W124 500E lean mixture

[GeorgiAtanasov]

Hello,
I recently found this forum and it seems to be really helpful, so I hope someone might give me some directions to go with my issues.
The car is W124 500E 1992`. It came to me with very rough idle and a tendency to stall after a cold start. First thing to do was to read fault codes and there were several important once - CPS, KPS and O2 sensor.

The CPS was replaced, because it literally fell apart when I tried to unplug it.
The KPS was replaced some time ago and the client insists that it was a good used one. The resistanse of it is a bit high ~1100Ohms, so I still suspect it.
The EZL was also replaced with a used one from W140 and this is also something that bothers me as I am pretty sure that some of the early m119`s does not have magnets on the flywheel and this EZL might have come from a later one with magnets - hence the KPS DTC.

The ignition cables, caps and plugs are brand new.
The insulators and the rotors are old.
I checked the fuel pressure and it was steady at 3.6bar during starting and idling.

I scoped the ignition waveform and it suggested pretty strong ignition(~2.1ms burning time), but lean mixture(spike at the end of the burning), so I decided to replace the O2 sensor and look for confirmation + had some hope the self adaptation would correct the issue. Replaced it with a compatible Bosch sensor and it confirmed the scope trace - the mixture is way off at idle and part-load. Next thing was to replace all vacuum tubes and crankcase ventilation. Had a bit of an improvement, but the ON/OFF ration still went to 100% once the O2 sensor becomes operational.

The MAF showed 20/h at idle which seems fine, but the mixture is still lean - at least on some cylinders. I am saying "On some cylinders", because despite of the lean overall mixture, it still smells like unburned dinosaurs from 10m away.



Here comes even the more interesting part:
At the previous shop the battery was flatted and they did not have cables to jumpstart it. What did they do - they putted a charged battery in the car, started it and then replaced it with the old one while the engine was running. We have some brilliant rednecks over here...

Now the fuel pump is permanently ON when the ignition is on. And this seems to be the required behavior from the LH module, because in the Actual values in HHT it shows that the FP is ON. There are no fault codes in the LH module except for the secondary air pump at the moment, but it does not want to enable the long term self adaptation. When I try to start the self adaptation manually, the HHT says "program status < 32" and drops the process. I checked the program status and it was 26, but I have absolutely no idea what this means.

I recapped the LH module, but nothing changed. I am really interested in this "program status" meaning, because i suspect it indicates some kind of an internal error.

At the moment I replaced the MAF sensor with one that I have adapted from m112 with a "convertor" electronics and I adjusted the transfer curve so that at idle the ON/OFF ratio hovers ~50-60%. The cold starts are pretty much purfect and the idle is good enough to place a coin op right on top of the valve cover(literally did this for demonstration), but I am still not happy with this solution. The car still misses every so slightly + it smells like unburned fuel even when warm. Enrichening the whole engine just masked the issue. I just wanted to prove that it is most likely a fueling issue and not an ignition one.

Another interesting fact is that the car lacks power in the mid-range. When I try to blip the throttle and look at the actual values in HHT, the cam adjusters are not enabled. This + the pump stayng permanently on leads me to the thinking of that this "program status" indicates some kind of an emergency running mode.

I am going to take one LH module from W140 with the clicking FPR issue and repair it for free in exchange to the customer approval to use it for test, but I still need to know what does this program status mean and why it prevents the self adaptation.

Any help will be precious!

 

[gsxr]

The CPS was replaced, because it literally fell apart when I tried to unplug it.
The KPS was replaced some time ago and the client insists that it was a good used one. The resistanse of it is a bit high ~1100Ohms, so I still suspect it.
The EZL was also replaced with a used one from W140 and this is also something that bothers me as I am pretty sure that some of the early m119`s does not have magnets on the flywheel and this EZL might have come from a later one with magnets - hence the KPS DTC.

All M119's have magnets on the flywheel. If the magnets are missing or not recognized, there will be a fault code on the EZL.

The ignition cables, caps and plugs are brand new.
The insulators and the rotors are old.
I checked the fuel pressure and it was steady at 3.6bar during starting and idling.

Does the engine EVER run smoothly, after warmup, etc? If not, triple-check all the ignition components. It's not unheard of to have a defective cap, plug, or cable new out of the box.

I scoped the ignition waveform and it suggested pretty strong ignition(~2.1ms burning time), but lean mixture(spike at the end of the burning), so I decided to replace the O2 sensor and look for confirmation + had some hope the self adaptation would correct the issue. Replaced it with a compatible Bosch sensor and it confirmed the scope trace - the mixture is way off at idle and part-load. Next thing was to replace all vacuum tubes and crankcase ventilation. Had a bit of an improvement, but the ON/OFF ration still went to 100% once the O2 sensor becomes operational.

Have you smoke-tested the intake tract to confirm all vacuum leaks are eliminated? Also verify the purge "MOT" valve near the EZL is not stuck open (common problem, causes large vacuum leak).

The MAF showed 20/h at idle which seems fine, but the mixture is still lean - at least on some cylinders.

20 kg/hr is normal at idle, it will be higher after a cold start, lower when at operating temp.

I am saying "On some cylinders", because despite of the lean overall mixture, it still smells like unburned dinosaurs from 10m away.

Are you sure this isn't a RICH mixture? Smelling of fuel out the exhaust is normal when the mixture is rich and/or misfires are preventing the fuel from burning.

Here comes even the more interesting part:
At the previous shop the battery was flatted and they did not have cables to jumpstart it. What did they do - they putted a charged battery in the car, started it and then replaced it with the old one while the engine was running. We have some brilliant rednecks over here...

Yeah, that was a Bad Idea.

Now the fuel pump is permanently ON when the ignition is on. And this seems to be the required behavior from the LH module, because in the Actual values in HHT it shows that the FP is ON.

This is NOT normal. When the ignition switch is turned on (but engine not started), the fuel pumps should run for 2 seconds only, then turn off. The pumps should run during cranking and after starting. Ignore the HHT indicator of fuel pump = ON.

There are no fault codes in the LH module except for the secondary air pump at the moment, but it does not want to enable the long term self adaptation. When I try to start the self adaptation manually, the HHT says "program status < 32" and drops the process. I checked the program status and it was 26, but I have absolutely no idea what this means.

Never heard of this before. Manual adaptation requires driving the car while doing what the HHT says on the screen. This is not your problem, just reset adaptation (will read 1.0004) and leave it there. Once you fix the root cause, adaptation will occur automatically after driving a couple hundred km's.

I recapped the LH module, but nothing changed. I am really interested in this "program status" meaning, because i suspect it indicates some kind of an internal error.

If you replaced all 6 capacitors, and the fuel pumps still run at all times, there's a decent chance the LH module is fried. However, also check the 38-pin diagnostic port as shown in this thread. Corrosion in the sockets can cause very bizarre problems. This is

At the moment I replaced the MAF sensor with one that I have adapted from m112 with a "convertor" electronics and I adjusted the transfer curve so that at idle the ON/OFF ratio hovers ~50-60%. The cold starts are pretty much purfect and the idle is good enough to place a coin op right on top of the valve cover(literally did this for demonstration), but I am still not happy with this solution. The car still misses every so slightly + it smells like unburned fuel even when warm. Enrichening the whole engine just masked the issue. I just wanted to prove that it is most likely a fueling issue and not an ignition one.

You are correct. This is NOT a good solution. Reinstall the original, correct M119 MAF and continue to investigate the root cause.

Another interesting fact is that the car lacks power in the mid-range. When I try to blip the throttle and look at the actual values in HHT, the cam adjusters are not enabled. This + the pump stayng permanently on leads me to the thinking of that this "program status" indicates some kind of an emergency running mode.

Lack of midrange power does sound like the cam advance is not working. This also points to a possible defective LH module.

I am going to take one LH module from W140 with the clicking FPR issue and repair it for free in exchange to the customer approval to use it for test, but I still need to know what does this program status mean and why it prevents the self adaptation.

No idea about the program status. Swap in a good LH module for testing. It can be from any M119, including 4.2L.

Any help will be precious!

When the engine is running rough after a cold start, observe the ignition voltage from the EZL on HHT live data. All 8 cylinders should be similar voltage, probably low/mid 30's. If some cylinders are much higher, like 60+, those cylinders are misfiring. This indicates a secondary ignition problem. If that is the case, replace the offending ignition parts. Again, you may have defective new parts installed, do NOT assume they are good because they are new. Also verify the plugs are non-resistor Bosch F8DC4 or equivalent, not fine-wire platinum/yttrium/unobtainum/whatever. Increasing the plug gap to 1.0mm is also recommended.

In addition to the above, post the part number on the EZL and other modules. Sometimes, incorrect/incompatible modules can cause strange problems.

 

[GeorgiAtanasov]

Hi, Thanks for the fast response!

All M119's have magnets on the flywheel. If the magnets are missing or not recognized, there will be a fault code on the EZL.

I did not know that. The early m104 definately have no magnets. That`s why they have additional KPS on the demper wheel. Will definately check it.

Does the engine EVER run smoothly, after warmup, etc? If not, triple-check all the ignition components. It's not unheard of to have a defective cap, plug, or cable new out of the box.

Tbh I am not sure. The customer said that he had the car for 15 years and this is a recent issue, but I cannot bet. There are too many issues to happen at one point of time.
I am aware of the issue with the new parts, but these look good. The waveform also looks good on all cylinders except for the lean mixture pattern.

Have you smoke-tested the intake tract to confirm all vacuum leaks are eliminated? Also verify the purge "MOT" valve near the EZL is not stuck open (common problem, causes large vacuum leak).

No, this is one of the next steps.

Are you sure this isn't a RICH mixture? Smelling of fuel out the exhaust is normal when the mixture is rich and/or misfires are preventing the fuel from burning.

100% sure. The ignition waveform has the characteristic spike at the end of the burning time + the ON/off ratio goes to 100% and the O2 reading is <450mV. The combination of lean mixture + strong smell suggests vacuum leaks near specific cylinders or fuel injectors that are not properly working(or not properly controlled). The overall mixture might be lean due to couple of cylinders running lean, but when compensated, the other cylinders run rich and hence the smell.
Rich mixture would hardly ever result in missfire. It has to be absurdly rich to flood the spark plugs.

Never heard of this before. Manual adaptation requires driving the car while doing what the HHT says on the screen. This is not your problem, just reset adaptation (will read 1.0004) and leave it there. Once you fix the root cause, adaptation will occur automatically after driving a couple hundred km's.

The adaptations were at 1.0004 when the car came to me, but I reset them anyway and drove it around. It does not enable the self adaptation at all. They stay the same.

When the engine is running rough after a cold start, observe the ignition voltage from the EZL on HHT live data. All 8 cylinders should be similar voltage, probably low/mid 30's. If some cylinders are much higher, like 60+, those cylinders are misfiring. This indicates a secondary ignition problem. If that is the case, replace the offending ignition parts. Again, you may have defective new parts installed, do NOT assume they are good because they are new. Also verify the plugs are non-resistor Bosch F8DC4 or equivalent, not fine-wire platinum/yttrium/unobtainum/whatever. Increasing the plug gap to 1.0mm is also recommended.

In addition to the above, post the part number on the EZL and other modules. Sometimes, incorrect/incompatible modules can cause strange problems.

The voltage stays ~39-41V on all cylinders and the burning time is ~2-2.3ms. Both values seem more than good, considering the lean mixture condition on at least couple of cylinders.
All cables have 2kOhms resistance and do not "zap" when touched.
I have not checked the resistance of the caps, but high resistance there shall be noticeable on the waveform.
The spark plugs are as per specification, no modern platinum/irridium/etc replacements.
I will take a picture of the modules next week. The compatibility is another suspect for the "program status" issue.

 

[gsxr]

I did not know that. The early m104 definately have no magnets. That`s why they have additional KPS on the demper wheel. Will definately check it.

Early M119 also have a crank sensor at the front / harmonic balancer, but this is only for diagnostic tools - it has no connection to the powertrain computers. The front sensor goes to the 9-pin (?) diagnostic port near the EZL, which was deleted for facelift models (E500).

Tbh I am not sure. The customer said that he had the car for 15 years and this is a recent issue, but I cannot bet. There are too many issues to happen at one point of time.
I am aware of the issue with the new parts, but these look good. The waveform also looks good on all cylinders except for the lean mixture pattern.

So currently, it never runs correctly, even at operating temp?

100% sure. The ignition waveform has the characteristic spike at the end of the burning time + the ON/off ratio goes to 100% and the O2 reading is <450mV. The combination of lean mixture + strong smell suggests vacuum leaks near specific cylinders or fuel injectors that are not properly working(or not properly controlled). The overall mixture might be lean due to couple of cylinders running lean, but when compensated, the other cylinders run rich and hence the smell.
Rich mixture would hardly ever result in missfire. It has to be absurdly rich to flood the spark plugs.

I'm still suspicious that your LH module is toast and miiiight be the cause of these problems.

The adaptations were at 1.0004 when the car came to me, but I reset them anyway and drove it around. It does not enable the self adaptation at all. They stay the same.

Adaptation can take a long time to adjust, possibly over several days including some cold starts. The time seems to vary depending on the software revision of the LH module. Some seem to adapt more quickly than others. I forget if early or late adapts faster.

The voltage stays ~39-41V on all cylinders and the burning time is ~2-2.3ms. Both values seem more than good, considering the lean mixture condition on at least couple of cylinders.

OK - similar voltage on all 8 cylinders at least rules out a misfire, but the voltage is higher than normal. Should be 30-34v at operating temp. Your burn time is slightly high but not bad, normal is around 1.8-2.0ms. See video below. Not sure what's going on but it could be related to whatever the root cause is.

All cables have 2kOhms resistance and do not "zap" when touched.
I have not checked the resistance of the caps, but high resistance there shall be noticeable on the waveform.
The spark plugs are as per specification, no modern platinum/irridium/etc replacements.

That sounds good. Caps and rotors should each have 1kOhm. Wouldn't hurt to check the spark plug gap but that will not cause the issue you describe.

I will take a picture of the modules next week. The compatibility is another suspect for the "program status" issue.

I would definitely obtain another LH module for testing, to start with, along with confirming the EZL part number.

 

[GeorgiAtanasov]

Okay, there is some movement in the right direction.
The lean mixture condition was fixed by replacing the LH module with one from a 4.2 W140. The self adaptation passed in just couple of minutes for lower part load and idle speed. 1.5kg/h idle speed correction and 1.023 lower part load correction seem reasonable. Far from ideal, but well within the limits. The car started just fine this morning. Most of the excessive unburned gas smell was also gone, but I was still not fully happy. The relatively high ignition voltages and the light misfires kept bothering me. Decided to replace one of the rotors and see if it fixes the voltage before jumping in with another 70 euro for the other one. It was pretty melted and decided it's worth the shot. And it was. The voltage on cylinders 2, 3, 5 and 8 dropped to the much more acceptable ~35V. The exhaust smell improved further. I ordered the other rotor and hopefully this will be the end of this saga for now. The ignition cables also need to be replaced, because at the moment there are 3 types of cables, but this will be done at a later stage.