NefMoto

Hi Guys,

I have been trying to overcome ignition retards issue (torque intervention) at low and middle rpm range for some time and at last decided to ask for your advices.

So I have AUQ engine with HN-box, GT28RS and S4 Bosch MAF, huge intercooler and non-stock intake. When the MAF was installed I calculated KFMIRL and KFMIOP for my requested boost 1.3 bar , moved MLHFM, MLOFS, KFMLDMX maps from A-box bin file. After MAF installation I noticed ignition retards at low and middle rpm range about 6-9 degrees. A lot of time was spent for analize log files and too much litters of gasoline were burned. And in order to overcome ignition retards I had to increase IRL map up to absolutely unreal values because of this now I have many other issues like unstable idle, too hight idle rpm when the engine is cold, hesitated car on the first gear etc. The table is here https://www.dropbox.com/s/jrqikkfb235gi48/kfmirl.xlsx

In addition: requested AFR and actual AFR is the practically the same on retarded ranges.

And the main question: Guys, does anyone have any thoughts why this happens? How can I overcome ignition retards (torque monitor intervention) without unreal IRL?

Thanks in advance!

How did you calculate IOP and IRL? You should leave the lower ~half of IRL stock and scale up the last few columns for your new load. Then re-scale IOPs axis for the new load.

If I understand you right, you need increase KFKEF and RKRMXN... but retardation is a good thing, it protects your engine from damage  :-\

catbed +1

I calculated IOP and IRL by myself and by "ME7 tuner wizard" as well the same result.

I also thought like you and at first I make IRL as you said "leave the lower ~half of IRL stock and scale up the last few columns" but I had a lot of retards and I had to increase IRL table at the lower half when I had a lot of retards.

Yea sure, I've forgot to write I rescaled IRL axis for my load.

I want to see the file you had lots of retards with.

IOP and IRL as you had it please.

It doesn't seem knock because knock sensors voltages were not high (absolutely not high) and retards disappeared when I increased my IRL.

I agree with you retards is good thing but only when it's needed. :)

Hi Daz!

Unfortunately I've erased old log file. I'll do it today.

Thanks in advance!

I don't want to see the log, I want to see your tune :P

Oh, I see :)
With retards https://www.dropbox.com/s/mcqilq3o9t8y1cc/TorqMon-with-retards.bin
Without retards https://www.dropbox.com/s/wiiazz5vt3y2lon/TorqMon-without-retards.bin
MAP pack https://www.dropbox.com/s/3mb5kboy92ub7ga/Skoda_Octavia.kp

LDRXN=193
IRL=190

Take a look at the values in a stock file.

What was the actual load that was logged? Desired? I find it hard to believe you weren't exceeding 190 with a GT28R. I mean, K04 cars do that.

How about inter/extrapolating KFZWOP/2 since you drastically changed the axis? That, and IOP is not great. Don't rely on the tuner wizard, it takes maybe 5 minutes to do it right.

Give this a shot. It's your file with a proper IOP/IRL relationship, recalculated ZWOP/2 and MDS tables in addition to fixing some other whatnot that I noticed. I honestly would have went further, but I had to draw the line somewhere. Shouldn't get any sort of intervention. It was a quick and dirty edit, not responsible......

You get the point, flash at your own risk.

In stock file LDRXN=135 and IRL=about 182.
In my file IRL little bit more than LDRXN as well :)

But I was pretty sure that  LDRXN cuts load and my LDRXN is more than IRL.

You are right my rl_w is more than requested load.
max(rl_w) = 204
max(rl) = 191.2
max(rlmx_w)=195

But I'd like to notice retards happen at middle rpm range from 1500 till 2800-3000 rpm.


I didn't touch KFZWOP/2 and KFMDS. I thought at first I should overcome timing intervention and after that adjust all other tables.

Could you explain why my IOP/IRL interpolation/invert wasn't correct?


Thanks a lot for your efforts! I'll move all your changes into my file and try.

Now I know that I was mistaken and I should adjusted all maps together IOP/IRL, *ZWOP* and MDS.

Please don't move any changes. Run the file I uploaded. It's YOUR file to begin with. If you have issues, how will I know you found all of the changes?

I use connect windows in WinOLS that's why I can't miss the changes. ;) Anyway I'll try file from you.
Again, thanks a lot for your efforts!

Could you explain why my IRL/IOP were wrong? Why did you change KFZWOP?

Thanks in advance!

ZWOP needs changing because it shares the load axis with IOP.

But retard appear at range axis which wasn't changed. I mean low load rpm. I just trying to find out how it can be related to retards.

I have no idea. You haven't posted any logs. When asking for help it's customary to provide the data necessary for us to provide that help.

Just posting:

"This doesn't work, why are there so many retards" (besides being hysterical I might add, thank you for that)

will not get much feedback I'm afraid.

Hi Daz!
Unfortunately all my log files were converted for logviewer program.
I'll use you tune file and write new log.
Thanks!

Hi Guys,

I've found one of my old log files with retards. Please don't pay attention to retards with AccellPosition=0%.

https://www.dropbox.com/s/0ia0ynqu42m9i48/winols_template_20130926_085028.zip

Daz, could you look through the file?

we <3 them retardz

What do you mean?

What makes you think that the intervention is resulting from torque management?

Does zwist == zwopt during the ignition retard?

I think so because of several things:
1. Knock voltage is too low during the reterds (I mean not due to knock);
2. Retards appear at low and middle rpm range and low accellpedal position;
And main thing - each time when I increased IRL and rescaled IOP retards were lower and lower until IRL raised to unreal values retards gone totally.

Practically all the time zwopt > zwist including during the retards.

Retards is a derogatory term used to describe mentailly handicapped people. The way you're using it is very funny.

I have apricated your joke.

Stupid question:

isn't not running at ZWOPT by definition torque intervention?

I'd say no,

there is no intervention occuring.  Zwopt is just used as a base for determining current torque and to reduce torque when required.

Torque intervention, for the fast path at least, is when zwist=zwsol

Rick

you mean fast path torque intervention is when zwist<zwsol ?

but technically, zwist<zwopt is torque intervention as well... just like maintaining torque reserve is technically torque intervention.. so its a semantic argument if you aren't using ME's definition of "torque intervention".

I.e. the OP is saying that ANY time zwist != zwopt ME7 is "retarding" timing.

I think his terminology leaves a lot to be desired...

I'm little confused...

As far as I understood I will have ignition retard always if zwist != zwopt but they always not equal.

Also I've noticed ignitions retard mostly appear when zwsol > zwist. The screenshot here https://www.dropbox.com/s/kvwet2t8rsnchlb/ign_retard.jpg

Hi Daz!
You helped me once again! Thanks a lot!

With "your" file my ignitions retard not higher than 3.75 degrees and it happens pretty less times.

Do you think my problem was in not rescaled KFZWOP/2 and MDS axises?

This is correct and what I meant to write.

zwist will never be equal to zwopt under any kind of load.  Zwopt is optimum timing for a non knock limited engine which rarely exists in practice, at least not on pump fuel.  It's there simply so that the ECU can know how much to retard ignition by in order to reduce torque if it requires.

Fast path torque intervention is when zwist does not follow zwbas but instead zwsol.  Optimum timing for the engine comes from correctly calibrated KFZW maps and if actual timing isn't at this point, and you have zero knock, then you have intervention.


Rick

Hi Daz and ALL,

I have retards ignition again. I really don't understand the nature of these retards.

Could you look through the log file and point me?
https://www.dropbox.com/s/s6elmjb0jxpolj2/winols_template_20131014_181622.zip

Regards.

please just post the log as an attachement, or use ecuxplot and post a shot of rpm vs zwist/zwsol

Hi nyet,

the log file is attached. I'll make screenshots later.

Faulty clutch switch

What do you call retarded folk in your language?

Ugh.

Can you just post a single WOT 3rd gear pull through a large rpm range?

That log is just a huge mess of data, unless you can pare it down to the areas that you think are suspect.

According to log file the clutch switch is working properly.

I mean "negative Correction Factor", "reducing ignition timing" etc.

I don't have any ignitions retard, boost or AFR issue under WOT. I have ignition reducing (ignitions retard) (I don't know how call this effect also :) ) only at low load about 20-30% accel position and 2000-3000 rpm range.
But anyway I've attached today's WOT log on 3rd gear.

I've attached cut log file and left there only retarded area.

winols_template_20131015_091906_ecux.zip - WOT file
winols_template_20131014_181622_retard.zip - with ignitions retard file

In addition: in the most times when I had ignitions retard zwsol < zwist

I think I have found solution. I will update the tipoff when I check.

It looks like your actual load spends almost all of its time over your requested load.

Also, at WOT, your load is significantly over 190 pretty much everywhere, which means you are probably on the last row of your timing table... which is probably near zero or negative.

Your MAF looks to be calibrated completely wrong.

Did you meant under WOT?

You are right my timinig is near zero and negative. I'll rescale KFZW load axis.

I have big frontal intercooler and free air filter. My MAF is 0 280 218 038 which is installed in S4 a2000 therefore I moved MLHFM, MLOFS and KFLDMX maps from A-box stock file.

MLHFM which I use is attached.

Could you advise what I did wrong with MAF calibrations?

You may also want to log ps_w to make sure you aren't maxing it.

you might need to underscale your MAF if you are.

Under WOT ps_w is 2559.96

Could you explain?

heh.

5120 or underscaled MAF time.

http://nefariousmotorsports.com/forum/index.php?topic=2747.0

Thanks!

Hi again!

Guys, thanks a lot for your hints and notices!

I've rescaled MAF calibrations and now rl_w is almost equal rl and also never bigger than rl. Because of this Requested AFR is pretty close to Actual AFR. But I'm still overcoming ignitions retard and I noticed several things:
1. Most of the ignitions retard happen when coolant temperature is less 80C;
2. In some area when rl_w > rlsol_w when WG is closed but turbo began spooling and ignitions retard happen.

So what would be better to smooth the difference with rl_w and rlsol_w by rescaling MAF calibrations or with IOP/IRL rescaling at these areas?

It seems related to optimal efficiency KFZWOP and something like that.

not an issue

not an issue, unless you are WOT.


No. Please study up on torque based systems...

Hi Guys,

Guys, I have to ask for your assistance again. I have remade many things but I'm still trying overcome ignition timing intervention. I haven't managed to find out with nature of the this issue. It seems to IOP/IRL issue.

DD, nyet, Rick - guys could you look through my WOT on 3rd gear log file and point me a little hint? I just stuck. :(

Thanks in advance!

In addition: In my understanding I need to reduce IRL with different efficients ETADZW*, DZWOLA* etc.

Have you tried disabling the torque intervention? That way you'll see pretty quickly if you kfmirl/IHOP maps are screwy

Yes, I've tried and I didn't have timing interventions. Yes, I know this is the answer for my question but I can misunderstand that's why I asked the assistance from guru.

I think I've found the issue my rl_w is always lower then rlsol_w so because of this my IOP most of the time is lower then IRL therefor I need to upscale back my ps_w in order to make rl_w higher.

Am I rigth?

Hi Guys,

Thanks a lot for your assistance!
The issue was fixed.
I've rescaled ps_w and rlsol_w and adjusted IOP table.