FWIW, I re-did my MAF scaling, and after about an hour of driving around without the car ever dying, I'm pretty confident that the lift-throttle-stumble-and-die problem is solved. RPM still drops a little below the set point sometimes, so there's still more work to do, but that's minor enough that I might not have noticed if I wasn't already paying close attention to the return-to-idle behavior.

I just copied the MAF curve from another tune that used the same MAF sensor and housing diameter that I'm using, so of course it isn't perfect, but it's made a bit improvement. I'll work on fine-tuning it next.

Since the MAF sensor is set to fail at 100hz, you wouldn't think that would matter. However I've heard from a couple people that even when configured that way the PCM will try to use the MAF sensor during rapid throttle changes - which apparently includes lifting your foot off the pedal.

Linson, are you using the stock MAF sensor and housing, or have you switched it out for a larger one?

 

Timing and fueling in the cells upon throttle let off have the biggest impact on return throttle than anything else. Using the real time tuning feature helps a lot. Do a log and take note of the cells it hits when you let off. Then tweak those. Too much fuel it’ll dip. Too little fuel it’ll dip. Too much timing it’ll dip. Too little timing it’ll dip. Timing and fueling react much faster than the IAC, remember that. Once you realize that the air adjustments don’t help much upon throttle letoff, the better off you’ll be. The only way they make a difference is if you give it too much air and the idle returns real slowly and the car will act like it has the cruise control on at low speeds. You don’t really want that.

 

Your info is not correct. The one time the pcm generally ignores the maf is during throttle transitions. You must have misread, or been given bad info.

 

Well, whatever the explanation, my car stopped dying when I lift my foot off the throttle pedal.

 

10 degrees overlap is no big deal.

I have a NW 102 cable throttle body and had all kinds of difficulty tuning idle. Long story short, no amount of tuning was going to fix my idle problems. I eventually figured out the throttle blade was cracked too far open in order to provide enough air flow for the engine. The swept area of opening increases dramatically as the blade angle increases. The blade angle was in a zone where very small changes caused a big change of airflow, which made idle tuning a real PIA. The warning signs were an extremely sensitive idle set screw, high IAC Hold Position needed (40% if I remember right), very low throttle position while cruising (1% TPS in 6th gear), and overlap in the fuel map between cruising and no-load deceleration.

The fix was to drill a small hole in the throttle blade so the engine had enough base air flow with the blade at a low angle. Suddenly tuning idle and cruise became a LOT EASIER. The fuel map parsed out differently too, so that cruise & deceleration were distinctly different. And I could cruise at 1200 RPM in 6th gear (that's incredible given my cam specs). It fixed the majority of my drivability issues and allowed me to follow up with a good tune that worked.

This escaped 2 professional tuners. They were talented on the keyboard and could make my engine work that day, but their changes didn't work on a different day. Anyway, it's an example how if the hardware isn't right, then no amount of tuning is going to hold.

I absolutely DID NOT want to tune my car and expected to be able to pay a professional to make it drive perfect. Just wasn't a realistic expectation. They don't spend enough time with the car to find all the drivability glitches. I eventually reconciled that job falls on me, and that's when I began to make the most progress.

 

its sad isn’t it? This is why I dont like tuning other people’s stuff. I’m very picky and people are always in a hurry. Last vehicle I tuned, the guy had to come back three times. First time he shows up with his whole family. Kids and all. Wife was getting impatient because how long it was taking. So I told him bring it back to me when he can and drop it off. So he does, but drops it off needing to pick it right back up in a few hours and I had a dinner I needed to attend. Plus left the fuel on empty. How am I to drive around tuning, with no fuel? Not my responsibility to put fuel in it. Lol. Then the last time I was very clear. Leave it all day. Make sure it has fuel. All is good now. But I’m not really wanting to deal with this stuff anymore. I’ll just tune my vehicles and close friends. Other than that. Sorry. Just takes too much time to do it right and people don’t wanna pay for my time.

 

Also, with the much improved control of air at idle, I was also easily able to lower timing from 24 degrees to 17 degrees. That gave the P/D spark timing control much more influence so it could do a better job too. (After all, it's hard for an engine to catch itself when you're already at 24 degrees base timing.)

I had a laundry list of drivability issues prior to that. The whole list got wiped out by having a proper throttle blade position. Tuning changes were needed too, but it was so EASY after that.

 

Ya, those professional tuners are way more talented than I am for sure, but they have just a handful of hours with the car. By contrast, GM engineers spend years with the car during development. The only other person that has that much seat time in my car is me. Whether or not I believe in myself to be able to do it, I'm still the best candidate to work out the glitches.

 

Exactly my experience tuning others cars. Often precluded with “fix all these things, then I’ll come back and get started”!

Never had anyone bring the whole family though! That cracked me up.

The swaps are the ones that get me. No one seems to think they need the narrow band O2’s......
Some think they don’t need knock sensors.
So they have the harness built w/o those unimportant items.....

 

I tested some MAF curve changes today and that will definitely make the problem come and go. Not sure if it's related to transitions or just idle or just return-to-idle, but the PCM is definitely looking at the MAF curve at some point during the return-to-idle scenario.

 

I'm pretty sure the maf is never ignored.....but the map is ignored at WOT above the rpm threshold for airflow cell.

 

i ignore my MAF all the time in SD mode

 

Cheater!

 

That's what I was told my car was doing too. I think I've been lied to.

Wanna try a science experiment? Multiply everything in your MAF scaling curve by 2, drive around town for 30 minutes, let us know how that goes.
Not every cell's value can be doubled, some cells will just get maxed out, but that doesn't matter. I'm really just curious about the idle, cruise, and return-to-idle behavior.

 

Look at your rpm threshold for airflow table and see what rpm its set to. Also look at your maf frequency fail table.

 

RPM threshold is 300.
MAF high frequency fail is 100hz.

 

Turn the MIL/report back on and see if the MAF is actually failing. That'll tell you for sure that it's being ignored. If not you have something else going on.

 

What if it’s completely unplugged or removed?

 

Ding ding ding!!!

 

for me at least, i have the SES light off, and report on first error.

set the entry to 300rpm and the maf fail hz at the lowest cell on the curve. zero the curve and then put 50 in the two lowest cells on the curve so it picks up and idles when you do a reflash/code reset