Is this a Lazy Titania O2 Sensor?

[Stephen]

[b]***UPDATE 1st April 2016***

• Oddly there is no secondary air system on my 1997 Z3.
• I had a tetra clean performed, it made no difference although my MPG increased by 5
• Crankcase breather pipe was a little clogged up so was replaced.
  [li]My long term fuel trim was steadily increasing, at one point it reached Minus -16%, so….

I installed a new pre-cat sensor even though reccomended replacement interval is 100,00 miles and mine has 60,000,
This made a huge difference and brought the LTFT down from minus -16% to minus -6%, I know ideally that LTFT’s should be zero.
Also the update intervals on O2 sensor is now reduced from approx 4 seconds down to 2 seconds.
anyway, UK MOT is due next week, so fingers crossed
[/b]

In an effort to diagnose a rich running car (fuel smell from exhaust & UK MOT Emission’s Fail), i decided to monitor STFT’s, LTFT’s & O2 data etc.

These readings/graphs were taken from a Titania PRE-Cat o2 Sensor on a UK 1997 2.8L BMW Z3 Roadster, Engine M52B28, Control Module=DME Siemens MS410DS3.
The car is nearly 18 years old with 57k on the clock

From what i understand the time it takes to switch from lean to rich should be a couple of times per second with Titania sensors but on my graphs it appears to be taking on average every 4 seconds,
but could anyone confirm this?
Or could it be that this 4 second switch is normal for the Siemens MS410DS3 control module?

The readings were taken during idle and after a 10 minute drive on the motorway.
There are no OBD codes, car seems to drive and Idle fine with no hesitations.

Ive attached 2 x pdf’s showing graphing of the one and only sensor on the vehicle which is a Pre cat O2,
also attached are 2 x pdf’s showing STFT’s.

I dont want to start needlessly throwing parts at this car where they are not needed, but before I continue to diagnose what seems to be a rich condition, I’m thinking the O2 sensor should be changed before I check any live data again and/or move onto the injectors, fuel pressure regulator etc.

Attachments:

• o2.pdf
• O2_2.pdf
• STFT.pdf
• STFT_2.pdf

Topic

[Daniel]

The o2 sensor seems to coincide with the fuel trim, so it looks to be working correctly.
Since the vehicle is so old, you may want to clean the throttle body with a tooth brush and cleaner. The most thorough way to clean it is to remove it from the vehicle because most of the gunk will be behind the butterfly valve. You will need to replace the throttle body gasket. Oh, and don’t use the tooth brush afterward. Lol
I’m guessing you are in Europe. There are a growing number of shops that use the “TerraClean” fuel cleaning system. I have heard spectacular things about this. Once it makes it’s way to the USA I plan on getting it done.

Home

[Stephen]

[quote=”thisisbuod” post=136034]The o2 sensor seems to coincide with the fuel trim, so it looks to be working correctly.
Since the vehicle is so old, you may want to clean the throttle body with a tooth brush and cleaner. The most thorough way to clean it is to remove it from the vehicle because most of the gunk will be behind the butterfly valve. You will need to replace the throttle body gasket. Oh, and don’t use the tooth brush afterward. Lol
I’m guessing you are in Europe. There are a growing number of shops that use the “TerraClean” fuel cleaning system. I have heard spectacular things about this. Once it makes it’s way to the USA I plan on getting it done.

http://www.terraclean.co.uk/%5B/quote%5D

Many thanks for your response…
I had a peek at the throttle body not too long ago and TBH it looked remarkably clean, I was shocked given the age of the car, but I guess removing it completely would be a better thing to do based on what you’ve said..

As the fuel trims respond instantly to O2 data received, then if the O2 sensor is lazy wont the fuel trims be slow as well?

I’ve heard of tettraclean, I saw it on a UK program called wheel dealers, its a bit expensive, but its on my list of things to do.

[Daniel]

That’s the program I saw it on too.
You’re correct about the O2 affecting the fuel trim, but (in my opinion)
There is no reason to assume the O2 sensor is acting oddly.
Keep in mind that the O2 sensor isn’t what is reacting, it is the computer. The O2 sensor only sends (very simple) information to the computer. The computer uses all the info from all the engine sensors to formulate the optimal measurements for the climate it is in at the time.
If the O2 sensor was not acting exactly as the OEM component It would throw a code. It just looks to me that it has no reason to react quickly to anything. It would be a bad sign if the graph was jumping all over the place. That would make me suspect a misfire or faulty injector.
Judging by the age and readings the intake probably just needs a good cleaning.

[Stephen]

[quote=”thisisbuod” post=136039]
If the O2 sensor was not acting exactly as the OEM component It would throw a code. It just looks to me that it has no reason to react quickly to anything. It would be a bad sign if the graph was jumping all over the place. That would make me suspect a misfire or faulty injector.
Judging by the age and readings the intake probably just needs a good cleaning.[/quote]

Indeed, I’ve never had a code and the graphed o2 voltages seem perfect, but I am curious about the switching frequency, I think I’ll invoke a lean condition by pulling the brake servo hose and see just how quickly the O2 responds.

Thank you for your advice, it is much appreciated…
Based on your advice it does make sense that I can rule out faulty injectors as a cause for running rich.
So, I shall remove the TB this weekend, give it a good clean and check the fuel pressure regulator also

[Daniel]

Cool. Keep us updated.

[Stephen]

throttle body surprisingly clean, but gave it a wipe down anyway.

Fuel pressure regulator is ok (visually inspection, Ive not done a pressure test on the fuel rail yet)

4 x electrode Spark Plugs removed, although they only had 2000 miles on them (Type-NGK BKR 6EQUP) and replaced with type-NGK BKR6EK

I compared the Z3 O2 graph with my Ford Focus daily driver, although the focus has a Zirconia sensor and the Z3 has Titania sensor, the response times were about the same, so I’m ruling out the pre cat O2 sensor being a problem although I might replace it when I’m flushed with funds, after all, its 18 years old and nearly got 60k on it.

I’ve noticed that the OEM exhaust has been replaced after the pre-cat sensor with a stainless system comprising of a sports cat and back box (No middle box),
seen as the car runs fine I’m concluding that being a sports cat maybe it wasn’t up to temperature during my 10 mile drive to the UK MOT Testing centre……the next time its tested I’ll drive the 10 miles hard in 3rd gear to get the CAT toasty.
As i don’t have an exhaust gas analyser I’ll have to wait until the test is due again next April.

Focus o2 graph is attached for anyone thats interested
–Ford Focus 1.6l Petrol MK2 2005 (UK Hatchback) with 46k miles–

Attachments:

• FordFocusO2Graph.pdf

[Daniel]

Was the throttle body clean behind the butterfly valve? Cause this is where most of the build up would be.
Like I said above: the O2 sensor only reads the O2 levels. As far as the speed of corrections, that is the computer. An O2 sensor is analog, meaning it doesn’t have a refresh speed. The speed that you are focusing on is how fast the ECU is refreshing NOT the O2 sensor.
As far as the Cat: I think you might be on to something with the temp. I don’t know much about performance parts like this, but it’s not OEM and, with cars like this, that can cause problems. Volvos are total buttholes when it comes to using non-OEM parts.

[Aldon]

Hey guys, I just wanted to bring something to your attention that I thought might be useful giving your circumstances and what you’re talking about. Now I’m not very knowledgeable about cars, so I don’t know how or why it works, but you mentioned a possible “rich” condition. Also given it seems like your O₂ sensors are working (based on what “thisisbuod” says), you might consider checking for a vacuum leak.

I currently have a code on my car for a rich condition in bank 1, and people were telling me to check the O₂ sensor, however based upon what Eric says in his video I posted about below (unless he happened to misspeak), It seems like a good thing to check. So I’m going to at least check out a vacuum leak when I have time since it’s really easy to do. Might be something for you to consider as well.

On this video of ETCG. (The statement in the time from 1:44 to 2:28 is what I’m referring to.)

Best of luck,
Aldon

[Stephen]

[quote=”thisisbuod” post=136720]Was the throttle body clean behind the butterfly .[/quote]
Yes, it was surprisingly clean….I can only guess that someone has been in there before me and kindly cleaned it.

[quote=”Aldon” post=136736]you might consider checking for a vacuum leak. [/quote]
This would invoke a lean condition, not a rich one…
or could it be that if it does run lean due to a vacuum leak then the system adds more fuel to compensate for the leak and emit the rich smell from the exhaust?……
I don’t know, my brain hurts! lol

[Aldon]

[quote=”sirius-c” post=136977]This would invoke a lean condition, not a rich one…
or could it be that if it does run lean due to a vacuum leak then the system adds more fuel to compensate for the leak and emit the rich smell from the exhaust?……
I don’t know, my brain hurts! lol[/quote]

Hey sirius-c,
I had a guy explain it to me on one of my posts. I have a code for a rich condition in my engine and was wondering if I should check for a vacuum leak based upon that video that I linked you. I’ve quoted his reply at the end of the post if you want to read it. I think I get it now and realize what I was misunderstanding.

Eric is only talking about the “fuel trims” trending rich in his second statement in the video. So a vacuum leak will give a lean condition, but if you look specifically at the “fuel trim” readings and they have increased, then you might have a vacuum leak. It’s only the fuel trims that change and become more rich, but they’re doing that to compensate for a lean condition. That automatic adjustment of the fuel system won’t cause a rich code though, but a lean code IF the system can not compensate enough to get the proper mixture. The fuel trims are just something else to look at to further diagnose the problem and know if you should look towards a bad O₂ sensor or a vacuum leak.

Sorry if I confused you with my post, I just thought I heard some useful information but I didn’t understand what it meant. At least I learned something and hopefully maybe you did as well. 🙂

[quote=”cap269″ post=136741]When there is an air/vacuum leak, there is air in the system that did not pass through the MAF, so it is not accounted for in the computer’s calculation. This typically causes a lean condition. The O2 responds to this and the computer enriches the mixture (positive short term fuel trim) to compensate. On the other hand, something such as a leaky injector will apply too much fuel. The O2 indicates a rich condition and the computer then tries to remove fuel or add air, reducing fuel injector pulse width (negative short term fuel trim), or maybe opening the IACV, or some other strategy, again trying to achieve stoich (the proper air to fuel ratio). If it is unable to provide stoich through the adjustment strategies it is programmed for, it sets a rich or lean code. If the compensation results in stoich, no code is set, and the new values are written to memory (long term fuel trim). This is called the “adaptive fuel system.” Only when the computer fails to compensate and produce stoich is a code set. This is the idea Eric was talking about in the video.[/quote]

Best of luck,
Aldon

[Stephen]

Hi Aldon,

Thanks for your reply,

Its not you that confused me …
I knew absolutely nothing about fuel injection systems a month ago, since then Ive watched lots of videos, strangely enough I found them very interesting (must be my age!), but what I do find is that the deeper you look the easier it is to get confused, well for me anyway :blink:

BTW some of the best videos I’ve seen came from this guy on Youtube:-
Schrodingers Box

My short term fuel trims (See attached files in my opening post) seem to trend more towards reducing fuel rather than adding it, so its slightly rich and trying to take fuel away.
From what I’ve read, the general consensus on STFT’s seems to be that +/- 5% is ideal, +/- 10% is acceptable and that the CEL/MIL light is normally illuminated and OBD codes thrown for anything above +/- 20-25% depending on the vehicle……anyhow….

The last test I need to do is a fuel pressure test and whilst I’m poking around under the Bonnet (or Hood 😛 ) I may as well use the propane method and check for vacuum leaks too, all diagnostic readings indicate that there shouldn’t be any leaks, but why skip such an easy test while I’m under there?

Does anyone know if I can use Butane instead of Propane?

[Andrew Phillips]

I think I will weigh in a little on this topic.

[quote=”sirius-c” post=137015]From what I’ve read, the general consensus on STFT’s seems to be that +/- 5% is ideal, +/- 10% is acceptable and that the CEL/MIL light is normally illuminated and OBD codes thrown for anything above +/- 20-25% depending on the vehicle……anyhow….[/quote]
That is mostly correct, however when looking at STFT% you also want to look at LTFT%. If the LT is, say, 5%, and ST is 3%, then the overall correction is 8% and that would indicate something is out of spec. Once a value is written to LT, the ST is supposed to go to 0. Then the adaptive strategies start over. If the LT continues to increase, by virtue of the ST continuing to increase, something is wrong. Personally, I do not endorse 10% correction as acceptable in most circumstances. When the overall correction gets into double digits the fuel economy is going to suffer.

The last test I need to do is a fuel pressure test and whilst I’m poking around under the Bonnet (or Hood 😛 ) I may as well use the propane method and check for vacuum leaks too, all diagnostic readings indicate that there shouldn’t be any leaks, but why skip such an easy test while I’m under there?

Fuel pressure tests are always good to run, just to get a feeling of overall health for the fuel system if nothing else. A clogged filter or pump beginning to fail will cause a lean condition due to fuel starvation. A faulty pressure regulator or leaky injector, for example, can cause a rich condition. As for vacuum leaks, these typically cause lean conditions, and remember that there can be other problems that can mimic a vacuum leak when there isn’t really a vacuum leak. A partially stuck open EGR valve is one such example among others.

Does anyone know if I can use Butane instead of Propane?

I’ve never used butane, but I can’t see why it wouldn’t work. You can also use carb spray, brake cleaner, starting fluid, etc. Just be smart when using flammables around a running engine. Blowing cigar smoke into the brake booster hose (without engine running) can help find vacuum leaks also, just watch for where the smoke comes out and you found the leak.

[Stephen]

Thanks cap269,

a good point regarding LTFT’s especially for others that might read this thread…
fortunately mine were changing between -1 and Zero while data logging on a long journey.

Ive seen the cigar method in action and IMO i regard this as the easiest way to do it, but I’m 1 week into kicking the cig habit so I think I’ll use the butane method instead, Id prefer this over the mess carb cleaner is going to create, it comes out of the nozzle with some force.

[quote=”cap269″ post=137036]remember that there can be other problems that can mimic a vacuum leak when there isn’t really a vacuum leak. A partially stuck open EGR valve is one such example among others.[/quote]

Interesting and noted, thank you

[Andrew Phillips]

[quote=”sirius-c” post=137037]fortunately mine were changing between -1 and Zero while data logging on a long journey.[/quote]
Yes, your graphs looked fine. The computer is supposed to alternate the STFT between rich and lean, this is done for the proper functioning of the O2 sensors. The O2 sensors should reflect this action, which your graphs do.

Ive seen the cigar method in action and IMO i regard this as the easiest way to do it, but I’m 1 week into kicking the cig habit so I think I’ll use the butane method instead

Good for you kicking the habit. There are smoke machines on the market for this purpose, and once I saw some plans online somewhere for a DIY smoke machine.

[Andrew Phillips]

What was the reason given for the emissions test fail?

[Author]

Replies