NefMoto

Hello again all,

This thread is in regards to some timing CF issues i'm experiencing.

I'm seeing timing corrections at IDLE, DECEL, and THROTTLE TIP-IN which of then continue on to WOT but start going down as the revs climb.

This was not the case when the MAF was underscaled but i do believe the timing map was much different then as well. I know because of this issue i see about 5deg less timing up top which is VERY noticeable. The timing map below about 80% load is pretty much stock or a little less for the sake of keeping that V8 down low drivability.

My tuner and I do not have an A2L specific to this ECU nor have found one so the variable locations for the one we do have do not apply. My ECU is as follows:

HWNumber           = "0261208443"
SWNumber           = "1037374043"
PartNumber         = "8E0910560K"
SWVersion          = "0010"
EngineId           = "4.2L V8/5V"

Locating necessary variables hasn't been successful either. (zwist, zwsol, zwbas for example)

I've attached a log of course for review. I forsee needing to have this ECU decompiled in the future :-\

log

If you expect to add FI to a NA ECU, you had better have the ability to trace ANY variable at will..

Yes, it'd be nice if i had that skill but i dont and there are others that dont. So here i am looking/hoping for help. :-\

Have you looked through my definition file project, may be of some assistance :)

http://nefariousmotorsports.com/forum/index.php?topic=8943.0

but of course.... Then I found a damos lol

Really, is there a DAMOS in existence other than mine for a 4.2 V8 S4????

If so, please post that up if you could.

the OP is crazy, there is no damos. Bosch of course never created one. yours is the only one, ever.

*snicker*

 :P

The DAMOS isn't in my possession. My tuner has it, i just paid for it. I'll ask if he want's to post it.

I'm going to start asking my clients if they'll buy me definitions.  ;D

But on the real, you should post it up!

no, he shouldnt post it up.

lol why? glass bead trading games rofl

at least you took the time to read all the posts in this thread correctly.  ;)

I went from left to right. I even did it twice! What did I miss?

you didnt miss anything, you were there.

Even I'm like double you tee eff lol

I think its fair to ask your friends to buy a damos when you are helping them out pro bono.

I whole heatedly agree to that

latest log of the timing pull issue.

Thoughts?

Don't post xls files. They're utterly useless for parsing.

sorry did that to drop the file size down so the forum would allow it. There is a bug that needs fixing.

The forum says: "Maximum attachment size allowed: 29000 KB" aka 29ish MB... but yet what that really means is 2.9MB come to find out.

Anyway here's a link to the log in csv for your pleasure:
https://drive.google.com/open?id=0B4w-boswX0J3XzdPeGVDLWVoVGs

So zip it and upload it.

next time

Looks good. Req load seems crazy high.

req load and fuelling is a mess, it's úber rich

yea it stopped meeting target a/f which makes no sense.

Req load needs to be dropped back down. It was that high from when we were trying to diagnose the MAF limit issue.

Log short term trims and try to log the various enrichment paths (ATR etc).

There's a lot of enrichment paths, ATR for one.  Log time :)

edit Basically what nye said lol.

Here's channel/block 001 from VCDS to verify fuel trims are working up top.

Something definitely off... Not sure what would dump fuel w/o altering req lambda.

The flat line before the cliff is definitely an indication that something went wrong BEFORE fuel started dumping.

For reference, this is how FKKVS looks on those previous files (which I had nothing to do)

(http://i.imgur.com/irCv37j.png)

A pro map  :P

I'm fixing things as I go but that timing intervention is quite annoying.

SMH

Do take note this was written when the car was with a returnless fuel system.

This log was taken on the way to a restaurant after a fresh flash and approx 7mile drive so the car was warmed up.

have a look, timing pull is all normal. no intervention. She feels great...

And this log was taken about 45mins later after eating. No changes other than a full belly.


WTF IS CAUSING THIS!? Clearly something is activating/triggering intervention.

Can you get a log where you are actually WOT for the whole run?

Its almost impossible to tell what is going on with you feathering the throttle through the run.

Literally EVERY one of your runs has a different throttle position, and a different start/stop rpm, making it impossible to compare apples to apples. It's driving me insane :P

Pick a consistent thing you are trying to log, and STICK TO IT

that's because it's not WOT I'm worried about. It's the fact I'm pulling timing at IDLE, PART THROTTLE, THROTTLE TIP-IN, AND DECEL.

If you look at the last log, when it starts, the ecu is pulling timing at idle. I rev the engine and the timing pull stops and stays at zero. I get moving and it comes back. Maybe that's a clue?

WOT has been consistent with every other run I've logged in the past where it pulls a degree or two up top. The only difference is when this lower load timing intervention occurs it results in a couple degrees less timing up top which is aggravating but AGAIN it's not WOT I'm focused on in this thread.

Delete.

You know it's kinfa funny, after spending so many hundreds (if not thousands) of hours making my own DAMOS. I almost don't want there to be a DAMOS for that file lol. It would seem awfully anticlimactic to see my work made obsolete.

Then again, it would be kinda cool to compare the two and see how close I was.

Odd that there were never any DAMOS files released for that engine though. Although it's a low volume production unit, it's not THAT rare. Hell, there are 2-3 different DAMOS files for the W12 engines, which are far more rare.

OP, you may want to ask your tuner where he got his 4.2V8 DAMOS, it may look awfully familiar lol [;)]

That looks like it is torque intervention doing exactly what it is supposed to; limit torque part throttle - but without a consistent methodology pull to pull (the EXACT same pedal position) you're going to have problems.

Every single log you have there is with throttle at a different position, so the torque limiter is acting differently, making it very hard to diagnose where to start.

And you're going to have to log the ENTIRE torque limiter path if you want to "fix" this problem, since it seems to happen past the area where a NA motor would be.

This is a gigantic problem when adding FI to a NA file. Unless you can literally log everything, and you have the entire torque path disassembled, you aren't going to get anywhere.

Yes, you're right. All that advice is fine and dandy but i'm still looking for someone to lend a hand. I'm not an engineer. I'm not a computer programmer. I'm just a customer reading and learning and i have a car that someone else is tuning.

Would you like to take a stab at disassembling the torque path so those variables i need to log the ENTIRE limiter path can be located?

Lol,

Take a stab?  I takes hours to understand wtf the code is doing.

Exactly. NA->FI is NOT something you're going to be able to DIY, even with help in this forum, unless you are really very very good at everything, not just tuning and cars, but programming, disassembly, assembly, etc.

The list is pretty long..

You should not have gotten into a NA->FI project w/o more due diligence.

Agreed it needs someone with a fair bit of time and skilled at all that or for the OP to learn it, like I did (cya in a yea or so).

MAF diag springs to mind, load limiters eveywhere, ATR will be low, cat stuff will be low, timing maps will be off, injectors, blah blah..

Raise you're lower lambda limit as a quick fix, you not getting smoke at that afr? lol.

Alright, i get it! LOL...poor choice of words ;D

Fueling was cleaned up last night. FKKVS needed a "haircut".

Anyway...well damn. This might take a while.

Guess i can at least start with logging. Here's what i have available to me currently:

;mifa_w; {}; {indexed engine torque driver command}
;mifab_w; {}; {Limited indexed Driver input torque}
;miges_w; {}; {Indexed desired engine torque for transmission protection}
;migs_w; {}; {Indexed desired engine torque for rapid intervention GS}
;miist_w; {}; {indexed engine torque high-pressure phase value}
;misol_w; {}; {Indexed resultant desired torque}
;misopl1_w; {}; {Solluftmoment, calculated back to lambda = 1 and zwopt}
;miszul_w; {}; {Maximum permissible indicated torque}
;mrfa_w; {DriverRequestedRelTorque}; {Relative driver's desired torque of FGR and pedal}

;B_mibeg; {}; {Condition} torque limitation is active
;B_krldy; {}; {Condition load dynamics for knock detection active}
;B_fev9; {}; {Condition long load exceeding rl over max value}
;B_fev10; {}; {Condition short load exceeding rl over max value}

;rl; {} engine load; {} relative air charge
;rl_w; {} engine load; {relative air charge (Word)}
;rlsol_w; {EngineLoadRequested}; {Target} filling

Need BTS, ATR, etc too, it will be one thing after another I think.

Throttle will need redone as well iirc.

Pretty sure those two are disabled. Pedal is pretty on point. No surging or limiter stage 2 like i used to get (especially at highway crusie and trying to pass in high gear).

3rd gear pull.

 ??? ??? ???

Until i find someone who can help break down/decompile/recompile/trace this ECU's load path then this is the crap i have to deal with.

Check the S4 Wiki, Nyet did a very comprensive write-up about the ME7 torque structure.

Also, I would highly suggest getting yourself a registered version of MegaLogViewer from EFIanalytics. It's only about $70, but I live and die by that program when it comes to tuning my cars.

It allows you to take the CSV and lay out trace graphs, scatter plots and histograms.

FYI: you're running into problems whenever your misopl_W value exceeds your mifa value.

If you look at the logs, you're running into timing interventions even at part throttle. But at WOT, the timing pull begins at 3200 as soon as you go WOT

Man, thank you very MUCH. I'll look into all of that! I've been using NYnet's ECUxPlot program to display all the data.

The part throttle pull has me concerned as it continues on to WOT despite the change in condition.

so..MRFA_W is going to 105% but my MIFA_W is staying gradual. It's not hitting MIMAX exactly, its a few percent below it but is that still indicative of intervention? Should MIMAX be raised to what MRFA_W (from KFPED) can go to?

This binary is significantly different than your standard ME7.1 so the wiki nor the circulating FR's help out much in this case. This ECU has a bunch of modules that are set up exactly like MED9

Well, here's the basic stuff for logging. Set up me7l cfg file with factors from "similar" known vars.

mizsol_w 0xCC96
mibas_w 0xCCAA
misolv_w 0xCC90
mizsolv_w 0xCC94
dmzms_w 0xCC82?
dmaufr_w 0xD298
etazwb 0x3809E0
dmllr_w 0xCAFE (tee-hee)
dmar_w 0xCBB8
zwbas 0xF535
dzwb 0x3809D9
miopt_w 0xCCAE
mimax_w 0x3853DE

B_zwvz 0xFDA8.4? or .3?
B_zwvs 0xFDA8.2

I also strongly suggest FFing TMAR at 0x10FD8, to take ARMD out of the equation.

Also I think that it's a good idea to log rlmax_w at 0x385272

Most of the addresses should be correct. I definitely need some sleep now, will update the post tomorrow with more vars.

I already tried that and it wasn't pretty. Without dashpot anti judder it was twitchy.

KFDMDARO is currently FF

wow that was quick. thank youuuuuuuu. wish i had check this before i left to go logging tonight!

From what I've seen, ECUxPlot is a great program, however, Megalogviewer is hands down, the best program I've ever run across if you're trying to hunch out an issue.

Seriously, go to this link and check it out https://www.efianalytics.com/MegaLogViewerHD/ and pay to register it. You will NOT regret it.

Well I do not propose always running it like that. For testing purposes only.
Orrrr, well, just watch the dmar_w - if it's zero, you're fine ARMD-wise.  ;)

I strongly suspect that the problem originates with rlmax_w since the car is originally NA.

alright here are the logs, i'm only 90% sure if i entered the correct factors in the fields for the new variables.

When in doubt - use "RAW" conversion factor of 1, compare with other vars and correct the value afterwards. ;)

I didn't like what I saw in your log happening with the variable I named "rlmax_w", which is why I went and picked another (NA) ECU to cross-reference stuff to.
Turns out, the NA var is actually called rlmxs_w, and funnily enough you have already looked that way (http://nefariousmotorsports.com/forum/index.php?topic=10472.0).

So, after rlmxs_w is formed, it goes through KFMIOP to form mimxl1_w, which is then multiplied by FMIVL to finally output mimax_w.
Anyways, looking again at your log, your WOT run doesn't seem to trigger any tq intervention?

I'll look at the part throttle one in a few.

Well, part throttle is not that happy, but nothing too severe. Some ARMD action here and there, etc.
Let's start with the basics: what are we trying to solve? What is the initial problem? WOT run seems fine torque-structure-wise. That "sawtooth" pattern you see on the zwbas/zwout plot is normal since you have different CFs in different cylinders - and the logger just samples them one by one. So it can pick 0 degree cylinder, then pick the one with -4.5 CF, then another one with zero. Boom, you have a dip on the graph.

If you're concerned about the lack of ignition advance up top - well, tune the KFZW tables then? Adjust the axis if it's not enough. but be careful with other maps that run off the same axis. Let me know, I can list those for you.

EDIT2: Nothing to be scared of. SRL11OPUW works only for KFZW/2 and KFMDS in this bin.

EDIT3 (I just love editing my posts, yeah): Might want to play some with ZWOP to make etazwb a bit higher, but that's optional, I guess.

Exactly this. Torque intervention is doing EXACTLY what it is supposed to part throttle, and it doesn't seem to be interfering with WOT.

That's what i wanted answers on and finally Alex stepped up and helped alleviate my concerns by providing variables.

My overall concern because i'd never seen so much intervention before was part throttle and decel. Why does the ECU pull MORE timing when i'm completely off the gas. It didn't make sense to me but if it's of no issue and is not affecting other conditions then i suppose this case is close.

My last few revisions ironed out some things. PT and DECEL is still weird but the car drives fine.

I believe we had a few issues with timing intervention earlier on but I think its been solved in the past 3 revisions or so. Let's see what happens when we go for more power

Thanks for the help so far