Jump to content

o2 sensor


Fern
 Share

Recommended Posts

I have recently bought an oscilloscope and have already measured the oxygen sensor of mine. Does anybody has a waveform of good working o2 sensor on discovery? I interested in discovery, not in the samples from the other cars, 'case I have them a lot, but would like to have from disco. The thing is that I want to absolutely sure that my waveform is wrong (o2 sensor is not working correctly). I attach a waveform which I got.

post-8324-1219517248_thumb.jpg

Link to comment
Share on other sites

Looks OK to me, when the engine is running closed-loop it will cycle back and forth like that. What makes you think it's faulty?

BTW here's a trace from MegaLogViewer, the Lambda sensor is the green trace at the bottom, the red trace being correction applied by the ECU in response to the Lambda sensor:

post-21-1219568851_thumb.jpg

Link to comment
Share on other sites

Looks good to me,the trace is quite clean - ie,going rich - lean and back with no dodgy 1/2 way squiggles.All you need from them,(with the engine up to temp) is to cycle at once a second or less the higher the revs whilst at cruise or idle.Stay at .8-1v,(rich) with heavy load and large throttle openings.and to stay down at 0v on overrun,(lean as the injectors shut right off)Scoping the 2 oxy sensors is how I check and setup LPG systems.(and faultfind petrol EFI) Its quite an interesting thing to play around with and is a really good way to understand how closed loop fuelling works.

What is the issue you have with your car ?

Link to comment
Share on other sites

My car has problems running idle, big fuel consuption, loss of power. Well, these symptoms may be caused of bad o2 sensor, you might all agree, that why I wanted to measure it before I replace.

THANX FOR THE WAVEFORM, FridgeFreezer! Do you have a more detailed waveform to see the time?

I've read lot's of docs, but they were mainly for BOSCH sensors, probably sensors in LR work a little bit different. Anyways, according to the docs here are the things I do not like in my waveform:

1. Voltage below 0v. I expected it to be at least 0, but not below.

2. Voltage above 0.9v. Never seen a waveform with a signal above 1v. Until now.

3. Frequency. The disctance between peaks must be not more than 1 second. The signal must change 1-2 times a second on idle running engine. But not really sure.

4. Attack. The distance from the low value giong to high must be not more than 120ms. And I have much more.

Look at this video file, it's in russian but you'll understand what I am talking about. Have patience to look at it:) http://injectorservice.com.ua/docs/lambda.swf

I will appreciate any comments and sharing of your knowledge.

Link to comment
Share on other sites

1) this can easily be caused by poor earth being used, so you are comparing against say 0.01v earth potential rather than 0v. It can also be caused and is quite likely, to be the capacitance of the probes, as the swing comes back you end up with overshoot/ringing, quite common to see.

4) Couldn't tell from the trace, but it looks somewhat suffering from low sample rate, are you sure that the max rise you see isn't being limited by your sample rate? I couldn't tell what timescale I was looking at, it looked like seconds along the bottom?

Don't know much more than that of help..

Link to comment
Share on other sites

As Ally said, you need to understand how closed-loop fuelling works - if you have problems at idle it's unlikely to be related to the lambda sensor as the ECU only runs closed loop at cruising conditions. If it's running rich and misbehaving I'd look at other things such as coolant sensor & airflow meter, also if it's hotwire the idle steppers can play up although that wouldn't make it run rich.

Also to add - some LR lambda sensors are different than most others, they are resistive so do not give the normal 0-1v signal. Voltage difference, frequency and attack are not really issues - your trace is still OK and certainly doesn't indicate a fault with the sensor.

Link to comment
Share on other sites

Your engine issues are unlikely to be the fault of your oxy sensors if both are switching as well as the scope trace you show.I would look elswhere,as stated you are only looking at a small electrical potential,so the fact that they are switching so cleanly is all you can ask of them.Poor plug leads,plugs,leaky injectors,disconnected fuel pressure regulator,inlet and exhaust leaks,poor compression or blocked exhausts are just a few reasons for poor performance.You would be suprised how poorly an engine can be running yet the ecu is still able to run closed loop and keep the sensors switching.Dont get too hung up on specific waveform times,your problem is elswhere.Good luck in finding it,its an interesting journey.

Link to comment
Share on other sites

The idea about bad ground is might be reason of negative voltage. Thanx, steve_a.

The samplerate is ok. Will take some more waveforms tomorrow with good samplerate anyways.

2FridgeFreezer: I always thought that o2 sensors take part in idle work. Sure stepper motor gives air on idle, but ECU still has to know how much fuel should it give.

About the LR sensors and waveform. Thats what I thought. Decpite BOSCH docs my sensor's waveform is looking good in general. I was confused that it does not work as I expected, but it works completely different, so much different that I began to think that it should be in this way and everyghing is allright.

My last open question. As I read there are types of o2 sensors zirconium and titania. But as I see here http://www.picoauto.com/applications/lambda-sensor.html I would expect from LR's sensor voltage from 0 to 5v, but I see only 1.2v max.

What's wrong? Nevertheless I found o2 titania sensor for Disco 3.9 http://www.lambdasensor.com/main/lrover.htm

Is it Zirconium or Titania? I suppose titania, but why does it work so strange?

Link to comment
Share on other sites

Your engine issues are unlikely to be the fault of your oxy sensors if both are switching as well as the scope trace you show.I would look elswhere,as stated you are only looking at a small electrical potential,so the fact that they are switching so cleanly is all you can ask of them.Poor plug leads,plugs,leaky injectors,disconnected fuel pressure regulator,inlet and exhaust leaks,poor compression or blocked exhausts are just a few reasons for poor performance.You would be suprised how poorly an engine can be running yet the ecu is still able to run closed loop and keep the sensors switching.Dont get too hung up on specific waveform times,your problem is elswhere.Good luck in finding it,its an interesting journey.

:) You just read my mind! Thanx!

Link to comment
Share on other sites

It seems I found an answer to my question

http://www.landroverforums.com/archive/thr...or-71115-1.html

Zirconia O2 sensor (narrowband/wideband) Creates an output voltage corresponding to the quantity of oxygen in the exhaust relative to that in the atmosphere.

Titania O2 sensor Does not generate its own voltage, but changes its electrical resistance (ohm) in response to the oxygen concentration.

Titania O2 Sensor Cable designation:

red - 12 V (heater)

white - 10.8 V return (heater) 1.2 volt stay always inside the sensor

yellow ground

black - sends resistance (ohm) signal to the ECU

So here it is: 1.2v.

Link to comment
Share on other sites

2FridgeFreezer: I always thought that o2 sensors take part in idle work. Sure stepper motor gives air on idle, but ECU still has to know how much fuel should it give.

The ECU knows how much fuel to give from the fuel map - the O2 sensors are narrowband type and as such only accurate for 14.7:1 AFR, which is too lean for a good idle. You cannot use the O2 sensors for normal closed-loop tuning at idle, high RPM or high load as you need to be richer than 14.7:1 at these points. The MegaTune, MegaLogViewer and Tune Studio software allow for very good tuning at all mixtures using data logs taken with a narrowband sensor, but this is different to anything the Hotwire ECU is capable of.

I would expect from LR's sensor voltage from 0 to 5v, but I see only 1.2v max.

What's wrong?

With resistive sensors you must generate a reference voltage in order to then measure the resistance - a lot of applications (for example coolant, air temp, throttle position) use a 5v reference, but it's just as possible to use 1v, 10v, 1.2v, anything. As Ally (once again) said - it is very unlikely that your O2 sensor(s) are faulty, so STOP WORRYING about them and look for the real fault. :rolleyes:

Link to comment
Share on other sites

....STOP WORRYING...

Thank you for giving some light on my doubts, 'cause I've been completely confused. I'm currently thinking of buying two additional sensors for my oscillo: pressure (to test intake pressure) and high-voltage (to test ignition problems).

Link to comment
Share on other sites

Thank you for giving some light on my doubts, 'cause I've been completely confused. I'm currently thinking of buying two additional sensors for my oscillo: pressure (to test intake pressure) and high-voltage (to test ignition problems).

Don't bother - you don't need to worry about intake vacuum, and ignition problems are almost always the low-voltage side of things -you can test the high voltage side with a cheap (£1) flash tester / indicator. There's nothing useful I can think of you'd do with an oscilloscope on the HT side of the ignition.

If your setup is Flapper or Hotwire EFi, all you need is a multimeter and the patience to understand how the system works - there is a very good fault finding guide in the technical archive that explains things pretty well.

If you're really bothered about all this, fit MegaSquirt'n'EDIS and then you can see everything with a laptop:

main_window.png

tune_fuel.png

Although with MS you're less likely to have stupid problems in the first place :lol:

Link to comment
Share on other sites

With a high voltage sensor I want to recon myself that everything is allright with wires, sparcs, distributor, coil.

MegaSquirt......, never heard of it before, will think of it for the future, but not now. Very seducing thing, probably will need to get divorced with my wife to install MS:)

And BTW, here is the waveform I've got today from two o2 sensors. Green is right sensor (which I had measured before) and pink is left (which I did not!). And as far as I understand it does not work properly! After thinking a bit I had cutted the black wire on "not working sensor" and hardwired it with a black wire of "working sensor" (you'll say it's stupid but I did it, let's say for experiment), to make ECU read signal from working sensor. So I supposed to get 1 working sensor instead of 2. Nothing particulary happened, I did not notice the difference in engine work.

Have to dig further. I think even if the o2 sensors are dead my main problem is not in them.

post-8324-1219785120_thumb.jpg

Link to comment
Share on other sites

OK, how many ways can I say this?

Your O2 sensors are not faulty!

If your engine was overheating you would not blame your temperature gauge, so don't blame your O2 sensors for rough running :rolleyes: Just because you don't agree with the reading does not mean the sensor is faulty :huh:

They are both switching up & down cleanly which is about as much as you could ever want to see. The output they generate is in response to the air/fuel mixture in your exhaust gas. Now, the fact that one reads richer than the other (purple trace shows more time spent rich than the green trace) could be as simple as a leaky gasket in the exhaust (air contains more O2 than exhaust gas does, so will give a false "rich" reading as air is drawn in by the exhaust gas pulses). It could be a sticking injector on one bank, a loose injector connector, a corroded injector contact, a dodgy transistor in the ECU, a spark plug, HT lead, oil leak/burning oil (oil will coat the O2 sensor and cause incorrect reading)... all sorts of things that could cause a different mixture on one bank to the other.

And, I will say again, a high voltage probe is not going to tell you anything useful about your ignition system - you are trying to apply rocket science to clockwork here! You can hear if the car is running on all 8 cylinders, if there is a misfire you can find it easily by pulling HT leads off one at a time to see which one causes no change. Everything else is low voltage, most ignition faults are with the amplifier.

Link to comment
Share on other sites

Fridge,totally agree with you that Ferns issue is not with his oxy sensors,but a couple of other things you have stated I am suprised at.Using a scope is really useful - esp on v8's that sound fine but wont pass emissions test or pick up roughly from low revs.Bosch injected Thor V8's with these issues I scope the plug leads before doing anything else,esp for garages on sales/warranty jobs where they want the quickest cheapest result.(Often just replace a single plug lead)

Narrow band sensors and stoich not being good enough for idle ? Where did that come from,they all cycle at idle,and stoich is fine - its good for emissions and rich enough to prevent stalling.Dont want to start an arguement or upset anyone,I'm just suprised at your comments.

Link to comment
Share on other sites

My car is a big toy for me right now. I've bought an oscillo to measure some things and this will probably help me in studying the engine work.

I agree now that o2 sensors are not my case. But let me explain why I began to think of them. When I've seen the injection timing the 1,2,3,4 injectors where 2.7-2.8ms and 5,6,7,8 where 2.4-2.5ms. The injectors themselves I've cleaned allready, no change. The same story running on LPG. So I'm still in search for answer why ECU gives more time for left set of cylinders and less time for right. The first idea was: it's because o2 sensors work differently and one of them might give infromation about lean exhaust and computer wants to make it reach. I was wrong.

Link to comment
Share on other sites

Ally - as I said, using a scope on a Rover V8 is overkill and liable to lead to further bouts of automotive hypochondria ;) sure they're (very) useful for some things but for basic diagnostics it's complicating the issue rather.

As for idle mixture, every V8 I've tuned has preferred being richer than stoich at idle, and if you enable O2 correction at low RPM it will hunt around and stumble from idle.

Link to comment
Share on other sites

I guess we'll have to agree to disagree about the scope - I'll carry on using it as a quick and effective way of sorting poor running/emissions issues.

Where you are talking about enabling oxy sensors at idle,what injection system are you talking about as I play with Mems up to Mems S3,14Cux, Gems and Bosch to current and have no problems in achieving a stable,reliable idle on any of them.What are you using to measure AFR ?

Link to comment
Share on other sites

Ally - I'm talking about MegaSquirt, and using the native NBO2 sensor voltage to watch the mixture. It's entirely possible to get a stable idle running lean, but all RV8's I've come across prefer richer, especially when bimbling round at idle off-road. You could just as easily idle lean and have more accel enrichment to avoid stumbling, but this seems to work.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...

Important Information

We use cookies to ensure you get the best experience. By using our website you agree to our Cookie Policy