Subaru Outback Forums banner

1781 - 1800 of 1819 Posts

·
Registered
Joined
·
31 Posts
I'm sure you're right but I have a lot to learn about it. The car came with a Cobb A/P that I unmarried from the ECU and sold on Ebay. I honestly thought you had to have an A/P to tune, and it sounds like I was wrong.

Whatever happens I'm thrilled to be putting this one back on the road.
 

·
Registered
Joined
·
142 Posts
@cardoc welp I thought my P0420 was gone for good after replacing spark plugs, using seafoam and cleaning the TB. It has come back now every time the readiness checks are fully complete (every 200-300km or so). This is with top tier Costco gasoline. I plan to properly asses everything using this thread so I can avoid loading the parts cannon. Currently I am just clearing the code as it comes up with my ultraguage that I always have hooked up. The engine runs fine and fuel economy is good but I want to avoid having a CEL come up and needing to be cleared every so often (my wife will be driving this most of the time).

My ultra gauge isn't quick enough to properly assess O2 sensor voltage but next week I will use my scanner and plan to check the following PID's:

MAF flow
LFT's and SFT's on both banks
Pre and post O2 voltage on both banks

My engine tends to hover around 200F at 20*C ambient temps with the a/c off so it appears my thermostat is working as expected. I have noticed recently that my LFT's tend to read +4% and alternates between bank 1 and 2. I have 260k km on the engine and it is a EZ36D.

Anything else you think I should be monitoring or would want to see to determine the issue? It probably isn't a bad thing to replace the MAF and all O2 sensors due to the mileage, but the Family Bank is tight right now and I don't have the funds to replace parts that are not definitely causing thee issue.
 

·
Premium Member
Joined
·
11,899 Posts
Discussion Starter #1,783
@cardoc welp I thought my P0420 was gone for good after replacing spark plugs, using seafoam and cleaning the TB. It has come back now every time the readiness checks are fully complete (every 200-300km or so). This is with top tier Costco gasoline. I plan to properly asses everything using this thread so I can avoid loading the parts cannon. Currently I am just clearing the code as it comes up with my ultraguage that I always have hooked up. The engine runs fine and fuel economy is good but I want to avoid having a CEL come up and needing to be cleared every so often (my wife will be driving this most of the time).

My ultra gauge isn't quick enough to properly assess O2 sensor voltage but next week I will use my scanner and plan to check the following PID's:

MAF flow
LFT's and SFT's on both banks
Pre and post O2 voltage on both banks

My engine tends to hover around 200F at 20*C ambient temps with the a/c off so it appears my thermostat is working as expected. I have noticed recently that my LFT's tend to read +4% and alternates between bank 1 and 2. I have 260k km on the engine and it is a EZ36D.

Anything else you think I should be monitoring or would want to see to determine the issue? It probably isn't a bad thing to replace the MAF and all O2 sensors due to the mileage, but the Family Bank is tight right now and I don't have the funds to replace parts that are not definitely causing thee issue.
For starters, the MAF doesn't go bad on these.

Looking at the rear O2 sensor, if there is a waveform at idle or cruising, the cats are bad. If one is steady and the other has a waveform, it's just one cat, the one with the waveform.

If the values are steady but show high voltage, over .8V majority of time, that means the cat is burning but not good enough. Or, you have weak O2 sensors that are stuck or have weak heaters in them.

AF sensors also get weak and with 260k km, you may need new ones.

LTFT under 5% is normal. The ECM makes adjustments according to ambient temp, MAP and BARO measurements and IAT. Higher IAT gets more fuel, lower MAP values get more fuel, and if the car is cold, more fuel. The LTFT also fluctuates while the car is operating. I wouldn't worry about the LTFT unless it goes =/-10% over a long period of time.

Wait to see what the O2 data looks like before you put parts to it.
 

·
Registered
Joined
·
142 Posts
For starters, the MAF doesn't go bad on these.

Looking at the rear O2 sensor, if there is a waveform at idle or cruising, the cats are bad. If one is steady and the other has a waveform, it's just one cat, the one with the waveform.

If the values are steady but show high voltage, over .8V majority of time, that means the cat is burning but not good enough. Or, you have weak O2 sensors that are stuck or have weak heaters in them.

AF sensors also get weak and with 260k km, you may need new ones.

LTFT under 5% is normal. The ECM makes adjustments according to ambient temp, MAP and BARO measurements and IAT. Higher IAT gets more fuel, lower MAP values get more fuel, and if the car is cold, more fuel. The LTFT also fluctuates while the car is operating. I wouldn't worry about the LTFT unless it goes =/-10% over a long period of time.

Wait to see what the O2 data looks like before you put parts to it.
thanks for your response CarDoc. That is good info to know about the MAF, I can pretty much rule that out now as I recently cleaned it and have no reason to think it has a problem.

On a long rural quiet road drive I was able to monitor both bank 1 and 2 downstream O2 sensor voltage pretty well. My Ultragauge shows the voltage adequately but it isn't fast enough to always capture or show the very brief low voltage lean condition you typically should see when you open the throttle. Most of the time it just shows the higher voltage you typically see after the original lean low voltage condition.

Here are the results of my recent drive only looking at the downstream O2 sensors not the AFR sensors:

During steady state constant throttle driving - 90km/h relatively flat:
-Coolant temp stayed constant between 200-203*F (26*C ambient temps)
-Bank 1 Sensor 2: typically sits around 0.700V but I noticed that it will randomly fluctuate down to 0.2V every couple seconds or less
-Bank 2 sensor 2: typically stays at 0.7-0.8V and stays very constant except fluctuating by 0.01V ( I believe this is completely normal)
-LTFT's and STFT's on both banks seem to be fine. Randomly here and there I will notice a LTFT on either bank will be + or - 2-5% and fairly quickly goes back to zero. As you stated this is normal and isn't a concern

After this I am assuming this shows that my Bank 1 Cat is toast due to the fluctuating values? This sort of makes sense as the code was a P0420 which if I am correct, is specifically for bank 1 where as P0430 is for bank 2.

I recently had to replace a coil pack on the passenger side (bank 1) so I am not sure if this was the cause of the fouled Cat or not. I never noticed it misfiring until the night before I changed it out. That being said the ecu also did not throw a code even when it was noticeably missing at idle. It was until it starting missing during acceleration that it threw a CEL, only to then clear itself 10 seconds later once the car was moving at speed and it stopped missing. It may be possible it was misfiring for a while just not as noticeable.

If this is the case then I plan to replace the cat with a Rockauto aftermarket one in the somewhat near future when the family bank gives approval. I just don't want to have the new one become fouled again as I plan to have this car for another solid 6+ years.

I will still be using my proper scanner once I'm back from vacation but these results so far seem pretty clear based on your response and previous posts. I have another drive tomorrow on the same road, I'll access different PID's such as MAF flow, and the upstream sensors to see if I notice anything different

Thanks for your help so far
 

·
Registered
Joined
·
142 Posts
New update, @cardoc please let me know what you think of my conclusions.

I checked a/f sensor lambda and they are responding very quickly and almost identical values. MAF values are fine at idle (as expected). Secondary O2 sensors are responding quickly as well, not lazy like I would expect to see if they were bad.

During cruising there is no doubt that the bank 1 O2 sensor occasionally (every 5 secs or so) drops down to 0.1 volts while the bank 2 O2 sensor does not. This to me shows that the bank 1 cat is toast.

During cruising the Bank 2 O2 sensor reads around 0.65-0.85V, seems slightly high but it seems to not be an issue.

What is odd to me is when at idle for 5 mins I could not get the Bank 1 O2 sensor to drop down to low voltage again. If I held it at 1500 rpm it would do it briefly at first and then stop doing it.

So in conclusion it appears that my bank 1 cat is not operating properly but for some weird reason it seems that is isn't always doing this, or at least not after idling for a long time (always tested when fully up to operating temps). Bank 1 sensor always fluctuates a biut lower than bank 2 does, but it doesn't always drop to 0.1V.

It is about $400CAD for new Denso O2 and AFR sensors so I think I'll hold on on ordering sensors and just stick with a new Bank 1 cat. Changing the sensors isn't a bad idea due to the mileage but with the funds lacking I don't think the data supports a sensor replacement.
 

·
Registered
Joined
·
4 Posts
Hi Cardoc,
  1. I have a 2005 outback with the non turbo 4 cyl. I've been fighting with the P0420 code for a while now. Can't pass inspection now and I've been trying to sort it out. Recently had a code for bank 1 sensor 1 code, so I replaced that as well as the bank 1 sensor 2 sensor. I can get a anywhere from 100-200 miles fore the P0420 code triggers the MIL light. I have logging capability, through Giotto (used to be Autoengenuity, got it for my ford diesel). I'm wondering if my coolant temps are too low, seems to run in the low to mid 180's. I also see fluctuation in the S2 O2 sensors, does that mean the CAT's are bad? I changed some of the logging data in the attached images.. The upper sections was monitoring engine rpm, manifold pressure, coolant temp, and timing advance. Then I changed those to monitor fuel trim. Any suggestions on what to track or what I can do would be greatly appreciated. I've been reading this thread but, i;m only a short way into it.
 

Attachments

·
Premium Member
Joined
·
11,899 Posts
Discussion Starter #1,787
New update, @cardoc please let me know what you think of my conclusions.

I checked a/f sensor lambda and they are responding very quickly and almost identical values. MAF values are fine at idle (as expected). Secondary O2 sensors are responding quickly as well, not lazy like I would expect to see if they were bad.

During cruising there is no doubt that the bank 1 O2 sensor occasionally (every 5 secs or so) drops down to 0.1 volts while the bank 2 O2 sensor does not. This to me shows that the bank 1 cat is toast.

During cruising the Bank 2 O2 sensor reads around 0.65-0.85V, seems slightly high but it seems to not be an issue.

What is odd to me is when at idle for 5 mins I could not get the Bank 1 O2 sensor to drop down to low voltage again. If I held it at 1500 rpm it would do it briefly at first and then stop doing it.

So in conclusion it appears that my bank 1 cat is not operating properly but for some weird reason it seems that is isn't always doing this, or at least not after idling for a long time (always tested when fully up to operating temps). Bank 1 sensor always fluctuates a biut lower than bank 2 does, but it doesn't always drop to 0.1V.

It is about $400CAD for new Denso O2 and AFR sensors so I think I'll hold on on ordering sensors and just stick with a new Bank 1 cat. Changing the sensors isn't a bad idea due to the mileage but with the funds lacking I don't think the data supports a sensor replacement.
This is all good. I would suggest you spray the AF and O2 sensors, the manifold nuts and the hanger bolt at the transmission with PB Blaster in advance of working on the car in case there's a lot of rust.

I concur, with the cost of the sensors, as long as they are reacting fast, leave them be. Make sure you get the snap when you plug them back in.
 

·
Premium Member
Joined
·
11,899 Posts
Discussion Starter #1,788
Hi Cardoc,
  1. I have a 2005 outback with the non turbo 4 cyl. I've been fighting with the P0420 code for a while now. Can't pass inspection now and I've been trying to sort it out. Recently had a code for bank 1 sensor 1 code, so I replaced that as well as the bank 1 sensor 2 sensor. I can get a anywhere from 100-200 miles fore the P0420 code triggers the MIL light. I have logging capability, through Giotto (used to be Autoengenuity, got it for my ford diesel). I'm wondering if my coolant temps are too low, seems to run in the low to mid 180's. I also see fluctuation in the S2 O2 sensors, does that mean the CAT's are bad? I changed some of the logging data in the attached images.. The upper sections was monitoring engine rpm, manifold pressure, coolant temp, and timing advance. Then I changed those to monitor fuel trim. Any suggestions on what to track or what I can do would be greatly appreciated. I've been reading this thread but, i;m only a short way into it.
From this, I would look for a vacuum leak. Possible weak AF or O2 sensor on bank 2. Weak or leaking spark plugs. Cracked coil tower(s) or leaking cables. Poor grade gasoline (like Murphy gasoline sold at Wal Mart and Sam's Club). The engine may be knocking due to some reason; detonation, misfiring, could even be a bad knock sensor. Knock correction would have to be logged.

The wave form is opposing on each bank. One is down, 2 is up, then vice versa. (It would be nice to have a 2 minute log on a spreadsheet so I can see TPS, MAP, MAF, knock and the AF and O2 sensors' output combined.) Since I can't see what your foot is doing, I will assume that when the bank 1 sensor drops, your off throttle or have light throttle applied. The LTFT is up on bank 2 which means the ECM is trying to compensate based on the front AF sensor feedback. There is also the possibility that the bank 2 sensor is lazy or has a heater condition that has not been picked up on by the ECM. Usually when a sensor stays low and then comes up in voltage on acceleration it's due to a poor heater in the sensor and it doesn't heat up until exhaust flow increases. The sensor has to be HOT to operate.

A more accurate log would be nice. Can your data logger save a spreadsheet of the data that can be posted somewhere and linked here?

Your temp is fine. The engine runs 180-190 usually and this is dependent on load, ambient temps and condition of the cooling system.
 

·
Registered
Joined
·
142 Posts
This is all good. I would suggest you spray the AF and O2 sensors, the manifold nuts and the hanger bolt at the transmission with PB Blaster in advance of working on the car in case there's a lot of rust.

I concur, with the cost of the sensors, as long as they are reacting fast, leave them be. Make sure you get the snap when you plug them back in.
Thanks for your confirmation!

I have my own home brew mix of 50/50 atf to acetone which does well in our salt belt. I have had that O2 sensor off before and with the proper sockets it wasn't too bad.
The exhaust bolts on the other hand will he a completely different story...
 

·
Registered
Joined
·
4 Posts
From this, I would look for a vacuum leak. Possible weak AF or O2 sensor on bank 2. Weak or leaking spark plugs. Cracked coil tower(s) or leaking cables. Poor grade gasoline (like Murphy gasoline sold at Wal Mart and Sam's Club). The engine may be knocking due to some reason; detonation, misfiring, could even be a bad knock sensor. Knock correction would have to be logged.

The wave form is opposing on each bank. One is down, 2 is up, then vice versa. (It would be nice to have a 2 minute log on a spreadsheet so I can see TPS, MAP, MAF, knock and the AF and O2 sensors' output combined.) Since I can't see what your foot is doing, I will assume that when the bank 1 sensor drops, your off throttle or have light throttle applied. The LTFT is up on bank 2 which means the ECM is trying to compensate based on the front AF sensor feedback. There is also the possibility that the bank 2 sensor is lazy or has a heater condition that has not been picked up on by the ECM. Usually when a sensor stays low and then comes up in voltage on acceleration it's due to a poor heater in the sensor and it doesn't heat up until exhaust flow increases. The sensor has to be HOT to operate.

A more accurate log would be nice. Can your data logger save a spreadsheet of the data that can be posted somewhere and linked here?

Your temp is fine. The engine runs 180-190 usually and this is dependent on load, ambient temps and condition of the cooling system.
Thank you Cardoc!
I can save a log of data, not sure the file format though. I'll give it a try tonight. The engine was at idle during those pictures, I was not touching the accelerator. I even turned off the A/C to remove the change while it cycled on and off. I replaced the spark plugs about a week ago. I've been trying different brands of gas over the past few weeks, even got some premium with 0% ethanol. Oddly I did notice that my mileage was up almost 2 mpg the last two fill ups. I usually average about mid to high 24 mpg, last two fill ups were 26.5 and 27.2. I realize that is not terribly accurate, but since I got two in a row, I found it interesting.
Thank you,
 

·
Registered
Joined
·
79 Posts
@UKPR1

EVAP - Not necessarily. The way the monitor runs, you could have an issue and it won't show a code. It takes several days of operation for the ECM to clear the EVAP monitor because of the way it runs it's tests. My car will run "incomplete" on the EVAP for weeks, and I have yet to find an issue with the EVAP system to explain why it takes so long, or locate any issues with the solenoids or level sensors. Until a monitor is "complete" you won't see a code and sometimes the lengthy time for a monitor to set would be an indication of a problem the ECM is watching. It can be "complete" and the ECM may still not set a code for an error and you could still have a problem. This is why every code set on any car has a detailed list of checks. I've had a lot of vehicles have lean codes, misfires, P0420 and others that relate to fuel trims and it would be a stuck open purge valve. Not too many Subarus, mostly Ford and Dodge/Fiat, but they have been in the mix.

"In other words: You're not allowing the potentially backed-up outflow to create that increase in pressure, the resulting temperature and then detonation because of the holes that are now in the exhaust. The only question I have, is that if the knock sensor is basically a microphone, then wouldn't all the spitting and backfiring throw the knock sensor off in picking up a whole bunch of chaos? While on the subject - I didn't test the knock sensor - could that be suspect and warrant testing/replacement? Of the things you've listed, and knowing that the problem only exists at higher rpms and not at idle, I would think that some of the possible reasons fall out of the equation. What would your money be on right now?"

You have holes in the exhaust now and when you ran the logs?

The knock sensor is a microphone. It picks up noise in the block/heads that travels through the metals. The noise created by catless systems, exhaust leaks, vibration from suspension components or rough idle due to old timing belts are not picked up as detonation or misfires. You can run a straight open header and nothing behind it and the knock sensor function won't be affected. The spitting and backfire, if it occurs, is outside the block/heads and the "noise" is not picked up by the sensor because it won't travel through the aluminum of the block in the same way.

The knock sensor can be cracked or corroded and not throw a code. Loose plates in the sensor would allow for resistance to change with the rpm and vibration of the engine and it's the resistance that the computer sees. Exacerbation of engine vibration would be created from "false knock" allowing for continued and improper feedback from the knock sensor causing continual adjustment of fuel and timing by the ECM. Plenty of times a knock sensor tested bad and there were no codes. The ECM test the sensor at "key on" and if the resistance is good at that moment, it runs as if the sensor is good all the time and doesn't check for fault with the engine running, only making adjustments based on feedback. It can't see a fault in the resistance while the engine is running when the resistance is what it bases knock/misfires on. It will assume knock and misfire. To test the ECM's ability to acknowledge a failed sensor, hard fail, unplug it and turn the key on. A code will pop up for knock sensor circuit. And when an engine warms up the plastics expand with heat and any deterioration or crack in the plastic will allow for easier movement of the plates.

At this point, I would look at the knock sensor and pull the AF sensors.
Cardoc,

Well it was meant to be 3 weeks, but ended up being 3 months...life has a way of changing things around on you. I'll be heading back up to PA on Thursday afternoon and will be able to make the above adjustments and start logging on Friday morning. I'll make one run with just the new knock sensor and then another with the AF's removed too. I hope you'll be able to see and make sense of the results when I post them on Friday morning. Thank you in advance for your help!
Cheers,
Geoff
 

·
Registered
Joined
·
72 Posts
Hey I need some guidance on what I should be looking for .. am I looking for the wave form or the actual voltage of the O2 sensors?

2013 Subaru Outback 2.5L, PzEV I believe. p0420

Please note the sensor 2 is top and sensor 1 is bottom.
The wave form looks bad but the voltage looks good? It's mostly around .7, and if the scale wasn't so small it'd probably be close to straight line... Do I have a bad cat?

I get tons of failed codes for the O2 sensor 1 but sensor 1 graph looks good too...
469832
469833
 

·
Premium Member
Joined
·
11,899 Posts
Discussion Starter #1,794
Cardoc,
Thank you for your help. I was able to record a log of data. I uploaded it to google docs. Please let me know if you have any issues accessing it or if a different site would be better.
Thank you,
Edwin
The AF sensors are working, as well as the O2s.
The fuel trim on bank one is a little high, but then it comes down. I would check for a leak in the intake on bank 1 or an issue with the PCV valve or hose that is allowing unmetered air in. Also check the oil fill cap seal for the same reason.
The engine seems to be running good at idle. I would like to see a log with it driven. About a 2 minute cruising at 45 then highway speed.
The cats are toast.
 

·
Premium Member
Joined
·
11,899 Posts
Discussion Starter #1,795
Hey I need some guidance on what I should be looking for .. am I looking for the wave form or the actual voltage of the O2 sensors?

2013 Subaru Outback 2.5L, PzEV I believe. p0420

Please note the sensor 2 is top and sensor 1 is bottom.
The wave form looks bad but the voltage looks good? It's mostly around .7, and if the scale wasn't so small it'd probably be close to straight line... Do I have a bad cat?

I get tons of failed codes for the O2 sensor 1 but sensor 1 graph looks good too...
The "Air Fuel" is the ratio the computer is seeing from the AF sensor. It is normal for the computer to switch back and forth incrementally from rich to lean. This helps control the catalytic converter temperature. The O2 Sensor 1/1 Voltage is the voltage value the computer sees and the waveform will match the AF trim, and yours does.

The rear O2 sensor should be steady somewhere around .7V. To see if a cat is bad or not based on feedback, you would have to put a load on the cat by driving it. The engine has to be up to temperature when you log it. If there's a waveform while cruising, the cat isn't working. If there's a waveform at idle and warmed up, the cat isn't working.

Now, what is making the cat suffer? The O2 monitor is showing the AF sensor is malfunctioning. You may need a replacement AF sensor before your condemn the cat. A lazy sensor, such as what you show, will cause problems. The sensor on these cars is a wideband and they need to work fast.

Also, what is the LTFT? Is the throttle body and PCV clean? What is the condition of the plugs? How many miles on the car? Where did the gasoline come from and what grade is in it? What does the knock correction look like under load?

I suggest you check the battery and grounding. The battery needs to be tested cold for ampere output and the grounds need to be clean and tight. No more resistance than 0.1 ohm is good. If the conductance is low and available current is low then the sensors and outputs suffer from lack of electricity.

I am assuming this snap shot was with the engine idling.
 

·
Registered
Joined
·
72 Posts
The "Air Fuel" is the ratio the computer is seeing from the AF sensor. It is normal for the computer to switch back and forth incrementally from rich to lean. This helps control the catalytic converter temperature. The O2 Sensor 1/1 Voltage is the voltage value the computer sees and the waveform will match the AF trim, and yours does.
The rear O2 sensor should be steady somewhere around .7V. To see if a cat is bad or not based on feedback, you would have to put a load on the cat by driving it. The engine has to be up to temperature when you log it. If there's a waveform while cruising, the cat isn't working. If there's a waveform at idle and warmed up, the cat isn't working.
Can you help me understand what is expected behavior? I was idling, not driving, but the car had made a 20 minute drive so she was warmed up. The line was certainly moving up and down like a wave form but it was moving from .7 to .65V, is that truly a wave form or should it be 100% stable at a consistent number with no fluctuation?

Now, what is making the cat suffer? The O2 monitor is showing the AF sensor is malfunctioning. You may need a replacement AF sensor before your condemn the cat. A lazy sensor, such as what you show, will cause problems. The sensor on these cars is a wideband and they need to work fast.
Is there further testing I can do for the AF sensor?

Also, what is the LTFT? Is the throttle body and PCV clean? What is the condition of the plugs? How many miles on the car? Where did the gasoline come from and what grade is in it? What does the knock correction look like under load?
I'll pull that together. Last time Ir ead this at idle it was .8%. I believe it was 0% last night. Plugs were replaced at 80k and the car has 115k miles. Gasoline was just filled with Chevron and is almost always filled with Chevron or some type of good tier gas. No idea about knock correction; Ill pull that. I had just ran seafoam last weekend too so I cant assume the TB is clean but Im hoping the seafoam got it clean. difficult to see since it faces away from me.

PCV was replaced at around 95-100k.

I suggest you check the battery and grounding. The battery needs to be tested cold for ampere output and the grounds need to be clean and tight. No more resistance than 0.1 ohm is good. If the conductance is low and available current is low then the sensors and outputs suffer from lack of electricity.
I was reading through the tool and I saw that the battery life or health was at 6%. Im new to reading these percents so I really dont know how to interpret but to me, 6% is bad. Also, the other weekend the car sat for 4 days and it was dead so I think I have a battery drain. Alt and battery have been replaced within 2 years. I think you may be on to something here. I would never imagine a battery could go this far downstream but perhaps youre right.
[/quote]
I am assuming this snap shot was with the engine idling.
Correct. I was finished with a 20 min drive and sitting in my driveway taking these measurements.

Are there other graphs I should pull up to help diagnose?
 

·
Premium Member
Joined
·
11,899 Posts
Discussion Starter #1,798
@Jaredmac11

So the waveform that indicates a bad cat would be large. The voltage would vary from say .8 down to .1-.4V. You have to look at the data in a long time frame in order to get at the issue. The cat may be working at one point in time but fails under specific conditions and that's where the ECM pops the code up. That would also be the point you need to see what the rest of the system is doing to maintain engine performance. And logging multiple points is the key here.

I suggest Rom Raider for logging or other software that can log multiple PIDs at the same time over time and record the data.

You need to see MAF, MAP, AF sensor, O2 sensor, ECT, IAT, TPS, Knock Correction, AVLS, CPC duty, Fuel Tank Pressure and RPM at minimum. Depending on what these show would depend on whether or not we need to see other points.

Sea Foam or any other poor in cleaner will not clean the throttle body. It only goes where the gasoline travels. Fuel tank, while it sits will dissolve the deposits from the potentiometer on the level sensor, clean out the pump and lines on the module. It will clean out the lines up to the engine. It will clean out the fuel rails and injectors. It then helps to dissolve carbon on the valves, piston and chamber walls. To get more off the chamber area, you would have to feed the cleaner directly in to the intake vacuum stream while the engine is running at about 1000-1200 rpm and feed it in slow, right up to the point the engine may start to stumble from the rich condition.

You would have to physically clean the throttle body of any carbon and oil deposit. To make it easier, have a throttle body gasket available and remove the throttle body from the intake. You can either leave the coolant hoses connected and just turn the body up a little to clean it, or, you can pinch off the coolant hoses and take the throttle body off completely to make it even easier. You can use a soft bristle tooth brush or similar and the choice of cleaner is up to you. Sea Foam or B12 is good. The B12 is a bit more stringent. When you put the throttle back on, make sure there is no gasket material bits stuck on the intake or throttle before you put it back on.

Before you bother with any more data logging, the battery and grounding need to be addresses. Get the battery replaced. I recommend you NOT get one from Auto Zone. They suck. Trim back some of the insulation on the ground cable an see how far up the corrosion goes and trim the wire to get a clean strand replacing the cable clamp with a good universal. You could go the route of replacing the cable altogether. Subaru sells the battery cables as a set and you'd have the pos and neg clean and new. Easy to replace both if you go that route.

While you are in there, clean the ground lugs everywhere. Intake, body, bottom of the heads, firewall, etc. All you need to do is take them loose, brush the lugs with a small wire brush to get them clean and put them back. If you run in to any that are corroded, just make a new ground strap or trim them back to a clean spot and put on a new lug. If an offending ground wire is part of the engine harness, it's best to trim it back and make sure the replacement lug is crimped tight and use a heat shrink protector at the wire to lug joint. Do not use any grease. If you want to help with moisture intrusion, just spray the battery clamps and ground lugs with a clear paint after you tighten them down.

Another option for you on top of cleaning all the grounds is to add a heavy ground cable to improve conductivity. You can get 4 gauge cable with eyelets on both ends at the part stores relatively cheap. Best way to start is to run one from the neg battery post to the alternator/AC bracket. Just pull one of the bolts from the AC bracket and feed the bolt through the eyelet and tighten it back down. This will also improve the alternator function because you are taking the battery ground directly from the source to the alternator.
 

·
Registered
Joined
·
72 Posts
This is realy great information. I hadnt realized we are at 1800 posts but something tells me you have said this before. If not, I hope others can see it!

Thanks!
 

·
Registered
Joined
·
142 Posts
This is realy great information. I hadnt realized we are at 1800 posts but something tells me you have said this before. If not, I hope others can see it!

Thanks!
FWIW my 3.6r OB experiences something similar to what you have - the downstream bank 1 O2 sensor doesn't constantly fluctuate at idle. I am able to monitor it when at constant throttle travelling approx 90 km/h. At idle it rarely fluctuates and looks fine alongside the bank 2 sensor. But when I am at constant speed it drops from 0.8V down to 0.1 or 0.25 and back up again. This seems to happen every 5 seconds or so and the bank 2 sensor doesn't fluctuate like the bank 1 sensor is. The P0420 is for bank 1 not bank 2 so this makes sense. All other PID's show normal while this fluctuation is happening so the likely issue is the cat.

Another way to "test" the cat is to give a good amount of throttle so that O2 values read rich (0.9V) and then let off quickly so that they eventually read lean (0.1 or 0.2). A good cat will store O2 and there will be a delay from when you let off throttle and when the downstream O2 sensor shows that lean value. A bad cat will drop voltage almost instantly. My Bank 1 cat drops down to lean right away where as my bank 2 cat had a second or two delay.
 
1781 - 1800 of 1819 Posts
About this Discussion
1.8K Replies
243 Participants
tefloff
Subaru Outback Forums
Welcome to the Subaru Outback Owners Forum, we have tons of information about your Subaru Outback, from a Subaru Outback Wiki to customer reviews.
Full Forum Listing
Top