Thank you for reply. as you said, I search for LTFT. problem. is that related with O2 sensor?

 

LTFT is a data trend on how the ECM ( engine computer ) reacts to fueling condition which is not 14.7 ( Lambda 1 ) , the target is as close to 14.7 Air Fuel Ratio.
It is a long term historical data of the last many seconds of changes of STFT.
STFT is instant, real time.

LTFT is responsible for the calculation during a cold engine start or any engine start after engine shut down.
It used the WAS condition to predict how to fuel mix your engine when you start it the next time or next day.
When engine running, the STFT then will take over if the sensing condition already goes into close loop.
Close Loop is the time when oxygen sensors are used as data input to TRIM the fuel + air mixture, hence the name Fuel Trim, Short Term and Long Term.

Engine start do not use oxygen sensor until some seconds later. So if both or single upstream and/or downstream oxygen sensor/s is/are faulty or slow, your LTFT data is possibly wrong and
when engine started the next day or next hour, its fueling calculation is then wrong and you have shaky engine etc yada yada.

I will get back to you later tonite. Me going to dentist soon.
I would like you to do a proper test again and no rush please. You are impatient, this is NOT good for troubleshooting and data collection.





Among other things you need to improve :
You did not start your engine from proper COLD condition.
You rev up to soon, your coolant temp is not ready yet. Start reving only when coolant hit 80C or higher.

In the mean time , learn some for fuel trim and oxygen sensor :
https://gtc.ca/blog/long-term-fuel-t...rim-fuel-trim/

http://aa-bosch-ecat-ap.resource.bos...gue%202013.pdf
From start to page 39 in pdf reader.


Find attached my test worth 28 minutes. I suggest you do the same and don't do too short.
The only thing I forgot to do during this test is at idle after 4,200 RPM and wait 3 minutes is to blip the throttle to 3,000 RPM fast once every 5 seconds for 10 times to excite the oxygen sensors.
Please do the throttle blip.

My downtream oxygen sensors is rather "tired" too and I am at 30,000KM only.
I can sense a very-very minute imperfection of combustion during first 12 seconds of engine start.

Will get back to you tonite yah.........

 

Sorry for late reply, I will go where test is available place. Within 2 hours, I will share logging data.

and thanks for infor about LTFT.

 

Hi Kim,

Based on your log, which is missing engine prior-start up data log and missing 8 seconds in the 1st 1 minute.......

Engine start up data is before you crank your engine, logging must be activated first at least 10-15 seconds ahead of time.
This is to read some available data before engine actually is running.
Ignition position 2 ( ready to crank ) where all the dashboard warning lights up, is at a stage where ECM is already awake and many data are avaible to log.


Below is your log which I summarized to see the seconds before engine start. Your start of logging was some seconds into engine ALREADY running. Maybe 10 seconds too late.
The red highlight with yellow text showed +-8 seconds data lost/jumped, seconds 31 to 37 lost... I wonder why.

I dont understand why you logger is so weird and to make a graph out of it, it needed a few step in excel to disregard the 7 missing data lines.
The App seems to create a 8 data line per second but it is only 1 data content , and is randomly placed between 1st to 8th line...weird.
If there is a section in the Apps called FAST DATA POLLING, turn that off and probably you can then set per how many seconds per 1 data log would take place.


01. You reported that the car shakes/rough only at engine start, be it COLD or HOT.
Please verify :
1A - Your car location altitude is still within 200 meters above sea level or close to sea level ?
1B - How long does the shake last till it smoothen out ?
1C - When you start driving the car after the shake gone, does it ever shake again for the entire trip and at traffic stop ?
1D - Let say into 10-20 seconds your engine have started and shaking, can raising RPM remove or reduce the shake ?

I am aslking above to narrow down the point in time of the shake.
Let say it is always within 3 minutes of engine start, that means cause of shake probably involved certain data/value of sensors or mechanical position of components, or temperature of components
which is specific ONLY when an engine has been dormant for sometime.

Let say your injector is a high suspect.
It could only be bad at lower temperature < 35C ( during engine start ) and is fine when operating temperature reached or certain temperature reached, say 50C.
Mild leaking injector tip can cause shake at start because fuel mix become too rich. This can kill your cylinder if occured too long. The fuel will wash off lubrication film from piston rings and liner.


Let say your MAP sensor membrane is a bit stuck and during engine stop it never could return to 1 BAR ( 1 ATM ) and was always 0.8 BAR.
The ECM would then calculate at 0.8 BAR / ATM, the car is at 1,800 ish meters and will set fueling and ignition scheme to this altiude, while actually your car is at sea level.
We start with RPM data as indicaton of engine start. Your data started late based on your coolant temperature first logged at 45C where your intake temp was 27C and the log is missing 8 seconds.
My data has 13 seconds logging before engine started. Both data below is worth 120 seconds.




Because your logging is weird and missing 8 seconds and was not including 10 seconds before engine start....... I can't even verify that your MAP
before engine start is reading normal atmospheric pressure or not ? 1 ATM = 1.01325 BAR , just assume as 1 BAR okey.
It has to read ambient* ATM ( *depending on your car/garage altitute ), because there is no vacuum yet by piston when engine is not running.

I am at sea level basically. So 1 ATM / BAR is what I should get when engine OFF.




Let say you have a stuck open Evaporator Purge Valve, but at under 40C component temperature only and when engine bay get hotter it is then normal. This will be equal to a mild vacuum leak.




Below is the high pressure fuel pump's pressure feeding the fuel rail. Since you data never started from STOP engine and is missing 8 seconds, it is difficult to compare.
My engine RPM at start, based on 27-32C ambient has always been 1,200 ish RPM for first 10-15 seconds and goes down after that. Your log started at 878 RPM, so comparing fuel pressure data is not good yet for now,
but I just want to show you that if a bad fuel pressure regulator can also be the culprit of misfire.




ALL ABOVE GRAPH ARE approx 120 seconds worth of data ( my car ), Kim's one is 120 + 8 seconds and unknown missing initial data.

==================================

ALL THE FOLLOWING GRAPH BELOW FROM KIM IS WORTH APPROX 27 MINUTES



So you needed to move your car some distance from your garage. fine.




NOTE : Idling fuel trim can never be the same as when car is moving ( more load ), even though the same RPM.


Below is telling us as follows :
Your engine's Long Term Fuel Trim value, indicated that the ECM on average has sensed a LEAN mix and have to ADD / POSITIVE TRIM more fuel. More so at Bank 1.
Running LEAN is more air than fuel condition, and the positive trim up to 10% of Bank 1 and positive trim up to 5% of Bank 2 is the additional fuel % commanded by the ECM.
ECM uses Fuel Table as a baseline, oxygen sensors and other sensors are the fine tuning and verification devices. http://www.formula1-dictionary.net/map_fuel.html








The responsible oxygen sensor for calculation of the STFT is the front Wide Band 02 Sensor ( AFR or Lambda sensor ). It is fast acting and wide range AFR reading. This cost like 300% of the narrow band 02 sensor.

14.7 Air To Fuel Ratio (AFR) is the target. Also known as Lambda 1. Lamda 1 is better to use actually as not all fuel is a 14.7 fuel to 1 of air.
E85 fuel's Lamda 1.0 is 9.8 air to 1 fuel.



Below is the downstream narrowband 02 sensor. Low resolution sensor.

{ICTURE ABOVE : The 02 sensor downstream ( Bank 2 ), narrow band. Above 0.45Volt and up it is reading RICH.
0.45 volt down it is reading fueling mix as LEAN. It is not a wide band 02 sensor , so its behaviour is like switching.
Here I find it odd. It is sensitive to throttle blip at 11,000 ish to 12,000 ish data points which will cause reading to go high and then low, this is the correct response.
But it doesn't do well during idling before the blip and after the blip. It should behave like mine.


BELOW : my engine's downstream 02 sensor behaviour.


ABOVE : My red line ( bank 2 ) sensor is rather lazy. It should be behave like the blue line ( bank 1 ). I am ordering a new one but have no news yet on availability.
At the same time I will be replacing all my spark plugs too, now used for 20,000KM.
The downstream 02 sensor is slow, but it can not stay too long at RICH condition, it would have informed ECM when it passed 0.5 volt and ECM would reduce fuel to make it touch LEAN ( down to 0.45V or less ) and then it will go up back
to RICH and cycle up-down that way, see the blue line. It will only work when its temperature has reached 400C or more. Refer to Bosch Oxygen Sensor catalog I linked yesterday.

The Bosch LSU 4.9 is the type M276 engine uses as upstream wide band sensor. 6/5 wires.
The downstream one is the simple narrow band one. 4 wires.


Supposedly the downstream 02 sensor is to make sure the CAT works well and also as a 2nd input to the ECM.
How and who is finally the decision maker* (*if readings are in conflict due to one bad sensor ) of Lamda 1 ( 14.7 AFR ) between upstream wide band 02 and downstream narrow band 02 sensor, I do not know.


My suggestion to you Kim, is to replace at the least the downstream narrow band 02 sensors of yours.
As for injectors, you also need to check that and do so using a Direct Injection Specialist. There are a few important caution when working on DI injectors. Some parts must be be replaced when injector removed from its rail.
I am sure that your issue is a mix of technical issues and not just the injector/s ( if indeed injector/s is/are bad ).
Make sure your MAP sensor can read 1 ATM ( sea level ) with engine OFF and you EVAP system is also shut close during some seconds of engine first start.

Do a mechanical gauge compression test and inspect liner, piston and valves with inspection camera/boroscope, if ever you removed your injector.
You might as well log all condition you can as part of the car's 125K KM Bday...

Happy troubleshooting.........

 

I'm glad to see suggestions what you are very care for.

thank you for your deep analysis for my car. thanks thanks

this is really thanks to me even if this problem not solved.

anyway you suggest that your suspicious points need to check asap.

I will consult with DI engineer who are good at diagnosing engines.

for your question I will reply as belows. today I almost run 100km. and shake occured irregulary but almost 80% cae happens. even if shaking happens, output power if I drive full throttle is still poweful.. I wonder that when shaking happens, my led lights of front map lamp is trembled..it is very weired.




01. You reported that the car shakes/rough only at engine start, be it COLD or HOT.
Please verify :
1A - Your car location altitude is still within 200 meters above sea level or close to sea level ?

--> I'm in seoul, altitude is 45m.

1B - How long does the shake last till it smoothen out ?
--> I sense that shaking is last now. It's get worse and frequent I think.

1C - When you start driving the car after the shake gone, does it ever shake again for the entire trip and at traffic stop ?
--> yes, that's correct. my shaking symptom at traffic stop not driving.

1D - Let say into 10-20 seconds your engine have started and shaking, can raising RPM remove or reduce the shake ?
--> when shaking happens, I push accelator pedal. RPM get higher than idle, shaking is removing. high RPM remove shaking symptom. So cuurently I usually push accelator pedal when idling.

 

A quick look, seems like a lean condition indicated by positive long-term fuel trims... possibly a slow or poor performing O2. Downstream O2s (post-CAT), usually, only are used for emissions purposes and don't affect engine performance since the A/F ratios are not applicable. Essentially if the rear O2 output mimics the front, then the CAT is "presumed" to be not functioning and thus a code is generated.

I would still focus on coil swapping to see if it follows. Also get MB plugs to insure their indexed correctly.

 

also I have old data for start engine logging data. I attached it and capture with fuel rail pressure status. thanks.

 

Hi Kim,

Is that most recent log this morning ? The file name/date is April 17th ?
The short 3 minute engine start looks okey, similar to mine.


Let's narrow down some more, your problem.

01. Is your car modified/tune in any way, engine wise, special mapping of the ECU, maybe piggy back, sports exhaust and those thing ?

02. Do you have any electronic device added close to or before the shake issue started ?
Example : The 10" Android Display aftermarket unit common today.
Or special aftermarket gauge which uses OBD2 port of the car.

03. Your OBD2 dongle, try your car without the OBD2 dongle.
02 and 03 is my thinking what if your CAN BUS is effected somehow...

So far this is what we know from you :KIM WROTE :
my car mileage is 128000km.
change all of spark plugs and coils at 127000km
My car shake at idle repeatedly. this symtom is not regulary. but when shaking happens, my room lamp's led light trembles and alternator voltage get dropped... at this time, I push accelator up to 1000 rpm. this shaking symtom is clear and light is no more trembing..also alternator voltage get to 14.7V

Even if I change all of spark plugs and coils, my car sometimes occur misfire on cylinder no.4 when driving over 100km distance.. this problem occured on cylnder no.4 repeatedly.
When I checked CAM magnet and shaft sensor, it looks fine without oliy.

========================

I change after market brand from BBT ( COIL ) and bosch ( spark Plug ).
-misfire problem occured before replacing and current. both are same misfire on cylinder no.4


========================

mechanic said BBT is just selling brand that plug is original part from delphi.

===============================

Here is my thoughts and I hope someone else with more experience will chime in.

01. We now can confirm that your car shake is at idling and not only at engine start.
Usually a vacuum ( air ) leak will cause shake at idle and upon higher rev shake improved.
Our engine uses MAP ( Manifold Absolute Pressure ) sensor, it is not using MAF ( Mass Air Flow ).
Being equipt with MAP, a minor vacuum leak can be tolerated by the ECU because the MAP can read it and assume there is more air. Of course we are talking of minor leak.
Different to MAF which reads air coming into engine very close to throttle body and it can not calculate air leak ( vacuum leak ) into engine if leak is after throttle body & MAF.
I do no see your MAP reading as crazy, but it who knows if the OBD2 dongle was slow to register....
The long term fuel trim indicated LEAN, that means too much air...like a leak or injector( fuel system a better word ) unable to deliver enough fuel at idling but okey at higher RPM.
Here we assume the upstream wide band oxygen sensor as healthy enough.


02. If say spark plug is no good, it will not burn fuel properly and RICH is what your engine will get as the oxygen is not consumed enough by the combustion event.
oxygen sensor reads oxygen level and not fuel vapor.


03. Shake so bad must be triggered by something significant.
We have FUEL, we have AIR and we have AIR TO FUEL MIX and TIMING. These 4 if all well, combustion goes well.
I don't exactly knows how the MB variable timing system works, so I can't comment.
When I was young and had my first car in 1983, my favourite prank to my friend was I like to swap 2 cylinder spark plug wires at the distributor. In the 80s even with CDI ignition, single big coil and the distributor was the norm for ignition.
So by swapping 2 sparj plug wires, the firing order of the engine went banana at 2 of its cylinder and engine shake like crazy.... , I was thinking of your engine shake last nite and I laugh remembering what I did in the past.
You know removing 1 spark plug wire, on a 6 cylinder engine will not make engine shake like crazy, shake yes....but not violent. This is why your case is unique, because your high RPM is okey.

Thats all for now, please answer the questions I asked and hopefully we can narrow it down before you send your car to the workshop.
Any information will also be useful for your mechanic.

Later......

 

Hi Prihadi

Thanks to you, I have some understanding of the vehicle problem

I reply per your question.

[size=16px]Is that most recent log this morning ? The file name/date is April 17th ?

--> yes, this log were made on Apr 2021. at that time, shake symptom was occcured as now.[/size]

[size=16px]The short 3 minute engine start looks okey, similar to mine.[/size]


[size=16px]Let's narrow down some more, your problem.[/size]

[size=16px]01. Is your car modified/tune in any way, engine wise, special mapping of the ECU, maybe piggy back, sports exhaust and those thing ?[/size]
-> no, my car is original version not mapping.

[size=16px]02. Do you have any electronic device added close to or before the shake issue started ?[/size]
[size=16px]Example : The 10" Android Display aftermarket unit common today.[/size]
[size=16px]Or special aftermarket gauge which uses OBD2 port of the car.
-> no, I do not install any elec components.[/size]

[size=16px]03. Your OBD2 dongle, try your car without the OBD2 dongle.[/size]
[size=16px]02 and 03 is my thinking what if your CAN BUS is effected somehow...[/size]
-> even if without obd2 dongle, symptom is same. actually, I bought obd2 dongle because of shaking symptom coming.

 

Kim,

Your trouble code is still showing ( check engine light ) , yes ?
Get your mechanic with a better scanner or MB official Xentry to retreive the Freeze Frame data on the trouble code, if you OBD2 dongle does not have such data.
Here is what Freeze Frame means :
https://www.samarins.com/diagnose/fr...was%20detected.

Your P0300 code group is a generic data. P0300 is a diagnostic trouble code (DTC) for "Random/Multiple Cylinder Misfire Detected".
P0304 means cylinder 4, but that is not a 100% guarantee.
For what threshold level to be exceeded till the ECU/PCM of engine will show such code, it depends on manufacturer set parameters and not all manufacturer uses the same formula.

If you really want to know a Misfire definition , see page 11 of the PDF version.
https://www.diva-portal.org/smash/ge...FULLTEXT01.pdf

.

 

currently I have no DTC and engine check lamp now.

Today I took my car for 200km, I guess that altenator is very suspective cause.

because when engine almost idle status like undet 700 rpm, fuel rail pressure was dropped 18000pa to 11000pa. at this time, shake was occured. and also battery voltage is dropped 14.5V to 13.5V. it means alternator is not worked at this time. and shaking is occuring.

is the fuel pump sensitive with battery voltage?

I attached today's logging data. based on my data, under 700 rpm, battery voltage was dropped. and fuel rail pressure also dropped at this time.

And my alternator is very weird acts. because charging voltage is very very fluctuate like under image. but your log data, alternator voltage is steady at 14.5V.

How about my opinion?

this is my car data






this is your car data

 

Hi Kim,

Our car is B03 ECO start stop. Its algorithm for battery charging when vehicle is moving or D gear is not the same as when it is in P parked.
In Parked and in my case when I tested 29 minutes for you, I reversed only 10 meters away to get the car out of garage to open air, is considered by car's computer as car never made any journey yet..... near constant 14V ish is the norm.

If car is moving, our car charging algo is the worst of the worst and will never get our battery charged to more than 80% and you can even see voltage down to 12.2V for ECU.
Follow this super long thread....
https://mbworld.org/forums/e-class-w...ml#post8337068


Fuel rail pressure at 118-120 BAR or 120,00Kpa at idle is same as mine, it is normal.
When you rev up, it goes up to 180 ish BAR.


This log at a quick look is zero movement of the car ... YES ?
I do not see any speed data. 10 minutes worth of stationary.

How bad and how long was the shake during this logging session ?

I will take a closer look tomorrow yah.

 

thank you for quick reply.

I see ECO algorithm for battery charging system.

I attached log data.

and captured inage. as you see, shake occured at low fuel rail pressure with low voltage and under 700 rpm, under 10km/h.

 

I'd suggest doing a smoke test into the intake and examine everywhere for any leaks.

 

yoyo below 12.5v

I see your concern with low voltage... simply DRIVING WITH HEADLIGHTS: ON will get you 13.5v to bypass all low voltage troubles.

This way you'll see if there is a strong correlation between misfiring and low voltage (90% chance misfire is there and low voltage is side contributor).

My guess is your engine has a lean base mixture (PCV diaphragm).

 

I sorted the lowest RPM during the listed time frame.
I never seen as low as 505 RPM on my car. Is this time frame the occuring shake you keep mentioning ?
- Your car is 4 matic or Rear Wheel drive ?
- Did you operate the ECO start stop when you experience the "shake" ?










I am worried your throttle body maybe dirty or closed too much after de-acceleration, and OBD2 may not read well.
The learning/adaptation of throttle body needs the MB Xentry diagnostic tool.
I can't even find reccomended idle speed from EPC/WIS, all seems to be locked to Xentry.

This is how Xentry teach the throttle body. Go to 9th minutes

Mud's advice for any potential leak using smoke test is also something you need to consider.

 

I forgot, what Bosch plug did u use ?
It should be this number https://germanparts.ca/parts/m91044-...lug-v6s113328/
It is EVO Spark Plug https://www.boschautoparts.com/en/au...s?partId=96347

 

sorry for late reply.

my car's VIN is WDDHF6HB1GB218267

my car is 4matic and sometimes I use ECO start/stop because of shaking is unpleasant.

I observe engine RPM during idle. Really I saw that my car engine RPM is too low (almost 500~600 RPM) to keep idle. low RPM makes shaking my car. also it makes led lamp is trembling. if I push accelator up to 700~800RPM, shaking is disappeared. or if I turn off blower or change gear D to N. idle RPM is little higher than before. this helps relieved shaking.

I think my throttle body is root cause of shaking my car. I will try to clean my throttle body.

 

thank you for reply.

first, I wil clean my throttle body. after then shaking is still occured, I will do smoke test.

 

I was change my spark plug and coil set in repair shop. I do not know partnumber, but plug is from bosch part and coil is from BBT.

as I said first thread, misfiring was occured before and after changing with same cylinder no.4.

 

Today DTC is occured P1193.
I will go to repair shop to check what exactly fault is.
I will share the check result.

 

I know that shake occured before plug and coil change, what I need to make sure is that your spark plug is the correct one and will not ADD more issues down the road.
Also the BBT coil if not up to MB spec, can lead to new problem, not now...maybe at high rpm if you drive agressive.

This trouble code you just got, are they new ? Have you ever seen these types of code ?
If such code never existed and now its popping out, if it is not related to previous code, whatever new stuff you put in can therefore be part of the suspect.

You need a good scanner with Mode 6 or better MB own Xentry.
OBD2 generic is not the best tool if you need more reliable data.
Also OBD2 itself has many different capability between simple ones to workshop grade ones.

Does your engine really have EGR ? mine does not.
Easy to check EGR.

This one without EGR ( Exhaust Gas recirculation )



This one has EGR



Does you car have NOX sensor ? Mine doesn't.

This car has NOX. So more sensors than 4 at exhaust system and located at the middle to the back, not the front CAT with 2 + 2 oxygen sensors. Me don't understand NOX, but they can add to complexity when comes troubleshooting related to combustion.

 

Damn. P1193 can mean different things from different car manufacturer
Go for MB Xentry please.

 

UPDATE :
This time your trip 26th June 21 Songpa to Dontan showed some VERY important VERY LOW voltage data, for OBD2 which means Rear SAM module being the source.
Such low voltage was not picked up during your stationary test of 23rd June 21.





.
.


During all my test, the worse or lower voltage one is usually the ECM/ECU and not the OBD2 /BUS voltage/Rear SAM.

No doubt the ECM/ECU for engine is powered by the Front SAM ( Battery>>F32 Prefuse via Q Diode>>Front SAM>>N3/10 ECU/ECM ), you must sort out the low voltage issue first above all else.
These electronic MAD car of ours is very sensitive to low voltage and things can go banana.

Read this and do as what I and Cali did.
https://mbworld.org/forums/e-class-w...ving-them.html


Take photo of your F32 Prefuse please. Your car is left hand drive and 4-matic, mine is right hand drive and rear wheel drive.
I will show you how to test K2 relay output, just in case the low voltage starts from there.

This is F32 Prefuse. The long black plastic, not the brass SHUNT my finger is at




Open up the cover, so that we can see the connections. Do not unplug anything.

 

Your threads always lead us to something interesting... 👏


skewed voltage measurements front/back

The voltage going to 11.3V now makes me wonder... this really looks like R-SAM is regulating with a 1Volt skew.
As in whatever module that is measuring voltage (*) has a bad GND reference that keeps it above reality.
The voltage is regulated at 11x because it's working of a measurement that's worse than 1 Volt different!
Internally the SW logic sees 12.6v when the reality is near 11.3v 🤣

(*): As in...:
Alternator regulator over LIN
Hyundai sensor over LIN
R-SAM VDC measurements<<<<
(F-SAM VDC measurements?)
(ECU measurements??)

I believe R-SAM is in charge of its local turf. Yet I don't know precisely WHERE it measures voltage and don't exclude R-SAM may be regulating with a voltage measured remotely somewhere else (ECU, F-SAM, ...)
- That being said, always "go to the obvious first!" : clean SAM'S GND in the trunk right side AND perhaps elsewhere shown on the R-SAM schematic .

Repair exit condition: "Tiny DROP VOLTAGE" under similar driving scenario.
To test after repair we need to use these exact same conditions to reproduce the issuewhen R-SAM is tasked with regulating the 12.6v float voltage.

Back to basic GND cleanings!!

"go to obvious 1st": old-fashioned painted R-SAM GND nuts.

Also keep in mind the skewed dropped voltage may be internal to R-SAM caused by the solderless pins... we'll soon find out as we still have a few unknowns >>> See Edits !!!!

+++++++++ I had this backwards !!
The module that is regulating (*) is using a "measured 12.6V" that is shown by to be 11.3v in reality.
The skewed voltage is the one presented as being "12.6V".
The true voltage is the one that is reading low!

-or there is a serious drop on the positive rail reaching R-SAM as 11.3v