OM648 ocassional ‘misfire’at idle

Page may contain affiliate links. Please see terms for details.

Richteo

Active Member
Joined
Aug 31, 2011
Messages
85
Car
W211 E320cdi 2003/ C218 CLS350cdi 2012
My 2003 E320cdi seems to be suffering from a slight misfire on ocassion when fully warmed up.

It is most prominent when idling and it can be felt through the car body. It makes no difference if in park, drive reverse etc. No fault codes are logged when scanned with an Icarsoft MBII scanner.

I’ve had the EGR off and given it a clean but I don’t think it is at fault (moves freely, EGR position reported by MBII seems to be operating normally).

I’m suspecting an injector. Car has done 202k miles now. Will a leak off test prove which one is at fault or is there something else I should be doing?

Are there any other potential causes I should be looking at?

Car is regularly serviced with new air filter and fuel filter in the last 5k miles or so. Issue has been present running BP and tesco fuel.

Thanks
 
find a code reader capable of (fast) live data reading... check first if it is rail pressure related or not.
 
Thanks for taking the time to reply. The MBII does do live data although I'm not sure that the refresh rate is particularly good. I will check using that first but wouldn't there be a DTC if the rail pressure was low?

I plan on doing a leak off test on the injectors as a test kit is only £20.
 
Could be black death on one injector, once warm and the crap melts, then starts spitting diesel and misfiring. Have you had a look around the injectors?
 
Could be black death on one injector, once warm and the crap melts, then starts spitting diesel and misfiring. Have you had a look around the injectors?

Yes, no sign of black death. I've replaced two of the copper injector seals during my ownership but both were only at the 'chuffing' stage before anything built up.

Its not a full on misfire, more an regularly uneven idle if that makes any sense!
 
I've had a bit more of a play with the icarsoft MBII. Still no fault codes but lumpy idle is still present. No limp mode.

When the car is idling the fuel pressure fluctuates between 316 and 336 bar.
Pressure control valve position is reported between 27.4 and 28.1% - I presume this is Y74 on the fuel rail.
Y94 Quality (or Quantity in MB speak) Control valve between 41.26 and 41.62%.
Injected quantity of fuel approx 9.6 and 9.8mm^3/Hub

With the car in neutral taking the revs to about 950RPM the injected quantity reduces(?) to 8.3mm^3/Hub, with rail pressure of about 375bar. I don't understand why the fuel quantity would decrease.

Annoyingly the MBII doesn't give injector correction values. Only 4 cylinders are shown, and all 4 report 0mm^3. Useless.

Are the above values normal?

There is some good info in the link below.
https://www.benzworld.org/forums/w2...8-running-perfectly-but-making-serious-5.html

If the info in the above link is to be believed Y74 regulates fuel pressure until the fuel temperature hits 20degC, then Y94 takes over. Y94 My idle seems to get worse as the car warms up.

The current plan of attack is see if I can tie the rough idle to a change in fuel temperature, indicating Y94 may be at fault. And run an injector leak off test when the kit arrives, probably at the weekend.

Is there anything I'm missing other than getting a proper STAR diagnosis? If the above tests don't reveal anything that will be the next step.
 

Attachments

  • upload_2018-12-10_19-50-42.png
    upload_2018-12-10_19-50-42.png
    139.8 KB · Views: 12
Some progress but I'm not sure its conclusive. Advice is welcomed.

The idle definitely gets worse once the car warms up. Not easy to tie down to a specific temperature of change over though.

I did a leak off test when the engine was cold and when the engine was warm (approx 15-20 miles of running). In each case I ran the engine for 3 minutes. I did have to stop the cold test after 20 seconds as I had fully clamped the return line resulting in a bit of leakage. The rust colour on cylinder 1 was just from the fitting on top of the injector and cleared very quickly.

The leak had one positive result - I was able to identify the start of injector seal failure on cylinder 3. I will fix that as soon as the weather improves. Would this been enough to cause an idling issue. From my experience of other injector seal failures I'd say no.

I'm inclined to think that cylinder 1 injector is faulty. am I correct? Car is booked into a Mercedes specialist on Tuesday so will update again then.


Cold (6-1 left to right):

leak off cold.jpg

Hot (same order):

leak off hot.JPG
 
Video showing engine shake at idle, leak at injector 3 seal and view of belt showing no obvious issue with alternator or water pump pulley.
To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
 
To close this thread, star diagnostics showed injectors 4 & 5 were outside of the fuel correction quantity tolerance.

I replaced them myself yesterday with refurbished items from PF Jones (Bosch approved) via their eBay store. £90 each plus £30 surcharge to be refunded on receipt of the old injectors.

Vibrations gone and all back to normal. Thanks to those that replied.
 
Great video you supplied there. Did you have injector 3 replaced also as that was leaking in the video. I have an injector leak on my new to me e320 2004. The cylinder head is very clean for 200,000 plus miles like your own. but with the engine cover off its obvioius the leak is injector 1. I have booke the car in with a good reputable indy just a mile or so up the road. |My usual indy is more distance away from me givin this new guy a go. Replacing the injectors doesn't seem to much of a task its having the correct tools I;e the slide hammer to pull the injectors from the head. I have well enough belief I could do this job but im going to leave this with a mechanic this time round. Did you need to code the new injectors to the car?

I plan on just removing the faulty injector to begin with and see if that resolves the issue of cabin fume smell as that's what highlighted the problem to begin with. With the engine cover in place you cant see or hear any injector faults is so sound proofed. Booked in for the 2nd Jan so will report back on findings.

Tom
 
When my engine was hot it was clear that the injector seals on both 2 and 3 were gone. I replaced these before taking it to the garage for the star session. Both of the injectors themselves were fine. Luckily as it was the very early stages of 'black death' they both came out easily by hand with a bit of twisting.

I use a 1/4" torque wrench set to 10Nm to undo the bolts (didn't click), a long screwdriver to dislodge the copper washer from the seat and a pick tool to fish it out. Then I used the Laser 6101 kit to clean up the seat. I meticulously cleaned the threads for the hold down bolt and made sure there was no debris at the bottom. To do this I used one of the old bolts with a groove cut in the threads parallel with its axis, cotton buds and a thin tube taped to a vacuum cleaner. This is the most critical step. Obviously it is essential that new bolts are used. Tightening torque is 7Nm, plus 90 degrees, plus 90 degrees.

Changing injectors 4 and 5 I obviously had to repeat the above process. I did the work myself firstly because it is easy, and secondly because I can spend as long as required to clean things up, without the time pressures that come with running a business.

Regarding coding, no I haven't. All of my injectors plus the replacements have a number 4 in a circle. Based on my engine number and the info below my engine should require a number between 1 and 3 coding into the ecu- the injector classification. Every image I've found of my injector part number has shown a number 4, with many references to not having to code them. IMA coding is not applicable to my engine number nor the part number for my injectors A6480700187. A6480700287 injectors do need the IMA coding.

This is what WIS says:
upload_2018-12-28_12-43-42.png
upload_2018-12-28_12-45-3.png
upload_2018-12-28_12-46-30.png
upload_2018-12-28_12-47-2.png
 
Good to know you resolved the issue and all is well. Time is essential with this issue 100% agreed.
 
Great video you supplied there. Did you have injector 3 replaced also as that was leaking in the video. I have an injector leak on my new to me e320 2004. The cylinder head is very clean for 200,000 plus miles like your own. but with the engine cover off its obvioius the leak is injector 1. I have booke the car in with a good reputable indy just a mile or so up the road. |My usual indy is more distance away from me givin this new guy a go. Replacing the injectors doesn't seem to much of a task its having the correct tools I;e the slide hammer to pull the injectors from the head. I have well enough belief I could do this job but im going to leave this with a mechanic this time round. Did you need to code the new injectors to the car?

I plan on just removing the faulty injector to begin with and see if that resolves the issue of cabin fume smell as that's what highlighted the problem to begin with. With the engine cover in place you cant see or hear any injector faults is so sound proofed. Booked in for the 2nd Jan so will report back on findings.

Tom

Just an update.

Had the faulty injector removed cleaned up new seal reseated new bolt no more chuffing running really well is the e320. No more fuel cabin smell. Air-con bombed also really fresh fragrance going on. Happy days.

Thanks

Tom
 
The car is up to 215k miles now and it feels as though another injector is starting to go. No fault codes showing using my icarsoft MBII that I updated a couple of days ago.

The MBII still doesn't show injector correction values.

Does anyone know of an affordable code reader that does? Does the MB v2 from icarsoft? What about the Autel MD802? I know MB Star is the correct answer but I don't really want to spend a fortune on a chinese clone system.

Thanks
 
After much inconclusive reading I took a punt on a cheap Delphi DS 150E clone. It does quite a lot but again all injector correction quantities are reported as 0! The Interface hardware can also be used with Wurth Wow diagnostic software so I tried that too. Unfortunately the result was identical.

Is it really on MB Star that can read the values I'm after?
 

Users who are viewing this thread

Back
Top Bottom