Another 0420 tale

[Steves72]

About a week ago while driving at a steady 60 mph the check engine light comes on. Not to worry I have a code reader. After finding out I have a P0420 code and no other codes, I stopped by here and began reading the myriad number of threads on determining just what is wrong with the emission system.

 

Short of throwing parts at the problem I did learn that the O2 sensors should operate in a specific range of values. And, since I do have a code reader I thought I would take this opportunity to learn a little more about what my code reader could do.

 

Now before you read any further understand this - my first attempt to make sense of the information on the code reader was not a valid test. I failed to fully warm up the car before staring at the code reader. So I will have to repeat this another day. Now my questions.

 

1) To test the O2 sensors in this fashion can I do so while the car is idling in the driveway or must it be driven while the readings are observed?

 

2) I observed two items that I believe are the O2 sensors but which one is the front and which the back? One reads O2S B1 S1 (V) and the other reads O2S B1 S2 (V).

 

What I observed while the engine was not fully up to temperature was the following: The S1 voltage varied from 1.4 to 1.6 volts. The S2 voltage varied from 0 to 1.2 volts. The second set of numbers mainly ran in the 0.3 to 0.75 range but did drop to 0 and did spike up to the previously mentioned 1.2 volts.

 

Obviously, I am new to scanners so if I made any bad assumptions I would appreciate being led down the correct path. I will repeat this test tomorrow weather permitting or Sunday.

 

Steve

2000 OBW Limited auto

[Scoobywagon]

This really going to depend on the brand and model of your reader. So as much as I hate to say this, RTFM. Having said that, it does sound like you're on the right track. No, it it is not generally necessary to be driving while observing operating data unless, of course, the complaint is related in some way to vehicle speed. This is not. Odds are, the S1 number is going to be the front. The front O2 sensor should always have slightly higher values than the rear unless your cat is shot.

 

That help?

[Steves72]

I have an Equus 3130 reader. The manual gives general instructions on how to use the scanner; such as plugging it in; scanning for codes; how to bring up the live display; how to bring up the freeze frame; etc.

 

What it does not talk about is how to interpret the information given by the scanner, and I can understand why. The number of variables involved with an OBDII computer would make such a guide massive. So I turned here for some idea on how to interpret what I am seeing.

 

I know I saw a thread somewhere - either here or on subaruoutback.org - talking about expected voltages from the O2 sensors. I have to do some more searches and see if I can find that thread again.

 

And, the weather is cooperating so I will give the scanner another go when I get home from work with the engine at temperature.

 

Steve

[Steves72]

OK. I did another test with the code reader in live data display and I think I see a pattern.

 

First, I now believe that the O2S B1 S2 (v) display is for the front O2 sensor. The O2S B1 S1 (v) is probably the rear sensor. I drove the car around the mostly empty streets in my neighborhood and noted that the S2 sensor mostly stayed around the .7 volt range under light throttle. The voltage dropped to 0 when I removed my foot from the throttle and allowed the car to coast. The voltage climbed to almost 1.0 volts once when I floored the throttle - I only did this once. The variations I saw on the S2 sensor lead me to suspect that this one is the front sensor. The other sensor - S1 - mostly ran around 1.5V varying only a little from 1.45 to 1.6 volts when I glanced down at the reader.

 

What was interesting was the S2 sensor appeared a little lazy, taking its time to drop to 0 on a few occasions. Other times dropping quickly to 0. Changing the throttle position was noticed quickly by the TPS % on the scanner (throttle position sensor?) but the S2 sensor took several seconds to react to the change in the gas pedal.

 

I believe that I have to replace the front O2 sensor.

 

Does my analysis make sense? Or, am I really off? Knowledgeable opinions sought. Thanks.

 

Steve

[WoodsWagon]

S1 is the front pre-cat sensor, S2 is the after cat sensor. B1 means "bank 1" S1 means "sensor 1". Since subarus only use 2 sensors, there is only one cylinder bank, so B1 S1 is bank one sensor one, AKA the first sensor in the exhaust stream.

 

Voltage readings of 1.45 means it's not a traditional O2 sensor, it's an A/F sensor, air/fuel ratio. The voltage on those means nothing, it's the current that it generates is what the computer works with. There should be an EQ ratio or something like that listed near by, which should be around 1.

 

P0420 codes are usually due to tired sensors, and it's a lot cheaper to put in two new sensors than replace the cat. If you need to get the code to go away, but you don't want to actually fix the car, then get a 18mm spark plug anti-fouler, with a flat face w/ washer, not tapered. Use a 1/2" drill bit to drill through the middle of the anti-fouler, remove the rear 02 sensor from the exhaust, thread the anti-fouler into the exhaust, and thread the sensor into the anti-fouler. You can have no cat, and the computer will still think it's working. That's because the sensor is in the exhaust atmosphere, but not in the stream, so it reads the right mix, but not at the sampling rate it used to.

[OB99W]

I also have an Equus 3130. It might be useful to post Freeze Frame Data for the P0420, since it will tell us what the ECU "saw" when it set the code. If you're not familiar with FFD, look at pages 26-27 of the owner's manual.

 

To better understand wide range (wide band) A/F sensors, see:

http://www.aa1car.com/library/wraf.htm

 

Look over the 3130 manual starting on page 50 for the PID (Parameter Information Data) list. You'll see that "EQ Ratio 11" is for the pre-cat sensor, for example.

 

Oxygen sensor (not wide band type) info:

http://www.aa1car.com/library/o2sensor.htm

 

P0420-related info (not all may apply to your car):

http://www.motor.com/magazine/pdfs/102006_09.pdf

http://www.aa1car.com/library/p0420_dtc.htm

http://www.catalyticconverter.org/news/news_page.cfm?Key=catalytic_converter-&News=120

http://www.import-car.com/Article/39019/Diagnostic.aspx

[Steves72]

Well, I did something that perhaps I should not have done. I cleared the error out of the computer last night. So the freeze frame data is gone. I drove the car around 150 miles today and the code did not return so far. All of my system I/M lights are also on except for the HCA indicator. I will check out the links provided to see if I can get a better understanding of my scanner. Thanks.

 

Steve

[BluetoE]

Actually, that's the first thing I would have done. Clear the code to see if it comes back

[Steves72]

Well, the 0420 code is back and the in the previous instance OB99W wanted the freeze frame data. Here it is from today.

 

Fuel Sys 1 ------------- CL

Calc Load (%) --------- 9.0

ECT (degrees F) ------- 183

STFT B1 (%) ---------- 0.7

LTFT B1 (%) ---------- -0.7

MAP (inHg) ------------ 10.9

Eng RPM -------------- 1812

Vehicle Speed (MPH) -- 45

 

That was all that was in the freeze frame data.

Steve

[OB99W]

There doesn't appear to be anything revealing in the freeze frame data -- the parameters are all within "normal" bounds. Of course, since the ECU was recently reset, the LTFT isn't all that "long term".

 

You can get a lot more data (which may or may not be helpful) by using the 3130 in the live data capture mode -- see beginning page 37 of the manual. Using "Record by DTC trigger" with the "Middle" trigger point option might provide more useful info.

[WoodsWagon]

So have you tried fixing it yet?

[Steves72]

There doesn't appear to be anything revealing in the freeze frame data -- the parameters are all within "normal" bounds. Of course, since the ECU was recently reset, the LTFT isn't all that "long term".

 

You can get a lot more data (which may or may not be helpful) by using the 3130 in the live data capture mode -- see beginning page 37 of the manual. Using "Record by DTC trigger" with the "Middle" trigger point option might provide more useful info.

 

There was eight days of driving between the reset and the repeat of the 0420 code. It represented around 600 total miles - about half of it highway.

 

If what I am reading about the "Record by DTC trigger" function, I will have to drive the car - possibly for days - with the scanner plugged in before another such event is triggered. Is that an accurate representation of the requirements for the next test?

 

Steve

[WoodsWagon]

The freeze frame data won't tell you much, as the computer is looking at average readings of the rear o2 compared to the front o2 over time. When it sets the code, it's just finished completing the catalyst monitor, which is a pass-fail set of tests. That's why it takes a long time to set the code. What freeze frame data is useful for is if you have a intermittant failure, and one sensor drops out at random points. You set it to trigger at DTC set, and see which sensors readings are way out of spec.

 

That's why it's worthless for a PO420, it's not an electrical componant that's failed. It's just comparing 02 sensor readings over time.

 

Lazy 02 sensors can skew the readings, so the computers comparison will be off and it may set a P0420 when the catalyst is still good. That's why you replace the 02 sensors first, even though it has not thrown a code for the 02 sensor having failed.

 

You can put in an anti-fouler, which mechanically skews the rear 02 sensor readings in the direction you want so that the computer will think the catalyst is working. Whether the catalyst is working well or not, a lazy o2 sensor spaced back out of the exhaust will make the computer think everything is OK.

 

So, either replace both 02 sensors, or put an anti-fouler on the rear one.

 

Don't bother playing with the code reader, it won't tell you more than you already know.

[OB99W]

The freeze frame data won't tell you much, as the computer is looking at average readings of the rear o2 compared to the front o2 over time. When it sets the code, it's just finished completing the catalyst monitor, which is a pass-fail set of tests. That's why it takes a long time to set the code. What freeze frame data is useful for is if you have a intermittant failure, and one sensor drops out at random points. You set it to trigger at DTC set, and see which sensors readings are way out of spec.

 

That's why it's worthless for a PO420, it's not an electrical componant that's failed. It's just comparing 02 sensor readings over time.

Besides using it to find obvious defects, freeze frame data can help in ruling out certain problems.

Steve reported:

Fuel Sys 1 ------------- CL

Calc Load (%) --------- 9.0

ECT (degrees F) ------- 183

STFT B1 (%) ---------- 0.7

LTFT B1 (%) ---------- -0.7

MAP (inHg) ------------ 10.9

Eng RPM -------------- 1812

Vehicle Speed (MPH) -- 45

 

From that we can determine:

1) The engine was warm (coolant 183 F) and the system was running closed loop (CL).

2) The engine wasn't being pushed very hard at the time (9% load - 1812 RPM - 45 MPH).

3) Fuel trims weren't compensating for something way out of whack (in fact, besides STFT and LTFT each being less than 1%, they averaged to 0% at the time the code was set).

 

So, the FF data did tell us something -- it told us that things it reported were within quite reasonable ranges, and we can look elsewhere.

 

 

Lazy 02 sensors can skew the readings, so the computers comparison will be off and it may set a P0420 when the catalyst is still good. That's why you replace the 02 sensors first, even though it has not thrown a code for the 02 sensor having failed.

The front sensor data is used as a basis for the ECU to constantly correct the A/F ratio, always aiming for the ideal 14.7:1 ratio. A lazy front sensor can allow the engine to run further away from that ratio, for longer times. That makes it difficult for the catalyst to work efficiently (the amount of oxygen it has to work with, among other things, is critical).

 

The rear sensor output, when things are working correctly, should be swinging much more slowly and to a lesser degree than the front. A lazy rear sensor shouldn't generate a P0420, and in fact is likely to not trigger that code even if the cat isn't doing its job well.

 

 

You can put in an anti-fouler, which mechanically skews the rear 02 sensor readings in the direction you want so that the computer will think the catalyst is working. Whether the catalyst is working well or not, a lazy o2 sensor spaced back out of the exhaust will make the computer think everything is OK.

Spacing the rear sensor further out of the exhaust stream will usually mislead the ECU into "thinking" things are okay with the cat. This is not dependent on that sensor being lazy.

 

 

So, either replace both 02 sensors, or put an anti-fouler on the rear one.

 

Don't bother playing with the code reader, it won't tell you more than you already know.

Replacing the front sensor might resolve the problem, as might fixing an existing exhaust system leak, etc. Its unlikely that replacing the rear sensor would impact P0420.

 

An OBD-II scanner is just a tool, and how valuable it is depends on whose hands it's in. If the scanner info isn't conclusive, when I need to evaluate signals that vary rapidly with time (O2 sensors, cam/crank position sensors, ABS wheel sensors, etc.), that's when I use a 'scope for better results.

 

My exhaust gas analyzer doesn't get much use since catalytic converters became common (did I just give a clue to my approximate age? ), but sometimes it comes out when P0420 shows up. It sometimes makes definitive testing for a bad cat a lot easier.

[WoodsWagon]

Repeated experience has shown replacing the rear 02 sensor to cure the P0420 code on subarus. I recommend replacing both of them at the same time because they've been in the same environment for the same time and probably have equal wear.

 

OBDii scan tools are pretty basic, and the amount of testing I could do with the HDS on the newer Hondas vs what I could do on the same car with an OBDii tool made for a pretty huge gap.

 

While it may be fun to spend lots of time obsessing over potential causes of the PO420 code, it's pretty cut and dried when it comes to Subarus. Replace the sensors, if it comes back the cat is cooked *very, very rare* and you put in an anti-fouler or replace the cat. Or, you skip doing the sensors because you just want the light to go out and want to spend a minimum, and put a fouler on the rear one and call it a day.

 

Other manufactures, it's a different story. Cats do fail on a regular basis, odd fuel trim/missfire problems destroy the cat, or pinhole exhaust leaks set off the P0420 code. Not so much on Subarus.

 

I see a lot of subaru cat's replaced because of the P0420 code, often with inferior aftermarket cats, and 10 months later the light is back on and the customer is out $600 or more. Spend the $200 on the pair of sensors, see better gas millage, and have the light stay out. That and the cat's being replaced because the flange rusts off the end of them when you could spend $20 on a split flange kit and fix it in your driveway.

 

I don't see the point in reviewing all the available data in a case that's pretty much open and shut and has been repeated on thousands of cars. Christ, between replacing sensors (the right way to fix it) and putting in anti-foulers (the wrong/cheap way to fix it) I've fixed PO420 codes on at least a dozen Subarus. I'm not a high volume subaru mechanic either, it's all been friends and families cars. My volume was all with newer Honda's, which are a different bag of worms. They rarely threw a P0420, and when they did the cat was most definiately finished. Mostly people that drove with a missfire for long distances.

 

As for the freeze frame confirming nothing else was wrong with the car, yes, of course the rest was fine, otherwise it wouldn't have been running the Catalyst efficiency monitor. It has to have a set of parameters met before it will run the self test, and if the fuel trim is way off, you will see a fuel trim DTC as well as the P0420 DTC. It won't run the monitor if the engine isn't up to temp, it's in open loop, or you're not a cruising speed. So none of the freeze frame info is usefull, unless you're confirming a DTC related to one of the sensors recorded in freeze frame.

 

I'm sorry to get snippy, but I don't see the point in wasting the posters time having him play with his scan tool so that you can look elsewhere when he could be fixing the car and moving on with life.

 

Oh, and one extra thought popped into my head. The subaru board and collective internet knowlege works as a TSB for a lot of common problems. Spending lots of time (and diagnostic money potentially) to definitively rule out a catalyst failure, when a "TSB" of sorts is available is a disservice to the customer IMHO.

Edited April 21, 2010 by WoodsWagon

[OB99W]

[...] I'm sorry to get snippy, but I don't see the point in wasting the posters time having him play with his scan tool so that you can look elsewhere when he could be fixing the car and moving on with life.

 

Oh, and one extra thought popped into my head. The subaru board and collective internet knowlege works as a TSB for a lot of common problems. Spending lots of time (and diagnostic money potentially) to definitively rule out a catalyst failure, when a "TSB" of sorts is available is a disservice to the customer IMHO.

You can be "snippy" all you want, but I reserve the right to similarly reply if I feel like it.

 

Let me address your last thoughts first -- this forum doesn't have "customers", it has "members". Steve seemed interested in using his scan tool, and learning about it's capabilities, which is why I went down that path. If he (or anyone else) feels that suggestions made here by me (or anyone else) are time-wasting, those suggestions can simply be ignored. As to "diagnostic money", that seems completely irrelevant -- I certainly didn't suggest that Steve (or anyone else) have someone who's going to charge for it do what we've been discussing. Besides, based on his previous posts, Steve seems to mostly do his own work.

 

As to the rear O2 sensor relationship to P0420, your argument is inconsistent. If using anti-foulers to pull the rear sensor back out of the exhaust stream eliminates the code, then the sensor wasn't defective to the point of being able to cause the code (since it's reacting to the change). Yes, rear sensors can fail, but your reasoning doesn't support your point.

 

You're also being overly dismissive of the value of FF data in this case. Just because the catalyst efficiency monitor runs doesn't mean that the fuel trims are near ideal. They can be compensating significantly (which could be a clue) and the monitor will still run.

 

The cause of Steve's P0420 seems to be a marginal situation, based on how many miles he's had to drive to get the code to retrigger after the reset. So, it's certainly possible that a lazy front sensor is the culprit, and changing it (I'd suggest OEM only) would be a quick way to find out if that's the case.

 

Just to throw one in from left field -- I've occasionally seen P0420 codes triggered by PCV problems. That can upset the A/F (sometimes momentarily, if the valve is sticking), or cause more oil to burn and contaminate both the O2 sensors and the cat.

[Steves72]

The back and forth on the 0420 code is fascinating, to quote an old SCI-FI show, and show my age. I read every link provided by OB99W a few days back. A was familiar with some of the information in those articles but they were good reading and filled in many gaps. All in all very informative articles and well worth the 1+ hour I spent reading them. Thanks.

 

Returning to the car, I will provide some additional information regarding my Subaru so you know a bit more about the situation. The car just passed our state's annual inspection process at the end of February. Annual inspections are required in Pennsylvania. A spacer in the O2 socket will automatically fail the car at the next inspection, so such a repair is not an option. Furthermore, as OB99W points out, it just fools the computer and solves nothing. The car required a gasket on the muffler flange - it was leaking but that is after the rear O2 sensor so I do not believe that it has any bearing on the current 0420 code. Prior to taking it to my mechanic for the annual inspection I had the car on a lift to check the brakes, tires, exhaust, etc. I replaced a failed headlight, changed the oil, and rotated the tires. When I checked the exhaust I did a visual only inspection as it was quite hot so I did not find the leaking flange gasket. The system appeared to be in excellent condition for its age - 10 years. My mechanic found no other issues with the car.

 

So, now what? Like I said when I started this thread I want to understand what is wrong and why. And, I feel that I am heading in that direction. The consensus seems to be to throw an O2 sensor (the front) at it and see if that fixes the problem. I do not have an issue with repairing the car in this fashion but I do have around 10 months before this MUST be fixed and I am learning quite a bit about how the car's computer works and about how to get the most out of my code reader.

 

So, I will proceed on the slow and steady course and drive the car to and from work tomorrow with the scanner attached and hope that another event is triggered and more information becomes available. This coming weekend, when I can put my car back on a lift, I will do a through inspection of the exhaust. Who knows - perhaps I, or my wife, hit some road debris and something was damaged. Or, another gasket has failed before the cat.

 

If I run out of all other alternatives, I will replace both O2 sensors first; see if the problem is resolved; and, if not, replace the cat. I can replace all those parts myself for about the same amount that my mechanic would charge me to a) diagnose the problem, and replace a single O2 sensor. Actually, it would cost me a bit more for all new parts but than I would have all new parts!

 

If I get a triggered event I will be back and post it.

 

Steve

[Steves72]

I was typing up my post and did not see this prior to my previous post. Just to let you know, I do appreciate the insights. I mostly work on older carburetor cars - my toy and my friends old stuff. The newer computer stuff left me scratching my head and the vast number of wires, vacuum lines, etc. under the modern hood makes me leery to do much more than change my oil, filters, and plugs. This issue has pushed me to to better understand the newer systems. Besides the undercarriage is far less complicated than the engine bay.

 

Steve

 

 

You can be "snippy" all you want, but I reserve the right to similarly reply if I feel like it.

 

Let me address your last thoughts first -- this forum doesn't have "customers", it has "members". Steve seemed interested in using his scan tool, and learning about it's capabilities, which is why I went down that path. If he (or anyone else) feels that suggestions made here by me (or anyone else) are time-wasting, those suggestions can simply be ignored. As to "diagnostic money", that seems completely irrelevant -- I certainly didn't suggest that Steve (or anyone else) have someone who's going to charge for it do what we've been discussing. Besides, based on his previous posts, Steve seems to mostly do his own work.

 

As to the rear O2 sensor relationship to P0420, your argument is inconsistent. If using anti-foulers to pull the rear sensor back out of the exhaust stream eliminates the code, then the sensor wasn't defective to the point of being able to cause the code (since it's reacting to the change). Yes, rear sensors can fail, but your reasoning doesn't support your point.

 

You're also being overly dismissive of the value of FF data in this case. Just because the catalyst efficiency monitor runs doesn't mean that the fuel trims are near ideal. They can be compensating significantly (which could be a clue) and the monitor will still run.

 

The cause of Steve's P0420 seems to be a marginal situation, based on how many miles he's had to drive to get the code to retrigger after the reset. So, it's certainly possible that a lazy front sensor is the culprit, and changing it (I'd suggest OEM only) would be a quick way to find out if that's the case.

 

Just to throw one in from left field -- I've occasionally seen P0420 codes triggered by PCV problems. That can upset the A/F (sometimes momentarily, if the valve is sticking), or cause more oil to burn and contaminate both the O2 sensors and the cat.

[Steves72]

Well, three one way trips to / from work and no triggered events. Thinking back to the two times that the check engine light came on, both times were fairly long stretches of driving at a constant speed, and constant gas pedal position. The first at 60 mph and the second at 45 mph. I am wondering if I have to recreate the circumstances that caused the the trouble code to be stored in the computer. My driving back and forth to work in traffic may not cause this event to occur.

 

I will run the scanner in the car on Saturday and see if anything happens.

 

Steve

[eppoh]

This is an excellent thread and thank you to all the pro's and others participating. I have a 97 Lego that has the 420 code for several months now. No smog inspections required here, so I can still drive it but I do want to fix it.

 

It is a southern car, and the exhaust system is clean and tight. I have cleared the code twice, only to have it return after a couple hundred miles. After reading this thread, I now know why it takes that long. So, I am waiting to hear how Steves72 gets his resolved.

 

I am considering replacing the front O2 first- but can't do anything now or for a while- recovering from back surgery.

 

So keep the thread alive and thanks.

[OB99W]

Well, three one way trips to / from work and no triggered events. Thinking back to the two times that the check engine light came on, both times were fairly long stretches of driving at a constant speed, and constant gas pedal position. The first at 60 mph and the second at 45 mph. I am wondering if I have to recreate the circumstances that caused the the trouble code to be stored in the computer. My driving back and forth to work in traffic may not cause this event to occur.[...]

For the I/M monitors to be ready, a certain set of conditions have to be met, referred to as "drive cycles". Part of that is a need to run for a period of time at highway speeds with very minimal or no change in TPS angle ("constant gas pedal position", as you put it). As you surmised, your trips to work likely don't fill drive cycle requirements.

[kami333]

Annual inspections are required in Pennsylvania. A spacer in the O2 socket will automatically fail the car at the next inspection, so such a repair is not an option.

 

I had my old Subaru pass inspection in PA with a spacer, once it gets dirty you can't tell.

 

Both of my current Subarus have P0420s fixed with spacers.

[live2fish]

What would be the "driving performance" effect of replacing the O2 sensors vs. using the spacer method? Is any long term damage averted by replacing the O2 sensors that would occur if an anti-fouler spacer is installed instead?

[Steves72]

For the I/M monitors to be ready, a certain set of conditions have to be met, referred to as "drive cycles". Part of that is a need to run for a period of time at highway speeds with very minimal or no change in TPS angle ("constant gas pedal position", as you put it). As you surmised, your trips to work likely don't fill drive cycle requirements.

 

Well it took this long to have the 0420 code trip once more, and I did not have the code reader hooked up when it happened!?!

 

However, the freeze frame stored data is as follows:

 

Fuel Sys: CL

Calc Load: 7.8

ECT (degrees F): 181

STFT B1%: -7.8

LTFT B1%: 1.5

MAP (inHg): 7.0

Eng RPM: 2600

Veh Speed: 67

 

I know the last time you were most interested in the STFT versus LTFT numbers. These are considerably different than the last time. It took about five weeks for the code to trip this time.

 

Any thoughts?

 

Steve

[dasaan]

This is an excellent thread excepting that there is no finish. Steve, have you fixed this problem, and what did you do?