'05 5.3l mysterious misfire

[LSXapache]

I've been trying to diagnose what feels like a slight misfire in my lm7 for awhile now. At idle you can feel it and see it if you watch closely, also the rpms drop slightly when it happens. It happens continuously but not in any kind of pattern. If the rpms are increased you can feel/hear it almost in a rhythm of running right, rpms drop runs poorly, running right rpms increase, etc. Can't feel it at cruising speed but can at idle and during acceleration. Acceleration isn't smooth like it should be, and it takes a slightly longer time to crank than it should. Does it in open and closed loop. Throws no codes. My torque app shows the fuel trims to be normal but slightly rich(+4 is the highest I've seen it go), no misfires detected. I realize torque app is no tech2. The exhaust is slightly black indicating a rich condition. Engine is stock internally, LT headers, Cold air intake, Frost tune, modified factory wiring harness, this powertrain sat in a wrecked truck for about two yrs before I purchased it and did my swap.
Things I've checked or tried:
New o2 sensors
cleaned MAF
new intake gaskets, also checked for vacuum leaks everywhere multiple times
New knock sensor harness, and cleaned the valley area of dirt/corrosion
fuel pressure stays 55-60psi
Fuel rails/injectors pulled and engine turned over to verify all were spraying
new ngk tr55s, plugs all looked the same(slightly brown porcelain, black soot on the rim of the threads, new wires
compression is 180-175 on all but one cyl which is 155-160
pulled valve covers and inspected springs and pushrods

I'm running out of ideas and have exhausted searing google, ls1tech, etc. Will a tech2 show misfires that the torque app will not? I feel like unless I have away to pinpoint it to a cyl or verify its a random/all cyl misfire I'm never gonna find it. Also read that I may need to do a crank relearn, or the misfire detection could've been turned off in the custom tune, waiting to hear back from frost on that question.

[LEEVON]

Yes the tech2 will be able to do a cylinder balance test, and will show if the misfire is random or not. That's where I think my techs would probably start. Seems like it generally leads to a bad coil pack.

[LSXapache]

I hope its something simple like coils. I just dont want to pay for a tech2 scan w/o knowing that the misfire detection is still enabled and the values haven't been loosened up alot.

[LSXapache]

The guy that did my mail order tune told me tech2 was a waste for me(I'm guessing his tune turns the misfire detection off). He told me to get it dyno tuned and the tuner would be able to find any problems. Being as I don't have an extra $600 and the closest shop that dyno tunes is 3hrs away that's not an option right now. I swapped the coil packs with a spare set I had from another engine and the misfire is still there so my next step is going to be to get an infrared thermometer and read the header tubes and hopefully find a difference.

[faribran]

"Infared thermometer" , I like that idea, maybe that will narrow down to which cylinder is the problem child.i will be watching.

[LSXapache]

Worked on the truck some more today. Pulled all the plugs they all look normal to lean(I attribute that to how much timing is in the tune). I leak tested the one cylinder that has lower compression and its coming out the oil fill so that's rings not a valve issue. Checked resistance on plug wires even though they are new, checked ecu grounds. Still haven't found a thermometer to buy locally to test the header tubes. Starting to think it might be injectors as there was some variance in the spark plugs, and I have no way to bench test them. I did however soak the tips in seafoam and cycle them all multiple times before the swap. Although without bench testing I have no way of knowing if they function consistently or have a proper spray pattern.

[Hart_Rod]

HPTuners scanner will allow you to log each cylinder for a misfire so you can isolate the cylinder. Why did your tuner turn off the misfire detection? Swap your injectors side to side and see if you see any difference. Autozone and Advance auto sell the infared thermometers at a decent price. I would also recheck all of your grounds and coil pack wiring for connection and wire breaks.

[BR3W CITY]

Misfire detection being completely off is bad, its a bandaid to hide less-than-stellar tuning techniques...been thru that before. Usually it happens on camm'd ls cars because the tuner can't get the misfire count in check (cam/crank sensors out of phase), and that causes lock/unlokc problems in o/d auto's. Thusly, they just turn it off.

Harbor freight sells a cheapy laser thermometer that does work in a pinch.

This sounds like a tune thing, unless maybe you have an injector sticking open randomly.

WHen you cold start, how does it run? Any change when the truck is hot/heatsoaked?

[LSXapache]

I wont say the misfire detection is completely off but the range is opened up so wide that its not detecting any. Engine runs exactly the same from first start up on through closed loop as far as the misfire is concerned. That's why I'm not pointing a finger at tuning is it does it cold or warm in open loop. I revisited the leak down test this morning with a fresh mind and realized I listened at the oil fill and TB but not the exhaust. I went to the exhaust on that side and sure enough it was leaking air. So I pulled the head thinking I was going to find an obvious burnt valve but what I found was 3,5,7 have leaking exhaust valves(leak solvent, can see flash light(stuck in port ) around valve in dark room). Gonna fix the valve issues(going to do both heads as the other's probably in the same condition) and see if that has any effect on my misfire. Kind of puzzled as to why compression could be 175-180 on 5,7 w/ leaking valves, probably just means I have a ring issue with 3 as I could hear air out of oil fill.

[BR3W CITY]

What kinda shape are the valve springs and PR's in? To get 3 cylinders on one bank leaking, you got a heck of an anomaly, unless something clear happened.

I've seen oil buildup on some early LS's, largely in the intake manifold, but moving into the chambers as it gets sucked in. I suppose enough deposit buildup could bind behind the valve's edge and prevent a full seal.

Can you see how drastic the leak is?

[Jon01]

I'd just put a big cam in it, then you'll never notice a slight misfire at idle.

[LSXapache]

Pushrods are true/straight, valve springs look ok and problems doesn't get worse in high rpm. I lapped all the exhaust valves, the intakes looked fine. honestly the exhaust even though they were leaking looked normal for todays fuel and sitting for an extended time, I don't think lapping the valves is going to fix my problem but at least I can eliminate it off the list of causes. Going to put it all back together tomorrow and check with thermo

[LSXapache]

Wanted to update this thread in case anyone else ever has this problem. The issue the entire time was tuning. My mail order tune wasn't cutting it. For the cost of in person wideband O2 sensor tuning I decided I'd go ahead and swap in an LQ4 before I shelled out the cash for a custom tune. Couldn't be happier with the results. Truck is a tire roasting monster now.

[LEEVON]

I like your diagnostic flow chart!

Misfire >>> Yes? >>> Replace with LQ4

Done

[faribran]

Lol, me 2

[littlevictories]

Originally Posted by Jon01

I'd just put a big cam in it, then you'll never notice a slight misfire at idle.

this is also solid hotrodding advice