NefMoto

hi,

ever since my ignition has these spikes and i cant figure out why.

is this some kind of correction or related to kfzwop maps?

Your load is quite a bit above rlsol

http://nefariousmotorsports.com/forum/index.php?topic=13929.msg112945#msg112945

this?

may maf is fine and honestly i didnt think about load at all. time to read and learn...

prj wrote "if it's only on high boost then KFPBRK/NW should be changed at higher load."

so i know what todo

will see what happens with maf vs. maf@tb
maybe it will be solved by rl_w vs. rlsol_w or maybe its a topic on its own.

WDKUGDN

Obviously pressure test from the turbo inlet with intake manifold connected and make sure nothing leaks before you mess with any of this.

15 logs and a lot fuel later i still have spikes in the ignition

what is rlsol vs. rl supposed to look like? in my mind i think i read here "keep it below rlsol" but im unsure.

i had to redo my boost profile because with the changes in KFPBRK/KFPBRKNW i noticed my load was way to low in irl/iop/ldrxn (220)
i dont understand where the spike in my desired boost is coming from because in ldrxn the rpm axit is 3000, 3500, 4000 and the "hock".
i hope its the value @ 3000 but i have to finetune this tomorrow.

my boost pofile is rebuild to ~1.6bar
my rl is below rlsol
my maf vs maf@tb is okay i hope

Might be ARMD now?

Set last % column of KFDMDARO to 50 and value to 50 also, and re-try.

Also your KFZW/2 is too advanced, you have knock on all cylinders = losing power.

There are 5 DARO Maps:
KFDMDARO_0_A
KFDMDARO_1_A
KFDMDARO_2_A
KFDMDARO_3_A
KFDMDAROS

I changed all 5 in the last row (50) to 50.
I tried to configure b_ar in me7logger like this

I am unsure about the b_ar but the ignition definately changed.
I still have the hook in my load/boost and dont know why.

Based on the DARO change to 50 i think i have ARMD interventions. can someone confirm this?

Yes, ARMD was the problem, but now it's off over 50% torque, and your ignition is normal.
ARMD is configured like this in very many OEM files too. Really, there is no need for it to be active on WOT, it's only important at lower load, especially when starting to move and clutching/declutching.
If you turn the entire function off then the car will kangaroo much easier.

Rest of your sawtooth you see there is because of knock retard.

Also, you are very lean and EGT is skyhigh. You are only making a 5 second pull and EGT is already 900C+ (!!!).
Stop with the ridiculous lambda, set 0.8 and reduce boost up top.
With this current joke of a tune even ATR won't save your engine in time.

Reduce boost and add fuel until you're at least at 15 degrees timing up top.
You've long ago hit the compression ratio wall and you're going right through it. Result will be a dropped valve and complete destruction of cylinder head and bore.
Also I hope you're not on stock rods, if you are I am 90% sure they're not quite straight anymore.

Thank you very much for your input prj.

I will try to lower boost and work with ignition instead.
Engine is fully build and egt probe is in the manifold pre turbine.
I leaned it out the last days just to see where my egt will go to. before this i drove it with 0.78 up top for the last year or so.

Is there a rule of thumb on how to set the ignition or boost profile when you havent access to a dyno?
1,6bar tapering to 1,3 maybe? or in other words 95 to 65 wgdc maybe.
What do you mean with this?


thank you very much

Dyno is not needed here, basic knowledge of engine is needed.

It means your CR is way too high for your fuel, so you can not run the turbo at max boost.
You are knock limited.

0.78 loses power on VAG 5V engines, and generally you don't want to be below 0.8. Peak power is around 0.81-0.82.
But you can't run 0.82 because you're gonna melt shit.
So ... all you can do is aim for 0.8 and turn the boost down so you don't have insane EGT.

If you want to use this turbo fully on this engine you need race fuel, WMI or E85. On standard fuel it is not possible, not safe.
Typical example of choosing the wrong CR when building engine.

Will play with wdkugdn to hopefully cure the dent. i havent changed anything in the load to boost path.
I lowered boost and i am almost at 15 degree up top despite the currently hot summer weather.
will try to enrich it to see if the -3 degree will go away. or will try to log with better iat.

thank you very much for solving my spikes aka. ARMD

Instead of screenshots post csv files if you want any comments at all.
Also IAT makes almost no difference to ignition timing.

sorry was removed because of the size yesterday

Don't run 0.9 on peak torque, it's not a NA engine.
You will make more power with 0.82 and more ignition. Source: "trust me bro, I've run it on dyno".
Since you have EGT sensor you can also 0.82 flat and let ATR take care of the rest.

Lower the boost in the midrange a little, set normal lambda, add ignition timing, and you will probably be able to add a little boost on the top end too.

Hi,

car runs much better now with the richer fueling. very responsive to the pedal and kind of aggressive.
before i waited and waited for the boost to build and for the car to move and now its just moving based on fuel and ignition and then boost is coming in.
do you think the overlap where the boost is rising and the lambda is falling (3000 to 4500) is good or would you desing the lambda curve different?
ignition is still a problem on the 3rd but is was always 30+ degree celsius when i logged.

pssol has this dent @ 3500 but rlsol dont.
KFURL is stock, KFPRG is stock, KFPBRK/KFPBRKNW is changed in the 165 row starting from  4000rpm
intake is 034 motorsport

thank you very much prj for the lambda hint. it makes alot difference.

log

You can (and probably should) move the VVT changover anyway

EGT looks much more sustainable now. You're probably not going to hit ATR unless you push hard on the autobahn, and this is good.
Ignition retard probably okay too... I think at lower rpms you can even add a little.
I'll tell you again, ignition is not affected very much by IAT. Mostly by cylinder filling.
30C here or there makes so little difference when you are filling the cylinder and compressing the air after that. Your car won't magically make 3 degrees more timing in winter or anything.

For valve overlap it is very hard to say. Your current one is quite aggressive already...
I'd say even switching at 5k is probably more than enough.

But the real answer what the sweet spot is can only be done on dyno.

Tbh your tune looks quite decent now. It is possible to make more power, but not safely.
Boost is on target, no excessive knock and EGT is also now under control.

Looks like you have some ethanol code in there as well - in that case if that is actually working you can add some ethanol content and blend boost based on that to fully use your turbo.
But on 98 RON (or whatever you are using now) I see no point to push anymore. Also, tuning with this giant torque in the middle is not what I would do, but if you like it like that, then up to you. If you have a built engine and it's not knocking it is not going to break anything.

These are the nicest words from prj I have read in a long time  :-*

Attached is a log where i added 3degree overall in my e100 kfzw1/2 and based on the e65 blend the ignition angle difference is 2 degree.
I have nearly the same ignition retard in this log as in the last log (which was good pumpgas).

I will try to log the car with more advance and will restock my ethanol to test with higher E content. 

Keep in mind that on ethanol you can usually run all the boost and all the ignition on E50.
So blending the KFZW/2 maps only based on ethanol content is not the right way to go. You need a map in between that, which gives blending factor based on ethanol content.

i have two additional kfzw1 and 2 maps where i put the theoretical numbers for e100 content. lets call them eKFZW1/2
and i have the stock petrol kfzw maps lets call them pKFZW1/2
based on the reading from the flexfuel sensor it will blend between my pKFZW1/2 and e100KFZW1/2
it will also calculate the krkte value for this ethanol content.
that is all what the custom code is capable of currently.

boost will stay the same

i will alter my variable names a bit and log with higher values in my eKFZW maps.

As I said, you need an intermediate map:

e content -> 1d  map -> e content for ignition -> and then use that to interpolate the maps.

Same for boost.
e content -> 1d map -> e content for boost -> and then use that to interpolate.

Ethanol influence is linear only on krkte, it is completely non-linear for boost and ignition.

i think i understand your point because my krkte is 10% too small with the e65.
the current calculation is linear.

the car and the retard behaves strange or interesting...
yesterday i had alot of retard with only +3° in my eKFZW1/2
today i added more and more to eKFZW1/2 and the ignition angle climbed and the retard is moderat

the +12° file has best ignition angle because of the least retard.
the real ignition based of the e65 content is ~7° higher than on petrol (xxx_12 logfile)

I will have to investigate why the content reading from the flexfuel sensor drifts a bit on high rpm

Time to read and learn some theory about ethanol and ignition °

regards

Turns out my car is very picky when it comes to iat.
This is my only explanation of why i can run 1.6bar flat with 23degree today at 18degree celsius outside.
Last week it had 30+ degree celsius outside and i could only run 1.6bar to 1.2bar with 21degree up top.

If this is how my engine "should" perform with 8,5:1 on fuel I will have to take it apart to find out about my cr and lower it.

I now have reverted the lower eKFZW1/2 rows back to stock and still have added 14degree up top.

Hi,

interesting find here!
My car is faster with more ignition because of better fuel.
My car isnt faster with more boost on same fuel.

edit: the 1.6 to 1.6bar log is fucked somehow

here is a better log for the 1.6 to 1.6bar file and the car is faster, injection time and maf reading is higher.
so no need to change my cams or anything.

i dont understand why the one log is broken.
how the afr/lambda current is looking is just impossible i think
here is a new log.

Is this again anti judder?

When driving in the rpm window where the boost builds my car drives wavy.
As you can see in the log the B_ar is 1, throttle plate is opening and closing, boost is high low high low and ignition is wavy too.
This is happening with light request (60% wped) and high gear (4th).

What is the root cause here?

I changed the bottom rows of the KFDMDARO maps and it behaves better.

Honestly, I don't know what I am doing with these numbers and there isn't much available on the forum.
http://nefariousmotorsports.com/forum/index.php?topic=1905.msg18422#msg18422

Also, I'm not sure if this is the root cause or if I work around it.

Your contribution is highly appreciated

Solved!

KFPLGUB was stock. I totally forgot about it since the new bigger turbo is mounted.
I reverted KFDMDARO back to stock+ so that only the last row for wot is modified.

Switched to Catcams 1013658 with 1.3mm and 0.95 lift and vvt. Nothing else was changed. Idle was perfect before.

Now my idle and low load is funny. I increased KFMRES several times and its getting better and better.
Is it okay to have like 150% increased KFMRES/K or is should i change other maps too?

Call me stupid... vvt was active on low load, during warm up and idle.
I reverted the following changes back to stock. KFMRES, KFMRESK, KFMSNWDK, KFWDKMSN
All I tried to fix the low load within this maps was pointless because VVT was the only problem.

Now my low load driveability is like stock but the idle is still bad.
Idle rpm is set to 960 but the car is still doing braap braap braap like a mazda rotor because the ignition is jumping.

Next thing I will try is KFZW and KFZWOP.
When I hit the pedal only a little bit the ignition doent jump anymore and angle is fixed.