02 legacy misfire and 02 codes

[Brent P]

I'm just not ready to start throwing parts at the car, but too stubborn
to pay a mechanic to fix it. Dammit, if I can swap cams in my race car
I should be able to diagnose this... but I feel like a blind man
describing an elephant. Maybe I shoulda bought a G6 or Malibu Maxx...
they're fwd and the 3.5 is 20 years obsolete, but at least I know how to
fix it.

To risk the hatred of some.... I'll post a conceptual look at it.

To look at it a different way, what if the misfires are occuring because
the fuel mixture actually is lean? That is the O2 sensor is operating
correctly and getting a result that the computer isn't expecting?

What if the computer is thinking that it is getting a certain amount of
air in but really is getting more?

I've read that it behaves well when driving, but not at idle. That the
idle bounces around.

How does this vehicle control idle air? It may be worth a look into how
that system operates and where it may need to be tested/cleaned/inspected
to make sure it's ok....

 

[Ray]

To risk the hatred of some.... I'll post a conceptual look at it.

To look at it a different way, what if the misfires are occuring because
the fuel mixture actually is lean? That is the O2 sensor is operating
correctly and getting a result that the computer isn't expecting?

What if the computer is thinking that it is getting a certain amount of
air in but really is getting more?

I've read that it behaves well when driving, but not at idle. That the
idle bounces around.

How does this vehicle control idle air? It may be worth a look into how
that system operates and where it may need to be tested/cleaned/inspected
to make sure it's ok....

Which is why I'm not yet throwing parts at it because I'm not 100% sure.
I'm going off a combo of gut feel, experience, and my Firebird shop
manual for troubleshooting P0130 and related codes.
One of the things in there for checking lean conditions is to check for
intake leaks and/or exhaust leaks upstream of the 02 sensor(s).
However, even then, you'd think that the value for the 02 sensor should
be changing as the computer is trying to richen it up, which is NOT
happening, which causes the second code for "too lean" (P 0171 IIRC)
And because it's not "major serious" like no oil pressure or a busted
rad hose, I'm not in any huge rush to throw parts at it on the
possibility it might fix it. They already changed the plugs and wires
on it...

and does anyone know how to clean an 02 sensor? Carb cleaner?
Brakekleen? Sandpaper?

Ray
(and I ordered a shop manual CD off ebay, so I should be in a much
better position to diagnose and repair this thing shortly.)

 

[Lawrence Glickman]

Which is why I'm not yet throwing parts at it because I'm not 100% sure.
I'm going off a combo of gut feel, experience, and my Firebird shop
manual for troubleshooting P0130 and related codes.
One of the things in there for checking lean conditions is to check for
intake leaks and/or exhaust leaks upstream of the 02 sensor(s).
However, even then, you'd think that the value for the 02 sensor should
be changing as the computer is trying to richen it up, which is NOT
happening, which causes the second code for "too lean" (P 0171 IIRC)
And because it's not "major serious" like no oil pressure or a busted
rad hose, I'm not in any huge rush to throw parts at it on the
possibility it might fix it. They already changed the plugs and wires
on it...

and does anyone know how to clean an 02 sensor? Carb cleaner?
Brakekleen? Sandpaper?

===================================================================
the sensor may be carbon fouled. It is sometimes possible to clean a
sensor in the car. Do this by unplugging the sensor harness, warming
up the engine, and creating a lean condition at about 2000 rpm for 1
or 2 minutes. Create a big enough vacuum leak so that the engine
begins to slow down. The extra heat will clean it off if possible. If
not, it was dead anyway, no loss. In either case, fix the cause of the
rich mixture and retest. If you don't, the new sensor will fail.

http://www.gnttype.org/techarea/ecmsensors/O2sensors.html

Lg

 

[Lawrence Glickman]

===================================================================
the sensor may be carbon fouled. It is sometimes possible to clean a
sensor in the car. Do this by unplugging the sensor harness, warming
up the engine, and creating a lean condition at about 2000 rpm for 1
or 2 minutes. Create a big enough vacuum leak so that the engine
begins to slow down. The extra heat will clean it off if possible. If
not, it was dead anyway, no loss. In either case, fix the cause of the
rich mixture and retest. If you don't, the new sensor will fail.

http://www.gnttype.org/techarea/ecmsensors/O2sensors.html

Lg

READ THIS FIRST:
http://www.kemparts.com/TechTalk/tt07.asp

Lg

 

[Brent P]

One of the things in there for checking lean conditions is to check for
intake leaks and/or exhaust leaks upstream of the 02 sensor(s).

Of course...

However, even then, you'd think that the value for the 02 sensor should
be changing as the computer is trying to richen it up, which is NOT
happening, which causes the second code for "too lean" (P 0171 IIRC)

Not changing is odd... Would expect to see some kind of response.

Is the post cataylst O2 sensor the same? If so you could swap them. If
the malfuction moves to sensor 2 or starts giving a code for a cataylst
malfunction then you'll know it's the sensor.

 

[Ray]

READ THIS FIRST:
http://www.kemparts.com/TechTalk/tt07.asp

Lg

thanks.


And next question... what's the difference between a pre-cat and
post-cat 02 sensor? Less worried about construction, more a question of
what "normal" output would look like for both.

Should they both hover around the 450mV mark under idle/normal load?
Will the pre-cat one move around more as the throttle is opened/closed
while the post cat one should remain fluctuating back and forth? From
what I've read it's not always possible to read this without a scope as
it switches back and forth too fast for a meter or even my autotap to
display correctly.

Ray

 

[Lawrence Glickman]

thanks.


And next question... what's the difference between a pre-cat and
post-cat 02 sensor?

The toggle rate of the switching between low voltage 0 point something
and a high of nearly 1 volt.

Less worried about construction, more a question of
what "normal" output would look like for both.

My scanner shows the voltage values. I don't know what a scope image
would look like other than a flat bar indicating a low voltage to high
voltage and then back to low...Direct Current

Should they both hover around the 450mV mark under idle/normal load?

That's the *toggle threshold* where anything below is considered low,
anything above is considered high.

Will the pre-cat one move around more as the throttle is opened/closed
while the post cat one should remain fluctuating back and forth?

IIRC, pre-cat toggles at about 2 cycles/second
post cat toggles at 127th that rate, as a general example.

From
what I've read it's not always possible to read this without a scope as
it switches back and forth too fast for a meter or even my autotap to
display correctly.

read it without a scope? My Actron 9145 catches it every time. Even
my Davis Drive Rite chip catches it, with a sampling rate of once
every 5 seconds.

2 hertz, 2 cycles/second, should show up fine on any analog voltmeter,
even a $3 volt meter from Harbor Freight, but you have to backprobe
the *output* wire to the ECM, and put the ground wire from the volt
meter on the engine block which is B-.

Me, I use the scanner and the Chip at the OBDII port under my
dashboard. It is easier for me, believe it.

Lg

 

[mylesv]

Assuming the car has typical mileage - yeah, front O2 is a candidate for
being lazy. I wouldn't rule out something as simple as a vacuum leak
(since work has been done, a hose may be missing or cracked) all the way
up to a MAF sensor ($$$ unless carefully cleaning the element doesn't
fix it).

Does it idle/start differently cold than hot? Engine Temp Sensors (not
gauge sender) can confuse the ECU if bad.

hopefully someone will chime in with more experience.

Carl

I don't think this engine has a "hot wire" type sensor. It has a MAP.

 

[AS]

Ray: Provide me with your email address and I will forward you the
diagnosis procedures from the manual for my 03 outback, they should be
pretty much the same. (e-mail address removed). Please put SUBARU in the subject
line so I do not toss the email as spam.

Good luck

AS

 

[AS]

Brent:

They are 2 different kind of sensors, they cannot be swapped and doing
so could lead to ECM damage.

The b1s1 sensor is a wide range sensor, it can continuosly change the
current output to the ECM. It is called and air/fuel A/F sensor.

The b1s2 sensor is a narrow range sensor and it is almost on or off. It
is the commonly named O2 sensor.

This link may be helpful: http://www.autoshop101.com/
Look for Technical Article No. 37

 

[Brent P]

They are 2 different kind of sensors, they cannot be swapped and doing
so could lead to ECM damage.

Had a bad feeling suburu might do it that way. And of course swaping
without checking first would be foolish.

The b1s1 sensor is a wide range sensor,

The b1s2 sensor is a narrow range sensor and it is almost on or off.

That is what I thought it might be in this case, but it's worth getting
the question answered. The swap option in diagnosis is closed off
unfortantly.

That said, from a manufacturing/design perspective the goal would be to
make those sensors the same part even if if one would not be used to
it's potentional. There would need to be good reason not to, such as the
front sensor needed to be considerably better because of how the ECM
works and there was a large cost difference. The wide band sensor is
proabably still too costly to be used in both locations.

Some vehicles only use a single sensor type for both locations because
it's all that's needed and doing so saves on costs from volume,
inventory, etc.

 

[aarcuda69062]

However, even then, you'd think that the value for the 02 sensor should
be changing as the computer is trying to richen it up, which is NOT
happening, which causes the second code for "too lean" (P 0171 IIRC)

At (IIRC) 23% fuel correction, it may be at its limits for adding
fuel. Fuel trims don't usually go much over 20%.

You can try adding propane from an unlit propane torch and then
watch the O2 sensor for response.

 

[Carl 1 Lucky Texan]

Had a bad feeling suburu might do it that way. And of course swaping
without checking first would be foolish.




That is what I thought it might be in this case, but it's worth getting
the question answered. The swap option in diagnosis is closed off
unfortantly.

That said, from a manufacturing/design perspective the goal would be to
make those sensors the same part even if if one would not be used to
it's potentional. There would need to be good reason not to, such as the
front sensor needed to be considerably better because of how the ECM
works and there was a large cost difference. The wide band sensor is
proabably still too costly to be used in both locations.

Some vehicles only use a single sensor type for both locations because
it's all that's needed and doing so saves on costs from volume,
inventory, etc.

The front sensor is exposed to a different, hotter mix of gasses. The
rear one is to monitor cat efficiency. I can see their functions being
different enough to warrant specific designs.

I dunno

Carl

 

[Brent P]

The front sensor is exposed to a different, hotter mix of gasses. The
rear one is to monitor cat efficiency. I can see their functions being
different enough to warrant specific designs.

The older 'narrow band' type certainly can survive and do the job to control
mixture well enough. Wideband O2 sensors have only been around since the
mid 90s or so.

Looking into this further it appears that wideband sensors have been put
into use for meeting the various LEV emissions standards.

Poking around some, it's the earlier OB2 cars where having the same
sensor type before and after catalyst appears more likely.

 

[Comboverfish]

At (IIRC) 23% fuel correction, it may be at its limits for adding
fuel. Fuel trims don't usually go much over 20%.

You can try adding propane from an unlit propane torch and then
watch the O2 sensor for response.

Neil,
This Subaru has an air/fuel ratio sensor for B1S1. The research I've
done has spotty results so I was reluctant to post it, but it appears
like the max range you would expect to see out of this sensor is from
..025 volts to about 1.1 volts. It should stay near the center and not
move much on a properly running car. I don't know if it responds in
the reverse manner like 3v/3.3v comparator A/F sensors (scantool A/F
parameter --> 5 volts is lean, 0 volts is rich), but I would suspect so
as it is an ion pump style sensor. The original post mentions -23%LtFT
*and* P0171 too lean. That doesn't jibe, so perhaps the minus mark was
a typo.

Toyota MDT in MO

 

[ray]

Neil,
This Subaru has an air/fuel ratio sensor for B1S1. The research I've
done has spotty results so I was reluctant to post it, but it appears
like the max range you would expect to see out of this sensor is from
.025 volts to about 1.1 volts. It should stay near the center and not
move much on a properly running car. I don't know if it responds in
the reverse manner like 3v/3.3v comparator A/F sensors (scantool A/F
parameter --> 5 volts is lean, 0 volts is rich), but I would suspect so
as it is an ion pump style sensor. The original post mentions -23%LtFT
*and* P0171 too lean. That doesn't jibe, so perhaps the minus mark was
a typo.

Toyota MDT in MO

It was a bad B1S1. And over the course of the weekend, it threw both a
P0130/P0171 and a P0130/P0172 combo with a P0301/P0302/P0303/P0304 in
between. When I first checked, LTFT was at "max lean" and then later at
"max rich." Basically, the pre-cat 02 appears to have been "stuck" at
one end of the range or the other, thus causing the computer to lean it
waaay out, or richen it waaay up, and freak when it hit the limit.
Anyway, they changed the 02 and it runs way better now.

So, yes, the -23% was a typo... that happened later when it threw the P0172.
Ray

 

[aarcuda69062]

Neil,
This Subaru has an air/fuel ratio sensor for B1S1. The research I've
done has spotty results so I was reluctant to post it, but it appears
like the max range you would expect to see out of this sensor is from
.025 volts to about 1.1 volts. It should stay near the center and not
move much on a properly running car.

Been a few years since the wideband class, IIRC, the current
varies with mixture, but it's a bitch to measure in the
conventional manner.
Scan tool re-interprets it into voltage as the PID.

I personally haven't seen any failures on that type of system,
although I'm sure they're out there.

My thought was to force a known condition and look to see if
there was an appropriate reaction...

I don't know if it responds in
the reverse manner like 3v/3.3v comparator A/F sensors (scantool A/F
parameter --> 5 volts is lean, 0 volts is rich), but I would suspect so
as it is an ion pump style sensor. The original post mentions -23%LtFT
*and* P0171 too lean. That doesn't jibe, so perhaps the minus mark was
a typo.

Roy posted that it was indeed a +23% trim value, not a -23%,
that's what made me think that the propane trick might validate
the failure before he committed the bucks for the sensor.

The whole thing reminds me that I need to get one of these
crittters in and do some hands on...