Running Rich After Injector Replacement

[Kwing]

My '07 Tahoe 5.3L FFV with 300k on the clock had a chronic misfire problem, which I narrowed down to the injectors (still the original injectors, which I was impressed by, LOL). Since I'm 6-months unemployed, I bought the new injectors on Amazon for $100 (for all 8).

First drive after the swap, misfire was definitely gone, but MIL came on for rich mixture on both banks. I've spent so many hours looking at telemetry data from my HP Tuner at this point and trying to reason my way through the problem that I'm pretty sure I've totally lost the plot entirely. The only consistency I can see is that the STFT and LTFT are constantly more than -18% and the LTFT spends most of the time pegged negative.

I can't find any air leaks, fuel pressure is steady as a rock at 58 psi, TPS and APS data looks clean, and given the raw fuel I can taste dumping out the pipe, I feel confident saying it's not false O2 readings. I've tried re-establishing the Delta P vs Injector Flow table based on the data, which seemed to mildly improve it, but not enough to make a difference.

Any suggestions on what to look at? Or anyone know a lot about the data than I do to try and get to the bottom of it?

 

[swathdiver]

My '07 Tahoe 5.3L FFV with 300k on the clock had a chronic misfire problem, which I narrowed down to the injectors (still the original injectors, which I was impressed by, LOL). Since I'm 6-months unemployed, I bought the new injectors on Amazon for $100 (for all 8).

First drive after the swap, misfire was definitely gone, but MIL came on for rich mixture on both banks. I've spent so many hours looking at telemetry data from my HP Tuner at this point and trying to reason my way through the problem that I'm pretty sure I've totally lost the plot entirely. The only consistency I can see is that the STFT and LTFT are constantly more than -18% and the LTFT spends most of the time pegged negative.

I can't find any air leaks, fuel pressure is steady as a rock at 58 psi, TPS and APS data looks clean, and given the raw fuel I can taste dumping out the pipe, I feel confident saying it's not false O2 readings. I've tried re-establishing the Delta P vs Injector Flow table based on the data, which seemed to mildly improve it, but not enough to make a difference.

Any suggestions on what to look at? Or anyone know a lot about the data than I do to try and get to the bottom of it?

I would put the old injectors back in 1 at a time until you get a misfire if you did not know which cylinder was misfiring in the first place.

What's your alcohol content? High negative trims at idle is often a sign of intake gasket leaks. You can try tightening the intake bolts at first before spending money on new gaskets.

 

[Fless]

I would put the old injectors back in 1 at a time until you get a misfire if you did not know which cylinder was misfiring in the first place.

What's your alcohol content? High negative trims at idle is often a sign of intake gasket leaks. You can try tightening the intake bolts at first before spending money on new gaskets.

^^ THIS. The injectors were very inexpensive and likely not worth the money spent. A $12 or $13 injector is not going to work properly or meter fuel correctly. Over-fueling is going to ruin the cats.

Verify that the alcohol content that the scanner reads is close to the fuel alcohol % that is in the tank. If not, reset the alcohol value with a scanner.

I'm curious -- how was it determined that the OE injectors were bad? Injector balance test? Scan tool looking at individual cylinder misfires? Or??

 

[Marky Dissod]

Since I'm 6-months unemployed, I bought the new injectors on Amazon for $100 (for all 8).

Here's the problem.
In Spanish: Lo barato, sale caro ... cheap turns out to be expensive.

Most cheap@$$ injectors:
are likely not actually octuplets (bad quality control, they do not behave identically)
are significantly different from what the tune expects

If the pcm gives two different injectors the same 'instructions', they will not behave the same.
If the injectors are not exactly what the pcm's expects, both initial fueling and further corrections will be skewered.

 

[Kwing]

Thanks for the replies! Some day I will remember to set the watch thread when I post, but it's anyone's guess WHEN I'll do that, LOL. I'll try to chip through the responses here...

I would put the old injectors back in 1 at a time until you get a misfire if you did not know which cylinder was misfiring in the first place.

I'm curious -- how was it determined that the OE injectors were bad? Injector balance test? Scan tool looking at individual cylinder misfires? Or??

All 8 were missing intermittently/randomly, but #6 and #4 were pretty much dead stick - based on live-read data from the HP Tuner. Having two standouts allowed me to move the plugs, coils, wires, and injectors one-by-one (from 4 to 2 and 6 to 8, then swapped 8 and 2 just to double verify) until my miss followed the moved part(s)... which happened reliably with the injectors.

What's your alcohol content?

Verify that the alcohol content that the scanner reads is close to the fuel alcohol % that is in the tank. If not, reset the alcohol value with a scanner.

The ethanol % (for this tank) reads a rock-steady 12.15% across all the data sets. The question is, how close does "close" need to be? For the next week, the blend in Arizona should be 10%, after which it should roll over to 15% until November. What the tolerances on those blends are, or on the FFV sensors/systems, or for that matter how far it can be off before it starts causing problems, are all things I don't know.

High negative trims at idle is often a sign of intake gasket leaks. You can try tightening the intake bolts at first before spending money on new gaskets.

An air leak is the first thing I thought would be at the root of it, but I can't find one. I just did a smoke leak-detection on it a couple months before I lost my job (so < 9 months ago?), trying to determine if a leak was causing the miss, and at the conclusion of fixing those leaks there were no others. This time I don't have my smoke machine, but the brake cleaner test isn't finding anything. I didn't pull the manifold to swap the injectors, FWIW.

Also, FWIW, the negative trims aren't just at idle. The bars are negative under all conditions except for when the accelerator is first pushed to rev it up, at which point it approaches zero trim for a second or two, then starts climbing back negative until the pedal is let up, at which point it pegs hard negative for another second or two before leveling off back at it's -18%+.

Other than a smoke machine and brake cleaner (and obviously visual inspection), what are some other ways I could try to track down an air leak?

 

[swathdiver]

The ethanol % (for this tank) reads a rock-steady 12.15% across all the data sets. The question is, how close does "close" need to be? For the next week, the blend in Arizona should be 10%, after which it should roll over to 15% until November. What the tolerances on those blends are, or on the FFV sensors/systems, or for that matter how far it can be off before it starts causing problems, are all things I don't know.


Also, FWIW, the negative trims aren't just at idle. The bars are negative under all conditions except for when the accelerator is first pushed to rev it up, at which point it approaches zero trim for a second or two, then starts climbing back negative until the pedal is let up, at which point it pegs hard negative for another second or two before leveling off back at it's -18%+.

Other than a smoke machine and brake cleaner (and obviously visual inspection), what are some other ways I could try to track down an air leak?

Smoke didn't get past my intake gaskets but the trims cleared up whenever I tightened them and after we replaced them.

Ethanol content at gas stations can vary wildly, it's almost never at 10%. The 5.3s seem to get roughly the same gas mileage from 0% to 30% and then about 23% less between 40% and 80%.

Oxygen sensors best calculate the percentage when they are less than 100K miles old and are GM Original Equipment. Aftermarket sensors, even the Denso, are known to usually not calculate the alcohol percentage accurately. That will mess up the trims too.