AF learning A-D minimum and maximum threshholds

td-d · **Posted:** Mon Feb 20, 2012 12:27 pm

They say imitation is the best form of flattery

I was looking through the COBB SD implementation, and there are a couple of useful tables in there in relation to AF learning and AF3 (rear O2) correction. I'm still busy scratching around for the rear O2 sensor tables, but I've found the AF learning minimum and maximum threshold figures.

In a nutshell, if you set the maximum threshold to lower than your D range in the A - D learning - no more learning carried through into open loop. Quite useful

Quite cooincidentally, fastblueufo PM'd me to find it for the '04 USDM sti (synchronicity!) so:

For A2ZJ710J:

Code:

<table name="Af learning Max threshhold" address="5bb10" type="1D" level="1" scaling="MassAirflow(g/s)"/>
<table name="Af learning Min threshhold" address="5BB00" type="1D" level="1" scaling="MassAirflow(g/s)"/>

for Az1g201G:

Code:

<table name="Af learning Max threshhold" address="C7AF8" type="1D" level="1" scaling="MassAirflow(g/s)"/>
<table name="Af learning Min threshhold" address="C7AE8" type="1D" level="1" scaling="MassAirflow(g/s)"/>

Happy to find them for any other roms - as a tip to any other IDA'ers on here - the values are basically adjacent to the AF learning ranges. Seems stock is 1.5999 min and 80 max on most roms.

fastblueufo · **Posted:** Mon Feb 20, 2012 5:26 pm

td-d wrote:

They say imitation is the best form of flattery

I was looking through the COBB SD implementation, and there are a couple of useful tables in there in relation to AF learning and AF3 (rear O2) correction. I'm still busy scratching around for the rear O2 sensor tables, but I've found the AF learning minimum and maximum threshold figures.

In a nutshell, if you set the maximum threshold to lower than your D range in the A - D learning - no more learning carried through into open loop. Quite useful

Quite cooincidentally, fastblueufo PM'd me to find it for the '04 USDM sti (synchronicity!) so:

For A2ZJ710J:

Code:

<table name="Af learning Max threshhold" address="5bb10" type="1D" level="1" scaling="MassAirflow(g/s)"/>
<table name="Af learning Min threshhold" address="5BB00" type="1D" level="1" scaling="MassAirflow(g/s)"/>

for Az1g201G:

Code:

<table name="Af learning Max threshhold" address="C7AF8" type="1D" level="1" scaling="MassAirflow(g/s)"/>
<table name="Af learning Min threshhold" address="C7AE8" type="1D" level="1" scaling="MassAirflow(g/s)"/>

Happy to find them for any other roms - as a tip to any other IDA'ers on here - the values are basically adjacent to the AF learning ranges. Seems stock is 1.5999 min and 80 max on most roms.

If you get a chance, post up adresses for all the usdm STI roms 04 to 12. That way I can add them to my own defs. Some of us older peeps suck at IDA. Everytime I open it I'm like wtf, scratch my head and close it down lol.

td-d · **Posted:** Mon Feb 20, 2012 7:11 pm

fastblueufo wrote:

If you get a chance, post up adresses for all the usdm STI roms 04 to 12. That way I can add them to my own defs. Some of us older peeps suck at IDA. Everytime I open it I'm like wtf, scratch my head and close it down lol.

Chuckle... I'm an old fart myself, promise, think mid life crisis

No worries - I'll start working through whatever roms I already have disassembled - a little hamstrung at the moment, busy resetting up my home machine after it (very expensively) got nailed by a power surge...

td-d · **Posted:** Mon Feb 20, 2012 8:15 pm

AZ1J500J : MIN - C94C0 MAX - C94D0
AZ1G702I : MIN - C920C MAX - C921C
AZ1G202I : MIN - C8BF4 MAX - C8C04
A8DH201X : MIN - C8574 MAX - C8584
A8DH200V : MIN - C6C98 MAX - C6CA8

Here's a trick that is likely to work for most of the roms. In ecuflash, open up the AF learning range tables, and check what the address is.

That address minus hex 4 is the minimum. That address plus hex C is the maximum. Use calc in 'programmer' mode.

td-d · **Posted:** Mon Feb 20, 2012 9:21 pm

A2ZJB10J : MIN - C5DF4 MAX - C5E04
A2ZJB11J : MIN - C5DF4 MAX - C5E04
A2UG000J : MIN - C89C8 MAX - C89D8
AE5IB00V : MIN - CC130 MAX - CC140
A2UJ000J : MIN - C89C8 MAX - C89D8

fastblueufo · **Posted:** Mon Feb 20, 2012 9:52 pm

BIG thanks and a hand shake. Pat on the back and a raise

dschultz · **Posted:** Mon Feb 20, 2012 10:01 pm

If I knew up front that I'd get a raise... well I would have done this a long time ago!

td-d wrote:

Chuckle... I'm an old fart myself, promise, think mid life crisis

You are not old according to your profile.

td-d · **Posted:** Mon Feb 20, 2012 10:16 pm

fastblueufo wrote:

BIG thanks and a hand shake. Pat on the back and a raise

No worries, only a pleasure.

dschultz wrote:

You are not old according to your profile.

Indeed... little did I see what I was getting into when I got this car in '09 !

td-d · **Posted:** Mon Feb 20, 2012 10:20 pm

Now to find those AF#3 limits - don't forget to send me the stock values for the 08 when you can, Ed, appreciate it.

td-d · **Posted:** Tue Feb 21, 2012 7:42 am

For AE5I910V:

<table name="Af learning Max threshhold" address="CA568" type="1D" level="1" scaling="MassAirflow(g/s)"/>
<table name="Af learning Min threshhold" address="CA558" type="1D" level="1" scaling="MassAirflow(g/s)"/>

td-d · **Posted:** Tue Feb 21, 2012 9:44 pm

I think I'm making some headway with the AF#3 (rear O2) tables - once I started thinking in terms lambda, it started making sense. I'm pretty sure I've found the AF 3 correction limits, working on the learning and sensor limits.

MFB · **Posted:** Tue Feb 21, 2012 10:18 pm

I'm sure a lot are looking forward to the rear 02 comp tables.

Can you check the address for AZ1G101N? Tia!

Btw what do you think is the advantage of this vs just changing the a to d learning ranges (10 40 80 )disabling OL learning ?

td-d · **Posted:** Wed Feb 22, 2012 7:43 am

AZ1G101N

Min: C7730
Max: C7740

MFB wrote:

Btw what do you think is the advantage of this vs just changing the a to d learning ranges (10 40 80 )disabling OL learning ?

Well, I'm not sure it's so much an advantage, as a different approach. Reducing the max threshold really just takes learning out of the equation in open loop, as opposed to shutting the whole closed loop system down. I guess the advantage is that you can maintain the stock approach to CL learning in terms of specific areas required (i.e. idle, low throttle deceleration and cruise) whilst not worrying that your OL AFRs are 'contaminated' due to environmental factors such as temperature on a specific day.

MFB · **Posted:** Wed Feb 22, 2012 8:58 am

Thanks for the add! Useful for Cars that don't get monitored regularly. With this You still get learning d corrections at the same time prevent ol fueling. However disabling this means ecu won't be able to adapt to fuel quality change for open loop. Of course will only work if maf was properly scaled with a particular fuel.

Eagerly awaiting for rear 02 comp tables. :-)

td-d · **Posted:** Wed Feb 22, 2012 11:27 am

Argh... I was requested to put up a Romraider definition, and I cannot get it to show up in Romraider, no matter what I do. Dale - could I ask you a favour, to post up a template for it to show up with the one of the roms above that people could cut and paste their addresses into?

RomRaider

AF learning A-D minimum and maximum threshholds

Who is online