How are you tuning your timing maps? I see there are are two base ignition tables which have the same values but 3 dynamic advance tables. It appears to me that the car seems to use the 2nd dynamic advance table most.
I'm analyzing logs from an 04 STi and the car seemed to be using "Ignition Correction A" yesterday which was the first day the map was flashed and now it is using "Ignition Correction B" today. The IAM is set to 1 in the map so as soon as the rom was flashed the IAM was instantly at 1.
The timing maps actually look pretty decent from the factory. The car can still be tuned but it's just makes it more time-consuming to look for knock. On WRXs, I set the knock correction tables reading all 10s so if at any point in the logs I don't see 10s, I know the ECU pulled timing. However, without knowing how exactly the ECU interprets timing I am willing to forego this and look for knock the old-fashioned (and tedious) way.
Let this thread be the beginning to fully understanding how the STi ECU calculates timing and which maps it uses when.
Ignition Timing and Correction - Maps A, B, C?
7 posts
• Page 1 of 1
Ignition Timing and Correction - Maps A, B, C?
by MRF582 » Wed Oct 11, 2006 7:22 pm
- MRF582
- Posts: 34
- Joined: Thu May 19, 2005 11:25 am
by Freon » Wed Oct 11, 2006 7:48 pm
So far I'm noticing it uses the first map. I have worked that map over quite a bit.
I'm trying to figure out how it determines which one it uses. I almost have the entire AJ243 ROM disassembled. I do have the portion where the ignition advance maps torn apart.
I'll post anything I find here. I'm dying to know as well.
I'm trying to figure out how it determines which one it uses. I almost have the entire AJ243 ROM disassembled. I do have the portion where the ignition advance maps torn apart.
I'll post anything I find here. I'm dying to know as well.
- Freon
- Posts: 700
- Joined: Thu Nov 17, 2005 5:50 pm
- Location: Indianapolis, IN
by fasterthanurwrx » Wed Dec 06, 2006 10:20 pm
What are you doing to figure out what correction table the ecu is doing?
- fasterthanurwrx
- Posts: 19
- Joined: Sat Sep 16, 2006 4:28 pm
by Freon » Wed Dec 06, 2006 11:46 pm
I've made some progress. I can at least say that switching or sliding is related between base and advance.
Base map 1 is used with Ign Adv 1
Base map 2 is used with Ign Adv 2
If you have sets of maps like this, they are averaged together based on a sliding or weighting value of 0 to 1. So both maps are pulled, and a certain weighting determines how much of each map is used.
Ign advance 3 is something else. I think it has to do with how advance is pulled back or general reaction speed to knock. That is, I think a higher number, the quicker it pulls timing in response to knock. So at high load, high rpm, you want higher numbers so the ECU will back off quicker based on a single knock event. I'm not 100% positive about this, but the knock correction maps are subtracted from one another sometimes. Still looking into that. I'll figure it out eventually, just a matter of time. I'm pretty certain once I figure it out once I can tell you how they do it on all the Subaru DBW cars. Even the 2007 code (which is obviously updated) isn't all that different.
There is one main weighting factor that determines most of the maps with multiple tables. So Ign Adv 1, Base Ign 1, Target Boost A or C, Ignition CT comp 1, etc are all "linked". This weighting factor has three possible "strategies". One is more likely to set it closer to 1, and use the second set of maps, another strategy more likely to set to 0, and another does a different strategy. I'm trying to figure out exactly what these strategies do to set the weighting. There a few things I need to trace to get a better grasp of it. I hope once I get a hold of it the possibility to use these extra maps and tweak the map selection strategies will be available. Maybe with a bit of work a driver controlled system could be implemented, or it could be made to work in a more desirable fashion.
If someone is willing to help out by logging, I can tell you the memory address of the weighting factor if you can point to a copy of your ROM. I'm only concentrating on the Subaru DBW cars for now, however. I need more logs from different cars to see how this value moves around.
Base map 1 is used with Ign Adv 1
Base map 2 is used with Ign Adv 2
If you have sets of maps like this, they are averaged together based on a sliding or weighting value of 0 to 1. So both maps are pulled, and a certain weighting determines how much of each map is used.
Ign advance 3 is something else. I think it has to do with how advance is pulled back or general reaction speed to knock. That is, I think a higher number, the quicker it pulls timing in response to knock. So at high load, high rpm, you want higher numbers so the ECU will back off quicker based on a single knock event. I'm not 100% positive about this, but the knock correction maps are subtracted from one another sometimes. Still looking into that. I'll figure it out eventually, just a matter of time. I'm pretty certain once I figure it out once I can tell you how they do it on all the Subaru DBW cars. Even the 2007 code (which is obviously updated) isn't all that different.
There is one main weighting factor that determines most of the maps with multiple tables. So Ign Adv 1, Base Ign 1, Target Boost A or C, Ignition CT comp 1, etc are all "linked". This weighting factor has three possible "strategies". One is more likely to set it closer to 1, and use the second set of maps, another strategy more likely to set to 0, and another does a different strategy. I'm trying to figure out exactly what these strategies do to set the weighting. There a few things I need to trace to get a better grasp of it. I hope once I get a hold of it the possibility to use these extra maps and tweak the map selection strategies will be available. Maybe with a bit of work a driver controlled system could be implemented, or it could be made to work in a more desirable fashion.
If someone is willing to help out by logging, I can tell you the memory address of the weighting factor if you can point to a copy of your ROM. I'm only concentrating on the Subaru DBW cars for now, however. I need more logs from different cars to see how this value moves around.
- Freon
- Posts: 700
- Joined: Thu Nov 17, 2005 5:50 pm
- Location: Indianapolis, IN
by Justin 05 STi » Fri Feb 02, 2007 8:37 am
Freon, if I can do any logging for you just say the word.
One of the things that has kept me from wanting to fine tune maps is the void of knowledge of the inner workings of the ECU's strategies between all the maps.
Justin
One of the things that has kept me from wanting to fine tune maps is the void of knowledge of the inner workings of the ECU's strategies between all the maps.
Justin
- Justin 05 STi
- Posts: 15
- Joined: Tue Nov 15, 2005 6:37 am
- Location: KC MO
by JMS77 » Tue Oct 02, 2007 2:09 pm
07 STi uses different timing base and advance maps for accel and deceleration.
BASE B- Accel Base-cruise/lift throttle
Advance C-Accel ADV D- Cruise lift
Not too sure about the others, doesnt seem to use them often... different algorithm?
BASE B- Accel Base-cruise/lift throttle
Advance C-Accel ADV D- Cruise lift
Not too sure about the others, doesnt seem to use them often... different algorithm?
- JMS77
- Posts: 2
- Joined: Fri Sep 28, 2007 2:45 pm
7 posts
• Page 1 of 1
Who is online
Users browsing this forum: No registered users and 1 guest