Hard start - Tried everything

[sdbuggyboys]

Throwing out a lifeline, I'm completely lost on this one. 10+ years mechanic, very comfortable with the LS platform. 2004 Yukon XL 5.3 VIN Z (wifes car actually) About 2 years ago, it started taking two times to start. Figured it was a fuel pump since it had been about 5 years since I put one in. Changed it out, problem kept happening, 6 months later it strands her on the side of the road with a dead pump (I think it was a precision brand). At this point I'm tired of dropping the tank so I bite the bullet, buy an ACDelco, put it in, and the problem remains. Fast forward a few months and we find out we're moving from TX to TN. The yukon has 300k on it and I'm not sure it'll make the drive towing horses, so I decide to put a crate motor in it. While im doing the motor I change out fuel filter and clean the lines. Go to fire up the new motor and the problem is STILL there. Drive it for awhile and I start getting lean codes and random misfire. Immediately think vacuum leaks, find a few, but the lean codes still stay. Then I think injectors since the manifold swapped with the motor. Send the injectors off to get flow tested and rebuilt, put them back in, and the lean codes go away but the twice to start thing is still there 80% of the time. Ive checked everything i can think of for fuel and air and don't see any culprit. It "feels" like a vaccum leak somewhere, but I can't find ANYTHING. I'm usually the guy people call when stuff breaks, but im out of ideas in this one and need some help. Currently, no codes, still takes twice to start most of the time. Almost every time it doesn't start, it'll give a little puff of a backfire out the tailpipe. Fuel trims are reading a little lean, but nothing crazy. Thoughts?

 

[mattbta]

If you have a non-OEM cam position sensor in it, try an OEM. Does it act like this?

 

[Fless]

The symptom is indicative of a possible cam-crank correlation issue. If the PCM can't tell what the relationship is on the first crank, it switches to a different (assumed) correlation for the second crank. The PCM needs to know how the two are paired, so if it hasn't been done do a CASE relearn just to rule it out. The new engine likely had new sensors, too, so the PCM doesn't know how they sync. This would also be true of the old sensors were used in the new engine.

Eric O describes the PCM strategy early (starting around 4:00) in this video:

 

[sdbuggyboys]

I can sure give that a try, but I didn't change sensors in the old motor when this happened the first time. The new motor got new sensors in it when I finished the short block, but there were standard brand. It does seem something like that, or it's not getting the initial fueling right and diverts to a different fuel map?

 

[S33k3r]

I can sure give that a try, but I didn't change sensors in the old motor when this happened the first time. The new motor got new sensors in it when I finished the short block, but there were standard brand. It does seem something like that, or it's not getting the initial fueling right and diverts to a different fuel map?

Any luck? I hate problems like this; I'm hoping you find and share a solution for the next guy that suffers it.