Oxygen sensor Tech Article
 

My latest article:


Oxygen Sensors


O2 sensor, oxygen sensor, air-fuel sensor, wide band sensor, lambda sensor. These are all different versions or names of the same sensor type, all with the same purpose. They are placed in the exhaust gas path to manage fuel delivery, manage emissions and monitor catalytic converter efficiency. There are two main types of design: O2 and air fuel. Often they are all generically referred to as O2 or oxygen sensors but they output a different signal in response to changing exhaust gas composition. The main job of the entire ODBII system and all of its sensors and components is to minimize emissions and it does this by keeping the air-fuel mixture as close to 14:1 as possible and then for the cat converter to be operating properly to remove certain emission content through a catalytic chemical reaction between the exhaust gas and the converter catalytic compounds.


For us there is always one upstream sensor for each cylinder bank in the respective exhaust header near the cylinders, sometimes one more pre-cat sensor just ahead of the cat, and always one downstream sensor after the cat. The first precat sensor on bank one (drivers side) is B1S1. The first precat sensor in bank 2 on the passenger side is B2S1. On older blazers with one more precat sensor just ahead of the cat this is B1S2 and the sensor after the cat is B1S3. On newer blazers without the third precat sensor the first two are the same and the sensor after the cat is either B1S2 or B1S3 depending on the year.


Air fuel sensors are used on newer vehicles and have better control of fuel by changing the voltage output in direct proportion to the air fuel mixture and the voltage hovers around this well controlled mixture. The graph is a slowly drifting flat line voltage around lambda or a 14:1 air-fuel ratio. That target voltage level varies by vehicle. We do not have such sensors; all of ours (3 or 4) are old style basic O2 sensors with 4 wires. These sensors flop back and forth between around 0.15V and 0.85V, switching between values in about 100 milliseconds and switch back and forth between 1 and 5 times per second. The output looks like a square or triangle wave generator. When the exhaust gas composition becomes rich the O2 sensor values slams high and the ECM reduces the fuel injector pulse width and when the exhaust becomes lean the O2 sensor voltage slams low and the ECM increases fuel injector pulse width. This switch dance occurs constantly and the ECM has to keep tweaking the fuel delivery to get the A/F ratio to hover around 14:1 which is the optimum value for minimum emissions.


When the truck first starts the fuel system is in open loop and the ECM does not use the O2 sensors to manage fuel delivery. After the O2 sensors get warmed up by their internal heaters and start doing their thing the system transitions to closed loop and the ECM stops using the other sensors to take an educated guess at fuel delivery and uses the upstream O2 sensors for fuel control. Also, the cat has to warm up to start doing its thing which happens naturally after a couple of minutes of run time. The downstream sensor is there to monitor the operation of the converter. If it’s working correctly, the downstream sensor will hover around 0.45v. If the converter has failed then the downstream sensor will mirror the output of the upstream sensors, switching back and forth.


O2 sensors fail in several ways. The most common is that the heater circuit fails and you get a corresponding code like P0135. This code will not cause severe problems but what does happen is that the sensor is slow to warm up and this impacts fuel mileage and emissions because the system is slow to get up to optimum performance. This can be due to a wiring or connector fault but usually results from the heating element failing and the only fix is to replace the O2 sensor. The sensor can also start reporting out of voltage range, responding slowly or fail to report altogether. Possibilies for codes are: the system is lean (P0171), the sensor is responding too slow (P0159), the circuit voltage is too low (P0157). There is a circuit malfunction (0162), among others. The sensor may be failing but there are no codes. As with all observed or reported problems the question: Is the sensor failing or is the wiring or connector failing or is the parameter being measured out of bounds and the sensor is reporting accurately?


It is beyond the scope of this article to cover every failure mode that may involve an O2 sensor but here are a few of the most common issues:


If you get a heater circuit failure, check the wiring and you can even check the heater power if you really want to confirm and then just replace the sensor.


If you suspect a bad O2 sensor or cat or may be having issues in fuel delivery (lean codes, running rough), look at the live waveforms in graphic format. The upstream sensors should be switching back and forth between 0.15v and 0.85v at between 1-5 cycles per second with about a 100ms rise and fall time.

Some additional tests include propane to see if the sensors peg high (rich) quickly and a major intentional vacuum leak to see if they go low quick enough, Also do a drive test and floor it under a load, the upstream sensors should peg high because the truck goes back to open loop with a rich command. Be careful with your conclusions because the sensors may be reporting correctly and they are not the problem. Use other measurements to confirm like fuel trims and misfire counters. The downstream sensor should be roughly flat lined at around 0.45 volts indicating a well regulated air/fuel mixture and proper cat function. If the cat is warm and the post cat sensor is tracking the upstream sensors then you may have a failed cat but it needs to be confirmed. Note that a cat can fail in two ways: it is no longer removing pollutants as designed but still allows proper exhaust flow or the internal honeycomb structure is getting plugged up and causing excessive back pressure.

Lastly, a simple method to rule out a bad sensor is if one looks wonky but another sensor looks good, swap them and see if the problem follows. If not then its not the sensor, its the system. Make sense?


George

Hi there. I have a few INC monitors including the O2 heaters and cat. If the O2 heater circuit is good, and the ECU is sending power to the O2 heaters and no codes present, is there something I should look for besides the average wiring...etc.?. I had replaced several parts on vehicle including bat, and scanned with a pro scan tool which showed everything was running as it should...again no codes present. ECU is sending power to O2 heaters...and O2 sensor heater continuity tests ok. It's about 80 degrees and I know EVAP is temperamental with changes in air and temp so that's not setting either. Tried the drive cycle with no luck. any help or suggestions would be great!

It may take more than one drive cycle to complete all readiness tests, including some overnights.

Did you also check for pending codes?

Next you could look at a graph of your O2 sensor outputs and see if they are working correctly. Do you have graphing capability in your scanner?

George

This would be especially true with the HO2S monitor, which I believe will only run when the engine is cold.

Thanks for replying. I was not aware my vehicle had 4 O2s till a day ago after seeing a photo online. Sure enough ."All this time" 2 on the Y and 1 before and 1 after cat. I replaced the 2 forwards...so I'm hoping this will set monitor clear. They are old and from what I have been reading if they get old they take longer to react and possibly never...but never spring a code??? The front 2 did. They coded for sure. So they got replaced. So far no codes and no leaks now at all and vehicle runs as it should. I'm thinking the 2 by the cat is suspect. I've driven the vehicle well over 500 miles and have not reset ECU since the repair. Reading the OBDII reader everyday for 2 months after repairs on injection...a new spider injection replacement. CAT, EVAP, and HTR are the monitors remaining. I've tested voltage and continuity of all O2s on all 4 circuits and they check out amazingly. The kind smog guy put it on his pro scanner and came up with no issues either...but he could see past codes embedded which are now gone. I will try to read the scanner again once I get the new ones...as I only did a non engine running test for voltages and a brief engine run to clear codes to drive it since the repair. I've heard some of this stuff is in series with each other as far as what the ECU reads??? At this point I think it's better just to have all new and go from there. I'm impressed with the will to live at 266,000 miles.

You can test O2 sensors to see if they are sluggish and if they are switching properly. You'll need live data graphing.

The post cat sensor should have a steady voltage that moves around slowly near 0.5V. If it tracks the upstream sensors when the cat is warm then the cat may be dead. You can verify with an IR temp gun and/or a back pressure gauge.

Your going to need a code to know where to start on the evap system.


George

I agree on the EVAP. No codes so far. I do have live data stream on my scanner so I will have to go back and recheck all 4. I hope it's not the cat as it's only a few years old. I had previously tried to get emissions test before all this and missed by one point...with several issues already started at that point so I do I hope it clears after all this.

For grins and giggles read through this thread (in it's entirety) on getting an EVAP monitor to actually run. Evap IM - Page 16 - Blazer Forum - Chevy Blazer Forums Just read it for the information about the monitors and what prevents them from running.

That had some really interesting info and started off as a good read but half way in my brain exploded.

Les - What is your current opinion about best scanners at various price levels including trying to get close to a Tech 2 at a value price point?

Thanks,

George

The following is not directed at George, but just some general info.

Any old code reader will read/clear codes and display universal OBD2 PIDs. If that is all you need, get Torque Light (free) for your Android Phone and a BAFX Bluetooth Adapter from Amazon ($25). Beware of the cheapie Bluetooth adapters - they can be VERY slow.

1. For routine stuff on 1998+ Blazers I still like Car Diagnostic Pro (Updated version of Car Gauge Pro). Bluetooth OBD2 Interface. Reason: It will read/clear codes for other systems like ABS, HVAC, SIR, etc, it displays virtually all of the GM Powertrain PIDs for no extra money, and it has the bluetooth interface. Does Cam Sensor Retard. Cheap. Note that Car Gauge Pro does not work on the latest Android versions (only Car Diagnostic Pro). Only bad thing: No apparent support for 1996-1997 GM PIDs and no support for GM PIDs with CAN Protocol started in 2006. Cost: $9

2. For routine stuff on 1996-1997 Blazers I still like Dash Command. Bluetooth OBD2 Interface. Reason: It will actually display the 1996-1997 GM powertrain PIDs if you pay the extra money. Still cheap. Does Cam Sensor Retard. There is also an iPhone version that requires a WiFi OBD2 Interface. Will do all the years of Blazers, but you have to pay extra for the GM PIDs for each calendar year. Cost: $10 + $10 for each year of vehicle for GM PIDs. Will not do anything with other vehicle systems.

3. I have to mention Torque Pro for it's very nicely done User Interface. GM PIDs without extra purchase, but does not have Cam Sensor Retard PID. Bluetooth OBD2 Interface. Cost $9 Will not do anything with any vehicle systems other than Powertrain. Not sure if it works on 1996-1997 Blazers. Worth having if you get a Bluetooth adapter (if nothing more, worth it for just for the ability to plot/watch multiple PIDs on same graph).

4. Scan XL Pro. Windows software. Made by same company as Dash Command. $150 including all years of GM PIDs. Does GM PIDs on all years of Blazers including Cam Sensor Retard. Hardwired cable OBD2 adapter plugged into USB is how I used it, but I think it will do wireless. Does some ABS functions I have read but not used. Haven't used it for a long time.

5. HPTuners VCM Scanner. Windows software. Diagnostic end of the tuning software. $300 IIRC. Hardwired cable OBD2 adapter plugged into USB (made for the specific software). Has bi-directional control for select things like Crank Sensor Relearn, Injector Balance test, transmission control, etc. This one I used for Crank Sensor Relearns until I found #6.

6. Vident I-link400. Hand held scanner. $159. This one gets reasonably close to a Tech 2 in the stuff it can do. Still figuring out all it can do. Know it does Crank Sensor Relearn and Brake Bleeding. Know it does HVAC and 4WD PIDs. I'm sure there are some things that it doesn't do but I haven't found them yet.

As far as the Tech2 clones...... I have never had one, so I don't feel comfortable making recommendations about them. It's my understanding that they are really clones and have all the functionality of the Tech 2.

As far as the other software that is supposed to do the same things as I described........ Autoenginuity, EFI live, Blue Driver, etc....... I have never had them so I cannot review them.

Les

Perfect, thanks.

George

Ok so all O2 sensors replaced. I have to know...if after market O2's are any good. Store bought should work but i keep seeing direct fit and the ones I bought is meant for my model but does not say direct fit. No codes...but if they are not working will they spring one??? but same monitors are still present. HTR EVAP.

The aftermarket question is a loaded one because there are crappy quality options as well as the OEM version in many cases. When I did a Toyota clutch replacement some years ago I found out that Asin made the OEM version, got the equivalent part number and paid 50% of Toyota prices for the same clutch plate. The replacement has 125k perfect miles on it and it’s “after market”. Once you pick a quality part then you need to match the type (O2 vs wideband) and the wire count. A direct replacement label usually means this plus the correct connector, for plug and play. If not then you may have to do some wire splicing. I have found that Rockauto has a very good part number match track record then sometimes I use Amazon customer ratings if I’m not familiar with a quality pick on that part. Rockauto also has good pictures to match connectors, location etc.

To confirm proper O2 circuit operation:

Any codes or pending codes?
Verify proper switching on a live graph.
Verify proper response by forcing rich then lean on a live graph.
Verify proper heater circuit power.
Verify proper signal ground and high side bias signal and wiring.

Let me know if you need any help with any of this?

George