Vx - Does knock detection cause spark misfire?

[JohnDee68]

Hi all,

I have just spent the past two days fully going over the wiring and sensors on my car. I keep coming back to the O2 sensors as still the issue.
I need some other peoples advice please on this before I go try to return the new sensors based on the strange readings I am getting (see images below).

In this Fig1 below, the two sensors react slow and don't seem to do as expected when at idle. They keep going out of 'ready' and then of course out of Closed loop mode. Even when they are sufficiently warm enough.

Fig1.jpg (57.24 KiB) Viewed 594 times

In Fig2 below, the sensors tend to cycle when revs are held around the 1600-2000 rpm. At what time they will stay in closed loop as expected. Cross counts were around 4 on one and 6 on the other during this time. Then as rpm returns to idle, the sensors go all haywire (i think) with what they are doing.

Fig1.jpg (57.24 KiB) Viewed 594 times

In Fig3 below, this is the same sensors as fig1 & Fig2 after idling for longer.

Fig3.jpg (52.98 KiB) Viewed 594 times

Now, with Fig4 below, this is what I get with one of my Old (BOSCH) original sensors (RHS) and one of the recent new ones (LHS) shown in Figure 1, 2 & 3.

Fig4.jpg (70.99 KiB) Viewed 594 times

Ideas appreciate, as always.

John.

[antus]

If the engine is warm in fig 4 thats what you want to see. keep those sensors.

[JohnDee68]

If the engine is warm in fig 4 thats what you want to see. keep those sensors.

Thanks antus, yes the engine for all four was warm and so were the sensors at running temp (if not a little hotter being stationary).

I really am confused (still learning). I thought that O2 signals should swing a more smoother sign wave like form and of a frequency about 1 hz or more depending on engine speed ?

I might need to run a tank of fuel and see how they go for economy then and for the PCM to self calibrate?

I wonder if the catalytic converter might be sooted up and not as effective due to the previous injectors running at about 25L/.100km for too long? (the exhaust is so badly sooted so it could be likely).

Would carbon burn off from the cat over a good run, or will i need to clean it physically with carb cleaner or something?

Thanks,
John.

[immortality]

If you had really good testing equipment you'd probably see a nicer wave.

Agree with Antus, fig4 is what you want to see.

A good clean running engine should burn off any excess soot from the system, especially once you get into lean cruise mode. The worst thing that can happen is having heaps of excess fuel in the cats that can end up melting the core and blocking it. Cat's used to smell of sulphur when they had lots of excess fuel in them.

[JohnDee68]

If you had really good testing equipment you'd probably see a nicer wave.

Agree with Antus, fig4 is what you want to see.

A good clean running engine should burn off any excess soot from the system, especially once you get into lean cruise mode. The worst thing that can happen is having heaps of excess fuel in the cats that can end up melting the core and blocking it. Cat's used to smell of sulphur when they had lots of excess fuel in them.

Thanks immortality for your feedback as well.

I was happier with the motor this morning that it was viable to try it on the road. Plus I needed to get fuel why cheapish. I filled up the car today to check both the refilling issue and the fuel gauge accuracy. Good to know the new charcoal canister has resolved the filling issue. While also the fuel gauge is still accurate and hadn't been upset while I had the fuel pump out last week or so.

So I did a couple of more trip logging today which I will look at later when I am up to it. Knowing that little more of what to look for with the o2's will help heaps. I might try doing a log with TunerPro again tomorrow. As it use to give a smoother sign wave form. But that was before the O2 sensors were replaced. At this stage over the past 3-4 days I have been trying to keep to one software as one of my available testing tools.

I think there is just one other issue to fault and then it may be back to reasonable and just a fine tune. Though I still cant yet see why the injector pulse timing goes from 1.x ms at idle, then to 2xx - 6xx ms when throttle is pressed. It is as if the injector timing is telling the injectors to stay open for nearly 0.5 of a second each time it opens.

I am glad to hear about the cat having a self cleaning method basically. So to clean the cat I could just take it for a country drive in cruise mode for an hour or more? I think I am lucky then that raw fuel didn't go down the exhaust at its worse time. (well It didn't seem it had). I wouldn't say the exhaust smell is of "egg gas". But it can be a bit harsh on the throat?

Thanks again guys,
I will document and put up all my findings once I get this finally resolved. Even if it might help only one other person in the future. It might help save time for others in the future.

John.

[The1]

ive never had much luck with bosch o2 sensors for narrowbands, especially in commo's, dephi's work much better and have good response time.

Ive had several bosch's out the packet do the same slow response as your seeing.

[JohnDee68]

ive never had much luck with bosch o2 sensors for narrowbands, especially in commo's, dephi's work much better and have good response time.

Ive had several bosch's out the packet do the same slow response as your seeing.

Hi The1,
These days its hard to know what are good, bad, or just down right going to be even adequate or even work out of a box. So I am glad for all peoples experience and suggestions.

One of the original Bosch ones in Fig4 (RHS) were in the car since 2004 (just on 539,000km now.) So I am happy they seemed to have lasted the life they have. Considering Bosch advise a new one these days have a rated life of only 50,000km on theirs now.
The other one in Fig4 (LHS) is a Goss brand. Other than their fuel injector flow rate chemical, it is the only other items I have used of Goss brand. I guess the local auto stores sell what they make the most profit margin on, rather than giving me option for quality? I am tempted (waiting Goss themselves to respond further to the sensor issue) to take both the new sensors back and go for something better? As the car even seemed to feel a lot more pokey just with the current set up of sensors in it today.

Is the Fuelmiser sensors actually Delphi ?

Does anyone else have similar high recommendation to use Delphi's O2 sensors ?

Thanks again,
John.

[immortality]

Other than my old VN I've only ever used Delphi O2 sensors in my cars.

However, it has been a good 6 or 7 years since I bought some so I can't say what the quality is like today but historically speaking I've not had issues with them and will most likely use them again in the future.

[JohnDee68]

Other than my old VN I've only ever used Delphi O2 sensors in my cars.

However, it has been a good 6 or 7 years since I bought some so I can't say what the quality is like today but historically speaking I've not had issues with them and will most likely use them again in the future.

That is good to hear. At least the ones six years ago have lasted 6 years so far. But yes, they may not be as good these days.
It seems a gamble with anything these days as we don't always know where they are manufactured in the world.

Thanks though, I will try seeing I they can be sourced easily this week if I end up needing new ones again.

[JohnDee68]

Ok, an update.

I have gone over quite a few data log files all the way back to 2016 (prior to current motor fitment).

The injector pulse times that go from approx 1.5ms to over 200ms on throttle, has been there all that time. Is it an issue, I think it could be, but not yet sure why and how.
Looking further in depth with the log files. It appears there are knock sensor tripping - ONLY when in motion. This knocking appears to be apparent on acceleration.

I am aware there is a known issue (vibration/shudder) which I believe with the transmission or just the torque coveter as it happens at predictable speeds and RPM. But I am yet to know for sure exactly what is the cause. As it is still the original unit with the full 539,000km on it and has only had the pump replaced a couple of years ago now.

I am wondering, could the knock sensors actually be detecting the vibrations from the transmission and thinking it is the engine with a knock?
What is a good and easy way to test this theory that the knock is just the transmission causing it?

Can I just turn off the knock sensors in the PCM and see how it varies in the log file?