Idle Stability / Airflow Tables

MFB · **Posted:** Fri Jun 07, 2013 6:19 am

AZ1G101N please :-)

Heide264 · **Joined:** Sat May 07, 2011 3:22 pm **Posts:** 144

td-d wrote:

Sounds like I need to provide definitions for the logger parameters

What parameters are you logging - I'll define them so that others can use the same strategy as proposed above.

Man... I'd love to see some logger parameters to help out with tuning these tables. I wish I had some more time to fool with these tables, but I just haven't gotten around to it.

I'm on AE5I910V, FWIW. No rush, but I don't mind being a test mule.

td-d · **Posted:** Fri Jun 07, 2013 11:39 pm

MFB wrote:

AZ1G101N please :-)

Code:

RAM:FFFF683C E_Engine_Speed_Delta_E157

The idle error is actually calculated from target Idle speed and actual idle engine speed, namely:

Code:

RAM:FFFF83A0 Ram_Target_Idle_Speed?
minus
RAM:FFFF8358 Ram_Actual_Idle_Speed?

So you would need to create a calculated parameter.

Easty · **Posted:** Sat Jun 08, 2013 12:43 am

td-d wrote:

MFB wrote:

AZ1G101N please :-)

Code:

RAM:FFFF683C E_Engine_Speed_Delta_E157

The idle error is actually calculated from target Idle speed and actual idle engine speed, namely:

Code:

RAM:FFFF83A0 Ram_Target_Idle_Speed?
minus
RAM:FFFF8358 Ram_Actual_Idle_Speed?

So you would need to create a calculated parameter.

Try logging this ram address FFFF83BC

td-d · **Posted:** Sat Jun 08, 2013 6:16 am

Easty wrote:

Try logging this ram address FFFF83BC

That's the result of the table - the MAP related correction that is applied.

MFB · **Posted:** Sat Jun 08, 2013 7:30 am

So ok to just use FFFF83BC or still need calculated parameter? What extended parameter address are still available?
Thanks for the quick response on this TDD :-)

Easty · **Posted:** Sat Jun 08, 2013 11:30 am

MFB wrote:

So ok to just use FFFF83BC or still need calculated parameter? What extended parameter address are still available?
Thanks for the quick response on this TDD :-)

I've been logging the equivalent parameter in my rom works a treat looks like this

Attachment:

Untitled1.png

No need to calculate

td-d · **Posted:** Sat Jun 08, 2013 3:22 pm

Easty wrote:

MFB wrote:

No need to calculate

Ah, of course - I missed the bit where fr8 is loaded back into this address - yes, this is the parameter to use.

Let dig through my PMs to see what E number was allocated to it.

td-d · **Posted:** Sat Jun 08, 2013 3:29 pm

Official allocation:

E159 - Idle Engine Speed Error (4-byte)*

Merp · **Joined:** Thu Jul 23, 2009 5:46 pm **Posts:** 863

A2UJ000J:

0xBD154 - big subroutine for target throttle
0xFFFF8088 - Gets added in to some calculations here, not sure exactly what happens after that yet, just assuming it impacts target throttle.
This ram address is only written in sub 0xBE7E4.

Its value looks to be derived from two tables multiplied together, but there is some funny stuff going on with an array. There are also two 'modes' in which the sub is called, each with different logic paths (determined by the incoming R4 'mode'). Essentially a function with a boolean argument.

One of the tables uses an axis lookup that reference back to some idle airflow parameters I was looking at. Multiplier looks like the axis may be in MAP??

Pull sub for the initial value (idle airflow related lookup): 0x4085C
Pull sub for the multiplier/compensation: 0x40924

Heide264 · **Joined:** Sat May 07, 2011 3:22 pm **Posts:** 144

td-d wrote:

Easty wrote:

Try logging this ram address FFFF83BC

That's the result of the table - the MAP related correction that is applied.

I would ideally like to log the MAP related correction that is applied, actually. It would give me a much better idea how to modify the table.

Any idea if that RAM address is only for a certain ECU? I have an AE5I910V over here.

I've found that with just a tiny tiny bit of pedal movement (I'm not sure if that results in throttle or not, actually... I'm sure it picks up a requested torque change though) my idle (specifically while in gear and rolling in traffic) goes from feeling like a roller coaster train to smooth as butter.

td-d · **Posted:** Tue Jul 09, 2013 9:16 pm

Heide264 wrote:

td-d wrote:

Easty wrote:

Try logging this ram address FFFF83BC

That's the result of the table - the MAP related correction that is applied.

I would ideally like to log the MAP related correction that is applied, actually. It would give me a much better idea how to modify the table.

Any idea if that RAM address is only for a certain ECU? I have an AE5I910V over here.

I've found that with just a tiny tiny bit of pedal movement (I'm not sure if that results in throttle or not, actually... I'm sure it picks up a requested torque change though) my idle (specifically while in gear and rolling in traffic) goes from feeling like a roller coaster train to smooth as butter.

There are also idle related torque tables - from what you are describing, likely the issue. I'll pull them out for your rom.

td-d · **Posted:** Tue Jul 09, 2013 9:25 pm

The result from the table from AE5I910V is stored at 0xFFFF8528.

td-d · **Posted:** Tue Jul 09, 2013 9:42 pm

I see earlier in this thread I already pulled out the idle target throttle tables for that rom!

Merp · **Joined:** Thu Jul 23, 2009 5:46 pm **Posts:** 863

Ah, I see them now. Will have to open up that rom to see how things translate over. A2UJ000J has a 2D base, a 2D multiplier/comp, and 1D min/max for the result of that calculation.

I'll have something up on git tonight/tomorrow for a tester.

RomRaider

Idle Stability / Airflow Tables

Who is online