the oil, maybe 100-300 miles after the change. Since changing that oil I have put 800 mostly highway miles on the car and the check engine light has not re-appeared curiously. I always have a hard time measuring oil if I'm not on a perfectly level surface, but I have reason to believe it's good and full.

The front and rear sensors are maintenance items on any car.
I didn't realize this, but I suppose it stands to reason given the conditions they operate under. Replacement sensors from Subaru are a bit pricey, any aftermarket suggestions? Rockauto has Denso, NTK, and Delphi listings for both front & rear O2 sensors. This car has lived in the northeast for its whole life, but the exhaust is actually in remarkably good shape iirc. How realistic is it that I'll be able to replace these sensors without breaking something?

Other things are battery power supply and the grounding.
The battery was replaced relatively recently after the old one was drained multiple times by a faulty CD player, so I believe the battery is good, though untested.

Conductance through the grounds should be no more than 0.01 ohm from the negative post to the bracket by the starter.
I'm only mentioning this because I recently learned this fun fact myself, but conductance is a distinct unit and it's the inverse of resistance (Resistance = Ohms, Conductance = 1/Ohms). The unit for conductance used to be called Mho's, and was represented by an upside down ohm symbol (℧). When I get a chance I'll make sure my grounding is good, this could be a contributor.

Correcting for carbon and oil deposit build up in combustion is easily done with an upper induction/combustion cleaner that you can do in your driveway. You can also add a cleaner to the fuel tank at a fill up and this will help dissolve gum and varnish build up in the fuel system and clean the injectors so they can spray more efficiently.
Any recommendations for brands?

If you can put a log up on Drive or similar and post a link to it here, I will look it over to narrow it down.
I'll collect a log with the data you listed and post it here when I get a chance, I'd love to have you look over it.


I religiously track my fuel usage/mileage at fill-ups and my fuel economy is surprisingly still good, averaging 27mpg with mostly highway and some city driving.
 

·
Premium Member
01 VDC, 05 R Sedan, 06 BAJA EJ257
Joined
·
16,253 Posts
Yes, conductance test and resistance is different. I think conductance test because that's what I do in the shop with a Midtronics.

Mobile1 may be a contributor to high carbon build up in the engine. I see alot of cars that use it and carbon on the valves is heavy. VW group engines are especially heavy, but the PCV systems on those engines is also a contributor.

Denso sensors from Rock Auto are good. I've not used NTK or Delphi sensors in a Subaru, so I cannot comment on their performance. I do know Bosch sensors are too slow for the Subaru ECM.

Berryman's B12 Chemtool, Sea Foam or similar cleaner run through the intake as an induction cleaner will dissolve alot of the buildup. Adding either to the fuel tank occasionally will also help keep it at a minimum in combustion. As I posted on your thread about the plugs and carbon on the pistons, if the deposits are heavy in the intake, removal of the intake to clean it and the intake ports on the heads is best. The only way to remove the carbon from the combustion chamber is heavy and repeated doses of the cleaner run through the intake with the car running, or remove the heads to clean it all out. Carbon on the piston heads is a contributor to knock, preignition and detonation because the carbon gets hot and stays hot and it ignites the fuel when it's sprayed in the cylinder before the spark from the plug. Think about charcoal in a BBQ. It gets hot and continues to create heat. If you spray charcoal with a fuel it ignites it. Same with the combustion chamber. Even after you shut the car off, the carbon is hot and it's also baking the residual oil deposits and left over fuel in the chamber and the build up continues to expand.
 
1 - 5 of 5 Posts
Top