AP and the ECU - how it operates

[bova80]

So by now you are all sore in the pants on how Cobb is dropping the Mazdaspeed development, oh well, it happens get over it. However, Christian has informed me he is not done working with us as he still has a lot to share with us on how the ECU operates. In this thread this will be discussed as well as things to look for in logs in determining if you have a problem. The first log I'm going to post is that of a bad 1st O2 Sensor. Other logs that will be posted will be as such:boost creep, boost spikes, uncontrollable boost, too much knock, etc.

as you can see in this log the sensor is only reading 14.85.

 

[MShreve]

Why is boost PSI showing negative numbers?

 

[bova80]

because its in vacuum

 

[ugadogs_11]

So I still have yet to figure out if they are done with everything Mazdaspeed? I was hoping to get a AP soon,and wanted to know if it is even worth it.

 

[tunersteve]

So I still have yet to figure out if they are done with everything Mazdaspeed? I was hoping to get a AP soon,and wanted to know if it is even worth it.

This isn't the thread for that, go to the 'News' section and read there.

 

[MontegoRx]

I see what you're saying about reading 14.85 exclusively, but think that saying "only 14.85" would lead people to believe that it's a low reading, and they should be seeing higher numbers. If someone takes that approach, they will be fast approaching problems.

That being beside the point, I'm interested to see the graphs showing boost creep/spike, as it seems alot of posts confuse them or just plain don't know what they mean. And since I'm new to this car/forum, I'm very interested in the LTFT and STFTs that I've seen mentioned. I see that your LTFT is rock solid, which I have to assume would be a very good sign.

The thing that catches my attention in that graph the most, though, is the Rel Throttle vs. your Timing. I have a bone stock 06 MS6 and can't log. If I could, I'd be pretty interested to give it a normal days' drive and plot those two against each other to get some more insight.

 

[tunersteve]

Seeing a consistent value for any parameter is usually a sign that something's wrong. AFR is a constantly changing thing, even to a small degree.

 

[MontegoRx]

better yet, here's the graph....

 

[MontegoRx]

and it apparently doesn't want me to embed pics, even after using the URL and IMG tags combined....

 

[MontegoRx]

Seeing a consistent value for any parameter is usually a sign that something's wrong. AFR is a constantly changing thing, even to a small degree.

I think you misread my post. I was talking about the LTFT being consistent as a good thing. I understand that AFR will be constantly changing, which is why it seemed weird that "understanding the ECU" would include something that basic. There's not alot of technical information on this forum, truth be told, so a thread like this would hopefully draw attention from folks who understand the basics of AFR and realize that a steady 14.85 isn't admirable.

 

[tunersteve]

I think you misread my post. I was talking about the LTFT being consistent as a good thing. I understand that AFR will be constantly changing, which is why it seemed weird that "understanding the ECU" would include something that basic. There's not alot of technical information on this forum, truth be told, so a thread like this would hopefully draw attention from folks who understand the basics of AFR and realize that a steady 14.85 isn't admirable.

I was referring to the OP's situation that he was explaining.

And yes, you're correct, you ideally want to see your LTFTs fluctuate as little as possible. Also, you shouldn't be seeing any knock either.

 

[MontegoRx]

I was referring to the OP's situation that he was explaining.

And yes, you're correct, you ideally want to see your LTFTs fluctuate as little as possible. Also, you shouldn't be seeing any knock either.

Sorry for misunderstanding your post then. Strange, though, how you look at just past 15 seconds on that graph and see the Throttle position vs. the Advance, and that's the ONLY time it registered knock....

I think I'm going to be looking into that correlation a bit more and see what I can figure out, but without real world driving logs, I won't be able to come up with anything.

 

[MontegoRx]

been a while with no updates. Do you have anything on the boost creep/spike issues? I'm interested to look at some more of these logs and see what kind of information can be pulled from them.

 

[ms6dj]

It would be fun to take this data and run it through a stats analysis program like Minitab (or even Excel's stats functions) and run a correlation matrix, multiple regression etc on all that data to see what is correlated and how strongly. Of course it would probably just tell the experts here things they already knew, but for someone new to the whole ECU and tuning world like me, it might be a really interesting way to learn how all this data interacts. Maybe after I get my AP tomorrow and start logging, I'll do just that!

 

[MontegoRx]

that was kind of my idea too ms6dj. I don't have the AP, and probably won't be picking one up this year, so I can't generate the logs myself.