Cant find cause of rough idle (warning...longwinded)

Hello everyone,

I have a 2002 Ford F150 Lightning (5.4L Supercharged).
I purchased the truck about a month ago with a rough idle. Doesn't stall, just a rough idle.
Sometimes it is rhythmic and kind of rocks the truck back and forth a bit...sometimes it feels more violent.
I would also say my mileage is down, but haven't had the truck long enough or ever running better to compare.
The engine roughness seems to smooth out as soon as I touch the gas pedal.
Seems to have good power, but again I have nothing to compare it against.

Truck is completely stock to the best of my knowledge.

With no history I started by pulling codes. Nothing CEL, but had a pending code for PO125.
My truck has the CHT sensor and to the best of my knowledge no coolant temperature sensor.

Decided to change the CHT sensor...and there is no way my hands are fitting under the lower intake, so stripped the intakes off.
Changed the spark plugs (old plugs looked to be burning pretty well) and the CHT. Cleaned the intercooler while I was there.
Vacuum line behind the lower intake (known to be problematic with these trucks) was in good shape, so put that back rather than plug it.

All back together...and it runs exactly the same...and I still have a pending code for PO125.
According to my scan tool the coolant temperature never gets above 160...thinking someone may have changed/gutted/removed the thermostat.
Changed the thermostat. It was a 180 in it that I am assuming was stuck open. Replaced with another 180 and the temperature gauge now goes halfway up the gauge (compared to only going up 1/3 on the gauge with the other thermostat) and the scanner now reports that it was running at 178 instead of 160.
Cleared the codes and so far no more PO125. Pretty sure that will take care of that.

Checked the TPS and made slight adjustment at idle from .96 to .99 volts.
Cleaned the IAC and made sure it operated freely.
Vacuum gauge says I'm idling around 18.
Cleaned the MAF sensor even though it looked perfect, no change.

Checked the voltage of the DPFE sensor. Read it should be under 1 volt...was reading 2.2.
Changed that out and have 1.04 volts at idle. Grounded out the EGR pressure solenoid and the DPFE voltage climbs as it should and the engine bogs down...so that seems to be operating normally now.

I read that the coil boots can sometimes cause issues, so I purchased and changed all 8 COP boots and also swapped the #8 COP with #5 since cylinder 8 had the misfire counters on it.
No difference in drivability and cylinder 8 is still reading between 80 and 245 misfire counters.

LTFT at about 11 on bank 1 holding steady at idle, but drops to 2 or 3 if I hit the gas. STFT going between -2 and +3.
Bank 2 LTFT is 3.1% and STFT is -0.8% which seems to look normal.
Both front O2 sensors switching between .1 and .8 volts.

I went through nearly a full can of throttle body cleaner trying to look for possible vacuum leaks, but no matter where I spray I don't notice a difference in idle quality or engine sound.

I put together a smoke machine to test for vacuum leaks as this seems to be what the fuel trims are pointing to.
Attached the smoke line to the brake booster line and found smoke coming out of the vent cap on the side of the IAC valve. Not sure why that little cap/vent is there? Either way smoke came out from there and if I spray throttle body cleaner in the vent it sets my O2 sensors to rich.

I also had smoke coming from under the top of the EGR valve. Not from between the intake and the valve, but from inside.

Changed the IAC last night...no difference in how the truck runs.


Sorry this is longwinded...just trying to walk you through what I have done so far.

Any help greatly appreciated.

Anthony

have you checked mode 6 for misfire data? also have you tried pinching off the brake booster hose and recheck fuel trims? to me when you say it runs better when you give it gas than at idle im thinking of a sticking open egr valve or compression issue.

also maybe try an injector balance test

cheryl hartkorn wrote: have you checked mode 6 for misfire data? also have you tried pinching off the brake booster hose and recheck fuel trims? to me when you say it runs better when you give it gas than at idle im thinking of a sticking open egr valve or compression issue.

Before the plug change Mode 6 showed misfire counters for 4 different cylinders...after a plug change only cylinder 8 had counters... and between 80 and 245ish for cylinder 8.

I pulled and plugged the brake booster line just to eliminate that as a possible cause of leak.

I blanked off the EGR to eliminate that and it made no noticeable difference.

Really hoping it isn't internal, but guess I will have to run a compression test to find out.

Thanks,
Anthony

cheryl hartkorn wrote: also maybe try an injector balance test

Thanks for the idea...will have to see if any of the "scanners" I have offer the balance test.

Anthony

probably not. youll probably have to get one of the old school injector pulse tools that otc made. there available on amazon.

try resetting the K.A.M? should let the fueltrims relearn after the repairs you made

Does temperature seem to make a difference in how bad the misfire feels? i.e. is it worse when hot?

Even if you don't have a scope, you can always do a relative compression test by listening to the engine during a no fuel crank. If you hear a 'skip' at regular intervals, it might be worth doing compression testing. Here's a video with a good example if you're unfamiliar:

How does the vacuum gauge needle behave, apart from inicating 18 inches HG?
Is it steady or flickering? A valve sealing problem will be notorious at idle and improve with rpm.

Any chance you can do a leak down test? Or do you have a scope??

cheryl hartkorn wrote: try resetting the K.A.M? should let the fueltrims relearn after the repairs you made

I reset this and the numbers quickly return to where they were.

Anthony

Tyler wrote: Does temperature seem to make a difference in how bad the misfire feels? i.e. is it worse when hot?

Even if you don't have a scope, you can always do a relative compression test by listening to the engine during a no fuel crank. If you hear a 'skip' at regular intervals, it might be worth doing compression testing. Here's a video with a good example if you're unfamiliar:

I dont notice the idle when I first start the truck, but when high idle settles down it becomes noticeable...dont know that it gets any worse with heat.

Anthony

juergen.scholl wrote: How does the vacuum gauge needle behave, apart from inicating 18 inches HG?
Is it steady or flickering? A valve sealing problem will be notorious at idle and improve with rpm.

Any chance you can do a leak down test? Or do you have a scope??

The needle moves around with the idle, so it is flickering.

Decided to run a compression test ... not good.

I don't know what PSI a good cylinder should run and don't know how accurate my gauge is...
This was run with engine cold ... Im too slow at pulling these plugs to run this warm.

1-113
2-104
3-110
4-101
5-62 dry, 80 with oil
6-115 dry, 130 with oil
7-50 dry, 65 with oil
8-100

Added a little oil to 5, 6, and 7 and got the numbers above.

Did a leak down on cylinders 5-7.
5 seemed to lose a lot through the crankcase and exhaust
6 was fairly quiet...just a little through the crankcase
7 had a little noise in the crankcase...but was surprised I didn't hear more elsewhere with the horrible compression number.

So ... now I have to figure out what to do next.

How horrible are 1, 2, 3, 4, 6, and 8?
My searches show people talking compression numbers from 150 lbs to 250 lbs per cylinder...and I have 115?

I was hoping cylinder head for all issues so I might not have to fight with the bottom end...but with compression numbers like that is it worth putting a head on this engine?

What should the next step be?

Thanks,
Anthony

I would go with a leak down test. See if the compression loss is from valve issues or worst case, piston rings. If you're "lucky" it'll just be leaking past the valves. Might not be the right next step, but that's what I'd do next.