NefMoto

What would happen if I took a stock .bin file from say ME 7.5 and only changed LAMFA so that AFR is where I want it to be at certain rpms?

 Will things richen up or lean out to what I want or would I have to play with actual fueling maps?

If you only change LAMFA, the ECU will do run with this values until you reach TABGBTS.
At this point the ECU runs with LAMBTS.
If you want a richer AFR at high EGT and WOT, you need to change KFLBTS.

BTS will only win if it is richer, afaik

That`s right... ;)

ok let me get this straight.

The ECU runs with LAMFA until the threashold (TABGBTS = 399C which is stock temp setting for my ECU) is reached. Once EGT reaches this temperature, the ECU goes into "protection mode" and uses KFLBTS until the EGT temperature falls below TABGBTS again. But if you go WOT and the EGT is below the TABGBTS threashold, it uses LAMFA. correct?

What is a safe TABGBTS temperature?
I noticed on some of the 2.7t ECUs files the TABGBTS is set way higher than 399C.

The ECU will always run the richer value of both

So then whats the point of tuning both maps piviting on TABGBTS? or having both maps at all for that matter? It seems like something is missing from the "equation" here.

Re read the fueling section of the s4wiki tuning article.

Hi Guys,

Is it correct to say that the stock TABGTS is a bit on the low side ? I have picked this up to 500 Deg C for normal day to day driving . I know with out logs its kind pointless , just thinking out aloud.

TABGBTS varies quite a bit depending on the engine spec from what I have seen...

On a Leon 180hp K03 for example TABGBTS is set at 400deg... on my AMK code S3 (K04, wideband lambda and EGT sensor) its set to 819deg... on an earlier APY S3 (narrowband and no EGT sensor) its set at 800deg...

LAMFA and KFLBTS are tuned very differently too... from what I can see the AMK/BAM setup as it has wideband and EGT uses lambda 1 as much as poss and relies on component protection and EGT modelling to enrichen accordingly... std settings on AMK/BAM KFLBTS is mostly set at 1... on the 512k APY maps its tuned to 0.74 on max load/rpm so guessing there is less intervention/protection models in these maps than the 1mb wideband ones...

<tuffty/>

Its getting weird to me.

I also thought, that tabgbts is the temperature treshold, wich starts the lambts to work for enrichment (if its chosen richer than lamfa etc).

But why are the tabgbts set at a level that cant be measured via egt-sensors in the s4 and rs4 models (499°C)? The lowest measurable temp is 940°C afaik.

Is a temp of under 499°C interpolated / calculated as signal, until the measurable temp of more than 940°C is reached?

Otherwise the system would enrich a lot, and "component protection" would protect in situations, where no protection is needed (500°C doesnt seem to be critical to me :D).

It's modeled EGT, not actual.

http://s4wiki.com/wiki/EGT

This is no answer.

Even on the wiki, were talking about the 980degrees, which make sense to start working with lambts.

But why the heck is tabgbts set at 500°C? When tabgbts is the temp point of starting point of letting lambts influence lamsel....
 

BTS is prophylatic.

egt nearly always exceed 500°C...

In all stock S4 tabgbts is set at about 500°C..

This would mean it would always run in lamBTS using kflbts.

Then Bauteilschutz isn correct as a declaration of this part of function imo.



 

I believe, you talk about two different things.

The measured 980 °C from the "TABGSS(2)" value is only the desired exhaust gas temperature for the EGT regulation.
It is the max. temperature target and not the starting point to run with LAMBTS.
If the 980°C thresold value is reached, the ECU will run additionaly the delta lambda mechanism "KFDLBTS" and "KFFDLBTS" to make a richer AFR.

At the internal modeled EGT threshold "TABGBTS" (S4= 500 °C, RS4=750 °C), the ECU runs with "KFLBTS".
The only exception is when the "LAMFAW" value is richer then the "KFLBTS".
In this case the ECU will run with KF "LAMFA" and not with "KFLBTS".

To run with a richer AFR makes sense to work against a high EGT.

No, it DOES NOT. Please reread the EGT article. AGAIN.

I m sure rs4boost and me are right.

rs4boost is right, you are wrong. There are two EGT temps, one is modeled, one is measured.

TABGBTS has nothing to do with the measured EGT.

This is not true.

[KF]DLBTS and KFFDLBTS are only used to offset Lambda for degraded ignition angle in BTS.

Lambda intervention from actual EGT is calculated in ATR and executed via dlamatr_w. In limp mode Lambda offset is calculated from DLAMTRNLN. During normal EGT regulation it is calculated using PID control based on TABGSS and tabg_w delta.

Yes, you are absolutly right.

I have just looked up the "ATR" description at the ME7.1 Funktionsrahmen on page 406.

The additionaly delta lambda mechanism is a PI regulation (P=ATRP / I=ATRI) with an influence to lambda over dlatr_w with a maximum of LATRO and DLATRMN.

Thanks guys. I will update the EGT wiki entry time permitting.

Ok, great! learned a lot!

I was also wrong with the modeled EGT with beeing interpolated.
Its calculated in ATM 33.50.

When I have some more time, ill try to find out, how fast tabgm reaches the s4 treshold of 500°C.

Did anyone log tabgm?

So to the topic, regarding tabgm_w (modeled) and the point it exceeds tabgbts, gives us an area, where we have to work out a proper transition.

Imo only tuning lamfa (eg to 0,83 lambda) doesnt make sence, because while exceeding tabgbts with stock lambts table, in several areas, its too lean. Therefore "tuned"  Lamfa might be richer and as result requested AFR stays at lamfa. This would run, but efficiency (fuel economy) is messed up.   

regards

All of this is covered in depth in the various fueling discussions here and in the s4 wiki.

thread:
http://nefariousmotorsports.com/forum/index.php?topic=141.msg9699#msg9699

LAMFA vs LAMFAWKR vs LAMBTS:
http://s4wiki.com/wiki/Tuning#Open_loop_AFR


yes


I can't make head or tail of this.

Regarding to the topic the answer makes sence. I dont know why its not clear that only changing lamfa into richer doenst make sence at an s4 with a stock lambts.

If i am wrong, i would be very interested wich positive affect changing lamfa an leaving bts stock would have except making fueling more predictable. For me there isnt any technical positive point.

Youre right by saying it is discussed in several other topics, but i just wanted to help with an answer at this topic here.

Regards

You answer makes no sense because it doesn't matter if modeled EGT goes over TABGBTS.

If LAMFA is leaner than BTS, that is what req AFR follows.

Please don't post misinformation!

That said, stock BTS is a bit rich in high load regions, and might have to be dialed back if you are attempting a LAMFA only tune.

What misinformation?

After exceeding tabgbts, the richer value of lamfa or kflbts will be choosen.

But eventually we only talk past each other. I consider not only the increase in power. The efficiency is also important to me. Here it is necessary to apply a good transition from lamfa to the component protection with reasonable values.  Changing only Lamfa into rich (eg 0.83),the efficiency would deteriorate significantly. But My opinion is, that enriching towards a powerful 0.83 or whatever you wich can be made nicely by using lamfa and kflbts together with a nice transition towards eachother.

And in my case modeled egt is always exceeding (in every wot pull) tabgbts´ 500°C.  (at least over 3000 and a bit rpms at 3rd gear)


 

This sounds good to me.  I couldn't figure out your previous posts, but this one gets it right, IMO.

thanks.

Ok, so I have a K04 on my car , TABGTS was always at stock , 300 deg C . I cranked this up to 500 dec C as a test and the car pulls much better when going from PT to WOT or during semi spool. Could it be that the EGT was climbing during normal driving , engaging component protection lambda control ? Take into mind that LAMFA has been correct for the "appropriate" AFR's , it doesnt sit at 14.7 across the range anymore .

Am I correct here - I swear the car drives way better

You're basically disabling BTS.

If the car drives better with BTS disabled, it means your BTS tables are screwed up.

Hi Nyet ,

BTS is stock

Stock BTS only works with stock boost.

If you alter boost in any way, you need to clean up BTS or it does some very bad things to your gas mileage.

I get this concept but what exactly do the numbers stand for in the text table of the map? I know it stands for AFR but I guess I'm asking how to convert the numbers I see in the map to AFR and vice versa like what AFR is .73? Is it the same as lambda voltage? So the text table is lambda voltage@ x=RPM and y=load? depending how you have the x and y set up?

Time to pick up an EFI book, son :(

AFR 14.7 is lambda 1.


No. Not for narrow band OR wideband.


No.

ok thanks for the recomendation fasha.

I know 14.7AFR = lambda 1...

What do the numbers in the text table mean then? Do they not represent AFR? How do I calculate AFR from the numbers in the text table of the map specificly? Mind you I already know how to calculate AFR via lambda. ex: 1 lambda = 14.7 AFR, 0.8 lambda = 11.76 AFR, 1.1 lambda = 16.17 AFR    Hopefully this is clearer than my first questions.

I'm now guessing they are not lambda values. ???
Here is a pic of an example KFLBTS map for ME7.5 HS:

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

"KFLBTS_0_A - requested lambda for component protection when calculated EGT is above TABGBTS, scaled by FBSTABGM."

BTW there is no such thing as "lambda voltage"

So FBSTABGM are the values in the text table of this map?? Come on dude how about a straight answer??
Please support your comment...

No. The output of the BTS table is requested lambda. The result is scaled by FBSTABGM. Is there wording you thing would be better in the s4wiki? Have you looked at the FR yet?


Is there a grams/sec voltage? Are MAF values represented as voltages in air mass calculations?

Ok, I will use "output" when talking about table content from now on. I know the output of the BTS table is requested lambda, my question is if the BTS table output that you see in the pic(ex: 1.0079@ intersection of x? and y? ::it's an example just pick one::..) is a direct representation of lambda or is it something else? So in other words if I have a value of 1.001 @ an intersection of the table, the AFR of that cell will be 14.7xx. (this would mean the table values are a direct representation of lambda(actually requested lambda)) what I mean by something else is: Do I need to perform some other kind of arithmetic to figure the AFR using the table output?
these questions are immaterial for this discussion. but yes and yes

 lambda voltage is the number represented as the lambda value
ex: lambda is at 1...aka 1Volt DC, lambda is at .83....aka .83V and so on..higher voltage=lean, less voltage=rich, 1V is ideal aka 14.7 AFR. this is how the ECU "reads" the 02 sensor's "info" by voltage. Almost all narrowband 02 sensors are like this.
some operate at a .5V range so ideal AFR is at .45-.5V. accurate reading are very dependent on EGT. widebands are measured the same way and accurate readings are not so dependent on EGT in comparison.

= Yes they are.

In the attached, you can see a pic of lambda voltage to lambda.

Regards










[/quote]

btw

The ECU calculates the actual lambda by the "LALIUSH (2)" maps.
This map convert from the measured narrow band O2 sensor voltage to a lambda value.

(http://www10.pic-upload.de/17.08.12/acjrmo2wqd.jpg) (http://www.pic-upload.de/view-15640750/LALIUSH.jpg.html)

Why would you need to do that? The z axis of tables is always the output of the table. Have you looked at the FR yet? If there are other conversions/scaling to be done, they are described in the FR.

Unless you are referring to the fact that the ACTUAL value stored in the ECU's memory (8 or 16 bit) isn't actually an infinitely precise floating point, and must be represented as a fixed point integer.... which is an entirely different discussion.

This terminology is NEVER used for a reason. Sensor voltage is converted to a lambda value. Depending on the sensor, the conversion is different. Same with MAFs.

Sensor voltage is sensor voltage.

After it is converted to different units (g/sec or lambda or AFR), it no longer has the "voltage" units.

Thus it is NEVER referred to as "lambda voltage" or "g/sec voltage", which would be confusing because there would be no way to know if the person is talking about the resulting (calculated) lambda value, or the actual voltage coming from the sensor.

It would be properly referred to as "O2 sensor voltage" or "MAF sensor voltage"

Lambda

0.73 * 14.7 = 10.73 AFR

Don't be so wrapped around AFR. Think Lambda.

I haven't dug too deep into Lambda regultation but I don't know if this gets used at all in NB ECUs.

Pre-cat control tries to maintain target voltage USR. IIRC LALIUS is not even present in NB ECUs.

LALIUSH is post-cat. I don't remember exactly if/how it gets used, but I don't believe it is utilized in Lamba regulation.

I am pretty sure in ME7 there is NO actual lambda calculation anywhere.

Thanks ,

I did a bit of diggin last nite , the 032BJ file has the threshold set at 300 deg c , where as the 180 PS files from the audi's for eg , are set at 900 deg C - all other fueling and component protection maps are the same .

LAMBTS is quite a funky map , when you say clean it up , you mean bring it closer to complement LAMFA ? In stock form it dips deep into the 10's at peak load . correct ?\

Thank you kindly

Yup, but with a stock chip it will never see that much load unless you have some sort of malfunction causing overboost.

So figure out what your max load will be (since you will be running more boost), and in those areas, bring it up to 11.5 or 12.5 or whatever you think appropriate for your tune.

another question:
wenn i change the load axis from LAMFA by 50 % to 100 %. What Value did the ECU calculating in idle (lower 50%)?

Great .. I did just that .

If I understand you correct , dont change anyhting below 50 % on the scale , I noticed that this plays a role with pt and idle AFR's . At 50 % I ran 13.7 and gradually dropped it down to high load high rpm . If you get the high load right , it will keep BTS in check - am I right guys ?

Your lowest axis values will be used during all conditions below that value. Same goes with your highest.

I meant to  say lambda sensor voltage, please excuse my typo. I see how this was confusing.

yes, I have read the me7.5 FR once. I know I need to read it again as you always miss stuff the first time around but my questions are in regard to interpreting the values on the maps using WinOls. Alot of the time while searching and examining maps I cant find what the z values actually stand for. (thats why I wanted to know if the z values of the map pic I uploaded(on pg3) were actual lambda values or the sensor voltage) That info isn't in the FR afaik.
 
Another thing is that a big difference between me7.1 and 7.5 is the implementation of the wideband 02 sensor. So what are the differences going to be in the maps that the both EMSs share? or do I NOT have to worry about that and just manipulate the maps as I see fit? Now that I know the z value is for sure actual lambda values I feel I can confidently change the map to what I want.

Sorry if I'm a pain in the @ss, I just want to understand and the S4 wiki doesn't explain everything (in some instances it raises more questions... NOT saying it sucks by any means!) and the FR prettymuch explains how everything works inside the EMS and what coincides with the maps and such. It also only shows what the x and y values are (maybe I missread something and missed where it shows what the z values are suppossed to be?) I dunno
 
Before I ask a question I look for the answer in these references and throughout the nefmoto forum using it's search box. Unfortunatly, there isn't a section in the forum on how to use WinOls or Tuner Pro including how to properly interpret the values that show up in the maps, or a legend of map axis symbols to values conversions (ex: %=load, U/pm=rpm, V=volts, z value of 1.0079=1.0079 lambda, etc) or which definition files are compatable with what versions of WinOls, or how to generate definition files, or how to locate maps in the hexdump without a definition file, or establish a new map previously undefined in a definition file.

A section with this info I believe, would be a nice addition and would clear up alot of confusion. Too bad I don't know the answers, otherwise I would write it up. :-\



edit: typo

All the answers come with experience.

It is impossible to cover every small detail. It just comes with time and understanding.

If there is something SPECIFIC that you don't understand...ask.

If you can help out with anything in my last post, you'll be my hero.