NefMoto

Hello, everyone!
Couldn't find similar question...
When we log frm_w (lambda control average) is it including the trims (32 group)?

For example in 32 group I have:
IDLE: -1.8%
PART THROTTLE: -5%

My frm_w is something like 0.95xxxx. Does frm_w include IDLE and PART THROTTLE trims or they're then added to this variable in later stage somewhere?

The way I understand it is that additive trim is applied before fr_w and multiplicative trim is applied after, but since is ~always runs in closed loop, it is always included.

And the only difference between frm_w and fr_w is in frm_w the ZA(forced amplitude) is not included.

Hello,

Trims are added/multiplied before Lambda correction.
frm_w is fr_w averaged over some time.
for trims calculation frm_w is used.

Hm... anyway to disable trims but leave o2 correction? (wideband o2)

 When I want to disable or limit adaption I use

FRAOMN  lower limit of correction factor frao
FRAOMX  upper limit of correction factor frao
FRAORN  reduced lower threshold for frao correction
FRAORX  reduced upper threshold for trao correction
FRAUMN  lower limit of correction factor frau
FRAUMX  upper limit of correction factor frau
FRAURN  reduced lower threshold for frau correction
FRAURX  reduced upper threshold fpr frau correction

 Make all of these =1.0 and you'll never see any changes on group 32 but still have normal O2 correction.

Thanks :) So subfunctions FRAU and FRAO are these trims? Which one is which (idle & part throttle)? Just looked at the FR... Wouldn't it be easier just to set B_NOFRA to 1 ? B_NORKAT?  ???

 If you can find it in your image it might be easier :) .

I have changed the codeword NOLRA to disable fuel trims.

I just did the same...
What is interesting is that originally these were disabled (set to 1):
B_norkaz (mixture correction - ignition timing)
B_nofrat (mixture correction - coolant temp)

Should I enable these? How are these set in your ecu file?

One more thing: since we disable TRIMS isn't it better to up a bit o2 sensor regulation boundaries (KFFRMIN & FRMAX)?
What would be good values for upper and lower limits for wb sensor?

BTW: I could easily find FRMAX map in ME7.5 but couldn't find KFFRMIN anywhere... Maybe someone could take a look in ASM and help me to locate this map?

Right after KFMIOP, 16602.

OK, just a few brainstorming things I was wondering about.
Totally not trying to play the n00b here, but let's pretend I'm totally new to tuning for a minute.

Why would I ever want to turn off the trims?
   Please correct me if I'm wrong, but trims can also be disabled by basic VAGCOM operation.

Are LTFT's disabled when the MAF is unplugged?


If the "optimal tuning technique" turns off LTFT's after KRKTE and TVUB are close, how do we define close?
Let's define close to be idle LTFT and partial LTFT +/- 1.0%.  Does that seem ok?

Thus, are we implying that a "refined tune" will clean-up all of the real-world fueling after disabling LTFT?

Most of this stuff is procedural and anecdotal.  I'm not a professional tuner, and that's why I'm asking all of this stuff.

Are you positive about this? In ols file that i have to compare there's one axis 8b and another 16b, at 16602 i see two axes in 16b... Could you tell me this map size? and factors for both axes and map values?

I personally don't turn off the trims ever on narrowband ECU's, and when tuning, I just specify Lambda 0.99 via LAMFA, so that it is open loop the entire time and unable to learn trims.
If tuning wideband however, it is a good idea to disable closed loop when tuning, otherwise it will constantly interfere. Basically the engine must run precise enough off of the MAF alone... the closed loop is mostly there to mask hardware problems and account for aging differences.

I have been turning off LTFT's but not closed loop, and have been doing this when I have been setting up for FKKVS/FRLFSDP/KRKTE on my returnless wideband 7.5.

It get reaaaally old to dial those maps when you have 1000cc injectors which are basicly non-linear until you get in boost, when you have partial LTFT skewing up the mixture. Does not really make much difference with them on or off now, as my corrections are +-5% under normal driving conditions, well at least over ~1ms effective PW's.

When im happy with my fueling model, im ofcourse turning them on again for the reasons prj mentioned.


IIRC its 4x5 or 5x4, I used the TT damos as reference, I can take a look after work.

I'm doing same for fuel adjustments... Basically with my ME7 Wizzard I'm generating FKKVS table with trims disabled. I'll check out TT damos, but if you could look too that would be awesome. Thanks

BTW: I've checked LRSHK function for postcat o2 sensor. Basically if anyone wants to disable sensor postcat correction but leave sensor plugged in then all you have to do is set bit0 and bit1 of CLRSHK to 1 (TRUE):

CLRSHK (bit0) = 1 => b_lrhkg = 0 => b_lrhkb = 0 => b_lrhkp = 0 => dlahp_w = 0
CLRSHK (bit0) = 1 => b_lrhkg = 0 => b_lrhkb = 0 => b_lrhk = 0
CLRSHK (bit1) = 1 => dlahi_w = 0

(single bank systems)

I know that in s4wiki it is said to set CLRHK to 5 but I don't know why it is needed so. 3 is enough

IIRC, I decided to set bit 2 to ensure that atv is clear.

Oh, I see... Phila_dot, have you checked the DLSSA function? I'm kinda lost here... How is system constant SY_STETLR set? :o When is DLSSA called? (Maybe we should increase DLAHISATO & DLAHISATU, when dhali_w = 0?) (I hate these ASCII diagrams...)

Also, when we disable LRSKA I think we still have to look at LAMKO... In LAMSEL when lamka_w < 2.0 then B_lamka is set and min value is taken between lamka_w & lamnswl_w.... Or am I wrong here?

One more small thing noticed... If you look at the ETALAM map you can find out which AFR is most effective to your car (axis is lambda x 14.7 = AFR)

I did small calculation and found out that at the ~13.2 AFR I get 101.5% of power ;)

Yes, an ECU value can tell you what AFR to tune to ....  ::)
Just a tip - you are very wrong.

I'm just saying how engineers in bosch thought and also programmed ecu. By using said AFR you'll get multiplier of 1,015 of your max torque

What calculation you did?
I didnt quite get you...

Look at the ETALAM map in FR and analyze it's axis. It is lambda val of 1V which is stoich...

Which means absolutely nothing in an actual real world scenario, as by using this uber AFR of yours you will reduce torque.