Why is the ECM turning off injector 4
2012 Mitsubishi Triton 4G64 Gasoline 4 cyl 2.4 lt engine
Car has to heat up and sometimes has to be driven for over one hour on the highway before misfiring. The technician working on this car has replaced the ECM Spark plugs and pulled his hair out over this one. This tech is no dummy and he showed me the evidences as to why he changed the ECM.
So I drove 100 miles (160 kms) to his workshop to try and help him out and after an hours highway driving the problem Mitsubishi Triton I notice cyl 4 injector was losing amps compared to the other injectors. So once back in his workshop and applying heat to the injector we discovered we lose all voltage and amps and the injector was open circuited. I was convinced the injector had to be replaced and the tech ordered a full set of new injectors and fitted them. The Tech drove the car again on the highway without an issue.
After handing back the car now with a new computer, spark plugs, and a new set of injectors and the owner had driven the car maybe 100 miles and you guessed it his misfiring returns. below is an an email the customer (Len)sent to the Tech (Darryl) that is working on this car.
Appreciate the efforts with this one. Just about ready to toss it. I got to Caboolture and its now running as good as it ever did. But it is in the 100km/hr zone and no hills to load it up it started to misfire again.
Will see again on the way home on Friday.
Now the part I do not under stand and would like some insight into this matter.
I have attached a capture of injector 4 being turned off while all the other 3 injectors are working (there are times that all injectors are off when coasting down hill without any throttle pedal on which is normal) But if you look at my capture you will see the 2 black arrows were it is clear that injector 4 has been switched off while all other injectors are operating. What I do not understand is if the ECM knows to switch off cyl 4 injector then why is a P0300 being set instead of a P0304?
The second thing I do not understand is I would have thought once the injector is turned off due to an engine misfire, I would not have expected it to turn back on until the ignition is cycled. Here we see it is only switched off for a very short time and then it is back on. Of cause all the injectors turn off as the car is coasting so maybe the software allows the injector to come back on once all the injectors have been switched off and then back on? I do not think this is the case, but I cannot get my head around this and it is the only thing I can imagine that might explain this, other then a fault in the ECM, but with a new ECM and good powers and Grounds I can not see how the ECM can be at fault. It also could be the computer is no longer seeing a misfire and so it allows the injector to come back on, but in my experience I thought the key needed to be cycled for the injector to come back on as long as there were no more misfiring.
The other thing that I am struggling with is under hard engine load conditions I can see multiple injector events and perfect cam and crank signals. The capture I have attached only has 2 extra events but I have caught up to 3 and 4 events happening between 720 crank rotations.
I have also uploaded the capture of losing number 4 injector when we applied heat to the injector and it opened circuited, hence the reason the injectors were replaced.
Thank you all
Can you do me a favor and send some screenshots of the freeze frame data for when the p0300 is setting? That would help tremendously. I want to know the engine run time, coolant temp, tps,rpm, and load pids during the ff event and if you have several events that would be great so we can compare and see if they all match. Also, how many miles are on this thing?
Just a thought, but heating the injector and duplicating a misfire I would check connections at injector and load test wiring. The ecm strategy could be looking at injector load.
Good point John if the connector is not a tight connection and by adding heat this would only serve to expend the connection even more. I will talk to the tech and get him to check this out. In fact this would explain why I am losing injection at times and not other times. How do you explain the multiple injection events I get some times?
Hi Robert, Thanks, I went through my scan tool and even though on the ay I did look at my freeze frame data it seems I did not record the image. I will contact the tech I was helping and get him to send me a copy, it may take a day or two.
Maurice, you have certainly done your homework. Fairly modern vehicle, could it be using a wide band O2? At cruise it is going to try to run as lean as possible. Maybe a bit too lean?
Thanks Robert, this is a theory I have thought of and when I went for a road test with Darryl the tech working on this car, I had my lap top on lap catching the scop patterns and my scan tool also monitoring what the O2 sensors were tell us. Part way in the trip I lose communication. I would regain communication and then lose it again. So in the end I had to give up with the scan tool. I am not…
Hi Maurice, I don't know a lot about this car but the loss of comms concerns me a little bit. Have you done a full car scan just to see what the other modules on the CAN line are doing? Crazy thought here may be that possibly when the the scan tool looses comms, maybe that is when the injector may cut out and then after comms comes back, the ECU may turn the injector back on? Just my two cents…
Maurice, when you note "perfect cam and crank signals", are these measured at the sensors? What if they look different at the ECM? Could there be some wiring issues or interference there? I look at the "heated injector" capture and see a lot of hash in the injector control, while the ignition control trace is clean as a whistle. Hence my thoughts about the interference.
Thanks Dmitriy we were taking all our captures from the ECM. And yes I notice the same, so many strange things about this case and the reason I have out this out to you guys.
Dmitry brings up the excessive "hash" in the voltage trace, in capture 3, in CH D. It appears to be a poor scope lead connection? Scope the injector amps again and apply the same heat to injector 4, the problem could be the injector connector or terminals.