[FIXED] Long cranking injector pulse width (60-80ms)

Spoiler - A new computer fixed it. No codes, no odd sensor readings, just too much fuel while cranking for some reason.


Have a 2003 PT Cruiser 2.4 non turbo, automatic. It is a no start, cylinders are super wet with fuel. After some poking around i did get it to start by disabling the fuel pump, and giving it a bit of starting fluid. When i say start i mean it ran on 4 cylinders until the "fuel" was gone. So i assumed at that point it was getting too much fuel, i back probed the ecm injector outputs, and i got 60-80ms pulse width while cranking!! (see attachments) The data PID for the "injector 1 pw" and "injector startup pw" are both ~76ms.. I tried resetting the computer with the scan tool (Autel MS905), no luck there.. None of the sensor inputs look to be out of whack, but i did notice both o2 sensors reading 5v KOEO, and the "bank 3" fuel trim was -100, but it didn't change with "resetting" the computer. I have checked powers and grounds to the ecm. At this point i want to say its a bad computer, but i want to throw this out there and see if anyone has seen this before.

Thank You!

Done

Out of curiosity what are the MAP or MAF reading while cranking? Just wondering if it's seeing a false reading of huge airflow. Just a thought. Also does the rpm pid look about right on cranking.

See attached image. I know the cranking rpm is a bit low, the battery isn't that great, i had it good and charged when i initially looked at it and the cranking speed was over 200rpm. This vehicle does not have a MAF, and i could not find any kind of estimated air flow value besides IAC estimated air flow, which i forgot to look at.. Might go do that.

If there's no Maf it'll just be off the Map. I'm almost certainly chasing something not there then, was just a thought. I've seen ecu's hold injectors on constantly, or not fire them at all, just seems odd to fire them fine just for too long.
My thought was if the computer was seeing a crank speed of 6000 or something crazy it may give a matching injection pulse.
Just a thought anyway.

I'm going to call this computer bad, reman on the way, I will post the outcome.

A few things. First it is normal to see 5v on the O2s on a Chrysler with the sensors cold.
Second, are you positive your ECT, IAT and battery temp sensors are reading accurately?
Is this a flex fuel engine?

I have checked them with the scan tool, and all look to be good, on dead cold startup they all three read very close. I suppose i should probe them and make sure the computer isn't making up values. I do not believe it is flex fuel but i will double check tomorrow.

Nothing i can find on the car or in the PIDs suggest it is flex fuel.

Here is what i got for ECT/IAT/bat temp

Sensor Actual/PID-Volts/PID-temp
ECT 4.1V/4.33V/41F
IAT 4.17V/4.1V/54F
BAT 2.988V/2.98V/43F

I installed the reman computer, started right up, i don't have time right now to drive it and make 100% sure its fixed, i will drive it some tonight and see what happens. On first start it had 50ms cranking, 3.?ms running, i forgot i had an open pcv hose, took care of that and we have 20ms startup now, will see what it does later.

jds8086 wrote: I installed the reman computer, started right up, i don't have time right now to drive it and make 100% sure its fixed, i will drive it some tonight and see what happens. On first start it had 50ms cranking, 3.?ms running, i forgot i had an open pcv hose, took care of that and we have 20ms startup now, will see what it does later.

Wow, congratulations! :woohoo: I gotta be honest, I've NEVER paid attention to the actual pulse width PID during cranking. I might glance at it to see that it's doing something during a no start diag, but not what it's exact value was.

I wonder if there's a good 'guideline' to go by? The 50ms number you found after PCM replacement sounds reasonable, but I'm sure IAT, ECT, and engine displacement go into that calculation, too...

I let it get up to temp, and took it for a drive, it looks like we can call this fixed. The "cranking pulse width" PID is around 50ms (at ~30F) which i guess is fine as it starts right up. Thanks everyone for your input!

I just ran across a similar post on IATN about an Escape with a similar problem.
Can I post a screenshot of that without getting anyone in trouble?

I blacked out the names, we can call them tech a and tech b I guess.

Here's the link for other IATN members.

"No start, Looks like a jumped chain"

autop.ro/eIv

Noah wrote: Here's the link for other IATN members.

"No start, Looks like a jumped chain"

autop.ro/eIv

Nice find! I DEFINITELY would have chased a timing problem if I'd seen that relative compression waveform.

The original poster mentioned the max cold injector on time is 20ms... I wonder if he was using that as a general guideline, or on this specific engine? :huh: Either way, it seemed like that was a critical piece of information in his diagnosis. Without it, I think it'd be easy to go down the wrong path.

As usual, I had to try this on my personal car. This is a cranking shot, engine cold:



Forgot to get the ECT in there. :blush: Figure about 40 degrees. This falls within the 50ms value that Jake found earlier, so I'm pretty happy with that.

Based on what we've learned, I think I'd be comfortable using a rough guideline of between 20 and 50 Ms on time while cranking. (For now of course. I realise that's quite a span of Time) I'll be paying more attention to this in the future for sure.
Thanks Jake!

50 m.s. sounds really high too! Had an old GM TBI with a 40 m.s. p.w. from a -40 ECT reading flood out the engine immediately. But then again, a big factor here is the flow rate differences of injectors themselves. I don't think the really is going to be a very good "ballpark" number.
Thanks for the update jds8086!
Nice job brother!