P0303 check engine light when cold.

I read through everything to be sure before I posted. FYI, on the 2.7 engine there are TWO different ECT sensors. One, the one located near the front of the cylinder head, is for the gauge. The second one (the one located on the BACK of the cylinder head, by the firewall) is the sensor that the computer uses for all fuel calculations. You cannot find that sensor data with a normal, handheld scanner, or a ultragauge.

I suggest replacing the sensor on the back of the cylinder head and go from there.

 

Thats the one I originally suggested. The one on the back by the firewall

 

Sounds reasonable, but given the ECM is reporting accurate engine coolant temps to the scanner is this sensor still suspect?

I'm using a laptop computer and ScanTool.net OBDLink interface cable with their ScanXL Professional software, so I'm reading PIDs right off the ECM data stream.

edit to add: given BamaToy's credentials, I'll be ordering up a new coolant temp sensor as my next step after the MAF/IAT swap is done.

 

As I mentioned, the sensor on the front of the engine is the one that you find on the data stream, and is used for data display and gauge reading. The one on the back of the head is the one that is used for fuel calculations. It is not uncommon for that sensor to fail and cause extreme rich, or extreme lean conditions depending on how it fails. I have never seen that sensor data on my diagnostic tool that I use in my shop.

 

Thanks for explaining. I took your previous advice and picked up a new sensor at the dealer this evening on the way home. $66 list price but I get a decent discount. Parts counter guy said "Your truck's running bad in cold weather, huh?" I described what it was doing, he said they keep 4 of these sensors on the shelf this time of year just for this problem. They didn't have the crush washer but he said they clean and reuse them without problems.

Just got done installing it...that job's a real wh0re...and burping the air out of the radiator. No visible sign of damage on the factory sensor but I'll test resistance some other time.

Test drive was good, but I had the batt disconnected during the sensor swap and the ECM reset. I took a test drive, monitored the data with no glitches, and parked it. Restart was immediate, no long cranking which is encouraging. Tomorrow morning's cold start should tell the story.

Thank you for taking time to share your expertise, and thanks to offrdmania for the original suggestion. I'll post an update tomorrow.

 

Good luck. Hopefully your truck will start fine for you in the morning!

 

Hopefully start fine...not set the Check Engine light...not spark knock up hills. Something to look forward to! Darn thing was misfiring so bad this morning the Check Eng light was flashing.

Thanks again, fingers crossed, MAF on the way just in case.

 

Update: It appears to be fixed! Engine started at the first bump of the key, no Check Engine light, no spark knock up hills.

It's snowing pretty hard here so it wasn't my usual work commute, had it in 4-hi and driving 40-45mph on the highway.

I hope this thread might help somebody down the road...I never would have found this coolant temp sensor problem without the help here and I appreciate it.

 

Glad its fixed OP

 

Me too, thanks again for your help figuring it out.

 

Anytime buddy

 

That is what we are all here for. Glad I could be of some help as well.

 

Evidently I spoke too soon...yesterday after work it was a long crank to start, rough idle, Check Engine Light with P0300, P0303, P0304 same as before.

I've discovered that resetting the ECM is the temporary cure for all this. Long Term Fuel Trims at idle are in the -15% range, but when I pull the ECM fuse to clear out the stored LTFT data it starts immediately runs like a champ at 0% LTFT.

I'm also getting audible spark ping up the steep & long hills again.

So the mystery continues. It looks like something is feeding bad data to the ECM, causing bad LTFT values. It sure isn't the Coolant Temp Sensor at this point.

Any suggestions continue to be appreciated.

 

bummer--here is a bump

 

Update: I installed the new MAF this evening, cleared the ECM by removing the fuse, and took a test drive. Of course it ran great as expected after clearing the ECM. Recorded fuel trim, intake air temp, etc. showed same as with the old MAF. Only odd things I observe on the scanner:

1.) Calculated Load is 25-30% at idle, in gear (auto trans), sitting still.
2.) Throttle position is 11% at idle, and with engine shut off.
3.) There is no Manifold Absolute Pressure sensor on these engines?? I'm a GM guy.
4.) Long Term Fuel Trim still shoots down to -11 to -15% at idle.

We'll see how it drives tomorrow, didn't pull the ECM fuse tonight so it'll be starting with stored ECM data instead of a clean slate.

 

It appears the front oxygen sensor may be the culprit. The symptoms returned this morning, same whacky low Long Term Fuel Trim at idle and spark knock up hills, with long crank time to start the engine. I put my old MAF back in and replaced the front O2 sensor...last piece in the Closed Loop puzzle...even though there was no error code and it seems to be working fine. We'll see in the morning, as I did not pull the ECM fuse and reset it this time. Long Term Fuel Trim percentages are more normal after a ~10 mile test drive at -7% idling and +2.3 to +3.6% cruising. Engine is smooth at idle. Fingers crossed, it's getting way too cold to be out in the garage turning wrenches at 18degF ambient temp.

 

Are you loosing any coolant?..with those cylinders being adjacent to each other..could be a slightly blown headgasket.you said they leak tested it,but did they pressurize #3 and #4 and check for leakage into the cooling system?

 

I had an O2 sensor go out too and it didnt throw a code either

 

I can safely say the O2 sensor replacement solved the engine behavior problem. Three trips totalling over 140 miles, not a hiccup.

The question now is what killed the sensor, as it was only 2 years old. Denso direct-fit replacement piece from Advance Auto Parts.

Two lines of thought:

1.) I had sprayed the frame rails with Fluid Film a week or so prior to the onset of the misfiring. Didn't spray the O2 sensor directly, but it was exposed to overspray from splattering off the frame rail. The reference side of the O2 sensor could have gotten coated, which could have affected the O2 sensor's reporting.

2.) Coolant contamination? I'm still settling my reservoir level from Monday's coolant temp sensor replacement. At operating temp, the level is *exactly* at the top line molded into the reservoir tank. I'm keeping an eye on it, hasn't changed since Tuesday when I topped it off at operating temp. Tonight I'm planning to pick up Auto Zone's loaner cooling system pressure tester kit to see if there's a problem before I kill another $160 O2 sensor. Plan is to test at cold temp in the morning, then bring the engine up to operating temp and re-test.

 

I snagged the loaner cooling system pressure tester from Advance Auto on the way home, popped it into the radiator neck and pressurized to 0.9 on the gauge while the engine was hot. Same pressure stamped onto the rad cap. It held 0.9 for just under 2 minutes while I stood there freezing. Terrific! I left it on, went back out at 26mins and it had dropped to 0.75 on the gauge...presumably due to the engine and coolant dropping temp. It's currently 19degF ambient, pretty cold. Checked again at 34mins and it was still just above 0.7. Engine is still warm to the touch but not hot. I'll leave it overnight, take a reading, then re-test cold.