Two questions this morning:

The trims actually have an effect on the demand% colder than 171F but only start to adjust themself at over 171... Why the target idle table is coolant temp dependant if the trims only start to change at 171F (idle mode)..... ?

Raptord, or Jamie if you have a IX to test, could you tell how the trims actually update versus time? In the IX ROM, there is a defined table called Trim adjust frequency that we don't have in 9417. The setting is 1000ms(1s). This is odd because my trims adjusts nowhere near 1second. Much more like 30 seconds I think (I have to check logs).

 

I don't have my laptop in front of me but I did recall seeing the deamnd% actually changing at the increments shown in the idle offset table. That table works the same as the boost correction. If the idle is off by 50rpm, it keeps adjusting the demand% by whichever value is listed in the offset table for 50rpm until the idle speed actually starts to change. The demand% acted like STFT while the demand trim acts like LTFT.

 

Can anyone find the address where we can log the "ISCV Demand Target Idle Trim" adjustment? This appears to be basically the same as Boost Error Correction for idle, similar to the STFT of idle I believe.

-Jamie

 

About Demand trims update times, according to my logs and the average LTFT udpate was between 46 seconds and 57 seconds. Oddly I logged once that it updated twice in about 5 seconds.
It seems to be a sum of "STFT" error and not a specific time parameter. I will have to compare some logs to find out the exact behavior of this, unless you already found this.

 

I believe its called "Idle Error-AC OFF" for the 9417 rom.

-Jamie

 

It is a great find Jamie. It makes sense a lot:
"Idle Error - Normal" and "Idle Error - AC ON" tables wouldn't be steps related like it's scaled from our xml.......

Instead of looking like this:

We'd use the IX ROM scaling:

Not sure about the RPM scale (250 or 500), I haven't any logs to check the correction strengh...I'm not at home. I'll be confirming it tonight if you are on something else.

 

Another thing to determine beside trims update is the Desired Idle RPM with AC on



There are 4 settings, which I suppose to be
1 and 2: When compressor is engaged
3 and 4: When compressor isn't engaged

The thing is that we haven't a Initial steps position - AC ON Drive. Maybe the AC OFF Drive, or AC ON neutral is used...
I will test by changing values drastically and logging demand / trims.
And again, the IX ROM isn't the same.

 

Yes, all of these tables are demand% scaling and not raw ISC steps actually. I have already proven that to be 100% true with my recent testing. The first thing we need to worry about has nothing to do with the rpm target tables. By disabling the correction tables (I.e; idle error), the ecu is running straight off the values in the demand% tables. This allows us to see what the ecu is doing with these tables without the ecu altering the base settings.

With every single table zero'd out (even the initial ISC steps-AC off) without touching the idle stepper table, the lowest demand% I have seen on the evo 8 was around 9-10% demand.

My guess is that we haven't found the most important table yet, which is the true BASE demand% target table.

-Jamie

 

The only table that I have seen to actually work as expected is the startup table, which only lasts a few seconds and then tapers off back to whatever demand% settings are in this missing BASE table.

-Jamie

 

Some demand% tables aren't defined right (CTS Adder post WOT, Baro adder, IATS adder) . Do you have some more defs, or you're in the same situation as me?

 

Here is another interesting thing. The idle target was 850rpm. During my testing with corrections disabled, idle never went below 900rpm. ISC steps were about 30. The demand Trim started ADDING trim, although there were no changes to the actual ISC step position since everything was basically disabled.

Once I **** the car off and immediately turned it back on, the demand trim was added and could be observed when logging Demand%.

It was interesting that the demand trim was ADDING demand even though the target rpm was too high.

Also, does anyone have the correct address for the idle IAT trim for the 9417 rom?

-Jamie

 

Even if they aren't defined right, if they are active and I set them to zero, a change could be observed. Some of these tables will be situational but right now we need to focus on idle control at operating temp with AC off at 0mph.

We still need to find the table that adds demand% when the fans or A/C are active.

-Jamie

 

Crappy iPhone browsing missed your big post

 

Hope you guys can get this all squared away and make a new thread with one set of proper definitions for all the idle stuff.

 

Amen!