ECU and maps - help me understand

[Mid_Life_Crisis]

Someone correct me where I am wrong, please.
This is my understanding of what is going on in the ECU.
There are lots of maps stored in there. My understanding is that they fall into two categories. I'll call them "Cause" and "Effect". Lets start with Effect.
This is what the computer wants to see on the different input sensors. Based on what it sees for example on the throttle opening, rpm and ambient temperature, it expects to see certain boost levels, oxygen sensor levels, etc.
These are what is actually happening. Under part throttle conditions it will tweak things like air/fuel mix, timing, etc to bring these values within target range.
Cause - these are starting points for things like timing, air/fuel mix based on what it sees for boost, rpm, etc. When you first start the car it uses these until it gets enough feedback from its sensors to tweak things. Under wide open throttle it stops tweaking and just uses these maps.
When you are driving around normally, the ECU learns that it is constantly changing particular values on the Cause maps to get the results it is looking for by referencing the Effect maps. So the Cause maps are constantly changing. This is what we talk about when we call the ECU a learning computer.
The Effect maps do not change. They are preset to what Mazda believes to be optimum. These are (I hope) capable of being changed by the Cobb AP, but not by the piggyback units, which is the main reason the AP should be superior. If the car is producing a mix of 10:1 and you want 11:1, then tell it to make 11:1 and let the ECU learn how to do it.

 

[Haltech]

Lets change the verbage. You have a Closed Loop map and an Open Loop map.


The computer is always looking at the sensors for information and programming is always changing as onboard conditions change.

 

[Mid_Life_Crisis]

Lets change the verbage. You have a Closed Loop map and an Open Loop map.

The computer is always looking at the sensors for information and programming is always changing as onboard conditions change.

I wasn't sure which was which, but other than that, my understanding is correct as far as you know? One set of targets and another of starting levels?

 

[JCell]

doesn't mazda use fuel trims for tunning?

 

[Haltech]

There a lot of IF>Than variables the computer is looking at. You have multiple tables the computer looks at depending on temperature, timing, boost levels,AFR, BAP. MAP and Heat. For instance, the Mazda ECU has a glitch that says if the weather is below 35F, cut boost and timing. Will it be fixed? Who knows but we can tune around it by manipulating the IAT values.You have different timing tables for open loop and closed loop. Different Air/Fuel mixtures for idle, part throttle, part throttle 50% load, part throttle 100% load, Open throttle 100% load, etc. The Narrow band 02 sensors mean nothing at WOT, however the Wideband Sensor is crucial at WOT.

 

[Mid_Life_Crisis]

There a lot of IF>Than variables the computer is looking at. You have multiple tables the computer looks at depending on temperature, timing, boost levels,AFR, BAP. MAP and Heat. For instance, the Mazda ECU has a glitch that says if the weather is below 35F, cut boost and timing. Will it be fixed? Who knows but we can tune around it by manipulating the IAT values.You have different timing tables for open loop and closed loop. Different Air/Fuel mixtures for idle, part throttle, part throttle 50% load, part throttle 100% load, Open throttle 100% load, etc. The Narrow band 02 sensors mean nothing at WOT, however the Wideband Sensor is crucial at WOT.

I just said "lots of maps".
I wouldn't be surprised if more like IF >X or <Y is more the case.
Serious multiple dimensional mapping going on in there, which is what Cobb was claiming was a major cause of the delay on the AP.
I wonder if it is a flaw in the ECU itself or just a screwed up map, in which case the AP could easily fix it.
What I don't like about the only IAT fix I have seen so far is that it appears to lock in one setting. I would like to see a design that works more like the MAP clamp, where it allows acceptable values in one direction but puts a limit on it in the other. In other words, instead of making the computer think it is always 50 degrees, it just never lets it get colder than 40. The way the MAP clamp lets lower pressure readings through, but blocks the higher ones.

 

[Haltech]

doesn't mazda use fuel trims for tunning?

Of course, but thats not the only thing that is observed from the computer to make those changes. You have long and short fuel trims.

 

[jp4130]

There a lot of IF>Than variables the computer is looking at. You have multiple tables the computer looks at depending on temperature, timing, boost levels,AFR, BAP. MAP and Heat. For instance, the Mazda ECU has a glitch that says if the weather is below 35F, cut boost and timing. Will it be fixed? Who knows but we can tune around it by manipulating the IAT values.You have different timing tables for open loop and closed loop. Different Air/Fuel mixtures for idle, part throttle, part throttle 50% load, part throttle 100% load, Open throttle 100% load, etc. The Narrow band 02 sensors mean nothing at WOT, however the Wideband Sensor is crucial at WOT.

Haltech I was under the impression that the narrowband was not used at all for fuel metering purposes, is this not the case?