OWNER a777fan (E420)

[nocfn]

I had a pulsing idle early on about 6 weeks after replacing caps. Turned out that the PS dizzy was loose enough to shake slightly when I grabbed it. I would get a rhythmic Idle surge at traffic lights 500-625 RPM (not specific but you get the gist) and I could also feel it in the pedal at highway cruising speed. I have not had it since and I think in my owner thread there is a video of it.

 

[a777fan]

The T/LLR is now re-capped.

I’m very glad I had the junkyard spare to try out first, as these units are a little tricker to disassemble than the LH modules. There is a rivet on the case in place of a screw that needs to be drilled out, and then inside there is a clamp which presses some transistors (?) into a heat sink on one of the case sides that needs to be gently pried at to remove it without breaking. I bodged both of these on the junkyard unit, but was able to apply my leasons learned to the unit the came with Üter and removed them without issue.

One observation. One of the caps (the largest one in the unit- 220uF/63V) is ‘glued’ to the micro relay thats positioned next to it. Its quite obvious that this was intentional in Üters unit, but the junkyard units adhesive appeared degraded to the point I thought it was cap goo. But it aint!

junkyard unit:


Üters unit:



The large cap was a Phillips unit and appeared in good condition. Given this and that it was attached to the relay i left it alone for now.
I did replace the smaller guys though. No brand or marking on them.

All ready to go back together:

Went for a toot and no oscillations thus far. We’ll see though as the problem reared its head quite intermittently.

 

[a777fan]

Well the idle oscillations are still here... less pronounced and less common, but they still do occur. (likely still in placebo mode with my assessment of 'less pronounced' and 'less common' lol)

The light came on yesterday. I hooked up DAS (which on my machine defaults to HHT-WIN - is that normal?) and looked at the codes. It was 19 - LH Adaptation at limits.

Here is a picture of the adaptations:



Does anyone know how to interpret these numbers? Is 1.1842 indicative of a +~18% correction? IE- The LH computer had to ADD 18% more injector duration under lower partial load than it was expecting to? Is ~18(%?) in fact the upper limit?

Also odd... the Upper Part load and the Idle Speed adjustment are at nominal values. Like COMPLETELY nominal (the values they take on when you reset the adaptations)... I didn't think anything of this until I went into the 'Set adaptations' mode in the HHT, and read the instructions on how to get these values to 'set'. The Lower Part load is pretty unexciting, all it needs is engine speeds between 1750-~2250 and a typical MAF value. The Upper Part load is a bit spicier with the criteria, needing 3500 rpm + and a hefty and specific MAF reading. The Idle adjustment is easy. Car needs to sit and idle with the MAF plugged in.

I went for a drive and attempted to get these adaptations to 'set'. When running the 'set' function the tool will display the test status of all three tests. Its either YES (Completed Successfully), ACTIVE (Adaptation in the process of running), and NO (Criteria for test to run not set).

I first re-set adaptations as instructed. I got the Lower Part load calibration to run just fine. I attempted to run the Upper Load test by skootching car around in B up and down a nice long straight. I got 'ACTIVE' to flash a few times, but it never switched to 'YES'. At first I just thought I wasn't keeping the car in the right rev/throttle range, but after a few runs, it stopped trying to run the test. IE- ACTIVE would not highlight any longer and NO remained 'lit'. The car was also getting quite hot at this point (which was surprising, but perhaps expected? It kicked the aux fans on High, and the temps came right back as usual), so I gave up on that one for today.

I pulled over, threw it in park and attempted to get the Idle correction to run. It had the same behavior as the Upper Load correction. ACTIVE would highlight for a bit, then it would flip back to NO. This repeated a few times, until again, it seemed like the adaptation just 'gave up'.

Now... circling back to the photo. It would appear that for whatever reason these adaptations (upper part and idle) cannot set with this LH module. Does anyone else know what parameters could be preventing these two from not being able to 'set'?

Side note: the Lower Part Load adaptation came back to close to the same value it was with the MIL on:



Scratching my head here.

 

[gerryvz]

Scroll down on this page until you come to the section about this problem.

Seems to be very endemic to the 119.975 motors.....

K6JRF Auto Page

My Mercedes S500 Coupe

www.k6jrf.com

 

[a777fan]

Scroll down on this page until you come to the section about this problem.

Seems to be very endemic to the 119.975 motors.....

K6JRF Auto Page

My Mercedes S500 Coupe

www.k6jrf.com

V. cool....

 

[gsxr]

The CEL light came on yesterday. I hooked up DAS (which on my machine defaults to HHT-WIN - is that normal?) and looked at the codes. It was 19 - LH Adaptation at limits. Here is a picture of the adaptations:

Does anyone know how to interpret these numbers? Is 1.1842 indicative of a +~18% correction? IE- The LH computer had to ADD 18% more injector duration under lower partial load than it was expecting to? Is ~18(%?) in fact the upper limit?

Yes. 1.18xx indicates the LH is adding 18% more fuel / duration. The FSM documentation says adaptation limits of LH-SFI are ±25%. The special "CEL 19" EPROM allows up to 32%, IIRC. However, the special EPROM is intended for cars that are running excessively rich and the LH is leaning them out, i.e. 0.82xx or below. That's the opposite of your problem. K6J's page won't really help with your issue.

Also odd... the Upper Part load and the Idle Speed adjustment are at nominal values. Like COMPLETELY nominal (the values they take on when you reset the adaptations)...

That most likely means they have not yet adapted. It can take quite a while of normal driving for this to occur after they have been reset, which may happen (but not always) when the battery is disconnected, etc. It is very, VERY unusual for any engine/module combination to remain at 1.0004 adaptation / 0.0 idle over hundreds or thousands of miles of driving.

I didn't think anything of this until I went into the 'Set adaptations' mode in the HHT, and read the instructions on how to get these values to 'set'. The Lower Part load is pretty unexciting, all it needs is engine speeds between 1750-~2250 and a typical MAF value. The Upper Part load is a bit spicier with the criteria, needing 3500 rpm + and a hefty and specific MAF reading. The Idle adjustment is easy. Car needs to sit and idle with the MAF plugged in.

I went for a drive and attempted to get these adaptations to 'set'. When running the 'set' function the tool will display the test status of all three tests. Its either YES (Completed Successfully), ACTIVE (Adaptation in the process of running), and NO (Criteria for test to run not set).

I first re-set adaptations as instructed. I got the Lower Part load calibration to run just fine. I attempted to run the Upper Load test by skootching car around in B up and down a nice long straight. I got 'ACTIVE' to flash a few times, but it never switched to 'YES'. At first I just thought I wasn't keeping the car in the right rev/throttle range, but after a few runs, it stopped trying to run the test. IE- ACTIVE would not highlight any longer and NO remained 'lit'. The car was also getting quite hot at this point (which was surprising, but perhaps expected? It kicked the aux fans on High, and the temps came right back as usual), so I gave up on that one for today.

This is difficult because you also need the engine load (MAF readings in kg/hr) must be in the proper range, along with the correct RPM range. The trick here is to stay in a lower gear (2nd gear with 2.24 diff) and ride the brake pedal as needed to boost MAF readings. A flat road with no traffic is a big help.

I pulled over, threw it in park and attempted to get the Idle correction to run. It had the same behavior as the Upper Load correction. ACTIVE would highlight for a bit, then it would flip back to NO. This repeated a few times, until again, it seemed like the adaptation just 'gave up'.

It SHOULD adapt at idle after a few minutes. If not, something is screwy.

Now... circling back to the photo. It would appear that for whatever reason these adaptations (upper part and idle) cannot set with this LH module. Does anyone else know what parameters could be preventing these two from not being able to 'set'? Side note: the Lower Part Load adaptation came back to close to the same value it was with the MIL on: Scratching my head here.

Do you have a '92 LH module installed, or a 93-95 module? Stock EPROM? When you choose option 1 in the LH menu to display control module information, what part number does it show?

For reference, our 1994 E420 with '92 LH module long-term adaptation values are in the 0.92-0.94 range for both upper & lower, with idle ±0.3 or so. Our 1994 E420 with '92 LH has the same 0.92-0.94 but is higher at idle, ~0.8 or so. I would expect yours should be similar, in the low/mid 0.9x range.

For another test - with the engine at operating temp, in park, with AC off... tell us what the MAF readings are at 650rpm idle, then 1000, 2000, and 3000rpm. Ballpark numbers should be 18, 30, 60, and 90 respectively. If your numbers are drastically different, your MAF might be inaccurate, or there's something else weird going on.

 

[a777fan]

Yes. 1.18xx indicates the LH is adding 18% more fuel / duration. The FSM documentation says adaptation limits of LH-SFI are ±25%. The special "CEL 19" EPROM allows up to 32%, IIRC. However, the special EPROM is intended for cars that are running excessively rich and the LH is leaning them out, i.e. 0.82xx or below. That's the opposite of your problem. K6J's page won't really help with your issue.

Excellent. Thanks. That makes sense.

That most likely means they have not yet adapted. It can take quite a while of normal driving for this to occur after they have been reset, which may happen (but not always) when the battery is disconnected, etc. It is very, VERY unusual for any engine/module combination to remain at 1.0004 adaptation / 0.0 idle over hundreds or thousands of miles of driving.



This is difficult because you also need the engine load (MAF readings in kg/hr) must be in the proper range, along with the correct RPM range. The trick here is to stay in a lower gear (2nd gear with 2.24 diff) and ride the brake pedal as needed to boost MAF readings. A flat road with no traffic is a big help.

Agreed!

It SHOULD adapt at idle after a few minutes. If not, something is screwy.

Yes... also my thoughts. Like... what signal is it also expecting that it is NOT getting?

Do you have a '92 LH module installed, or a 93-95 module? Stock EPROM? When you choose option 1 in the LH menu to display control module information, what part number does it show?

For reference, our 1994 E420 with '92 LH module long-term adaptation values are in the 0.92-0.94 range for both upper & lower, with idle ±0.3 or so. Our 1994 E420 with '92 LH has the same 0.92-0.94 but is higher at idle, ~0.8 or so. I would expect yours should be similar, in the low/mid 0.9x range.

For another test - with the engine at operating temp, in park, with AC off... tell us what the MAF readings are at 650rpm idle, then 1000, 2000, and 3000rpm. Ballpark numbers should be 18, 30, 60, and 90 respectively. If your numbers are drastically different, your MAF might be inaccurate, or there's something else weird going on.

Module is a 92 LH with WOT. P/N: 014 545 15 32

I do have the original the car came with that I can try out as well.

Thanks for those MAF numbers. I was wondering what they should be, as obviously the readings from that play a big part here...

 

[Joncf]

Nice night for a walk

 

[emerydc8]

Well the idle oscillations are still here... less pronounced and less common, but they still do occur. (likely still in placebo mode with my assessment of 'less pronounced' and 'less common' lol)

Speaking of the placebo effect, I have been chasing an intermittent pulsing idle on my 95 E420 (non-ASR) for a number of years. I've replaced pretty much everything I could think of that might affect it. It happens more in colder temperatures (70F or less), but it's also shown up intermittently this summer for no apparent reason. When I take it out of D into N the pulsing immediately stops.

A few weeks ago I replaced a bad voltage regulator that was causing, inter alia, a hunting idle. Maybe I'm getting the placebo effect here, but since I replaced the voltage regulator I haven't seen any pulsing while in D. I'm not saying that a bad VR was causing my problem, but I've driven it a few hundred miles now and I haven't seen any pulsing. If you have a spare VR, you may want to swap it out and see. In my case, a bad VR has been proven to affect engine idle, albeit the idle went up to 1200 RPM and pulsed there. For $11 from AHAZ, it might be worth it to install a new VR. Even if it doesn't solve your idle issue, if you haven't replaced your VR or if you did like me and installed a crappy Meyle VR from China, it's not wasted time.

Of course, now that I posted this, the pulsing problem will return the next time I drive it!

 

[gerryvz]

V. cool....

I'm not sure why I was awarded a CY award for providing legitimate documented information that pertains to this DTC, but I guess I'll take it. It's been a while since I got a Yoink, and keeping up appearances is a good thing.

K6JRF also offers some advice about ensuring the integrity of your vacuum system, as well as fuel pressure. I know you did a top-end refresh a few years ago, but a visual check for vacuum system integrity would be a good thing. Also a quick check on the fuel pressure regulator (and also to make sure that the internal diaphragm is not ruptured).

A question -- is your MAF working correctly, seeing as it is a key input to the adaptation? I wonder if a different MAF would help. It may well be your problem here.

Another question -- what is the status of your oxygen sensor? Has that been replaced? It is another key input into the system.

My belief is that one of these two items is your issue. You have a bad input that is skewing everything upward.

 

[a777fan]

I'm not sure why I was awarded a CY award for providing legitimate documented information that pertains to this DTC, but I guess I'll take it. It's been a while since I got a Yoink, and keeping up appearances is a good thing.

K6JRF also offers some advice about ensuring the integrity of your vacuum system, as well as fuel pressure. I know you did a top-end refresh a few years ago, but a visual check for vacuum system integrity would be a good thing. Also a quick check on the fuel pressure regulator (and also to make sure that the internal diaphragm is not ruptured).

A question -- is your MAF working correctly, seeing as it is a key input to the adaptation? I wonder if a different MAF would help. It may well be your problem here.

Another question -- what is the status of your oxygen sensor? Has that been replaced? It is another key input into the system.

My belief is that one of these two items is your issue. You have a bad input that is skewing everything upward.

Oh! Sorry. I figured ANY reference to a site that sold KOOL harnesses HAD to be a yoink! Lol.

I have a hard time believing that MB simply revised the sw to open up the adaptation limits as an actual solution to the described problem. Something in the system was adding too much fuel… what was it?! In modern obd2 systems, anything over 10% indicates an issue. I can’t believe these LH systems (while older tech) would be THAT much less accurate. If in good nick and fully operational.

I will have to smoke test the car. In fact, since I did not have my smoke tester when I did my top end rebuild, I don’t think I have ever performed one on Üter.

I ‘parts cannoned’ the O2 with a gen-u-ine MB part a year or two ago. I doubt it has more than 10k on it. The MAF however is original. I’ll need to check the numbers against what @gsxr posted above. I was in the process of doing so when my laptop battery died.

 

[a777fan]

Speaking of the placebo effect, I have been chasing an intermittent pulsing idle on my 95 E420 (non-ASR) for a number of years. I've replaced pretty much everything I could think of that might affect it. It happens more in colder temperatures (70F or less), but it's also shown up intermittently this summer for no apparent reason. When I take it out of D into N the pulsing immediately stops.

A few weeks ago I replaced a bad voltage regulator that was causing, inter alia, a hunting idle. Maybe I'm getting the placebo effect here, but since I replaced the voltage regulator I haven't seen any pulsing while in D. I'm not saying that a bad VR was causing my problem, but I've driven it a few hundred miles now and I haven't seen any pulsing. If you have a spare VR, you may want to swap it out and see. In my case, a bad VR has been proven to affect engine idle, albeit the idle went up to 1200 RPM and pulsed there. For $11 from AHAZ, it might be worth it to install a new VR. Even if it doesn't solve your idle issue, if you haven't replaced your VR or if you did like me and installed a crappy Meyle VR from China, it's not wasted time.

Of course, now that I posted this, the pulsing problem will return the next time I drive it!

Your pulsing issue sounds exactly like mine. Pop it out of drive and it goes away.

I also believe my VR is original (i at least have not done anything with it).

Random ?: is your NSS original?

 

[gerryvz]

There was actually a Technical Service Bulletin issued by MB about this very issue (adaptation limits being exceeded). I'm sure the @gsxr has it somewhere on his web site, but it is difficult to find as he has no site-search capability via Google Site Search, Elasticsearch or other usable search engine. Regular Google doesn't work that great for a specific site-interior search.

 

[gsxr]

I have a hard time believing that MB simply revised the sw to open up the adaptation limits as an actual solution to the described problem. Something in the system was adding too much fuel… what was it?! In modern obd2 systems, anything over 10% indicates an issue. I can’t believe these LH systems (while older tech) would be THAT much less accurate. If in good nick and fully operational.

The "Code 19" issue only occurred in unique climate conditions and was generally isolated to certain areas of Florida, IIRC. Only 4.2L engines were affected. @Klink told me the story years ago but I forgot to take notes, and have since forgotten the details.

The TSB specifically calls out the 'Jacksonville Florida Region" and issues related to high temperature with high humidity, along with looking for adaptation values lower than 0.92 for upper or lower part load. Jacksonville 4.2L M119.97x vehicles with CEL #19 would get the adaptation checked and module replaced, without additional testing.

Vehicles in other regions (outside Jacksonville) required testing for vacuum leaks or other issues first, before installing the module with special EPROM that expanded adaptation limits to ±32%.

 

[gerryvz]

Just as an FYI on this -- the purge (MOT) valve on the driver's side inner fender can also be a cause for this code #19. I found that my own purge valve had failed when I did my Top-End Refresh last year (I replaced it with a new one), and the spare I had on hand was also failed. As I've said here on the forum, it is likely that 90% of people on this forum are driving around with DEAD purge valves, and don't even know it. Nearly all of the time, a failed purge valve DOES NOT throw a code / trigger a CEL. These valves are overlooked by 98% of owners.

You may also want to swap in a spare LH module to see if that solves the issue. I wonder if the later modules (particularly facelift cars) had the modified software installed from the factory? Certainly the factory replacements did.

Be careful with those fan shroud bolts, too......

 

[gerryvz]

For future reference, attached is the Mercedes-Benz TSB relating to the Jacksonville, FL area computer replacement due to adaptation limits being reached, and the associated DTC #19. I was able to find it in my TSB library.

 

[gsxr]

Yup - I'd swap in a different LH module for grins.

If the purge/MOT valve is stuck open, that will cause problems as it's essentially a big vacuum leak. You can temporarily block the purge/MOT valve plumbing if needed, the car will run normally, it just won't be huffing gasoline vapors out of the carbon canister while disabled.

 

[gerryvz]

Yup - I'd swap in a different LH module for grins.

If the purge/MOT valve is stuck open, that will cause problems as it's essentially a big vacuum leak. You can temporarily block the purge/MOT valve plumbing if needed, the car will run normally, it just won't be huffing gasoline vapors out of the carbon canister while disabled.

The EGR valve can also cause problems, and not throw a code (although some EGR faults DO throw codes), that can cause a code #19. However IMHO, EGR valves are not a common failure, and I would recommend pursuing the aforementioned checks first:

1) Check for vacuum leaks
2) Check for proper MAF values
3) Check for proper O2 sensor operation
4) Check for proper MOT / purge valve operation
5) Swap out LH computer
6) Check for proper EGR valve operation

 

[emerydc8]

Your pulsing issue sounds exactly like mine. Pop it out of drive and it goes away.

I also believe my VR is original (i at least have not done anything with it).

Random ?: is your NSS original?

I've replaced the NSS more times than I can remember.

 

[emerydc8]

If the purge/MOT valve is stuck open, that will cause problems as it's essentially a big vacuum leak. You can temporarily block the purge/MOT valve plumbing if needed, the car will run normally, it just won't be huffing gasoline vapors out of the carbon canister while disabled.

How can you check this valve, Dave? If you hear it ticking, does that mean it's working properly?

 

[gerryvz]

How can you check this valve, Dave? If you hear it ticking, does that mean it's working properly?

The MOT valve is checked with a 9-volt battery, across the two terminals of the valve. If you hear a click, then you know it is good. I detailed this in my Top-End Refresh thread, last summer.

 

[emerydc8]

I detailed this in my Top-End Refresh thread, last summer.

Missed that part. Thanks Gerry.

 

[gerryvz]

The MOT valve is checked with a 9-volt battery, across the two terminals of the valve. If you hear a click, then you know it is good. I detailed this in my Top-End Refresh thread, last summer.

Missed that part. Thanks Gerry.

Detailed here.

 

[emerydc8]

Detailed here.

Not to hijack this thread, but I just checked both MOTs with a 9v battery. Neither is clicking. Is this a factory only part?

 

[gsxr]

How can you check this valve, Dave? If you hear it ticking, does that mean it's working properly?

Jon, if it is ticking, it is probably OK. HOWEVER! I have recently learned that for a full test, you need to remove the valve from the car and blow through it. When closed, it should be 100% closed and allow zero airflow. When open, you should be able to blow through it easily. Clicking it on/off with a 9V battery, it should alternate between fully open and completely closed, which is obvious when blowing through while clicking.

Most bad ones either won't click at all, or will allow some airflow when closed... both of thse failure modes require replacement.

Not to hijack this thread, but I just checked both MOTs with a 9v battery. Neither is clicking. Is this a factory only part?

If they don't click at all, they're probably dead. AFAIK it's an MB-only item. Try the airflow test described above. If they are completely closed/blocked at all times and don't allow airflow... replacement is optional. I've found it's a pretty common failure mode to have them stuck partly-open which results in a permanent, small vacuum leak.

I also recently discovered that if completely closed, there is no detrimental effects other than the carbon canister never has the vapors ingested by the engine, they would be vented to the atmosphere.

OH, and also, make sure there is no trace of carbon particulate at the defective valve. Bang it against a clean surface, if carbon bits fall out, that means a replacement valve may fail in short order, plugged up with bits of carbon. Don't ask how I know. If this happens, I believe the carbon canister needs replacement... or the valve could be removed/disabled and lines plugged so there is no vacuum leak.

 

[emerydc8]

Thanks, Dave. I'll try taking them out and inspecting as you suggested. You know the part number off hand?

 

[Jlaa]

Thanks, Dave. I'll try taking them out and inspecting as you suggested. You know the part number off hand?

P/N here. RESTO PROJECT: M119 / W124 / E500 Engine Top-End Refresh | Member Roadtrips and Project Journals

 

[emerydc8]

P/N here. RESTO PROJECT: M119 / W124 / E500 Engine Top-End Refresh | Member Roadtrips and Project Journals

I know -- I'm lazy: 000 470 21 93

Valve - (000-470-21-93) | Get Mercedes Parts

Valve 000-470-21-93

www.getmercedesparts.com

But for $93, I might just leave it if it's not leaking.

 

[gerryvz]

A classic Steve Brotherton article that talks about adaptation and the vaunted code #19.

Mercedes-Benz Engine Diagnostics

I’ve always loved the puzzle. Check Engine warning lights have become an everyday chore for me over the last five years. It’s a challenge that extends my interest and creativity in the auto service profession. Related Articles - Autel Adds EV Diagnostics and Maintenance to Roster of Classes -...

www.brakeandfrontend.com

Part of this great collection documented on this thread.

Collection of good background articles | Troubleshooting and Diagnostics

Over the past years, some of you may have come across one or more of this collection of articles that was written and published by Steve Brotherton, the owner and proprietor of Continental Imports, an indy Benz repair shop in Florida. These articles have been posted for years on the firm's web...

www.500eboard.co

 

[gsxr]

A classic Steve Brotherton article that talks about adaptation and the vaunted code #19.

That article has some good info. However it fails to note the TSB was only for 4.2L engines, and MB never released a program with expanded adaptation range for any other engines besides the 4.2L M119.

Also, the article implies that under normal driving, the engine RPM and load must be within specific parameters before adapting, same as during the manual/forced adaptation procedure. I'm not sure this is correct. Although he mentioned an M120 that require two weeks, the number of miles driven was not reported. I've seen M119 5.0L modules adapt in <40 miles, or one round trip to work for me. In other cases it can take longer, depending on the engine and how much adaptation is needed. I've also done the forced adaptation and found the values would then change noticeably over the next few hundred miles, indicating the forced quick/adaptation was a "rough" adaptation, which was adjusted over time.

As I mentioned previously, a dyno is not needed to meet the require engine load during forced adaptation. Simply ride the brake and adjust pedal pressure to simulate the required load. It only needs 5 seconds or so of this, and if done at lower speeds, it doesn't overheat the brakes. I've done this many times.

 

[a777fan]

Still need to tackle the fuel trim issue… but my new zoom tubes came in!



old zoomies for reference

 

[a777fan]

Some additional research tonight. I swapped in the original LH module and reset the adaptations. I then ran the quick 'adaptation' program and got the Lower Part and Upper part adaptations to set. Oddly, I could not get the idle adaptation to set. Results weren't identical to the WOT module, but similar:



So this would seem to absolve the WOT LH module. The positive values would indicate that the O2 is seeing a lean mixture, and the LH compy is asking to add more fuel to compensate. As Gerry and Dave pointed out above, this could be either a vacuum leak, letting in un-metered (measured) air into the system, or a failing MAF under reporting air volume. I did a quick driveway MAF test tonight and came up with the following:



The red dots are the target readings provided above by @gsxr . You can see my collected data trends pretty well with what Dave was indicating is proper. I took multiple readings by having my phone camera ready, and taking multiple shots as I worked the accelerator to achieve roughly my target rpm. Worked out pretty well, and I feel better having several readings around the points of interest rather than just one. Short of the one data point at 3500 rpm (which I am fine chalking up as an errant piece of data), the MAF seems to be reporting accurately.

Purge valve is also happily clicking away. With the fuel trims being positive however, its certainly not as if the purge is stuck OPEN. That would result in a negative fuel trim. Alternatively, if the purge line was blocked somehow, would the absence of fuel vapor be such that a 12% correction would be needed? I don't know, but my gut says it's doubtful.

I also removed the air box to listen for any massive/obvious vacuum leaks. None that I could find, but I have a smoke tester coming tomorrow to find out for sure.

So what does that leave?

• Vacuum leaks (will test with machine)
• O2 sensor operation - I have not scoped the waveform, but HHT-WIN shows lean/rich oscillations as expected, and its less than a 15k old OE MB sensor.
• EGR Valve operation - not entirely sure how to check this yet, but will figure it out
• Fuel Injectors - These are new 'rebuilt' units. Perhaps they (or one of them) is a bit short on fuel delivery?

More to come!

 

[Kyiv]

Still need to tackle the fuel trim issue… but my new zoom tubes came in!

View attachment 137536

old zoomies for reference

View attachment 137537

Your old zoomies look almost new. Want to sell them?

So what sportline components are on your car currently?

 

[a777fan]

Howdy!

Long time no see guys… still here and so is Uter, but sad to say he hasn’t gotten much attention recently, as I have been distracted with his brothers from Munich.

The car was last struggling with misfires, so today I removed the caps to perform the @JC220 ’vent’ procedure which I am happy to report went well. I re-installed the modified caps without issue.

I then went to the trunk to ensure the battery cable corrosion removal I performed a while back was still effective. To my surprise, I found a really really damp trunk:

Any ideas? The trunk seal doesn’t appear to be leaking. Perhaps the gaskets around the fuel filler door and its components have gone?

I removed the battery and re cleaned the cables, and dried everything else out as much as I could. Re-connected everything, and then fired it up. I wanted to let it warm up to see if I could get it to act up. It warmed up fine, but the second I put it into reverse to back out of the garage, the fish biting started. It got progressively worse and worse.

Oh well. Back to the basics I guess. V frustrating tho.

 

[gsxr]

Jon, can you get SDS live data on video from the EZL when the misfiring occurs? Specifically, the voltage display for all 8 cylinders.

 

[nocfn]

is the drain clean for the gasoline access? you can pour water from a bottle with precision in the drain hole and it should run rapidly out. If it isnt hitting the ground as fast as it goes in that might be a contributor. Moisture is getting in there, is it hot enough yet to get condensate (think sweaty beer can on the back porch in August) or is water getting in from the fender vents and not draining, the window seal from above or missing / plugged wheel well drains. I am sure you thought of all of this stuff, perhaps remove all carpet pieces, pull the grey light switch and let things dry out in the garage.

 

[a777fan]

Jon, can you get SDS live data on video from the EZL when the misfiring occurs? Specifically, the voltage display for all 8 cylinders.

I'll have to lug out the SDS 'CUBE' and take a looksee! That is next on my list.

 

[a777fan]

is the drain clean for the gasoline access? you can pour water from a bottle with precision in the drain hole and it should run rapidly out. If it isnt hitting the ground as fast as it goes in that might be a contributor. Moisture is getting in there, is it hot enough yet to get condensate (think sweaty beer can on the back porch in August) or is water getting in from the fender vents and not draining, the window seal from above or missing / plugged wheel well drains. I am sure you thought of all of this stuff, perhaps remove all carpet pieces, pull the grey light switch and let things dry out in the garage.

Read my mind! I did a quick check yesterday with a garden hose, and all looks well.

Currently in the garage, trunk lid open, gray switch pulled, and airing out!

 

[nocfn]

Carpet out, those side vent fins in good order - drainage grommets in good shape (how about that seal for the antenna). I cannot remember but the bumper attachment points inside the trunk? They usually have rubber washers that could fail and lets not forget the tail-light lenses have sealing properties too. Any rusting under the parcel shelf (as a w126 owner its part of the checking) lastly the sunroof also drains out the back corners. Same drain setup on the 126, pour the water slowly on the trax with the car on a slight upslope. Water should run out the back really fast..... drain down C pillar near the trunk.

 

[kegmankipp]

I had a small leak that I could never seem to find until I went to replace the rear window lower trim strip that was peeling up on the corners. Turns out the rivets that hold the clips to the body were broken and leaking. Simple fix, see link below.

HOW-TO: Replacing Rear Window Lower Rubber Trim | "HOW-TO" Tutorial Articles

The lower rubber trim piece on my rear window was curling up on the sides so I tried to bend it back but noticed the riveted in place clips were broken. So I decided to see if I could replace the rubber strip, clips and rivets. Amazingly the parts are still available from the dealer and a lot...

www.500eboard.co

 

[sheward]

I was pulling gently on that trim strip just a couple weeks ago to clean and broke 2 of those rivets. They can be a source of leaks. We find the same clips around the front window, and I found some of those weak.

 

[gsxr]

I was pulling gently on that trim strip just a couple weeks ago to clean and broke 2 of those rivets. They can be a source of leaks. We find the same clips around the front window, and I found some of those weak.

Excellent point - check those rivets. They are easy to replace if needed. May not hurt to dab a thin amount of flexible sealant on the new rivet head too.

 

[a777fan]

I had a small leak that I could never seem to find until I went to replace the rear window lower trim strip that was peeling up on the corners. Turns out the rivets that hold the clips to the body were broken and leaking. Simple fix, see link below.

HOW-TO: Replacing Rear Window Lower Rubber Trim | "HOW-TO" Tutorial Articles

The lower rubber trim piece on my rear window was curling up on the sides so I tried to bend it back but noticed the riveted in place clips were broken. So I decided to see if I could replace the rubber strip, clips and rivets. Amazingly the parts are still available from the dealer and a lot...

www.500eboard.co

Excellent tip Keg-er-ator! I have fiddled with that trim strip previously, and found those clips loose. I must say... thats my leading candidate at the moment!

 

[TerryA]

Excellent point - check those rivets. They are easy to replace if needed. May not hurt to dab a thin amount of flexible sealant on the new rivet head too.

I might add, use stainless steel rivits for your replacements. They won’t ever deteriorate again.

 

[nocfn]

Aluminum are the ones I use.

 

[sheward]

I believe MB may have used aluminum. Dissimilar metals will eventually cause corrosion; however, you and I will be using power scooters by then.

 

[a777fan]

A misfire saga in 9 short parts:

Now for a quick drive. Will report back.

 

[a777fan]

Short toot achievement unlocked…. For now.

So frustrating that this moisture thing is just continuous.

 

[gsxr]

Excellent diagnostics! This is exactly what I was expecting to see, abnormal voltages all related to one cap, likely the passenger side. The initial voltages of high 30's and low 40's were an early indicator even without misfiring. The ~70v readings will appear with a misfire.

Could you post a close-up photo of the slots you added to the cap? I'm wondering if they are not quite deep enough. @JC220 might have some input as well. With adequate slotting, AND after the engine is driven for 30-60 minutes at normal operating temp after the slots are added, maybe the issue will go away.

BTW, the strong odor should not occur unless there are active misfires. If you have abnormal odors with good ignition voltages, there may be a separate, unrelated issue (would need some LH live data when it smells funky at operating temp).

 

[JC220]

The slots have been cut there that's good and thanks for sharing the videos!

The slots have not been cut deep enough however.

The cuts need to extend further like the factory ones on the lower edge. This is so they can bypass the O ring on the insulator cup behind and also the cylinder head alloy.