oxygen sensor

Sure, air leaks in the manifold upstream of the sensors can cause problems. If you are sure the cars is properly tuned (timing belt, plugs, etc), and you are sure there is no air leaks then it may be tired oxygen sensors. A bad downstream sensor usually sets P0420. A P0136 may indicate you have a physical problem with the sensor circuit (gone bad) and not a range problem. You should go through the basic test procedure that can be found in a typical Haynes manual at the local auto parts store, but know that these can only be properly tested in a laboratory.

Yes, downstream sensors can affect fuel trim by typically 1-3% depending on the manufacturer. If you have more than 60-80K miles on stock Denso sensors then consider replacing them with Bosch (planar type preferred). IMO Denso heated sensors are not very good given by all the TSB and converter replacements needed. Denso makes good alternators but when it comes to sensors they shouldn't quit their day job.

I'd recommend Bosch planar type sensors. Best in class stuff. I managed to rid the P0420 code with new Bosch sensors when TSB suggested replacing the cat (and leaving the sensors alone!)

See:

If you have a volt meter, with the engine warmed up, rev the engine to between 2,000 and 3,000 RPM and check voltage output from Bank 1 Sensor 2 O2 sensor. It should fluctuate from less than .4 volts to more than .5 volts. If it doesn't, the O2 sensor is bad. If it does, check for an open or short in the circuit between the O2 sensor and the computer. Bank 1 Sensor 2 is the sensor located after the cat. If you have a single exhaust and cat., it is the only sensor. If you have 2 exhausts/cats, it is on the side that contains cylinder #1.

Crap!!!

the number two oxy sensor is usualy used to determine how well the cat is working . for example if you could see the sensor flow grafs the number one sensor would bee totally diff from the # 2 sensor if working right if they look the same or close then the cat is not working at the price of the sensors i would just replace the defective one you'll need it for emission testing. don't forget your engine management runs through a comuter so garbage in garbage out

The way to learn about how something works is to read or listen to what someone who is familiar with how that something works has to say, as you are doing now. For example, if you were a copier repair technician at one time and my experience is only with fixing the copiers I've had in my office, I would probably learn more about troubleshooting a copier by listening to what you have to say than by arguing that changing the drum will fix a problem with static buildup. ;-)

The OBD II numbering system is determined by the Society of Automotive Engineers. This site explains the numbering system:

Codes are grouped by system, not by component, and components do not generate codes. Most sensors are on-off switches like speed sensors; vary resistance like temperature and pressure sensors and throttle position sensors; or generate voltage like an O2 sensor. The ECM is programed with operating parameters for each sensor, so if the voltage is outside of what the ECM is looking for or the on-off frequency is incorrect or absent, then the ECM will generate a trouble code for that sensor or component.

As you can see from the site I posted above, a P01XX code has to do with fuel trim (air:fuel mixture), while a P04XX code has to do with secondary emissions controls. "Secondary" means after combustion has taken place.

The system is very logical if you think simply instead of deeply. Sensors send a signal, i.e., voltage or frequency, to the ECM. The ECM is programmed with a bunch of if-then statements. For example, IF the voltage from bank 1 O2 sensor #2 circuit is less than or equal to .4, THEN illuminate MIL and store trouble code P0136

You have to get away from the commonly held notion that a component generates codes and that code groupings are related to components. Since the ECM generates codes based on what it expects to see from a sensor, the proper way to diagnose what caused a particular code is to understand how a system works and find out what the ECM is expecting to see from a sensor and then checking the sensor.

Fortunately, most automotive sensors generate voltage, which is relatively easy to check with a volt meter, or frequency. The proper way to check frequency is with an ocilloscope, but fortunately, in my experience, a sensor that generates frequency like a speed sensor or O2 sensor rarely generates a frequency that is out of spec, and the problem is usually zero voltage - no frequency, so you can still check with a volt meter.

In the case of conditions that cause the ECM to generate P0420, you have to understand that while O2 sensor #1 and O2 sensor #2 work the same way, they serve different purposes. You also have to understand what systems being monitored are doing and what the signals from O2 sensors mean.

The the voltage from O2 sensor #1 should fluctuate at a frequency a little greater than 1 cycle per second. If the catalytic converter is working, it will burn the unburnt and partially burnt fuel in the exhasut gas and even out the fluctuations in O2 content so the frequency from sensor #2 should be lower than the frequency from sensor #1. If the frequency from sensor #2 is the same as sensor #1, the cat is not doing anything so the ECM sets P0420. In other words, the presence of a signal, not the absence of a signal, generates P0420.

If it possible for P0420 to come up with a good cat if sensor #1 is not reacting quickly enough, so its frequency fluctuations are too low and so the frequency of sensor #2 matches the signal from sensor #1.

In the case of P0136, the ECM does not see any signal from sensor #2 or the voltage from sensor #2 is too low.

Good post, Ray. May I add a caution, when measuring the volts by a voltmeter? If the voltage is fluctuating and the wave shape is peculiar, the value shown can be misleading. Imagine a wave that comprises brief spikes of 1v among a steady 0.1v -- the meter can not show that accurately, whereas a 'scope would (as you imply).

Good point. Since I do not have an oscilloscope, I make do with the volt meter. Fluke makes a very nice meter that has a built-in scope, but I'm having a hard time justifying its purchase since the $20 Radio Shack meter purchased about 30 years ago still works.

system:

Wow...this is a lot of great information. Thanks very much for your posts. I reset the codes last Thurs. and as of yet I've not seen any more check lights and my OBD II scanner has not picked up anything either...mind you, my scanner says the vehicle has not completed it's cycle of tests yet. I'm not sure if the system will generate codes before it reaches the ready state. If I don't get a code I think I'll swap the first sensor (pre-cat) out anyway, just because of my mileage. If I do get another code I'll likely do both...but I'm curious to hook the meter up and check them out before I do anything. (It's just a little too cold out lately to get out and work in an unheated garage, last night was -20 Celsius ;-)

Now that is an unacceptable answer!

You mean you can't justify an new toy?

Say it ain't so!!!!!!!!!!!!!!!

Good question. I do not know if the OBD II system will generate codes before it reaches a ready state, but I suspect that it will, because the system reaches a ready state by checking each sensor for output.

Let us know what you find out!

I'm saving up for the Aston Martin

Ah ok...then.....

That's a good attitude but, if you want the machine guns behind the side lights, you're going to have to persuade the licensing people you have a level-headed outlook. Denying yourself a few essentials of life on the grounds of mere cost is... not a good start. Or will you be satisfied with plastic guns and flashing LEDs, and a grotty sound-maker that goes "budda-budda-budda"?

[Head shaking in disappointment} OK, the machine guns behind the side marker lights might not be necessary but I would still want the tire shredders