Welcome to the DoItYourself Forums!

To post questions, help other DIYers and reduce advertising (like the one on your left), join our DIY community. It's free!

oxygen sensor test


Chris Oakley's Avatar
Visiting Guest

Posts: n/a

06-17-03, 05:12 PM   #1  
Chris Oakley
oxygen sensor test

1997 Mercury Grand Marquis,v8,auto,4door...check engine light came on.Used OBDll code reader and code po141 came up.It says bank 1 sensor 2.Which side of the vehicle is this and which sensor(front or rear)?I have Haynes repair manual 36012 for this vehicle.On page6-14 it doesn't explain in much detail how to test the sensor or the wiring leading to it.I tried these tests but seem to be doing something wrong.Can you tell me in more detail how to do these tests?I just got rid of my 1988 Caprice and there is alot of difference with newer computer.Thanks Chris

 
Sponsored Links
bejay's Avatar
Member

Join Date: Dec 2001
Posts: 2,538

06-17-03, 07:03 PM   #2  
bank 1 is always the side of the engine with #1 cylinder which on this car would be the passenger side bank, sensor 2 means that it is the second sensor on that exhaust bank, sensor one would be the closeest one to the engine and sensor 2 is usually after the convertor.

 
billys68ss's Avatar
Member

Join Date: May 2003
Posts: 1,459
WV

06-18-03, 09:40 PM   #3  
My wifes 96 Nissan Sentra has 250,000+ miles and still has the original O2 sensors. The check engine light has NEVER came on on this car. I checked it a few months ago and everything was working correctly except for a suspected TPS. hmmmmmm

I have seen O2 sensors listed as maintenance items. For instance we did a BMW a few weeks ago. It has 4 sensors and they are about $250 each. The manual states to change them EVERY 30k. The car has a little over 100k and this was the first set. I would say the customer got his moneys worth out of em. The sad part is only one of the 4 was failing, but it was a CYA kinda situation.
Billy

 
videobruce's Avatar
Member

Join Date: Sep 2002
Posts: 204
NY

03-31-04, 05:41 AM   #4  
Can you use a 'scope' instead? And how?

 
carguyinva's Avatar
Member

Join Date: Nov 2003
Posts: 1,147
VA

03-31-04, 06:10 AM   #5  
another opinion...

unless you have a really good scanner AND the computer in the vehicle monitors rich/lean switch times...you need a labscope to properly verify correct operation.

heated o2 sensors are usually alot more than non-heated (see billys post). as far as those high mileage vehicles that run perfectly...sure...they run well...but tailpipe emissions and peak performance ain't what it could be because your o2's are old.

when testing an o2 with a labscope you should be most concerned with amplitude (how high and how low does it go) and expect a good one to read above 800mV and less than 200mV AND switch times of less than 100ms. (50ms is REALLY nice)

the better the o2 works according to those two parameters, the better the catalyst works

some OBD II systems monitor switch times...some don't

the o2 behind the cat is there to help determine catalyst efficiency

when i had o2 heater codes in my Tahoe, it had chafed wires causing a short to ground and blowing the fuse...so if you check your fuses and find a blown one, that's the symptom...not the cause of your problem.

 
carguyinva's Avatar
Member

Join Date: Nov 2003
Posts: 1,147
VA

03-31-04, 06:15 AM   #6  
oops...

i should've mentioned about using a labscope for this...

a DSO (digital storage oscilloscope) works best and you set it for .2 volts/division and 200ms/division

backprobe the sensor return wire (you'll need a schematic to ID which of the 3 or 4 wires is the return) run the engine at operating temp and you should get a fluctuating pattern

 
videobruce's Avatar
Member

Join Date: Sep 2002
Posts: 204
NY

03-31-04, 06:35 AM   #7  
My OBD I reader ('89 Dodge Spirit) shows the voltage off the computer and the state of the sensor. Both do change, but it seems the engine is burning rich in the winter time (more so than before). A repair shop said to change the sensor after a diagnostic was done.

The manual says the heated sensor isn't used when the engine is warming up. Only after it is does the computer take reading from the sensor. The sensor from the reading of the OBD I reader usually shows 'rich'. It will switch to 'center' ands sometimes to 'lean', but very rarely. The voltage reading from the sensor does change. Does any of this indicate a 'bad' sensor as that diagnostic test showed?

 
davo's Avatar
Member

Join Date: Apr 2002
Posts: 2,417

03-31-04, 02:40 PM   #8  
Originally posted by videobruce
Can you use a 'scope' instead? And how?
Is there a reason you went back that far to dig up an old thread?I mean instead of a new post?

 
mike from nj's Avatar
Visiting Guest

Posts: n/a

03-31-04, 10:43 PM   #9  
mike from nj
about 90% of the failed sensors i've seen were stuck 'lean', meaning they weren't putting any voltage out, .00-.04 volts max.

the fact that the sensor is putting out voltage, above .6 volts usually means it's ok.

unlike any other sensor on the vehicle, the oxygen sensor produces it's own voltage, it's is a mini voltage generator. so, if it puts out nothing----it's bad. if it's doing something (making voltage), you can assume it's 'might' be good.


try pulling off the vac hose to the brake booster and see if the sensor goes to low voltage (lean), if it does, you have a rich running engine, and it's not that sensor. oxygen sensors don't like to live in that condition too long (very rich), and will very soon fail if it isn't already.


i have had a few sensors 'fail' at center, .5 volts, the car was nearly undrivable. a new sensor fixed that one.

with the price of gas today, a new sensor will pay for itself rather quickly.



to answer your other question, the computer will not start using the oxygen sensor's values until it sees it switch a certain amount of times, from .2ish to .8ish. that and a few other things(like coolant temperature) have to happen for the car to enter 'closed loop'.

let us know what you find

 
videobruce's Avatar
Member

Join Date: Sep 2002
Posts: 204
NY

04-01-04, 06:44 AM   #10  
Originally posted by davo
Is there a reason you went back that far to dig up an old thread?I mean instead of a new post?
I did a search and this one came up.
Why start another thread on the same subject?
You have to be the FIRST moderator that complained about someone NOT starting a new thread!

 
videobruce's Avatar
Member

Join Date: Sep 2002
Posts: 204
NY

04-01-04, 06:47 AM   #11  
try pulling off the vac hose to the brake booster and see if the sensor goes to low voltage (lean), if it does, you have a rich running engine, and it's not that sensor. oxygen sensors don't like to live in that condition too long (very rich), and will very soon fail if it isn't already.
Where is it exactly on this '89 Dodge Spirt (2.5 4 cyl.)?
What does that have to do with the O2 sensor, BTW?

 
Desi501's Avatar
Member

Join Date: Oct 2003
Posts: 2,207

04-01-04, 12:37 PM   #12  
Originally posted by videobruce
I did a search and this one came up.
Why start another thread on the same subject?
You have to be the FIRST moderator that complained about someone NOT starting a new thread!
This sounds familiar. How many people have to tell you?
Old discussions don't help anybody unless the questions are being repeated.

 
davo's Avatar
Member

Join Date: Apr 2002
Posts: 2,417

04-01-04, 03:31 PM   #13  
This thread has been edited by Davo.

 
videobruce's Avatar
Member

Join Date: Sep 2002
Posts: 204
NY

04-01-04, 07:56 PM   #14  
You have to be different!

Lets see, 40+ forums and all don't want new threads started when another thread covers the issue/problem!

But here, lets open another thread about the same or similar problem!

Yep, makes sense to me!

 
mike from nj's Avatar
Visiting Guest

Posts: n/a

04-01-04, 10:10 PM   #15  
mike from nj
your 89 dodge spirit 4 cylinder with one pre-cat sensor is very different from an OBDII ford V8 with 4 sensors, and a code for the post-cat sensor.


quote:
--------------------------------------------------------------------------------
try pulling off the vac hose to the brake booster and see if the sensor goes to low voltage (lean), if it does, you have a rich running engine, and it's not that sensor.
--------------------------------------------------------------------------------

Where is it exactly on this '89 Dodge Spirt (2.5 4 cyl.)?
What does that have to do with the O2 sensor, BTW?





that's how you test it, with the scan tool hooked up, you introduce a vacuum leak and see if the sensor responds or how long it takes to compensate.

are you asking about the brake booster or the oxygen sensor?

 
davo's Avatar
Member

Join Date: Apr 2002
Posts: 2,417

04-02-04, 04:46 AM   #16  
Originally posted by videobruce
You have to be different!

Lets see, 40+ forums and all don't want new threads started when another thread covers the issue/problem!

But here, lets open another thread about the same or similar problem!

Yep, makes sense to me!
If you feel the need to discuss this further we can in the proper place,PM or email not in the forum.The forum is for automotive questions and answers not how the forum is run.Any non automotive posts will be removed.

 
alung's Avatar
Member

Join Date: Feb 2003
Posts: 164

04-07-04, 03:57 PM   #17  
Re: another opinion...

Originally posted by carguyinva

when testing an o2 with a labscope you should be most concerned with amplitude (how high and how low does it go) and expect a good one to read above 800mV and less than 200mV AND switch times of less than 100ms. (50ms is REALLY nice)

This is something I don't quite understand... is the O2 sensor *supposed* to toggle every 50~100mS? Wouldn't it stay constant if the exhaust is constantly lean or rich?

or is it because newer cars make more frequent fuel/air adjustments than say my 84 Toyota Celica?

If I scope my celica, it toggles between 200-500mV approx every 1.5sec, (not milliseconds!) I've got about 80K miles on the sensor (187K on the car), so it appears this is on its way out in terms of output amplitude, OR, my engine is really running lean. My mileage is still 26-27mpg, so I don't think I have any problems.

I'll try attaching a scope shot to this reply, but the last time I tried, it didn't show up for whatever reason. Top trace is at idle, lower trace is at about 2000rpm

Attached Images
     
 
Desi501's Avatar
Member

Join Date: Oct 2003
Posts: 2,207

04-07-04, 04:46 PM   #18  
The O/2 sensor is supposed to go up and down constantly. It's called cross counting. Most scanners don't report data quick enough to see how many times it's actually crossing although some data streams had a PID for crosscounts. Whenever the sensor reports below .450 the PCM commands to go rich. Whenever the sensor reports above .450 the PCM commands to go lean and this ritual goes on forever. The faster the response and the wider the range of the sensor, the better condition it's in. The amount of time it takes from a cold start to get hot enough to start signaling is also an indication of it's condition. A quick short squirt of carb cleaner into the intake should drive the voltage immediatly to .800-.900 for a few seconds until the PCM responds to lean it out.

 
carguyinva's Avatar
Member

Join Date: Nov 2003
Posts: 1,147
VA

04-08-04, 04:56 AM   #19  
now...to muddy the water...

... a bit!

some manufacturers use altered control strategies for the many operating modes of their own FI system. some asian makers didn't use o2 input at idle and so the vehicle ran in open loop...so you might not see the toggling back and forth of the o2 output at idle. try running it at 2000 rpm...or tapping into it and driving it to see what it does. the other thing to note here is the difference between heated and non-heated sensors. one wire sensors are not heated and rely on the exhaust stream to get them up to operating temp (about 600 degrees). the exhause stream at idle can be cool enough to allow a system to revert back to open loop...esp if the sensor is located way down in the exhaust stream. a heated sensor usually has 3 or 4 wires and will stay operating even at idle (unless the heater circuit has failed...in which case, the sensor would operate like a 1 wire sensor) the purpose in heating the sensor is to get to closed loop faster to help control the performance of the catalytic convertor sooner, thereby improving overall emissions.

btw...the reason for the toggling back and forth is because of the catalytic convertor. for the cat to operate well, the exhaust stream needs to be BOTH rich and lean...some richness produces CO and this is necessary to handle NOX emissions...some leanness is neceaasry so the that the cat can store oxygen molecules to catalize the CO and HC emissions. doesn't matter whether it's a 2-way or a 3-way cat...the alternating rich and lean thing is the same.

as desi said...the more the amplitude and the faster the sensor switches, the better condition it's in (keep in mind that decentfuel mileage isn't the only related factor in the performance of your o2...there's also emissions to consider). and the better the condition, the better the performance all around.

 
alung's Avatar
Member

Join Date: Feb 2003
Posts: 164

04-08-04, 01:13 PM   #20  
ok, I think we're all saying the same thing but in different contexts. So let me clarify to see if I'm understanding this :-)

An O2 sensor is simply a voltage source whose output is supposedly proportional to the difference in O2 levels between the exhaust and surrounding ambient. Based on some curves I saw, it looks more like a step function with respect to the stochiometric point. It does not oscillate on its own.

The output will change in response to changes in O2 levels in the exhaust since ambient O2 levels stay constant. How quickly a good sensor reacts to O2 changes should be in the order of high-tens of milliseconds. Worn out sensors (sounds like mine) or contaminated units will be much slower to respond.

Exhaust O2 content will change only as quickly as the car's fuel system (carburated, throttle body EFI, or multiport EFI) is designed to do. The fuel system constantly leans/enriches the fuel/air in a closed loop fashion to straddle the magic stoichiometric 450mV level, hence the transitions. 450mV happens to be dead-smack in the middle of the 900mV max sensor output level between lean and rich indications.

Based on some quick googling I did, at 2500rpm, the transition rate is slowest on engines with feedback carburetors, typically once per second. Engines with throttle body injection are 2 to 3 times per second, while engines with multiport injection are the fastest at 5 to 7 times per second . Therefore, the switch rate of a good sensor won't be any faster than the fuel management's inherent "loop response" time

Note, the catacalysmic converter's sole purpose is the oxidize and reduce any CO, HC, and NOx components. The engine management does not intentionally increase/decrease any of these components to help facilitate that process!

 
Desi501's Avatar
Member

Join Date: Oct 2003
Posts: 2,207

04-08-04, 02:18 PM   #21  
Originally posted by alung

Note, the catacalysmic converter's sole purpose is the oxidize and reduce any CO, HC, and NOx components. The engine management does not intentionally increase/decrease any of these components to help facilitate that process!
I think you were real good right up until this statement.

 
alung's Avatar
Member

Join Date: Feb 2003
Posts: 164

04-08-04, 02:46 PM   #22  
aside from the intended humor, it's a true statement!

 
Search this Thread