Cummins ISL9 CM2350 L101

Hey there everyone. I'v got a problem going on and was hoping to get some insight from someone that is familiar with diesel injector waveforms.
The engines this is specifically happening to at the moment is Cummins ISL9 2250 engines. These are all Gillig busses and this is the fourth one now.
Cummins and Management are all happy and fine with the ECM replacement being the fix but no one can explain why it is fixing it. I just found out this is the 4th one they will be replacing the ECM on so I wanted to do some testing first just to verify the problem for my own knowledge

The details

Code 1659 Engine Misfire Cylinder 6 - Condition Exists its usually associated also with a
Code 3714 Engine Protection Torque Derate

I only found out about it toward the end of the day. Here's what Iv checked.

The Engine does feel like it has a miss. Cylinder cutout test (unloaded and loaded) all 6 cylinders are contributing.
If I clear the codes and try to run an injector performance test it will rev up the engine then immediately stop the test. The 3714 derate code will be the only code stored.
I Cleared codes and took it on a test drive. It took about 15 minutes and the 1659 code with no check engine light.
When I got back I took the valve cover off and checked the valves. The intake valve was pretty tight by my feel but I was able to get the feeler gauge in it.
I had enough time to get waveforms of all 6 injectors. Voltage, control side and amp draw.
The only real difference I found was that the number 3 injector amp draw was around .9A while the rest are around 2A.
They all have around 7V bias on them.
They all pull to ground when turned on
They all have around a 50v spike on the voltage side and the ground side
They all seem to have good current waveforms, except that number 3 injector is low. But I don't think that is causing this problem.
All injector coils measure at 0.0 ohms

Things they’ve done before replacing ECM in the past was an injector reset in Insite and of course….a recalibration

Monday I am planning on monitoring number 6 while I first,
: try to run the injector performance test and see if that shows anything and then
: monitor it while I have someone drive it to see if that shows anything when it randomly sets.

I just wanted to post this here to see if anyone has had this problem before. Any help would be appreciated
Im attaching a pic of the number 6 injector waveform just for a quick look. If anyone has any input or any other questions just let me know.

The pic didn’t attach is the first post. So here it is here. Thanks again

Here are some known good waveforms from a test ecm. the only difference I am seeing is that the problem ecm is cutting off the voltage way to early

and here is another known bad injector waveform from the problem ecm. I have a lot more info and more captures if anyone is interested in a conversation on this problem

I guess right now the part I am confused about is the resistance. If you have 50 volts and ohms you would have a straight up amperage waveform. Seeing as how your amperage waveform still has a little bit of a slope that would indicate that you must not have zero ohms. I wonder if number three having a little amperage draw is a little bit of an indication on the number 6 condition. Are three and six companion cylinders on that? An issue with number three could cause an issue with number 6 and vice versa if that's the case. I know I'm asking more questions than I have answers but I'm definitely intrigued. Seen that the ECM is fixing them it could very likely be a capacitor in the ECM causing the issues.

Thanks for the reply!! I had a long reply to your comment but I’m having trouble posting it.
Just know Iv seen this and I will get back to you with other info!

Bryan Pollock wrote:

I guess right now the part I am confused about is the resistance. If you have 50 volts and ohms you would have a straight up amperage waveform. Seeing as how your amperage waveform still has a little bit of a slope that would indicate that you must not have zero ohms. I wonder if number three having a little amperage draw is a little bit of an indication on the number 6 condition. Are three and six companion cylinders on that? An issue with number three could cause an issue with number 6 and vice versa if that's the case. I know I'm asking more questions than I have answers but I'm definitely intrigued. Seen that the ECM is fixing them it could very likely be a capacitor in the ECM causing the issues.

So I thought that was weird too. But I measured 6 different brand new injectors and they all measure between 0.0 and 0.2 ohms. So I’m assuming that’s normal.
I know what is going on and can see what is going on but I can’t figure out how to see exactly why it is happening.

You can see a big difference between the cleanses of all the signals between the bad ecm and the known good test ecm pics I posted above.

It’s weird because if you look up the code description it would lead you to think it was a mechanical failure. And that’s the path I went down on the first bus.

I can also tell you that cylinders 1,3 and 5 are all on 1 “bank” and 2,4 and 6 are all on one bank.

The code it’s setting doesn’t make since you me because cylinder six looks and measures exactly the same as the others, mechanically and electrically. I wonder if it’s some default code for some sort of internal failure.

If you have any thoughts on another test that would be great!!

These are the things that keep me up at night

Bryan Pollock wrote:

I guess right now the part I am confused about is the resistance. If you have 50 volts and ohms you would have a straight up amperage waveform. Seeing as how your amperage waveform still has a little bit of a slope that would indicate that you must not have zero ohms.

Just like an ignition coil current ramp, the ramp is present in this diesel injector because there's a coil of wire inside (how many turns I dunno. it can't be too many turn or there's going to be too much resistance I dunno...) it's just a property of current going through a coil of wire, if this coil was laid out in a straight line then yes the current would be a straight up amperage waveform.

ScannerDanner explains it better for the layman - current goes through a coil of wire and the magnetic field builds slowly, and as the magnetic field crosses the coils inside the injector, it induces a voltage and current that will counter the incoming voltage and current and henceforth you have a ramp. I think that's how it goes lol.

Check out this pdf link at the end as i found that easier to read with some pictures: www.motor.com/magazine-summary/fuel-inje...a-thin-january-2005/

kerry.kozma wrote:

Here are some known good waveforms from a test ecm. the only difference I am seeing is that the problem ecm is cutting off the voltage way to early

At first one wire is positive 50V and the other wire is 0V and that's to open the injector, then the computer pulses it on the power side to maintain the current and to also limit the current. If the computer gave this 0 - 0.2 ohm injector full 50V and 0 volts on the other wire it would burn up quick. So it's pulsing it. Then at the end, the wires switch potential, again it shoots to 50V on one wire, 0V on the other. I'm assuming it's closing the injector when the computer is switching the voltage around like that. Because the voltage is switched around, current is still going up, but it's going down in the waveform because the amp clamp just needs to be flipped over.

So basically, 50 volts to open this sucker and 50 volts again to close it (close it fast?). Logically one would think if the computer is faulty, why is it closing the injector. It's like reading a book, getting a stroke and keeling over, do you have time to close the book? Um, no. Then again I never had a stroke and I don't know a thing about diesel injectors...

It’s getting a spike on the end because the magnetic field is collapsing. The ecm is using that induced voltage at the end of the event to recharge the capacitors for the next injector firing event.

How the injector works is not the question.
The question is if anyone familiar with this system knows why it flags a 1659 cylinder 6 misfire code when everything about 6 is mechanically and electrically identical. And why an ecm fixes the issue.

I thought the ohm reading on the injectors was odd as well, but checking 6 different new in the box injectors and they ohm the same.

Plus, if it was an injector problem, the other busses that had ecms thrown on them would have been back by now.

I understand how the current ramps work. The induced voltage when the magnetic field collapses is used to recharge the capacitor bank for the next firing event.

My concern isn’t the injectors. 6 new in the box injectors test the same.
If it was shorted injectors the other busses would have been back by now that have had this exact same problem and ecms put in them.

I’m just curious if anyone has seen these codes with the ecm fixing them and what is happening.
Mechanically and electrically the number 6 cylinder is exactly the same as all the other 5