When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Scope waveforms:
it sounds like now would be a good time to involve an automotive pico-scope to troubleshoot the wave forms on both your spark-coil and injector-coil.
Data PID sanity check:
Before you scope electrical signals, I'd take a good long look at live data PID for anything obviously odd (blank field) or meaningful but skewed (highly positive LTFT on lean engine).
Camshaft timing:
There must be a significant detail we are all overlooking, right?
Eventually you may want to revisit factors affecting engine timings.
The whole question about camshaft position, welding the sensor plate,... may not be satisfactory or something else in relation with that topic.
Last edited by CaliBenzDriver; 05-25-2023 at 01:07 AM.
Cams are in time, each cam sensor wheel has a factory line on both the cam and the sensor wheel, if those line up it is in time. All 4 are perfect. Could it be a tooth off? I double and triple checked. The sensor wheels are dead center in the view window so I moved past that.
When #1 fires the car runs perfectly, then it starts mis firing so a random checking out of #1. I think any mechanical issue would be static with no change at all running.
Wish I had the ability to check wave forms.
I am in process of checking the wires looking for a hidden issue. Trying to get to a point where I can trace from the computer connection back to the #1 Cylinder, both COP and Injector wires. The computer has the function of telling the COP when to fire, that circuit may be at fault also.
Thanks to the great information shared above I have a lot of the schematics to trace. Sure wish my WIS would open and function again.... It was two years ago now that I last had it open, maybe it expires?
My Scan tool is a Launch X431V+, it allows 2 way communication to the car computer. It does allow me to count misfires in each cylinder. I have not found a way to watch wave lengths as injectors fire. It may have the ability but I have not found it yet. Since my car is a 2012 E350 same engine and runs perfectly I have been able to go back and forth between them looking for norms. My cars injectors all show 156 -150 V while running on the high side power measurement. The Black car on #1 only shows the 172 V measurement.
I did not program the new injector, had no idea coding was necessary or programming. I thought it was a drop in and run it part?
Still fighting COVID so not working in the garage much.
First, get well soon. Don't physically overwork yourself.
Launch for MB is better than Autel in terms of capability , based on my baby cheapo Launch Creader Elite for MB vs mid level Autel MS906BT.
West wrote : 01. I have not found a way to watch wave lengths as injectors fire. It may have the ability but I have not found it yet.
02, My cars injectors all show 156 -150 V while running on the high side power measurement. The Black car on #1 only shows the 172 V measurement.
01. Do you mean the waveform ? That is from a scope, not a scanner.
02. So this 172V #1 injector is the new one yes ? Did you measure the old one ?
The injectors swap test can shed some light later on, as you should be seeing supposedly max 160V, that is when you have recover from the Covid.
If all test done and still same issue, you may have injector driver issue at the ECM, pray that it is not the case.
No wonder potential free or non-contact measurement is required, the injector does get its voltage REVERSED as part of its function.
This can done witth crazy expensive differential probe , which I do not have.
Therefore I can only measure 1 injector voltage at any given time, even on my (limited) floating ground Picoscope 4425A. This thread is VERY good for me.
Multiple DMMs will work but they can't see such fast signal of 1/1000 second, aka basically useless.
I think the injector drivers in the ECM are like a stand alone isolated circuits with no ground ties to common ground of the car or ECM.
I never knew piezzo expansion and contraction can be reversed via reverse polarity : https://www.piezosystem.com/piezoped...d-performance/
Here is something new to learn.
Our injector uses piezzo stack as the activation mechanism. I did not know piezzo stack can have failure on its insulation too ...wow.
3 very good videos on piezzo injector insulation test. Dang, I am drooling over the Megger 400 series these guys is using, it is more advance than my Fluke 1587
Above shows a regular ohm meter ( DMM ) which uses only like 5 volts, can read OK or 220 kilo ohms ( or 0.220 mega ohm ) if for our injector, but when a Megger insulation tester is used which injects pulses of high voltage DC, the piezzo stack failed. The Megger 400 series is really good on its auto test and can show lowest voltage at which the insulation failed. On my Fluke I got to do it manually in increment of 50V>>100V>>250V>>500V or 1,000V. Insulation test usually requires 2x the operating voltage of the device or wire used.
A scope reading voltage and current waveform, can also show by comparison to other known good injector/COP what's happening to a problem injector and COP.
But still, no device will work properly if the wiring to it is not guaranteed 100% OK.
It was two years ago now that I last had it open, maybe it expires?
If it is hacked version, sometimes the licence simply expired. If you have the password/key generator, you can run it again.
On my 2020 WIS only , no EPC , hacked version, the WIS red square icon will become dark grey when the licence expires.
On my Ebay source via DVD, it can do WIS and EPC but 2018 version and it is stable, no need to do anything to it.
But it is slower for its WIS compared to the 2020 WIS.
NOTE : I am quite an IT dummy. , I am poor with PC software thingy.
Ordered a new copy of the WIS, seems like prices came way down since I bought last time. Hope the new copy works. Since 3 different injectors have been installed in #1 and ran exactly the same, tested the same I still think the problem is wires or the computer on this E350. I wish I could just plug my ECM from my 2012 in and try it but the ECM would need programming to try that test.
When you checked the connections on the ECU, did you clean them anyways? say some electronic cleaner sprayed.
You did mention in post #25.
Have you load/resistance tested the pins from the ECU connector towards the coil on injector #1 and #2? I mention both so you can compare: supposedly good (#2) vs not great (#1). Continuity only tells us is not completely broken, Load test will tell us if it can support the amount of current when active. I think Surya highlighted the pin numbers required to be tested.
Cleaned all connections with electronics cleaner and blew them out with air. Made no difference. Electronics is my weakness. I will have to study the terms to make sure I am describing the tests done. I have a nice Fluke Multi Meter, I used it to check for OHMs //resistance and to see if wires are good. I have yet to determine what wires or pin connectors to test at the ECM Harness. Once I know the locations at the ECM harness I can test more and check for loose pins.
So I haven't read all of the details posted here. But it sounds like the ECU is turning off the injector based on a variation in the RPM. Could the RPM signal be the culprit?
You're going down the rabbit hole.....not good. You'll need to do (should have already done) a running compression test. I suspect you have a problem with the rocker arms or valve springs.
So I haven't read all of the details posted here.
But it sounds like the ECU is turning off the injector based on a variation in the RPM.
Could the RPM signal be the culprit?
yep, I think you're in the right ballpark with that thought of RPM sensing issue.
I called that "timing related".
Not the the crank sensor but cam sensors.
Not "goofy sensors" but bad corrective ECU data!
As far as I understand the cam sensor plate was loose and has now been welded back. The ECU has learned some crazy corrections that are no longer valid with new normal conditions...
Xentry is required to look at camshaft correction factors. Generic scanner is useless here.
BTW, the high LTFT is earned from the #1 injector being shutdown. At least this prevents melting cats with raw fuel.
This highlights the multiple injection shots are timed with the variable camshaft position (not so much the crankshaft position that has always been reliable. CPS is in the mix for timings but out of question here).
> "#1 Cyl. and not others" :
There's something on the cam sensor wheel that tells ECU what cylinder is which.
To expedite starting up with few crank turns, the ECU needs to identify #1 easily during crank. That's where the pain is.
ECU has learned a corrective factor that no longer applies to when #1 position triggers. The old correction with now good position simply fails timing validation checks.
Retrain camshafts sensors!!
✌️
Last edited by CaliBenzDriver; 05-26-2023 at 12:20 PM.
You're going down the rabbit hole.....not good. You'll need to do (should have already done) a running compression test. I suspect you have a problem with the rocker arms or valve springs.
Help me understand this, No noise from valvetrain, Leak Down Test showed near Zero leakdown in the #1 cylinder which tells us all 4 vavles are sealed tight, the piston and rings are sealed and not leaking into the crankcase.
I understand you need compression to fire a cylinder. When the valve cover was off I did observe the cam lobes, all looked good, I did see the cams opening and closing the valves as I rotated the engine by hand so rocker arms and lifters/cam followers were operating.
I know I am missing something or it would be working so I appreciate the feedback for sure.
Engine is asking for more fuel on bank one so the intake valves are opening and closing letting in extra air. The Fuel Injector is not giving it proper fuel. Computer has it shut off when the check engine light is on.
When I tried to take a compression test my tool would not reach deep enough into the plug hole. I did buy an extension for my compression tester but have not used it to check compression yet. My experience has always been a leak down test which takes more time and set up gave more information?
IF my leak down tools and compression guage used the same connections I would have tested both but the leak down hose will not connect to my compression guage.
I am exploring the Coil On Plug Wiring, I am thinking the COP is not firing correctly and causes misfires so the computer shuts off the fuel injector? Trying to trace the #1 Coil wires back to the engine computer but so far I have not figured out what Pins on the computer harness operate my #1 Coil to test. Next I want to trace the Injector wires back to the Computer. I think a pinched wire, loose connection at one of the splices or even a bad ground might be my cause. I also think it is possible the driver inside the computer for either the COP or the Injector might be the problem but no solution yet. I have tried loading the WIS on my Mac and on a WIndows 10 machine and it will not work so far. 2 years ago it worked but cant get it to work today. I have another copy ordered, will be in next week.
I do appreciate the feedback as I obviously do not know what the problem is at this point.
When I first received this car it had a check engine light. Could not pass smog. Ran Terrible, dead miss in the engine. Cams were not in sync so it had the camshaft error code. Found not one but both intake cam adjusters were bad, made the ugly snap noise when you rotated it by hand. Found 3 of 4 cam sensor wheels were off a little. I think any time a cam adjuster fails and starts making the loud rattle on start up the rattle makes the cam sensor wheels move on the shaft, my theory anyway. All 4 cam sensor wheels are now in sync with the cam, verified by the factory lines etched on the cam and sensor wheels. Replaced both cam tensioners, added the anti drain back valves to the oil supply holes at the tensioners. Replaced the Crankcase Vent at the rear of the valve cover because there was oil in the intake manifold.
On start up engine ran quietly but had misfire code #1 cylinder. Injector voltage is off compared to the other 5 cylinders, not sure why but it reads higher than the others. Injector has been changed, New Teflon Seals, Programmed new injector flow code to the computer with scan tool. No Change, So you are right the problem is #1 cylinder, Compression, Fuel or Spark, one of the three is missing to keep this cylinder running.
Now one more consideration, when the #1 cylinder is not misfiring the car runs fantastic, the O2 sensors read normally but when misfires start the computer shuts down #1 and the engine shakes and runs on 5 cylinders till the code is cleared. I think we understand the computer shuts off the fuel supply to #1 to keep the Cat from being ruined by raw gas.
A couple times the engine has run so good I thought it was repaired. Even took it on a 30 mile test drive in the hills and it ran perfect until I accelerated hard, let it rev to 5,500 RPM or so and it tripped the check engine light. Always due to misfires in Cyl #1 but it has run even climbing hills with no misfires showing? Before it tripped the engine light it had checked off all the Smog Check devices and was ready for a smog test. Then the problem came back. I don't think mechanical problems would do that, I think it is electrical but I have been wrong many times before. I should also share, 3 different Coils have been on #1, 3 different Spark Plugs have been in #1, 3 different Injectors have been in #1 but the problem is still at #1.
Last edited by Westlotorn; 05-26-2023 at 04:50 PM.
Main battery is a near new AGM in great shape, Aux battery was installed by me in the middle of this mess because I did not like getting the dash telling me the Aux battery was bad. Fixed the Emergency brake release while in there replacing the Aux battery. Checked all grounds I could find right away early into this project. It has drug on for a while, the owner and I separately had deaths in the family to deal with and then this COVID issue hit. 25 people that attended the funeral service are sick today. It spreads really easy, We did attend the funeral and not one person wore a mask thinking COVID was over. Not wanting to go political at all just stating, the car is just one issue I am working on but I would like to see it to completion.
Replace all coils? It has had coils replaced and coils that worked perfectly in # 2 and #4 were installed. The #1 coil works fine while in #2. The other coils work fine in the other cylinders but not in #1. Done with the Coil exams for now but need to verify the #1 coil wire harness is good and the computer feeds are operating correctly but I need to learn more and get the WIS to work to see how the wires are supposed to run and trace them. A real possibility is the ECM not working properly for #1 but I need time to get in there and trace it. I do connect a 25amp battery charger before testing to make sure battery drain does not ruin tests. With the key in the on position and the engine off the car drains over 20 amps while scanning it so adding a battery charger is a must. All modern cars are this way.
From earlier note: Retrain camshafts sensors!! Did that with my scan tool, repeated the process again to make sure. Scan tools shows a simple Yes or No for Cam validation Both banks show YES Cams validated. Shows the actual degrees of advance or retard while running also. Looks normal. Identical to my 2012 which runs fine. Sorry if I forget to report steps done, I have been into this engine bay many times over the last few weeks. I guess I could study the cam sensor wheel as it rotates and see if the sensor wheel was maybe bent a little throwing something off? A new idea for something to check. The sensor wheel has specific spots for sure where the sensor picks up the signals but no teeth for all 360* like you see on a crank sensor wheel. I will go back and check again because I know there was an option in my scan tool to use if a new camshaft had been installed.
West,
What DTC ( full info please ) is now causing the check engine light ? I forgot to ask that since the start.....
DTC may tell us if COP or injector is the one deemed as the first occuring problem by the ECM, if COP a bad dude, injector being turned off is part of the strategy.
I see you have not load test your wires and check for female terminal pin fitment, using that welding torch cleaner rod, since you do not have MLK 1.2 and SLK 2.8 male terminal.
On the M271 ECM, I got tricked by the female terminal ( harness side ) poor pin fitment towards the male one at the ECM side. I did a load test on the wire harness but because
I do not yet have MLK 1.2 and SLK 2.8 male terminal at that time. I used something fatter than MLK 1.2 male terminal to poke the female terminal and load test was hence good result.
I can't rip open ECM and dont want too, if it were other simpler module which is not glued, I would open up the module cover and access the internal side
of the male pin, something like below : To really know if the mating of the female and male terminal can take proper load, the green arrow is where best to sense/use, but must be very careful when injecting power at the green arrow, battery must be unplugged both + and minus and read schematic properly as to not damage other device/sensors/semiconductors.
Here is an example of bad intermittent contact which visually can not be seen and driving the senior techy nuts, till they use a set of jumper wires.
Actually the same happened to my OBD2 port on my car, but visually I can see the damage on the female terminal because its design is not like MLK 1.2 with internal "tounge" which we can't see much.
The OBD female terminal uses exposed twin tounge.
At under 300 milliamps, shake-shake/vibration wont produce power loss, but at 600 milliamps above, the bad female terminal simply cant send enough juice to my Xentry-Ninja at slight shake-shake of the scanner/xentry side of the male connector. https://mbworld.org/forums/e-class-w...or-repair.html
West wrote :
I am exploring the Coil On Plug Wiring, I am thinking the COP is not firing correctly and causes misfires so the computer shuts off the fuel injector? Trying to trace the #1 Coil wires back to the engine computer but so far I have not figured out what Pins on the computer harness operate my #1 Coil to test. Next I want to trace the Injector wires back to the Computer. I think a pinched wire, loose connection at one of the splices or even a bad ground might be my cause.
I have given you the COP and injector wiring schematic and point out which connector's pin numbers they are at the harness connector into ECM, also which X26 connector and
the Zs splice and the Ws.
COPs. Only the ZUE1 to ZUE6 is at the M connector ( harness side ). The Zs you need to find them within the harness. The W is screwed down somewhere on engine block.
The X26 is a stand alone connector. All documents to find them I or Juamnor have attached them earlier.
The X26 is a stand alone connector.
The harness side connector is M, the bigger/longer one, 96 pin one. Schematic called ENGINE CONNECTOR
INJECTORS, also at M connector.
Get that welding torch cleaner rod and start checking female terminal pin fitment first.
Sorry to hear that you and your family got Covid and thank you for sharing the circumstances that you did. My neighbor's sister and 3 friends just got it while attending a concert. So it is a reminder to me to remain vigilant.
I can remember years ago my Dad's '67 250SE would start to miss when I would get on it. This was an indicator that it needed a tune up. When I replaced the plugs and points it would run cleanly and even chirp the tires on the first to second gear shift .
So you are actually able to feel the misfire before it shuts off the no 1 injector?
Damn, had a response all typed and it vanished.
MB Nut1. Yes you feel the cylinder drop out. Engine runs smooth till the misfire happens, misfire triggers the check engine light and cyl 1 is dead till the check engine light code is cleared.
Codes are P030185. And P030085. Both indicate #1 has misfire condition.
Prihadi, Thank you again for your time sending valuable schematics. I have printed them and they help for sure. My second WIS program has not yet arrived. I am still trying to determine where each connection is and where the wires from the Coils plug into the ECU pins.
Unfortunately I have not had much time working on the car due to being ill but I should be able to get back on it today a little.
I did not have the welding tip cleaning tool in my assortment of tools so ordered one today. Looks like it may help with cleaning and sizing the pin connections.
The Dodge wire video posted above is a very nice training tool for electronic tracing, unfortunately I do not have a tool to scan the system like they did. Very interesting case for training though.
Last edited by Westlotorn; 05-27-2023 at 03:09 PM.
The more I dig into piezzo injector, the more I realized it is super interesting and VERY DIFFERENT from traditional magnetic solenoid injector.
Below document is the very same overall design GDI piezzo injector BMW is using as our M276, M278, M157 and M177.
It describe load test or voltage drop test on the wires too. https://automotivetechinfo.com/2020/...s-and-recalls/
Now I understand better what M157 document ACTUALLY meant :
Piezzo stack behave like a capacitor or easier to picture it also as a generator and can be a load too, but a capacitive one.
It produce voltage when pressure applied to it, or it can also change "shape" when current and voltage applied to it.
So for the voltage applied to it by the ECM ( BMW call it DME aka Digital Motor Electronic ) , it doesn't change polarity.
However the current "seems" to reverse polarity, when the piezzo being a capacitor storing power ....is discharging itself.
If on traditional magnetic solenoid injector, we get that reverse polarity voltage spike when the magnetic field collapse from this inductive design, but the current wont reverse direction,
the voltage is the one reversing its polarity because electromagnetic solenoid is also storing power and is discharging.
Damn, fuel injector driver is still a very high load for ECM.
It used to be 2 wire COP type having its driver inside ECM failing, we still need to worry now for injector drivers too, which is inside the ECM.
I sure hope your issue is not ECM internal, but more of the wires or its connection.
ONE WARNING FOR PIEZZO.
DO NOT ever disconnect the connector to a piezzo injector when engine is still running, to simulate a dead injector.
The reason is, if the piezzo is in FUEL OPEN state and you remove its connector, fuel flow will not stop and it will behave like a full flow open injector and fuel will
flood that particular cylinder up to liquid hammer damage. The piezzo has 200K ohms discharge resistor to dump the power/current the piezzo stack produce,
but it wont be instant. It needs the ECM to accept ( or dump ) its discharge current and then piezzo stack change shape and FUEL CLOSE happens.
Basically it needs current flow ( receive or discharging ) to open and also close. Unlike traditional magnetic solenoid based injector, if we remove power to it,
it will guaranteed fuel flow will be closed fast by the spring.
Our piezzo injector works very much like diesel one, but we do 200 BAR, diesel does crazy 1,000 BAR plus.
So if you want to study them, look at the Bosch ( we use Bosch ) common rail diesel piezzo injector, while different a bit in fuel flow region, the piezzo stack electric effect on is the same.
Master Tasos has once open up our type of injectors, but he was not aware of piezzo electric effect.
05-25-2023, 01:02 AM
, which I do not have.
, as insulation test on injectors will be on my preventive maintenance list.
, I am poor with PC software thingy.
.
