Westlotorn

Yes, I pulled the #1 coil removed the #1 spark plug, by turning the coil I had access to hook my timing light to the coil wire and picked up the signal needed to run the timing light. Engine running spark plug sparked very well, no issues strong regular spark, timing was 18* BTDC so all good. Now I need a working injector.

Westlotorn

Tightened the pins on the #1 injector. No joy. #1 cylinder still totally dead. I guess a new PCM is in my future for this car and no guarantee that will fix it. If these cars did not have the anti theft feature I could take the PCM out of my 2012 and test with it. That would be too easy.

I guess I could take apart the PCM harness and check to see if signal is being sent out and if the injector is getting 12V from the computer. More reading to prepare for that test.

chassis

Did you confirm #1 injector is fully functional?

juanmor40

Post #1, injector is new OEM, coded later in another post.

Two bad injectors? If true, something must be damaging them fast.

Still valid question after all the disassemble and reassemble

Bizarre failure

S-Prihadi

At this point, if both female terminal MLK 1.2 at injector #1 connector both are good and tight ( assuming shims works ) and also female terminal at the ECM harness side, pin 2 an 50 ( injector #1 set ) both tight and good,
the next step is to scope the injector voltage or do current clamp measurement also using scope at injector #1 connector, like what I did last nite...... to test if indeed ECM send signals/power to fire injector #1.

That would be the last test before ECM is to be condemned.


Side Note :
Now I realized why my injector voltage signal is so noisy on the scope, my test pigtails is not twisted wires where the schematic pointed out all injectors wires are twisted like CAN BUS ones to reduce electrical noise.

CaliBenzDriver

Me other PLAN B :

-- Read Bosch design info: available WIS docs help us understand what input is causing the ECU software to decide to shutdown Cyl. 01. Once you have true design expectations, it helps kill inapplicable assumptions to a smaller number of "what if's scenarios..."


-- Good ECU Hardware: as much as I understand this trouble, Cyl-01 works great until the ECU logic flips a switch to deactivate it. Find out what drives that logic... bad sensor input, bad data... Nothing here says bad MOSFET COP driver to warrant ECU hardware replacement


-- Wasted Gear : would it be really too far fetched to replace the previously known bad "sensor plate" ?? Engineered part failure is intended to be replaced with new spares.
Field bush-fixes sometimes deliver well, other times not so much.

Hope good assumptions help see new light.
🤞


timing relationships to CPS

MBNUT1

So I am trying to reconcile this with post #78 reporting the #1 plug as being wet.

Westlotorn

.
When this project started I found lots of oil in the intake manifold. I poured out maybe a half pint of oil. There is still some oil in the manifold. Each time I pull the intake I still find wet intake valves from the oil being pulled in. I did replace the PVC or crankcase vent which is under the High Pressure Pump so the source is dry now but the intake is a little wet if you will. I see a few drops of oil in each intake port each time I remove the manifold. The engine does not have blow by so I think the new PCV is working fine. I think the wet plug was oil not gas. I remember smelling it and nog smelling gasoline. Probably should have flushed the manifold with gasoline or solvent and followed that with a warm soapy water wash but with all the other issues to repair I thought I would let this part self clean as it ran. My 2012 also had lots of oil in the manifold when torn down at 87,000 miles for a cam adjuster repair. The PCV vents in these is not good enough. I know others have installed catch cans but I did not.

MBNUT1

Ok thanks. That clears that up. Can you do a "bench exercise" with the # 1 injector. IE take a spare injector (or better yet pull the current injector and replace it with the original one) and connect to the #1 injector connector and see if it clicks? This assumes that the motive force for the injector action is electrical only and not dependent on the fuel rail pressure.

Westlotorn

Today I went back to the beginning. Hooked up my scan tool and started searching for anything that might help diagnose the issue. At this point #1 injector has been dead. It was dead on start up. I drove it out of my garage on 5 cylinders. There is an option in my scan tool to use after a new engine installation. I forget the wording but it clears the old engine history and lets it re program on the new start up. It only allows this test if the engine is cold. So I ran the Re Learn test and the #1 Cylinder starts firing again. Check engine stays off, I watch the misfires and #1 still misfires but not often and only 1 at a time. Not consistent like the other day. So with it running on 6 cylinders I went into the fuel injector test mode and started tuning #1 on and off. I was able to command it off and command it on each time. The power drop off seemed the same as all other cylinders?

I start thinking progress. I verified the cams were in sync and checked the fuel injection voltages, #1 is still running higher voltages for the tests I have available in my scan tool. The low voltage is higher than the other 5 injectors. The high voltage is higher than the other injectors. The smooth running of cylinders test shows the #1 is trying to inject fuel much longer than the other 5 cylinders while the O2 sensors show the right bank is running lean and keeps increasing the LTFT. Note: After doing the camshafts I had it running exactly like this but after a while or hard acceleration the check engine came on and #1 would be shut off till I cleared the codes again. So all work so far has netted no progress in my mind but I know we have solved some problems in this car, just not my #1 injector issue.

After about a one hour run time it set the check engine light again for misfires in #1. I could not get it alive again by clearing codes. On the next cold start I can run the reprogram option and I bet it comes right back to this state. By blipping the throttle while parked I was able to introduce #1 misfires each time so the problem exists and I think it is lean fuel supply in #1. Since I was able to command #1 off and on it seems the computer is functioning till it gets ordered off by the check engine light. I know the wires to the injector are good, I know the connection at the injector is good. I do not know if the connection at the computer is good enough or if the computer might have an issue causing this problem. You guys must be worn out by now. I am certainly tired of it but I keep plugging thinking I have never bumped into a car issue I could not fix, can’t let this one stump me.

MBNUT1

Keep going. What is LTFT? Which bank is left or right? So help me understand the implication of the high voltages of # 1 injector. Does this imply a high resistance to ground or is the injector driver outputting higher voltages and are higher voltages more fuel? And if that is the case why is the O2 sensor indicating lean?

At this point in the investigation I think of a guy I used to work with who would say there must be a pony somewhere in this pile of ****.

S-Prihadi

01 and 02 in red is a good news.
This reminds me of the M271 after new ECM install and the mild misfire occurred later on after engine tested on the road well the day before and then been idling 30 over minutes at the garage the next day.
The reason was that MLK 1.2 female connector at the wire harness side need "reconditioning" because it was not biting enough towards the male terminal on ECM side.
Now the M271 have done 15,000KM on the new ECM and that MLK 1.2 female terminal I DIY reconditioned....and so far so good.

The difference between your M276 and my friend's M271 was that, the M271 issue was from COP driver inside the ECM itself , erratic , will not power up to 3 COP after being driven some distance.
So indeed that M271 had a bad ECM but erratic bad .
For now, your issue seems to be injector related.

Until today, I never could get the true answer as how would an electronic semiconductor inside the M271 ECM be erratic as in OK and then NOT OK.
I mean I would love to see it simply fail and NOT OK permanently, as to not confuse me. Maybe bad solder joint inside, I don't know.
I am an analog guy, I can see if switching is done by a relay or a contactor , the usual culprit would be the contact points no more perfect after so many arcing events and causing bad contact sometimes.... at least I can see visually.
With transistor semiconductor, we can't see if the gate ( equal to contact points in a relay ) is OK or not OK.....unless the physical body of the transistor melted or visually damaged.
I hate electronics too.

The second kinda erratic OK and then NOT OK problem was with my own tranny pedal shift module called N135 which can shift down on its own.
This is a super low power device, so simple and no sign of bad soldering or anything visual. I replaced with a new one, but never could figure out what actually went wrong at them semiconductors on that board.


West,
I ran my Launch Creader Elite scanner, it does not have the injectors voltage data. Dang......
If you don't mind, can you take photos of the pages in your Launch scanner to the point you can see injector voltage data ?
ACTUAL DATA >>>what page/s next ? >>>> Injector Voltage


I know our engine is different, but I ran the Xentry simulator for M276 3.5NA and it is also using MED177 ECM like mine, and I can't find injector voltage data.


Thanks

Westlotorn

I actually took some video today, can that be loaded? I do not know this sites file sharing limits. The video was too large, cant load it. I have never done youtube I see some guys just upload it there and share a link but not a skill I have learned. I did not take photos just the videos which I see I can't share easily. Forum has a 20mb limit for uploads. My video was near 100mb. Posted this note last night but somehow I did it in another discussion. The other guys are probably scratching their heads.

chassis

Create a YouTube account

Upload the video at YouTube

post a link here to the video

Westlotorn

Went to re take pictures this morning, got the car cleared of codes, did the engine new install reset once again. Fired it up and it ran really good. I did not try to make it misfire, just let it idle. Everything was perfect to start getting the pictures and my scan tool announced a new update is available, I ask to do that later, it locked up. It has been in a loop refreshing ever since. 3 hours so far. It did this to me before so I shut off its internet connection so I could control when it got its update but I forgot to shut it off again after that update. Dang. With it running perfectly, no sign whatsoever of misfires I used my older scan tool just to check what it offers. It shows the Long Term Fuel Trim (LTFT) on bank 1, passenger side, cylinders 1-3 running at +18%. Tells us the computer is trying to richen that bank up, it is running too lean. I think I am still not getting enough fuel from the #1 injector OR, the intake manifold has an air leak into #1 cylinder. 3 injectors have been installed, I do not think this is an injector issue. Either the wires are not getting the proper signal to the injector from the ECM, the fuel rail is restricting fuel to #1 only, or I have a large air leak into #1. I had looked closely at the manifold and did not see anything odd. It would take a large leak or a ripped intake gasket to do this. I have had the upper manifold off numerous times, maybe I need to pull the lower and inspect those gaskets but they are brand new with only one install?

MBNUT1

MBNUT1

Thanks for making it all clear. Are you able to determine if the voltages at the injector are different for #1 than 2 and 3? Does the ECM use the same driver for all of the injectors and distribute the output or are they individual drivers? The reason I ask is because if it is the latter it could also be a driver within the ECM as I think has been previously mentioned.

Westlotorn

I sprayed brake clean all around the #1 cylinder/intake manifold while running, idle did not change so no air leak. Dang. I fired it up again and let it idle for 30 minutes. It ran perfectly, zero misfires but after 30 more minutes running it is still trying to richen up bank 1 cylinders 1,2 3. So a problem exists. I got the scan tool update done and looked at several running scans. I will show pictures to show what the tool allows. I will try and post the menus available to choose from also, this is only a small part of what the Launch X431V+ can do.

chassis

I don't see major issues with the data.

You say you have it running well now.

How about letting it alone for 3-5 days and see how it goes?

Westlotorn

The LTFT is so high it will throw a code but I do plan to do a test drive and hope the #1 injector starts working better but I do not see that happening. I did pull the PCM connector and tightened up Pin 2 and Pin 50 with zero change. Runs the same. I have some injector cleaner in the fuel so who knows maybe something will get better!

S-Prihadi

I am drooling over this data capability of ur scanner..........

S-Prihadi

The fuel trim re-learn may help you speed up the correction to the LTFT +20.1% on Bank 1, assuming now car combustion at cylinder 1 is already better.

MBNUT1

So you insightfully mentioned a potential blockage in the injector rail. While this seems unlikely it may be worth a flow rate test where you remove #1 and then another most convenient injector and run a tube into some sort of measuring cup and compare the filling times. This off course would be done with the engine not running but the key on. I am assuming that the pump will provide pressure with the key on.

juanmor40

, and before disassembling measure fuel rail pressure at the Schrader valve with the key ON.

MBNUT1

Really good point because if the overall pressure is marginal then the last injector along the rail will see the lowest pressure especially at high flow rate, Though just to say it they will all see the same pressure when not running and one would think the losses would be minimal at idle flow rate, which tends to point away from a rail problem if the LTFT is being set by idle performance.