Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[V8Cowboy]

Hi,

I've read numerous topics here concerning my problem with returning fault codes 0420/0430. I must say, replacing the diagnostic lambdas behind the main cat was absolutely new to me, very innovative.  It seems that there are far more 3.2 or 2.6 Omegas in the UK than here, so you should know it!

But I don't think this would solve my problem. To the history (I'll try to keep it short):

Got the fault codes again and again on my 2001 Omega 3.2, 180.000 km, so I checked Lambdas with an ODB-Scanner, even crawled under the car and measured the cables according to Opel testing instructions. Front lambdas worked, mixture cycled between lean and rich. Post cat lambdas cycled only when reving and got stable after a while when idle, I think that's how it should be. But always indicating lean mixture, at least the OBD-Scanner told me this!

So I took the car to our automobile club, the guy did an emission test: Everything fine, but when idling, lambda was about 1.150. At app. 3000 rpm everything's fine.
He read the lambda measuring blocks... seemed fine
He checked the lambda heating... worked
He checked the exhaust visually... seemed ok
Coolant temp... ok
MAF voltage... did hardly reach 3 V @ 6000 rpm, this seemed to be too low for the guy (MAF is about 1/2 year old and original Bosch).

So I think it could be a) the MAF, b)still a leak in the exhaust, or c) the ECU. Any other suggestions? 

ECU because I suppose I "insulted" the ECU when I once connected the OBD-Scanner. Since then, the fault lamps do rarely indicate a fault, although there are some stored in the ECU. And when turning ignition on, after a while the throttle motor clicks and chatters for a few seconds... which does not sound healthy. But maybe I'm just hyper sensitive. 

Sooo... Does anyone have some hints or a straight tip? I really would appreciate it. 

[Kevin Wood]

The lambda value at idle does seem a little high. Are you sure it was in closed loop at the time? They can slip into open loop at idle if the sensors cool off. Did you check lambda at a fast idle?

Everything else sounds fine to me. Very unlikely that you have damaged the ECU IMHO. They all do a calibraton of the throttle servo at startup and this is what the chattering noise is. Also, not all fault codes will light the EML.

I would say you've got the usual post-cat lambda issues, tbh.

Have a look at what the fuel trim values are doing to see if you've got a metering or front Lambda sensor problem.

Kevin

[V8Cowboy]

Good morning Kevin,

Lambda values at idle are definitely too high, I'm afraid I won't pass the test next time when I have to go to the annual ... MOT? I don't know if that's correct 

Well, here are all of the exact emission values:

RPM: 650 2790
CO: 0.00% 0.01%
HC: 13 ppm 19 ppm
CO2: 13.3% 14.4%
O2: 2.17% 0.62%
Lambda: 1.115 1.028

Ok, Labda was lower than I remembered  .

The long term fuel trim is -9% and -6% (I know for sure), short term hopping between -4 and +2 (I think I remember right). These long term values resulted within half an hour, because I had to recharge the battery, so the ECU should have been resetted after 1 day without juice.

I thought the post cat lambda issue is only of significance when the emissions are 100% ok, maybe I understood something wrong?

I can't tell you, if it was in closed loop while the test was done, but I know that my scanner shows a closed loop when the engine is warm, except when reving up heavily or at an overrun fuel cut-off.

About the ECU: This calibration sounds very odd. I once had a closer look at the throttle when this happend. It seems, that the motor does not find the "zero" position, it moves the throttle backwards, towards the engine. Or put it that way when you imagine a section view: 0 is closed throttle, 7 is idle throttle, -4 is where my throttle goes during calibration. I know that's confusing, but I can't describe it better  :-/

At least disconnecting the MAF should result in activating a warning light, shouldn't it? Well, it doesn't in my car.

[Kevin Wood]

In the UK, the Lambda reading is only part of the emissions test at a fast idle, not natural idle however, your fast idle Lambda looks a little high, although not as bad. I'm surprised it's idling OK at 1.115 as that's quite lean.

Fuel trims aren't too bad. I would expect them to be railed at 20% or more if the lean running was due to a MAF problem, air leak, etc. for example. Maybe the Lambda sensor(s) are a bit tired?

I've never actually looked at the throttles when they don this. I'm normally holding the key.

If there was a problem with the throttle servo it would raise a fault code and be in limp-home mode.

I would indeed expect disconnecting the MAF to light the EML. I assume the light works?

Kevin

[V8Cowboy]

Idling is really ok, no problem. That's why I think there might be a leak somewhere after the front lambdas, as they do work between appr. 50 and 800mV. With higher rpm the effect of the leak is minimized. At least this is my theory. 

Tomorrow I get a new OBD-Scanner which I hope will be able to record and show the values over time. Maybe we get smarter then.

The lights work definitely. After I turn on the ignition I got 4 lights, while one goes out after a few seconds (the little car shaped). Now, this only happens, when there is not a single fault code stored. As soon as this light stays on, I know there's something wrong.

The EML (emission light? - the one with the engine block and the flash) only rarely glows, even with 0430 and 0420 stored. But occasionally it does. Weird, isn't it? So you might understand why I fear that I have damaged the ECU somehow. Before that, this worked fine. And with damaged ECU, this could be an explanation for the rest of my problems... 

[Kevin Wood]

OK. A leak after the front Lambdas would make sense.

The Lambda value measured at the tailpipe is not necessarily what's going on in the engine, of course. It is an estimate of the original fuel mixture deduced from the proportions of the 4 gases measured, based on the assumption that they were all present in the cylinder! Add some air after the combustion and all bets are off.

In fact, given that the CO and HC values are fine, and that the idle is OK I would say that the engine is actually burning a good mixture. The cat(s) are clearly working OK, and they would be saturated with O2 if it were really that lean.

So, a leak after the cats perhaps?

I have had scenarios where an 0430 or 0420 has been stored but not lit the light. Maybe it has to persist for longer to light the light than to store a code? However, something like a sensor open circuit should light the light (not the same light though!). You should have 2 lights driven from the engine management ECU - one with a picture of an engine (emissions light) and one with a picture of a car and a spanner (which indicates non-emissions faults).

Kevin

[V8Cowboy]

Hi,

sorry, the company I'm working in is currently in the budget phase and I had absolutely no time to deal in detail with my problem. I got the new scanner and could record the engine data, but I have to visualize it yet. Can I attach an *.exe file by the way?

But I realised, that the fault codes are stored as "non present"... so you might be right and it needs a constant condition to light the light. 90% of the 0420/0430 cases do not light anything in my dashboard. But both lights work well.

[V8Cowboy]

Hi Kevin,

I'm sorry it took such a long time for a reply, but the budget phase was long and afterwards I was on vacation too, meanwhile the family car had to be fixed as well. But I could'nt really get on with my stinky little problem.

What I managed to do until now is an *.xls file, where I can load my livedata into and which visualizes it. So what I found out, is..... well actually everything seems fine except the post cat lambdas indicating a lean fuel mixture, while at the same time (this is another measurand btw) the mV ranges of the same sensors seem ok. No idea of what's going on or why this is as it is... But it would be great if someone could have a look at these files, maybe 4 eyes see more than 2. 

Once I also saw the short term fuel trim was about +-17%, just to mention it.

The fault codes come within 2 weeks after deleting them. When checking them with the scanner they are stored as non existent or more seldom as intermittent (one bank).

Thank you for your help and patience anyway. 

[Kevin Wood]

PM Sent.

Kevin

[V8Cowboy]

Hi,

after a long time I felt that I must inform you about the results. Well the sensors have been removed behind the cat and the temperature sensor has been replaced as well. Since this (well already 2-3 months), no FC has appeared again.

But still, lambda is at about 1.03 when idling, but it was right enough to pass the test. Maybe the aging cat is responsible for the rest, I don't know. Thanks for the tip, it seems that it helped!