NefMoto

had a look in the R32 definition files, and i realize im not very familiar with the way these ME7.1.1 works regarding fueling.. (022906032CN)
any pointers as how to sort boost fueling using i.e KFLBTS for these ecus that are set up for n/a?

i.e the stock KFMIOP map dont exceed 100%..?
the KFMIRL map on the other hand is up to 120ish..
and also the KFPED is 120ish %, which i never see on other me7s.

as of right now, theres no fueling at all once the boost sets in, lambda raises untill missfire.

any pointers/ideas would be appreciated :)

i have a file that works with turbo on this ecu if you want to have a look in at it pm me

I'd be interested in hearing how to fuel a turbo car on an NA ME7 ECU as well.. would be an interesting discussion I think.

I don't see how fueling is any different NA to FI in this case.

well.. obviously the n/a ecus are set up differently regarding KFMIRL, KFMIOP, KFPED.. LAMFA..
theres no PID present.. nothing regarding boost or boost control.. no MAP-sensor etc.

im curious if its as simple as undercaling the MAF to make 240hp worth of air represent 50% load instead of 100%, and then scale the fueling by the same amount..

OR, if its possible to extend the KFMIOP beyond 100% load capability.. (and lots of other maps obviously)


any info/ideas on the topic would be appreciated ddillenger, and anyonbe else with experience on the topic ofc :)

Well you had asked about fueling in the OP, which is the same between NA and FI ME7's. Load on the other hand....

The R32 OLS I have is missing LAMFA. Defining it how I normally would is obviously not correct. If you have the proper info I can help.

But, the mainstream tuners appear to be rescaling the BTS axis and using that as primary fueling.

well.. the fueling strategy is obviously not the same here.. and the LAMFA map is nowhere to be found.. hmm..
a definition for this ECU would help tremendously.

adjusting the BTS map and axis is obviously easy.. but wont help a thing as the fueling runs dry as soon as boost kicks in.
the problem here is how to get the ECU to accept more airflow then ~190ish g/sec.. (100% load), and how to extend this so it accepts and adds fuel up to 380g/s+, as said, any ideas are welcome =)

I haven't looked at your file, but I've found LAMFA in every R32 file I've seen relatively easy. That, and how does BTS enrichment fail when boost kicks in?

How about you start by fixing the hardware on the car before you go to the ECU?
Log lamsbg_w, more than likely there is nothing wrong with it.

There is absolutely no difference between NA and FI tuning for this, BTDT many times.

This is what I was saying. BTS doesn't just stop working. These are wideband cars, fueling is the easiest part here.

So.. We're just doing a gt35r turbo conversion on the r32 ...R32 has 4 lambda sensors.. 2 for each bank.. How to get it running 1 set of sensors or even 1 lambda on the single downpipe turbo manifold? Obviously code them out but, is it best to run 1 or 2 or all 4?

I'll find out which ecu its running, as this thread could be very helpful!

not my car, but i will look into the hardware and do some logging.

what i dont understand is how this is supposed to work when KFMIRL and KFMIOP are set up the way them are..
can i just increase KFMIRL to represent the reality, and then extend/adjust KFMIOP load axis accordingly? (beyond 100% as in other FI ME7.x)

atm the KFMIRL in this guys file requests ~125, BUT, the KFMIOP only stretches as far as 100 in the axis, which would cause major torque interventions.. which would explain things.

LAMFA/BTS should be the desired fueling strategy.. will search for more maps.. =)

What maps do you need? I defined the 032EG file fairly well. Certainly enough for what you're doing.

As for IOP, it's a 16 bit axis, rescale it to suit actual load and interpolate the table.

With that being said, I think an underscaled MAF would make this a bit easier.


Dan: I can help you remove the exhaust valve, the rear o2s, and tune the car, but I haven't had any experience with removing one of the primary o2's yet. Email me the ecu number, I'll set you up.

DD

http://s4wiki.com/wiki/Tuning#Mono_O2_sensor

shouldnt this be applicable even to these ecus?

thanks ddillenger, i needed that IOP confirmed.. i really think this is whats causing troubles right now.

as for maps.. well.. i think i can fetch the maps i need for now, else ill shoot you a PM  :)
LAMFA would be great tho.. the bts, timing maps etc im fairly sure i can sort.


underscaling the MAF, how would i do this, easiest way? i was about to, but dont think i ever tried this with the FI me7s.
i had a look at another r32 turbo file, and KFMIRL and KFMIOP are left stock in these, but the MLHFM is scaled.. dont know what other maps, if any..

Pretty sure most of the tuners use a larger MAF housing which would help with scaling.

rescale tables for the load you want, use a boost controller for the turbo, patch code for single Lambda sensor.

Rick

How does this relate to the topic title of "boost fueling"?
This has absolutely nothing to do with it, now you are talking about the torque model...

not everyone can be at your level of me7-knowledge prj, you need to try and think backwards ~10 years and go from that standpoint..  ;D
sure, the fueling works just fine running a MONSTER turbo and three s/c at the same time, without even touching the software, but as soon as you reach even 0.1bar boost everything falls flat and goes to crappers, missfiring, lambda skyrockets, throttlecut etc.

with "sort the boost fueling" i was asking for the most BASIC ways to make sure the engine still have fuel at 0.1bar boost and above.
i wasnt talking ASM coding, specific systems/paths in the ECU/code etc..
i.e why wont it add fuel once the load goes past 100%, and why is everything setup the way it is in the n/a ecus, could it be altered? and yes, so it seems.. even tho most major software developers for R32-turbo kits DONT do it this way.


so, back to the questions.. what kind of power will the R32 MAF support? and how do i go about to underscale it? (if i need to do this..)

Absolutely nothing happens to fueling as long as you don't max ps_w, the MAF and have injectors dialed in right.
It does. You're wrong, fix the hardware problems. Log the car instead of just typing and typing and typing and typing.

What kind of power your MAF supports - boost it and see where uhfm_w hits 5V, then you know it's time for a bigger one...
So much incorrect theory and pointless yadda yadda... tuning a car is not writing fantasy fiction, it is a game of hard mathematical numbers, so start providing concrete numbers and logs.

Trying to sort things in my head before i even approach the car in question, this based on the very little info i have from the project.

dunno if ps_w is involved in this at all? since theres a lack of map-sensor, pressure maps, pid and other boost-related stuff.
MAF should support the flow, and injectors are dialed in the same as in a few others cars that runs spot on.

ofc ill take a look at the hardware first chance i get, and start logging.

ill take your word for it that fueling works above 100% load, im sure it does.
so i will have to tune KFMIRL/KFMIOP to allow more then 100% load to begin with.. since it looks like thats the problem atm.

this is no fantasy fiction, gathering other peoples opinions/experiences before approaching, troubleshooting and tuning a vehicle is called preparation ;D


anyone with info on how to underscale the MAF?

Use a larger MAF housing, or decrease MLHFM proportionately to the amount you wish to underscale. Ideally, try to find an application that uses the same sensor part number in a smaller housings and steal the transfer from that file for a true underscale.

ah, ok, thanks :)
then the fueling need to be scaled up by the same amount? KRKTE?
(and ofc timing maps etc calibrated for the new load)

In a nutshell, yes.

the file i sent you is with std MAF and it is running around 470bhp

in my other file iam running this maf 0 280 218 015, 750cc injectors running the HGP manifold with split wall for running 2 lambda sensors before the turbo and 2 after the cat`s

What does ps_w have to do with a MAP sensor? The answer is - nothing.
You have a long long way to go, you need to look at how filling and fueling works in ME7, and then you will understand why what you posted earlier is ridiculous and why I called you on it.

There is abslutely no need to underscale the MAF. Not until you hit ps_w limit anyway.

As for torque management - IRL and IOP works the same way as on any ME7. On some implementations there is a hard cap at 100% load which will need a code change though, on others there is none...
MDMAX and MDFUE is where you need to look.

so, there is indeed a hard cap in some cases.. perhaps thats why HGP never touched the IRL/IOP tables.. and scaled the MAF.

yeah, i know i have a long way to go with understanding all this, as of now i feel quite confident i can tune most "stage 1-3" on the ME7s by manipulating things, but thats it.
making this n/a ECU work with boost is perhaps the next step, and its indeed time to do some more reading into how the filling/fueling/load works in the ME7.

i was under the impression PS_W referred to 'absolute intake manifold pressure', which is why i thought it might have nothing to do with this since the engine is "blind" to manifold pressure.

So, time to find an english FR and do some reading.

ps_w is indeed absolute manifold pressure, and all ME7 works based on pressure.
The charge pressure sensor you are on about is not a MAP - it is only used for charge pressure control, not for calculating fuel outside of limp mode.

As I said before, less talk, more facts. Read, read, read and figure out how things really are, not what gets regurgitated on the forums over and over again often with no other base than "someone at some point said so" :)

Well without even asking any specific questions I just got a ton of info as usual from prj.. Thanks!

Actually I do have a question... Is it right to assume that KFWDKMSN should be carefully looked at as well in this situation?

Yes, and not only that, entire throttle control needs to be looked at.

Absolutely understandable all things considered.  Thanks!  Maybe I will have a chance to play with one soon... just need to find a stock ECU for the R32.