NefMoto

Hello,
I've been reading NefMoto as much as possible so I can tune my own file. However I have a GIAC stage1 tune installed which is preventing me from reading/writing anything to the ECU. I running a Frankenturbo F21 which seems to be producing way too much boost and airflow for the current tune. I need to ground the pin and try reading/flashing in boot mode.

Anyways, more details about the car:
- ECU 8E0909518AK 0261208230 1037368072 0003 1.8L R4/5VT (AMB engine)
- Frankenturbo F21 including high flow manifold, turbo inlet pipe, test pipe, front mount, diverter valve, etc..
- 386cc injectors, stock maf+housing, GIAC stage 1 tune

Could someone please look at the logs and let me know if the fueling/ignition/temps I recorded are within reason? I am asking because the frankenturbo makes so much airflow and power that the motor instantly detonates or knocks when the boost spike hits even on part throttle. The GIAC tune boosts to 26psi and tapers down to 22psi which is murder on the engine right now. I've adjusted the boost to 10psi until I can figure out what the problem is, but looking at the logs I'm not sure if it's safe enough to do a full WOT pull to redline. I don't think that the GIAC tune is even designed for the 386cc injectors and the maps are only meant for k03s, so I need some input ASAP.

If anyone could give me input in regards to lean/rich AFR and timing/fueling that would be great.

Best wishes!

You are insane to be running that file with the wrong injectors.

Here is a CPUxPlot representation of the logs I posted.

Also, I managed to dump a copy of the stock ECU before it was crippled with the GIAC tune. I have attached the full ECU BIN in case anyone needs it!

Thanks!

Your car has returnless fueling meaning you are going to need larger injectors for that turbo like 630, 550s or larger. You are ruunning out of fuel the way you sit right now boosting 26psi on the tt injectors. So to do this setup proper youll need a larger maf (tt 225 or vr6) housing. Youll need larger injectors or convert to return fueling for overhead. The stock nap sensor only sees max 22-23psi so boosting over that the ecu has no idea how much pressure its seeing it just pins it at 23psi. With sll thr other stuff you got going on right now you can easally launch a rod threw your block. Until you get a better understanding of efi and form a plain to tune your car id just run wastegate spring pressure so your internals stsy in your engine

Thanks so much for the info. I have taken a few measures to keep things from blowing up while I wait for new Bosh EV14 750cc injectors and new MAF + 3" housing.
1. I have setup a boost controller along with the N75 and limited it to 20psi max. I've also set the tune to spike at 20psi max and hold boost at 15psi max.
2. I've switched fuel to 100% E85 and adjusted the primary/secondary fuel enrichment channels so the injector duty cycles are at 100% and flowing as much fuel as possible.
3. I've retarded the ignition timing angle additive offset by -3.75 degrees and logged AFR+knock sensor voltage to make sure there is proper fuel under all boost conditions.

If someone could please take a look at my new log files with these adjustments that would be great. The motor is running extremely smooth and knock-free now.

Thanks again for the help!

Plot RPM on the x axis and enable the filter.

I would not drive this car, period.

Thanks so much for getting back to me. I understand that it's very stupid to drive a car that's detonating on the wrong tune with a big turbo and wrong injectors, but I had to do whatever I could so I could at least limp it around with a safe air/fuel ratio at low RPMs.

Also I'm having some trouble plotting the chart with RPM on the X axis and turning on the filters. Could you please let me know what variables you'd like on what axis? I believe the chart is a little incomplete because I was logging regular driving and not WOT pulls.

Thanks again and please let me know how it looks.

Omg what are you doing?! Your injectors are too small for your turbo on petrol but yet your running e85 ???. You can not just switch tou need 40% more fuel whem running e85 to reach lambda of 1 vs petrol. Really you need to take a step back before you blow your car up. Id hate to see what your exhaust valves look like at this point

Using leminiwinks or unisettings is crap.

Pure insanity. How on EARTH will this prevent you from grenading your motor?

Why the hostility?

I've retarded timing all around, lowered boost from 28psi to 13psi max, tweaked fueling to 100% injector duty cycle under high engine load and mass airflow conditions, lowered redline to 4200rpm.

Then I logged AFR/lambda and confirmed it's at stoich and tapering more rich to around 12.6 AFR under boost and checked IAT/EGT, knock, ignition retard, and misfire events under all these new conditions.

So please enlighten me, how will I throw a rod given these conditions?
I've eliminated all harmful conditions I could see from the logs by doing these tweaks.

Thanks

386cc injectors are hardly large enough to max out a stock turbo on pump gas. E85 takes on about 40% more fuel flow for the same lambda value as pump gas. Si you have s larger then stock turbo with injectors tjat are too small anddd youre running a fuel that requires 40% more flow.... see now. Also you max out a stock maf with that turbo too. Plus all the other stuff i said before

Thanks for your input. I have made several more changes and the car is running absolutely great (compared to before).

I removed the GIAC tune via boot mode due to their ecusafe protection and flashed my own custom bin that's based off the information I found in the Stage 1 community project B6 A4 tune thread. I was able to compensate for the injectors, control desired boost via n75 and wg duty cycles, fix the lambda to a progressive range 0.92-0.77 during WOT under boost (it's ok?), retard and advance timing as needed, and bump up the BTS values a bit.

I still have a few questions that I need help figuring out:
1. I notice that the ECU will not read above 2554mbar (exactly) of boost pressure (minus 1000mbar for atmospheric pressure), so around 1.5bar of boost. Is this the limit of the MAF itself or the MAF housing, or some other software variable I'm not aware of?
2. What is te maximum airflow in terms of grams/sec that the ECU will read? Or is the limit purely based on the MAF readings?+--
3. Is my AFR of 0.92-0.77 safe? It starts with 0.92 at 6psi and ramps up to 0.77 by 19psi at redline in a lin``ear fashion.
4. Is the injector duty cycle the best variable to monitor if I want to determine whether any are running at their maximum flow capacity?
5. I am having trouble with massive boost spikes during the initial onset of boost or during high wastegate duty cycles. The actual boost is following desired very well except for this massive spike which is triggering overboost conditions as well as MAF signal too high (over 22psi).  Is there a way to tune the ECU as to prevent these spikes? Or do I need to either use a boost controller, remove the n75, or adjust the wastegate? Any chance it's a hardware problem? If so, how can I check?

Thanks a lot for all the help

Imx and drl for boost control pull some out where you have issue

Pro Tip: If you post under Diagnostics attach file and logs.  ::)

1. learn basics, read wiki, its the limit in the code, search for 5120hack
2. maf
3. depends on the graph, logs?
4. injector open time in ms or duty in %, logs?
5. file, logs, what marty wrote

If you ask for help, attach everything we need to help ;)
Your attached files will stop newbies from asking the same again and again.
Do it the right way.

Uh, there's a little more to it than that.   ;)

If you want proper boost control, do the linearization runs and use the LDRPID Tool.

Hey there,
Thanks for your input. It always helps. I always take logs so I've attached a sample run from 1800-6500rpm WOT. The boost spikes are usually a lot higher FYI.

To answer your questions:
1. Awesome, thanks, I read about it and setup the hack on my tune. It works even though MAF can't read that much flow.
2. I am using the factory MAF with factory size housing. I keep hitting the 5v limit and would like to go 3" housing and implement MAF scaling. I hope I can do it since I'm just learning basics. Maybe one of you can help.
3. See my attached logs and graph
4. Got these logged as well
5. See attached ;-)

Thanks guys. Looking forward to your responses!

hmmm, i cant deal with your logs.
Prefer the oldschool logger and ecuxplot  ::)

It is the "old school logger", just not using the ecuxplot UI.  VisualME7Logger just wraps ME7Logger.  Seems he just modified the logs so they are un-parsable.

except he has posted modified csvs.

I don't know how many ways I can remind people not to do this.

Sorry about that. I was just trying to post the log data from the run in the screenshot rather than the whole logging session. I've posted the whole session instead and you can find the run at 492 sec. Thanks and my apologies.

PS: I can post my ECU BIN if anyone thinks they would be able to make some quick improvements.

no wot run in your log
log again with ped100 from 2000 to nmax  ;D

Sorry once again for my mistakes. It's very hard to get a good WOT run because the auto trans will downshift even in tiptronic mode if I push the pedal too far

I've been busy trying to do the bare minimum ECU tune so I can drive the car without blowing up the motor. I had to start from scratch since I made too many changes that had adverse effects and caused all sorts of problems including rod-popping boost spikes. My current file has been modified to run the 386cc TT225 fuel injectors at low boost and not much else.

If anyone could help me make some very simple changes to the file, I would really appreciate it. I just don't trust myself to do more than I already have with the injectors.
The next few changes I'd like to try to make is
1. Increase fueling a little for the E85 mix (enrich to ~0.82 under boost or high RPM)
2. Raise boost to ~16psi until 85% injector duty cycle
3. Any timing changes that should be done for #1 & #2
4. Code out the O2 cat codes P0420 and heater circuit code P0141

If I could get a little help on the changes above then I could take my time to do further development on this tune myself at a nice slow pace.

Attached is the mod/orig ECU plus a few logs with WOT runs. If you need the XDF to help with making some of these changes, just let me know and I can post it. Many people have already posted the 518AK XDF so it should make things easier.

Thanks for the help with everything!


PS: In case you need my hardware details, here they are again:
2004 AMB 1.8t B6 A4 Quattro Auto
ECU 8E0909518AK
HW# 0261208230
SW# 1037368072 v0003
Frankenturbo F21 Hybrid (28psi WG)
386cc TT225 injectors
2.5" factory MAF/housing
3" catless down pipe w/ OEM O2's
RaceTec front mount IC w/ 2.5" piping
Frankenturbo 3" Turbo Inlet Pipe
CNC piston style DV w/ 10-25psi spring

Hmm,

there is one spot in 3rd gear, open throttle but rpm from 2400-4100rpm only.
So based on your logs, its still hard to help (at least for me).

Lets wait for someone to tell us how to log a automatic transmission car.

I would start with adjusting LAMFA and FKKVS(based on logs and fixer tool)

Thanks for your input. I've done a few more WOT runs to redline in 1st and 2nd; see attached. These logs should be a lot better and on point. They all look nearly the same with some funky stuff going on in 1st gear (it could also be normal, perhaps the turbo spooling/lagging or lambda adaptation).

Also what is this fixer tool you are talking about? I'd like to take a look. I will go ahead and try to apply the changes you recommended and report back.

ECUxPlot may give you more useful plots.

I plot in ECUxPlot after I finish logging, but since everyone has a different opinion in how to represent the data on which axis and so on I just let each person load the logs themselves.

I need some feedback in regards to the strange boost/correction behavior in 1st gear. It seems the throttle plate closes after current boost exceeds desired by too much, and this doesn't happen in the other gears. Also, is a lambda of 0.89 acceptable for 7psi boost between 4000-6500rpm? Also how exactly should I tune the timing? I am able to add about 7 degrees across the whole board at any RPM without any sort of adaptation, retardation or BTS kicking in. Would it be safe to add 5 degrees or so under all conditions? I want to make sure everything is working smoothly before I raise boost, timing and more fuel.

Thank you!

Almost worthless because the time you log the rpm range in such a small gear is too short.
The load is low and you may have wheelspin. Log 3rd or 4th gear to have a detailed log and give the engine the time to produce boost.

I tried to do this, but unless I start at 3500rpm or higher, the transmission will downshift and ruin the pull. I'm not sure if there's anything I can do like disable traction control but even when I am in tiptronic mode it will downshift at 100% PER.
Does anyone have suggestions?

I've successfully made all the changes I need in terms of load, boost, timing, throttle, fueling, etc...

Fuel trims are great, no CF and adaptation is minimal, and actual is following requested on almost everything except boost, which is 200mbar higher than requested at any given time. Since it's less than 300mbar over rrequested there are no codes being thrown.

Now my final and last issue that needs to be resolved: I need to fix the boost spikes on 75%+ throttle to avoid damage/knock. I believe the best way to fix this would be to limit the WGDC to 65% maximum at all times, since these spikes happen as a result of 95%+ wastegate duty cycles.

Can someone PLEASE tell me where I can find KFLDRL? I've been looking forever and cross-referencing other XDFs to try and locate it. I've learned a lot since starting off a total noob and I've located several maps, but not the N75 DC map which I desperately need. Original BIN is in this thread.

EDIT - Do I need KFLDRL or KFLDIMX for WGDC max?

Thank you so much!

KFLDIMX is integrator maximum WGDC of the PID control.
KFLDRL is the linearization map and takes the result of the PID module and outputs the actual WGDC command
Hope that helps you :)

Thank you. Could you or anyone please locate this map in my file, or at least give me the dimensions of the map or an address of a similar ECU so I can find it? I have not been able to find any XDF with this KFLDRL map defined and its driving me crazy. It's the last piece to the puzzle to get my tune finished.

EDIT- I found KFLDRL on another fully defined ECU. Address is 0x1EDA0, 3 decimal places, LSB first, 16x10. Will my BIN have the same rows and columns as this one?

If it's from the same type of ECU then the map should be similar, just located elsewhere :)

imx is 28CFE  8X16


          DRL IS 28E50 16X10

You may want to look at the LDRPID Tool to calibrate the PID.   Take the guess work out of the process.  I did this for my F21 and the results are awesome.  I thought my PID was fine before, but I was blind to the fact that it could be much better... thus the tool.

OMG, I just gave up waiting for an answer after checking the thread this morning, so I installed WinOLS, found a common variable in all N75 maps for 1.8t, which is "36B0 3E80 4A38" in hexadecimal. And BANG, found the map myself after a few hours of learning. I will be adding all the stuff I find as maplocator XMLs for everyone to use.

Anyways, thanks so much Marty for trying to save me the time and headache, but now I can finally do it myself and it feels gooood ;-)

Also thanks SB_GLI, I read a lot of your posts over and over, and I actually used your definition file and original BIN to practice locating maps for the first time. Thanks so much for the advice regarding tuning the PID. I had no idea there was a tool for this. I can't say I 100% understand the whole PID system yet, but since my German is better than my English it shouldn't be hard to figure out. Please stick around so I can bombard you with lots more questions ;-)

You guys all rock; I look forward to creating a noobs guide to tuning your own BT 1.8t project when I'm all done.

EDIT:
So I figured the Y axis for KFLDRL is RPM, which I got from 0x1EF7E (/4), which may be the wrong place but my ECU references RPM from there a lot. But what is the X axis? It's not super important for tuning, but I am just wondering for future reference. Thanks