Whew! Done.

Thanks again to bez_bashni for being the first one to mention this.

My car was overrunning the 260% load column on a regular basis. Rescaling the tables has helped eliminate the 'mystery' timing that was being pulled by the ECU.

l8r)

 

Ludikraut, nice work...this should be made a sticky. Why do you have 12 knock sensor filter map while I only see 1 on mine?

 

Thanks.

I'm running ECUflash in "Developer" mode ( options -> options -> User Level ), which is you see all of the maps in my screen shots.

l8r)

 

It's really cool that took the time to do this for all of us. Thank you so much. Super helpful!

 

Just to clarify. People who would want to resize would be the people with larger turbos and running high boost. This would include those of us running the stock turbo and alky/meth with 25psi or more boost, correct?

 

Ludikraut - great write up- STICKY THIS!

 

Yep, for the most part anyone with aiflow readings that go outside the bounds of the stock maps would need to rescale, what you describe is definition a condition..

 

MJ, that is what I was thinking, but wanted to clarify for myself and possibly others.

 

you will know when you need to rescale if you are using a logger like evoscan. when you see no knock count and the timing is lower than any cells on your maps, you are out of bounds.

The only mods I have are a turboback and little higher boost than stock and I was regularly hitting 280 on 3rd gear pulls from 2,000 rpm. the 03 8 rom only goes to 260...

 

PDF has been hosted and posted! ...see first post in thread.

l8r)

 

I have a question for the ecu-flash guru in here, about changing resolution of the stock ecu map.

I've play around with the map editing and found out the other method but not very sure will it work in real life.

Ok, here it is. Click on high octane map, click on any portion of the map and press ctrl M, edit metadata-high octane fuel map will pop up and from there change the value on the x-axis load elements number from 18 to 20, change the y-axis rpm elements number from 15 to 17. Now you have 2 more extra colums to work with, on the load and rpm colums, will this method work??

 

Interesting SGEVOTX. I tried it and it gives me columns though I have to go thru and do each map and change all the 255896 numbers it generates back to something useable.

 

There is a value in the maps just before the first byte that gives the length of each section. Without changing that, and other unknown parameters that define table dimensions, you're at best making areas of map that won't be accessed, at worse translocating a load of information or overwriting other important data. You're swimming in shark infested waters. You can edit what you like but it won't have the effect you're expecting unless you've redone all the code that sets the table dimensions.

There are large areas of unused memory that could hold alternative maps or code, but the above is just like filling them up and then not changing the code to use them.

There is plenty of resolution in the stock maps - just sacrifice areas that have gradual or even steps and make the new areas you want.

 

That will not work... The extra columns have to come from somewhere, and usually its the next portion of map or code that you are destroying..

These maps are currently fixed in size, and would require quite a bit of work to relocate them (its not as easy as just changing that value) since you would then have to move every other map and every reference to every other map that the ECU makes..

In other words, don't do it! It will not work.

 

so it is possible for me to hit a 260 load cell at only around 21-22 psi, because that seems to be the timing cells i'm hitting at peak boost, but i thought 260 would be closer to 23+ psi. maybe it's time to upgrade the old MR boost guage to something more accurate, i might be accidentally running 23 psi while thinking i'm only running 20-21 psi. but if indeed i am hitting 260 cells at 22, would this be a good idea to add a 280 column? thanks.