Well, I got everything in good shape in the lower VE tables.

BLM's at 128.

The problem is with the O2 toggling back and forth, it causes surging. Much less prevalent now that things are in line, but happening just the same.

How do I go about slowing down the toggling or changing sensitivity?

Also, on decel, RPM's fluctuate up and down, and truck will almost die when near a complete stop.

 

Might try reducing the O2 voltage rich, and lean limits.
Sounds like the DFCO, lower VE table area still needs a little work. Might turn off the DFCO for getting the low end of the VE table right, then turn it back on and play with it's settings. Doing both at the same time can leave you chasing yourself. Also, look at the DFCO min rpm, and VSS stuff.

 

Reduce the proportional gain term(s).

RBob.

 

That's where I'm at right now. After replacing the o2 sensor with a fresh (even though it's bosch) sensor I had that problem with a warm engine idle surge. The sensor reacts SO FAST it's not even funny. I'm going to be watching the miles vs o2 sensor x-counts to see what kind of milage a sensor can last.
The prop gains are definatly needing tuning now that I'm not so rich that it doesn't do anything . It's amazing how much tuning you can do to get around mechanical problems like vacuum leaks, bad sensors, bad spark plug wireS, dirty air filter, lol. I've compiled enough notes to be "ASE" certified 10 times over.

 

.


Aint that the truth. I even tuned around a fried ignition system. Didnt know I had a problem untill it died alltogether. The only thing I noticed is that it had a serious lack of power under load. Other then that it ran fine, jsut seemed to smoke a bit...

 

Bosch and O2 sensor should never be used together in the same sentence.

 

I'm using TunerCat, so I'm not quite clear on the terminology.

Are you referring to the "Mean/Rich/Lean Off Idle Thresholds"?

Also, DFCO has several settings: "DFCO MPH", "DFCO MAP Limits", and "DFCO RPM".

 

Would these be referred to in TunerCat as "Proportional Value vs O2 Error", and or "Proportional Duration vs O2 Error"?

Thanks to everyone for the patience and info

 

Not knowing how the ECM/Mask you are using sets up the terms, the '747 uses two tables and then separate idle terms. Probably similar to the code you are using.

The two tables are Proportional gain vs (slow) O2 and proportional gain vs airflow. A lookup is done on each table, the results mulitplied together and then divided by 256. The end result is the proportional gain used to adjust the PW.

RBob.

 

I'm using the 7060, $85 mask

Would "Slow O2 Filter vs Airflow" be the first table mentioned, and "Proportional Multiplier vs Airflow" the second?

Thanks for the function explanation

 

I had the best results when:

the motor VE tables are tuned really close, then I decrease the effect of the prop and int terms. I do this until (thanks to RBOB for this advice) the cross counts still remain. If you reduce to prop gains too far you'll see a drop off in x counts, then you know you've gone too far.

 

Then go take the tests! I'm sure that most of the guys around this board can easily pass the tests, but the only thing standing in the way of an actual vertification would be the experience requirement. They require 2 years of working in a shop for certification. Actually, I don't think all that much of it, other than an accomplishment. I had passed the auto transmission test before I had ever even had one apart!

What has been more helpful to me has been the classes we had gone to where we learn about what the OEM needs the computer system to do, and some of what they did to accomplish it. Unfortunately, it's not an advanced enough class to be of much real help in PROM stuff, and it really wasn't geared to be such, just to learn what the computer expects, etc.... I would love to sit down with the instructor and show him some source code, and what the ECM IS doing!!!!

JPrevost's commment also reinforces what I've learned in auto repair: Make sure the basics are all good, tuneup, fuel supply, mechanicals, etc.... Then make sure what the computer is seeing is correct (i.e. sensors are good). I'm sure we've all chased our tails on some stupid little problem that made the problem seem huge!

Now, back to topic.....
So the proportional gain zcontrols how fast the ECM reacts to what the O2 reads? Or is it how much it reacts to what the O2 is reading?

 

Thanks for the tip. I have a question though.

There are three columns in the table mean, rich, and lean.

Do I lower all 3, or adjust the lean and rich toward each other?

Bill

 

I wouldn't lower any just yet.
Everyone that hasn't already read Robs paper on PID fueling should.

http://www.diy-efi.org/gmecm/papers/747fuel.txt

 

RBob throws some good info around in this post:

https://www.thirdgen.org/techbb2/sho...391#post945391