Xantia - Is Oxygen sensor OK

[lancia58]

Hi

I changed the OE Oxygen sensor with an aftermarket one. I hooked a Lexia to monitor its voltage, it is changed from 320mv to 640mv. The original changed from 240mv to 820mv and it was much more erratic. The measures was taken when the car is parked. Are these readings OK or the 320mv to 640mv range is too narrow ? I guess the response time of the new one is better and this may explain the narrower range
Thanks
Zohar

[lancia58]

Are readings of 320mv to 640mv OK ?

[CitroJim]

It could be but I'd have expected it to make full slow excursions from 100mV to 900mV really...

How are you measuring it? If you are using a digital voltmeter it will tend to average the voltage rather than showing it continually varying - the same as a Lexia or ELIT tends to do.

Ideally you really need to look at the sensor output on an oscilloscope to be absolutely sure Zohar...

[Mandrake]

About 100mv to 900mv as Jim says is normal I believe, however a couple of points -

One is that the ECU will ignore the oxygen sensor when the engine is cold - so test it with the engine up to temperature and idling. You can tell if the oxygen sensor is being used (ECU in closed loop mode) when the oxygen sensor status is 1 on the Lexia. If oxygen sensor status is 0 the ECU is currently in open loop mode, which will be the case when the engine is cold or you are accelerating more than moderately. In this case the oxygen reading will not cycle but will tend towards rich. (high voltage)

The other thing to keep in mind is that a cycling voltage doesn't necessarily mean that the sensor is working properly, at least with an old sensor. Oxygen sensors can have subtle faults and this is more common than outright failure.

It's quite a bit of reading but I'd strongly suggest reading the material on the following website for a better understanding of how the oxygen sensor works, and how to determine if it is working properly or not:

http://www.lambdapower.co.uk/technotes/techindex.asp" onclick="window.open(this.href);return false;

One question, what prompted you to change the oxygen sensor in the first place ? What were the symptoms ? Which engine does your car have ?

[lancia58]

The car has 100K (miles ) and I read that Oxygen sensors are tend to be less "efficient" i.e. wrong voltage or slower response time, this what prompted me to change the oxygen sensor. It’s a 2.0 16v petrol engine. I use a Lexia to measure it, which I used with the previous sensor that showed 240mv to 800 mv. The Lexia should be quite accurate as it retrieves the reading from the ECU that I guess uses an A to D that has high input resistance for accurate measurement.

Zohar

[CitroJim]

The acid-test Zohar, is to have your local garage carry out an emissions check as used for the MOT to measure Lambda for you. That will prove just how well the ECU is holding the mixture.

Generally, when using a Lexia or an ELIT I see the sensor swinging over the full range although it might take some time to see it fully as it depends on when it is sampled.

Watch the oxygen sensor with the engine idling at normal running temperature and you should see the full range displayed eventually...

[Chlorate]

Lambda sensors like that generally speaking aren't very clever bits of kit, all the sensor usually says is either the mixture is rich or lean and no points in between. It's true that they don't last forever, but if you were having trouble with it the ECU would probably tell you (by means of the engine management light) and/or an MOT emissions test might flag up something.

What is the difference between the OE sensor and your after market sensor? It could be returning a different voltage simply because it's a different sensor from the standard Bosch (I'm assuming) sensor, and then one would wonder if the ECU would make any sense of the readings its getting.

Alex

[CitroJim]

As far as I can tell Alex, any 4-wire sensor will give the correct results. I installed a cheap (£28) plugless sensor in my XM and it works a treat. For how long though is another matter.

You are quite correct Alex, in how they 'switch' at lambda = 1* going rapidly from a low to high or high to low voltage. The ECU uses the reading from the oxygen sensor to constantly 'trim' the mixture around lambda=1. They are extremely narrow-band devices and even at full 900mV output indicating a rich mixture the mixture isn't actually very rich at all and conversely at 100mV the mixture is not very weak; they work only over a very narrow range and the mixture has to be nearly right before they can function at all so if you see them switching at all then it's quite near the mark but not necessarily near the right mark for the following reason:

Air leaks on the exhaust system upstream of the sensor can adversely affect their output through it seeing 'false' oxygen. A bad manifold gasket or manifold to cat joint can play havoc. The sensor and ECU will still be apparently working normally but the mixture will be 'trimmed' wrongly due to an excess of oxygen in the exhaust from the leak. Burned valves and duff EGR valves (not an issue on XU engines) can also allow false oxygen in.

This Wikipedia article on them is quite interesting and explains how they work.

*Stoichiometric (14.8:1 air:fuel mixture), the ratio at which cats work best but is not necessarily the best for power or economy