TD_D mod journey. From stock to the 'bastardmobile'

[Td_d]

Interesting to see what it's doing on the open loop now that I have a wideband - you'll probably lecture me, but I could see that it was running rich, not lean - opening her up WOT in 2nd gear (so limited load) is running super rich - 9.6AFR (whereas the target AFRs in the primary open loop fueling are in the 11's in that range). Now's when I get confused as to how the meth is factoring in :lol:

 

[HolyCrapItsFast]

The meth is what is doing it. Remember that this MAF scaling is a derivative of the stock calibration and we have not considered open loop yet. So the ECU is calculating fuel as if there is no meth so it will run rich. I bet if you turned of the meth you will see you AFR is much closer to what the table suggests.

I will have a look at the logs you sent but I won't be able to get back to you until tonight. I have to replace the wide band sensor on my car today and iron out it's open loop calibration.

 

[Td_d]

I thought as much - since the whole point of meth is to raise the octane / make it more knock resistant.

No worries - when you have time! Thanks.

 

[Td_d]

Another question of interest - the tip in enrichment tables - for lack of a better way to describe it, they basically ensure that if you plant it, the sudden change in g/s air increase does not cause a momentary lean condition, right? So if my reasoning is right, you're more likely to see the impact of this in the lower RPMs, where you are likely to change throttle angle significantly to accelerate, i.e. the 2500 to 4000 RPM range, right? And the last question - surely if the injectors are larger, the pulse width compensations for tip in need to also be different?

 

[HolyCrapItsFast]

Yes this is all correct. :tup: I was going to focus on tip in as the final closed loop thing we do.

This is why I keep asking for smooth throttle inputs so we log as little form the effects of tip in for now... When I ask you to shift allot and drive without concern to throttle input you will know we are logging for tip in.

Looking at the logs right now. :tup:

 

[Td_d]

^ yeah figured as much - I think you'll find the data more consistent this log - it should have been a no brainer, but I made sure I drove in S mode not S#, keeping the throttle transitions a lot smoother.

 

[HolyCrapItsFast]

Fuck me!

I screwed up! The last one I gave you I shifted the entire calibration one cell to the left. I'm very sorry.

I corrected my mistake in this rendition. Could you load this one and go for one log?

http://www.drift-ready.com/Files/TD_D/750RC_3-01-2011_Rev-A.rar

Again I apologize.

On another note how is the stumble. I still expect it to be there but I expect it to be greatly reduced.

 

[Td_d]

Chuckle, hey, it happens - especially in Excel!

Stumble was much reduced, yes - still there, but not as noticeable.

 

[Td_d]

Believe or not, I managed to sneak out for a midnight drive, 12 minute drive, but should provide enough data!
http://dl.dropbox.com/u/3099895/TD D 01 Mar 2011 Log A.rar

 

[HolyCrapItsFast]

Okay here is the next one. I labeled it rev B.

It's another rough correction because I want to get the A/F Learned column mostly 0.

http://www.drift-ready.com/Files/TD_D/3-01-2011.rar

 

[Td_d]

Here's the log: http://dl.dropbox.com/u/3099895/TD D 02 Mar 2011 Log A.zip

Crunching it now (only on MAF vs Learnt correction) and it's practically flatline around the zero axis. The largest individual correction is also -5%, so I think the learnt factors are pretty much dialed in. Just a heads up - I'm going to be out of town from tomorrow for 2 weeks, without the car - so will be able to do another 2 runs today or so, and then will have to park it till I'm back. Will still be online to shoot the breeze though :lol:

 

[HolyCrapItsFast]

Okay. I'll see if I can get you something out real soon!

I'm not going anywhere so at your leisure man. Just let me know when you want to pick this back up after your return! :tup:

 

[HolyCrapItsFast]

http://www.drift-ready.com/Files/TD_D/3-02-2011.zip

Here ya go. Tell me if the stumble improves at all.

I was noticing that there was some knock that was corresponding to lean AFR's in the 3000rpm range. The closed and open loop targets are good in those areas but I richened the fuel in those areas anyway by just a point or two and I made an adjustment to the load compensation tables for the corresponding RPM/Relative Pressure cells.

I also made an adjustment to ignition timing for those Load/RPM cells as well. I removed 3* of timing.

 

[Td_d]

Thanks Holy, I'll see if I can load and drive. Damn it, I've got a 6-7 hour drive tomorrow, and of course, I'm sick as a dog. Bronchitis FTL, love coughing up bits of lung.

I noticed that knock - never translated into FLKC, but pretty consistent around the same range - I thought it might be a tip in issue. Drove around on the tune the whole day, quite a range of conditions - felt quite lumpy on the bottom end, especially in S# (and stumble very evident again...)

I'll be on tomorrow, just without the car :lol:

 

[HolyCrapItsFast]

Okay. When ever you have a chance please give me a very short log of only the stumble condition.

I hope you feel better soon!!!

 

[Td_d]

Thanks Holy, that's the plan. When I'm back, you've got to take me through the comp tables - I've done a whole load of reading on the Romraider forums, so I'm getting the basics, but man... it makes your head hurt!

 

[HolyCrapItsFast]

http://www.drift-ready.com/Files/TD_D/750RC_3-02-2011_Rev-B.zip

Oh... Next time you log use this revision instead. I zeroed out the Load Comp tables. We should customize these tables to your setup instead of using something that was derived from stock. This is something that needs to be specifically tuned to the hardware being used.

 

[Td_d]

Ok, running out - but managed to get three short logs - feels much smoother!

http://dl.dropbox.com/u/3099895/TD_D_3_3_2011_A - .rar

 

[HolyCrapItsFast]

And that was done with Rev B? :tup:

Thank you, I'll have a look!

 

[HolyCrapItsFast]

Things are looking much better lately. We virtually eliminated the knock but the A/F learned still isn't where I want it to be. Maybe I'm just being over anal about it.

Anyway this time I applied the new Load Compensations to the map. This will either be awesome, or it will have horrible results.

This time to get a better data scatter I took a different approach when compiling the errors. First I applied the standard filters to the the logs but then added one more filter to filter out everything with a maf voltage less than 1.50. I applied this to the new maf scaling. Then I applied the same logic to maf voltages above 1.50 and ran it again. I wanted to isolate and improve the resolution and data scatter for those regions individually and it seemed to work. It was allot more work but it wound up being more accurate.

Let me know when you have a chance to apply these changes. I will be waiting the two weeks with baited breath! :lol:

http://www.drift-ready.com/Files/TD_D/3-03-2011.zip