NefMoto

Hi all,
I'm just getting started tuning my 1.8T.  I've been messing around with a few different stage 1 files floating around.  After increasing KFMIRL and LDRXN i'm seeing requested boost and load about where i want it.  I don't have the logs in front of me, but for most of the RPM range about 20psi was the requested boost and actual boost was hovering right at 17.5psi.  Load numbers were similar (i forget numbers) -- something like request 205 and actual 190 (those numbers could be way off, but it was very consistently not achieving the request).  This could just be a limitation of the turbo, or a boost leak or something.  My question is, if i'm logging a requested PSI of 20psi, then there's nothing else that could be holding it back from the ECU, right?  E.G. -- if the ECU were to intervene, requested boost and load would go down, right?


As i'm writing this, i'm wondering if maybe the ECU was pulling timing.  Would that cause a drop in boost possibly?  I'll have to check the logs...  Although i never saw a boost spike.  4th gear pull it went right to 17.5psi and held there most of the RPM range.

Thanks for any help!  (and sorry if these are dumb questions)

KFLDHBN

Someone needs to read the wiki.

http://s4wiki.com/wiki/Tuning

Thanks for the info.  I think this was set at 3.89 across the board (again, don't have it in front of me), which i don't think would cause a problem, right?

You said you increased IRL and LDRXN, but did you shift IOPs axis to accomodate higher load?

Yes, i haven't changed KFMIOP yet.  It's my understanding (and correct me if I'm wrong), that basically a change to KFMIRL for full throttle will not necessary need a change to KFMIOP as anything past it's axis limit will just follow the highest values.  Is that correct?  I was having part throttle problems with the other stage1 1.8t file in the tuned files section and i think that's because the KFMIOP axis wasn't scaled.  I don't think that's what's limiting me in this case though, but obviously my understanding could be way off.

BTW -- this is the file i'm working off right now:
http://nefariousmotorsports.com/forum/index.php?topic=1272.0title=

Yes

Thank you! -- I think that indicates a mechanical problem (e.g. boost leak) or simply that the turbo can't hit requested boost. 

How does WGDC react?

Don't have logs at the moment, but i'm 99% sure it was at 100WGDC.  Would that make sense?  I've been very confused in my reading as to whether 100% WGDC would be completely shutting off pressure to the wastegate or allowing 100% of pressure to wastegate. 

95% is shunting all pressure from wastegate.

<10% and >95% are generally invalid values.

thanks very much for the clarification!  I'll have to look at the logged values again.  Is the 95% limit associated with KFLDIMX -- e.g. if KFLDIMX is at 100, can the WGDC reach 100?  I thought i saw 100 in KFLDIMX, but i've been looking at a load of files recently and may be getting some things mixed up.

which ecu do you have?

06a906032dl running 06a906032hn bin. Though I believe it's actually HS disguised as HN according some info in the tuned file thread.

From what I have read I believe it is indeed an HS file sw version 0003. I'm looking for an HS ori with the 0007 software revision. I just want to see what the differences are between the 0006,0007 and 0008 versions. The 6,7 and 8 versions have the same addresses, just curious what Bosch changed.

 :)The max boost is always limited by the highest value in KFMIOP load axis. Reading 17.5 psi boost means 191.5 load.
You need to rescale load axis in KFMIOP for moore boost (20 psi  ~ 208 load) and adjust torque values to match KFMIRL request loads.

No.

ya i didn't think that was right...

Correct me if I'm wrong, but KFMIOP values are Throttle position, so the only way KFMIOP would limit full throttle (assuming your full throttle KFMIRL load is higher than max axis value in KFMIOP) is if the max values in KFMIOP were less than 100% (i think it actually ends up at like 99.7 or something).  Any load from KFMIRL over the max axis value in KFMIOP takes the highest point in KFMIOP, correct?

Values of KFMIOP are %torque, not throttle.

i see -- sorry, I'm sure that's well documented here (and I'm sure i've read it 1000 times already).  Thanks for the input!

The output of KFMIOP does not "limit" req boost in ANY WAY that people are saying it does in this thread.

It is used to calculate *actual* internal torque, not a *desired* torque limit.

If you set 100% column in KFMIRL and LDRXN over 208 (allow 20psi) rescale KFMIOP load axis and recalculated %torque according to KFMIRL map! The maximum load axis value must match with max load in LDRXN  if you want to see 20 psi in your boost gauge. Max %torque of your engine is set in last column of KFMIOP  map if LDRXN allow that. This torque (in your case 99.7%) is made with the last value of load axis. In all logging sessions you'll see this.             

Can anyone confirm this?  I don't think this is right.....

I agree, recalculating the KFMIOP table is definitely a good idea , but from what I've read, it doesn't really matter at 100% torque request (provided, of course, the highest values in KFMIOP are set to max).

Again, sorry for my ignorance -- I'm very new to understanding the ME7.

BTW -- if i remember correctly, the highest axis value for KFMIOP in the file i'm working of is somewhere around 160.  This is well below my requested boost in logs (and actually, my requested boost looks fine, which indicates a boost leak/hardware problem/physical limitation). 

Just to get back to the point of this thread a bit -- i think the reason i'm not hitting target is a mechanical limitation.  What i haven't had time to investigate yet is wether I'm getting timing retard or something like that due to knock.  Could that cause a significant drop in boost? 

Perhaps there is a limitation in other versions of ME, but in ME7.1 the IOP axis *definitely* has no bearing on rl_max, or any other req load limit.

Yes, if you calibrate it wrong, you might see *torque* intervention.

Perhaps there might be resulting error conditions that cause slow path intervention via load request, but I cannot identify it in the FR. Perhaps prj or phila know if such a thing exists.

In MDMAX, rlmax_w and nmot_w are input to KFMIOP giving mimax_w.

mimax_w does limit mifa_w and mifa_w does limit mifal_w before milsol_w.

HOWEVER, KFMIOP's load axis does NOT reflect max load. The map values can limit the torque request, but the axis itself is not a load cap.

If the load axis for KFMIOP maxes at 160 and is filled with 99.2 in all rows, then anytime rlmax_w meets or exceeds 160, the torque request will be limited to 99.2. The torque request gets translated to desired load by KFMIRL and capped by rlmax_w.

If your desired load is good and WGDC behaves correctly, then your problem is hardware.

Sounds like regurgitation.

Please, can you explain how and why?

Ok! I could be wrong but in my car , a VOLVO S70T5 with ME7.0 max load axis value on KFMIOP was set stock to 180 (~16 psi). LDRXN allow max 165 load. After I changed this value to 195 (KFMIRL >200 in last column) boost did not exceed 16 psi until I made changes in KFMIOP (max load axis value set to 200 and recalculated %torque according to KFMIRL).
Btw , in oldcarguy85 file max LDRXN is set to 160 (~13psi). How can his engine make 17.5 psi ???
  

                    

Look at this tutorial on pages 19,20,21 (logging data and remarks...point "d").     

It's a very pretty document, but that doesn't change the fact that the IOP axis does not limit load. If your spec load is higher than the axis, it will use the last available column.

And look at module MDMAX. The variable for the map KFMIOP is used to calculate maximum allowed indexed torque . So if in last column of this map     we have max 85% and load axis value is 180 everything above this (set in LDRXN) make more torque and we have operation/torque intervention .This can be avoided by scaling the load axis and recalculated % values according to KFMIRL .
Ex:
KFMIRL.....    100%.........4000RPM..........210load (if LDRXN allow this)
KFMIOP....       x%..........4000RPM..........200load (last load axis value)

            x=(100*200)/210=95.24%    Set 96% on map and this is the uncorrected max torque.

If we let the maximum load axis value 180 ,   x=(100*180)/210=85.71% and this is max allowed uncorrected torque even if LDRXN is set to 210.
Load             Torque
180    .....      85.71%
190    .....      85.71%
200    .....      85.71%  
210    .....      85.71%  
and so on...    
Over 180 ME7 limited max torque to this %value because is torque-based system.            

"FB MDMAX 1.30 Detailed description of function

The maximum volumetric charge rlvugd at unthrottled engine operation is taken from the characteristic RLVMXN resp. from the
characteristic RLVMXSN dependent on the position of the manifold switch-over. Thereafter the conversion into the charge rlugd w
is performed by altitude- and temperature-correction. For configurations with turbo charger (SY TURBO=1), the maximum charge
rlmax w is predefined directly. The maximum charge now is input variable for the map KFMIOP, in which the optimum torque is stored
over speed and charge. The thus obtained torque mimxl1 w is corrected by the factor FMIVL, which indicates the torque increase at
full load and the maximum torque mimax w is obtained."

                                  

Just to be clear, I never even mentioned LDRXN. It was/is set plenty high to achieve the correct boost.

Rufus,
Did you read my first reply on page 2?

If you did, read it again. You have the FR, so check it out.

Then explain to me how your theory operates beyond MDMAX.

Ill try to post a log at some point that clearly shows KFMIOP does not impact load requests greater than its highest axis value. I know in the file I originally referenced in this post KFMIOP maxed out at 160. KFMIRL highest points were all around 210. LDRXN was around 200-210 for most of the RPM range. As I originally noted, the ECU requested around 19psi (maybe more, I forget). If KFMIOP was limiting this, it would have requested 13psi(160 load request)

That was written by a member here based on the s4wik Tuning page and discussions here a while ago.

I do not think the author really understood IOP at the time, since tuning it properly wasn't discussed that in depth at the time he wrote it.

While we were all grateful that somebody went through the trouble of writing a tutorial, I expressed concern that his PDF was not editable, and that it, like proprietary software that becomes abandonware, would suffer obsolescence if the paper was not properly curated.

Unfortunately, the author did not understand my concern, or share my ideology regarding publishing information.

It seems my concerns were well founded.

what else in that PDF is not worth reading nyet?

By and large (IMO) it is pretty good. There are areas that could use some work, but I don't think i'm qualified to be specific.. basically, thats why the s4 tuning wiki is more of a guide to maps rather than a tuning guide; I don't presume to know how to tell somebody exactly what should be changed and how..

I would love to have an experienced (and knowledgeable) tuner go over that document and really do it right..

Well, I value your opinion sir. Thank you. I feel I'm still at the novice level as far as ECU tuning. Without you, I would be damn near clueless.

Add me to that chorus too nyet. Without your wiki I had no frame of reference... Much appreciated!